Establishment and inheritance of minichromosomes from Arabidopsis haploid induction

We have discovered that 1–2% of phenotypically normal arabidopsis haploids resulting from crosses to the CENH3-based GFP-tailswap haploid inducer carry minichromosomes (minis) derived from centromeric and adjacent pericentromeric regions. We detected the minis by cytological methods, genotyping, and dosage profiling genomic reads. We demonstrated that the minis originate from the haploid inducer genome using sequence analysis. We found that mini1a is circular with a junction between two sites that flank the centromere of Chr1, which was confirmed by cytological methods. In addition, a mechanism for generating minichromosomes for specific chromosomes was developed, and we showed that minis could be identified easily when the haploid inducer carried a selectable marker in the centromeric region of Chr4. These elements ranged between 3 and 10 Mb in size. The line carrying mini1a appeared as a single unpaired circle with a distinct DAPI-stained knob at prophase and metaphase (Fig. 3). Occasionally, two unpaired circles per cell were visible. In self crosses, mini1a was transmitted to ~ 25% of the progeny, which is consistent with the observed ~ 0.12 transmission rate through male or female gametes. Trisomics of Chr1 display very similar transmission rates upon selfing (Koornneef and Van der Veen 1983). This similar efficiency, however, is coincidental because, compared to trisomics of Chr1, mini1a displayed lower female transmission and higher male transmission. Considering the parental transmission rates, the predicted frequency of inheriting two copies of mini1a is 0.125 * 0.107 = 0.013, a relatively infrequent event. For certain plants, we observed a very low transmission (Fig. 1), which may be explained by mitotic instability and loss of the minichromosome. Murata et al. (2006) described an increase in transmission when their Ler-derived Chr4 mini (referred to as mini4S henceforth) was backcrossed into the Col-0 ecotype. This suggested a genotypic background effect on the transmission of a minichromosome. In addition, similar to our Bar-derived mini4b and mini4c, Murata’s mini4S contains the entire short arm of Chr4, which is known to carry the ribosomal RNA genes and function as a nucleolus organizer region (NOR). The presence of telomeres may favor associations of NORs during meiosis (Murata et al. 2006), which could lead to different outcomes when compared to circularized minis such as mini1a and mini3b that are derived from isocentric chromosomes that do not contain NORs.

Minis from genomic instability

In CENH3-mediated haploid induction, about \(^\!\left/ \!_\right.\) of the progeny are aneuploid, and about a \(^\!\left/ \!_\right.\) of these aneuploids carry chromosomes rearranged by chromoanagenesis (Comai and Tan 2019). Typically, aneuploids carrying these shuffled chromosomes are developmentally abnormal and sterile, probably due to imbalance of many genes and possibly to the action of novel gene fusions (Tan et al. 2015). We propose that minis arise when a mis-segregated chromosome is captured in a micronucleus, and after fragmentation, a chromosomal segment carrying the centromeric region circularizes and is restituted to the nucleus (Fig. 2E). Potentially, formation of telomeres may also stabilize certain centromeric fragments. Instability is commonly associated with haploid induction (Maheshwari et al. 2015; Tan et al. 2015; Kuppu et al. 2015; Amundson et al. 2021; Sun et al. 2022) and with wide crosses (Madlung et al. 2005; Shibata et al. 2013), and minis are likely to arise in these crosses (Seymour et al. 2012; Tan et al. 2015; Kuppu et al. 2015). One question is whether minichromosomes require mutations in specialized loci of DNA metabolism for their formation and maintenance. The frequencies observed are inconsistent with those expected for mutations, even if multiple loci could be involved. Furthermore, transmission in outcrosses (Table 3) indicates that no recessive mutation is required. In many instances, minichromosomes may not be associated with detectable traits, likely because of the reduced number of genes that are imbalanced. We have no direct evidence of gene expression for mini1a, but successful selection for mini4a, 4b, and 4c demonstrates that the transgenic marker is expressed. Identification through genomic analysis is also challenging in most species because of the difficulty in detecting CNV in the background of complex, variable, and uncharacterized centromeric regions (Hardigan et al. 2016; Hufford et al. 2021). Nonetheless, reduced-size, additional chromosomes can be found both by sequencing (Shibata et al. 2013; Amundson et al. 2021) or by cytological analysis (Shibata et al. 2013). The mechanisms by which minis can originate are illustrated in Fig. 6. Linear or circularized minis that retain centromeres during periods of genome instability are capable of germline transmission and maintenance as extrachromosomal entities.

Fig. 6figure 6

Formation of minichromosomes as a result of genome instability. We propose that minichromosomes arise from genome instability leading to fragmentation. Chromosomal fragments that contain a centromere or form a neocentromere can be stabilized by either formation of telomeres or by circularization. The resulting chromosomes are typically unstable due to their small size or circularity

Chromosome fragments that contain the centromere can either heal their exposed ends or circularize. Acentric fragments can persist through a number of divisions but are eventually lost unless they form a neocentromere. Linear or circular minichromosomes with functional centromeres can persist through meiotic generations, but their redundancy and suboptimal structure will eventually result in loss.

Method for isolation of minichromosomes

We demonstrate that minis can be identified readily when a haploid inducer carries a marker in the pericentromeric region (Fig. 5). Among the phenotypically normal haploids, those expressing the marker typically carry a novel mini. The formation of circular vs. linear minichromosomes may depend on local context and features. The method is applicable to any species where instability results from haploid induction and provides an approach toward studying and exploiting minis. Mini’s have been described for nearly 100 years, starting with the pioneering work of McClintock on maize circular chromosomes (McClintock 1932). Their potential for vectoring stacks of valuable genes and avoiding linkage drag has drawn considerable attention. One challenge is the loading of genes on the mini, for which multiple solutions based on transformation and induced recombination have been proposed (Birchler 2015; Kumar et al. 2016; Anand et al. 2019; Dong and Ronald 2021). Our experiments indicate that dominant markers in the centromeric region suggest an additional approach: minis carrying valuable genes could be generated if these genes are located in pericentromeric regions. It is also possible that minis combining distal segments of euchromatin with a centromeric segment could be selected in vivo if these genes have been previously marked with a suitable transgene.

Remaining challenges and opportunities

Specific information on the mitotic and meiotic instability of a mini, while not a specific objective of this study, should be obtained before its use on a commercial scale. Mitotic instability would affect the use of a minichromosome vector by resulting in chimeric plants. During mitosis, an uneven number of crossovers between sister chromatids can fuse replicated circular chromosomes and initiate cycles of breakage-fusion-bridge events that result in loss or duplication of DNA as well as missegregation. Another malfunction that can increase missegregation is early loss of sister chromatid cohesion.

Instability is also likely in meiosis (Han et al. 2007). The type of mini characterized here is derived from the normal chromosome complement and, therefore, it resembles trisomy. During meiosis, trisomy is a naturally unstable state when a full-length chromosome is involved (Koornneef and Van der Veen 1983). In addition, our minis do not appear to pair efficiently with their full-size homologs. The small size of mini1a, its reduced content of euchromatin, and the lack of telomeres likely contribute to this property. Lack of homologous pairing causes premature mobilization by the spindle, leading to increased missegregation. Instability during meiosis may be less damaging to vector utilization. Lack of pairing, even when two minis are present, destabilizes minis at metaphase I of meiosis and likely contributes to reduced transmission in gametes.

Both meiotic and mitotic instability may be overcome if the mini includes a selectable marker gene (Han et al. 2007). Mitotic chimerism could be overcome if the mini carries a gene essential for cellular proliferation. Incomplete meiotic transmission could be ameliorated by the presence of a detectable marker, although it would increase seed production costs. Notably, for certain uses, instability is not problematic but useful. For example, provision of genome editing transgenes on a mini could be advantageous: after genome editing, transgene-free progeny could be easily identified.

留言 (0)

沒有登入
gif