Portal de Periódicos da CAPES

T-DNA-Induced Chromosomal Translocations in feronia and anxur2 Mutants Reveal Implications for the Mechanism of Collapsed Pollen Due to Chromosomal Rearrangements

2014; Elsevier BV; Volume: 7; Issue: 10 Linguagem: Inglês

10.1093/mp/ssu062

1674-2052

Colin Ruprecht, Andrew Carroll, Staffan Persson,

Chromosomal and Genetic Variations

Dear Editor, Approximately every fifth Arabidopsis T-DNA insertion line contains chromosomal translocations caused by the inserted T-DNA (Clark and Krysan, 2010Clark K.A. Krysan P.J. Chromosomal translocations are a common phenomenon in Arabidopsis thaliana T-DNA insertion lines.Plant J. 2010; 64: 990-1001Crossref PubMed Scopus (68) Google Scholar). Albeit broad use of T-DNA lines for mutant analysis, little is known about the consequences of these chromosomal rearrangements and only a few studies describe chromosomal aberrations in the mutant lines. While plant growth in general is not affected by such chromosomal translocations, defects in gametophyte development have been observed in lines that are heterozygous for the T-DNA insertion (Ray et al., 1997Ray S.M. Park S.S. Ray A. Pollen tube guidance by the female gametophyte.Development. 1997; 124: 2489-2498PubMed Google Scholar; Curtis et al., 2009Curtis M.J. Belcram K. Bollmann S.R. Tominey C.M. Hoffman P.D. Mercier R. Hays J.B. 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.Planta. 2009; 229: 731-745Crossref PubMed Scopus (26) Google Scholar). These gametophytic defects are probably caused by interaction of two non-homologous chromosomes during meiosis resulting in the formation of tetravalents and subsequently unbalanced gametes after meiosis (Curtis et al., 2009Curtis M.J. Belcram K. Bollmann S.R. Tominey C.M. Hoffman P.D. Mercier R. Hays J.B. 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.Planta. 2009; 229: 731-745Crossref PubMed Scopus (26) Google Scholar). Importantly, the transmission of the T-DNA insertion through male and female gametophytes is not affected in the mutants, and plants that are homozygous for the translocation do not show aborted gametophytes anymore (Curtis et al., 2009Curtis M.J. Belcram K. Bollmann S.R. Tominey C.M. Hoffman P.D. Mercier R. Hays J.B. 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.Planta. 2009; 229: 731-745Crossref PubMed Scopus (26) Google Scholar). These results suggest that chromosomal translocations are easily overlooked when analyzing T-DNA lines. Here, we describe two previously uncharacterized T-DNA insertion alleles for the receptor-like kinase FERONIA (fer-6 and fer-7) and one T-DNA line for its pollen-specific homolog ANXUR2 (anx2-1), which comprise chromosomal translocations as assessed with genetic mapping experiments. These experiments also suggest a tentative mechanism for the deformed pollen phenotype. We were initially interested in the CrRLK1L-family member FERONIA (FER) (Boisson-Dernier et al., 2011Boisson-Dernier A. Kessler S. Grossniklaus U. The walls have ears: the role of plant CrRLK1Ls in sensing and transducing extracellular signals.J. Exp. Bot. 2011; 62: 1581-1591Crossref PubMed Scopus (117) Google Scholar) due to its tight co-expression with the primary wall Cellulose Synthase (CESA) genes (Supplemental Figure 1). Therefore, we obtained a T-DNA insertion mutant for FER (Figure 1A, fer-6, SAIL_320_C11), but were unable to obtain homozygous plants (Supplemental Table 1). fer-6 showed deformed pollen thus phenocopying primary wall cesa-deficient phenotypes (Figure 1; Persson et al., 2007Persson S. Paredez A. Carroll A. Palsdottir H. Doblin M. Poindexter P. Khitrov N. Auer M. Somerville C.R. Genetic evidence for three unique components in primary cell-wall cellulose synthase complexes in Arabidopsis.Proc. Natl Acad. Sci. U S A. 2007; 104: 15566-15571Crossref PubMed Scopus (413) Google Scholar). Staining anthers and mutant pollen with Alexander stain revealed non-viable pollen grains (Figure 1). In addition, electron microscopy analyses revealed collapsed fer-6 pollen that lacked cytosolic content, but that held relatively normal exine structures (Figure 1D and 1E). This phenotype was more severe than in primary wall-related cesa mutants, in which aborted pollen still held cytosolic structures (Persson et al., 2007Persson S. Paredez A. Carroll A. Palsdottir H. Doblin M. Poindexter P. Khitrov N. Auer M. Somerville C.R. Genetic evidence for three unique components in primary cell-wall cellulose synthase complexes in Arabidopsis.Proc. Natl Acad. Sci. U S A. 2007; 104: 15566-15571Crossref PubMed Scopus (413) Google Scholar). Nonetheless, the deformed pollen in the fer mutants initially suggested that FER was involved in regulation of primary cell wall formation. To confirm this result, we ordered additional T-DNA lines for FER and also lines for its pollen-specific homolog ANXUR (ANX) 2 (fer-7, SALK_017600; anx2-1, SALK_127359). Heterozygous plants of these mutants again displayed the collapsed pollen phenotype. However, also homozygous plants could be obtained for fer-7 and anx2-1 (Supplemental Table 1), which surprisingly did not show the pollen phenotype anymore. A second site mutation causing the deformed pollen was very unlikely, because fer-6, fer-7, and anx2-1 were each backcrossed twice to eliminate unrelated mutations and the pollen phenotype of the mutants was always strictly dependent on the T-DNA insertion. To investigate this further, we performed reciprocal backcrosses to wild-type plants. Surprisingly, the transmission of the T-DNA through pollen was not affected (Supplemental Table 2), and the heterozygous plants in the progeny showed the deformed pollen phenotype again. The amounts of deformed pollen were similar in the heterozygous parent plants and in the F1 progeny, regardless of the direction of the cross (Supplemental Table 3). These results were reminiscent of previously described reciprocal chromosomal translocations that cause aborted pollen only in heterozygous plants (Curtis et al., 2009Curtis M.J. Belcram K. Bollmann S.R. Tominey C.M. Hoffman P.D. Mercier R. Hays J.B. 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.Planta. 2009; 229: 731-745Crossref PubMed Scopus (26) Google Scholar). To test this possibility, we performed a mapping approach for our mutants (Clark and Krysan, 2010Clark K.A. Krysan P.J. Chromosomal translocations are a common phenomenon in Arabidopsis thaliana T-DNA insertion lines.Plant J. 2010; 64: 990-1001Crossref PubMed Scopus (68) Google Scholar). In brief, we crossed fer6/+, fer-7, and anx2-1, all in Arabidopsis ecotype Columbia (Col), to Arabidopsis ecotype Landsberg erecta (Ler) to obtain F1 plants that were heterozygous for the respective T-DNA insertion and displayed the deformed pollen phenotype. These F1 plants were crossed again to Arabidopsis Ler and the segregating F2 generation was analyzed. Plants containing the respective T-DNA insertion were used to determine the recombination frequency of the T-DNA insertion with various loci on the five different chromosomes. For this analysis, we utilized 19 simple sequence length polymorphism (SSLP) markers that produced differently sized amplicons in Ler and Col (Supplemental Table 4). Here, low recombination frequency of a certain SSLP marker indicates physical association of this locus with the T-DNA. As expected, several SSLP markers on chromosome 3 were associated with the T-DNA insertion in fer-6 and fer-7, since FER is also located on chromosome 3 (Figure 1). However, also the upper part of chromosome 4 and a large region on chromosome 5 displayed low recombination frequency with the T-DNA insertion of fer-6 and fer-7, respectively. This indicated that reciprocal chromosomal translocations had occurred in these two lines. For anx2-1, we did not detect a chromosomal translocation across two different chromosomes. However, the tight association of SSLP markers 17 and 18 on the lower arm of chromosome 5 with the T-DNA insertion of anx2-1 on the upper arm of chromosome 5 suggested a reorganization of plant DNA within chromosome 5 in this mutant (Figure 1). Interestingly, this mutant also displayed a lower amount of deformed pollen compared to fer-6 and fer-7, indicating a less severe effect of a chromosomal rearrangement (Supplemental Table 3). Previous reports hypothesized that the chromosomal translocation might be caused by recombination of two unlinked T-DNAs (Nacry et al., 1998Nacry P. Camilleri C. Courtial B. Caboche M. Bouchez D. Major chromosomal rearrangements induced by T-DNA transformation in Arabidopsis.Genetics. 1998; 149: 641-650PubMed Google Scholar; Laufs et al., 1999Laufs P. Autran D. Traas J. A chromosomal paracentric inversion associated with T-DNA integration in Arabidopsis.Plant J. 1999; 18: 131-139Crossref PubMed Scopus (56) Google Scholar; Curtis et al., 2009Curtis M.J. Belcram K. Bollmann S.R. Tominey C.M. Hoffman P.D. Mercier R. Hays J.B. 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.Planta. 2009; 229: 731-745Crossref PubMed Scopus (26) Google Scholar). To test whether a second T-DNA insertion existed in our lines, we performed thermal asymmetric interlaced (TAIL) PCR to detect additional left and right border junctions. Indeed, in addition to the original mutant allele in fer-6 (located on chromosome 3, 19121007), we could identify left borders of two additional T-DNAs: one adjacent to the fer-6 allele (Ch. 3 19216575) and one on chromosome 4 (9092178). For anx2-1, located on the upper arm of chromosome 5 (10722287), we found another T-DNA on the lower arm of this chromosome (Ch. 5, 255568169, Figure 2). The loci of these additional T-DNA insertions corresponded to the chromosomal translocation assessed by the genetic mapping experiment. We were unable to amplify the two T-DNAs in anx2-1 using respective gene-specific primers adjacent to the T-DNAs. However, using the gene-specific primer upstream of the 5′ junction of the anx2-1 T-DNA and the gene-specific primer upstream of the 5′ junction of the second T-DNA, we were able to amplify a 2.5-kb fragment. Sequencing of this fragment verified that these two junctions were adjacent and separated only by a truncated T-DNA. This indicated an inversion of the approximately 15-Mb fragment between the two T-DNAs in anx2-1. To gain more insight into the mechanism that leads to the observed pollen phenotype, we determined when pollen abortion occurred in fer-6 and fer-7. Interestingly, the pollen collapsed shortly before the first mitotic division (Supplemental Figure 2), which is at the same stage as in other described mutants with gametophytic defects due to chromosomal translocations (Ray et al., 1997Ray S.M. Park S.S. Ray A. Pollen tube guidance by the female gametophyte.Development. 1997; 124: 2489-2498PubMed Google Scholar; Curtis et al., 2009Curtis M.J. Belcram K. Bollmann S.R. Tominey C.M. Hoffman P.D. Mercier R. Hays J.B. 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.Planta. 2009; 229: 731-745Crossref PubMed Scopus (26) Google Scholar). However, a relatively small percentage of mutants showed defects at the first mitosis compared to other stages of gametophyte development (20% or 2%, depending on the screen: Drews and Yadegari, 2002Drews G.N. Yadegari R. Development and function of the angiosperm female gametophyte.Annu. Rev. Genet. 2002; 36: 99-124Crossref PubMed Scopus (165) Google Scholar; Pagnussat et al., 2005Pagnussat G.C. Yu H.-J. Ngo Q.A. Rajani S. Mayalagu S. Johnson C.S. Capron A. Xie L.-F. Ye D. Sundaresan V. Genetic and molecular identification of genes required for female gametophyte development and function in Arabidopsis.Development. 2005; 132: 603-614Crossref PubMed Scopus (440) Google Scholar). This therefore casts doubt on the previously assumed reason for aborted pollen due to imbalanced chromosomes after meiosis leading to a lack of essential genes for gametophyte development. Instead, we propose that a post-meiotic or pre-mitotic checkpoint might recognize aberrant chromosomes, which initiates apoptosis. Although this hypothesis needs further testing, the much stronger pollen phenotype of fer-6 compared to cesa mutants implied not only that the fer-6 pollen underwent developmental cessation, but that the cytosolic compartments were actively degraded. To characterize the relationship of the pollen phenotype with the chromosomal translocations further, we crossed fer-6 into the quartet (qrt)1 background. In qrt1, the four pollen grains originating from one pollen mother cell stick together to form tetrads (Francis et al., 2006Francis K.E. Lam S.Y. Copenhaver G.P. Separation of Arabidopsis pollen tetrads is regulated by QUARTET1, a pectin methylesterase gene.Plant Physiol. 2006; 142: 1004-1013Crossref PubMed Scopus (191) Google Scholar). Surprisingly, in fer-6/+qrt1-2, we found mainly normal tetrads (4 normal:0 deformed), tetrads with two deformed and two normal pollen (2:2), or tetrads in which all four pollen were deformed (0:4, Figure 1J–1M). These data also show pollen without the T-DNA insertion abort. Quantification of the different tetrads revealed a ratio of about one 4:0 tetrad, to two 2:2 tetrads, to one 0:4 tetrad; however, only very few tetrads showed one or three deformed pollen (Supplemental Table 5). To account for these observations, the existing model for pollen abortion due to chromosomal translocations (Curtis et al., 2009Curtis M.J. Belcram K. Bollmann S.R. Tominey C.M. Hoffman P.D. Mercier R. Hays J.B. 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.Planta. 2009; 229: 731-745Crossref PubMed Scopus (26) Google Scholar) can be modified to allow also independent crossover of the two sister-chromatids that include chromosomal translocations. Depending on the segregation after meiotic crossover, this would lead to unbalanced chromosomes in several cases resulting in our observed ratio of one 4:0 tetrad, to two 2:2 tetrads, to one 0:4 tetrad. In summary, we have identified three independent T-DNA insertion lines for FERONIA and its pollen-specific homolog ANXUR2 showing deformed pollen phenotypes caused by chromosomal translocations. Moreover, our detailed characterization of the mutant lines revealed a potential mechanism of pollen abortion due to the chromosomal translocations. Given the high frequency of chromosomal translocations in T-DNA lines (about 20%; Clark and Krysan, 2010Clark K.A. Krysan P.J. Chromosomal translocations are a common phenomenon in Arabidopsis thaliana T-DNA insertion lines.Plant J. 2010; 64: 990-1001Crossref PubMed Scopus (68) Google Scholar), our results highlight the potential of misinterpreting deformed pollen phenotypes in T-DNA-transformed transgenic lines. Hence, robust analyses are needed to distinguish between pollen lethality due to chromosomal translocation or to mutations of pollen essential genes. Supplementary Data are available at Molecular Plant Online. This work was funded by the Max-Planck-Gesellschaft and an IMPRS fellowship for C.R.