Summary: Heavy‐ion irradiation is a powerful mutagen that possesses high linear energy transfer (LET). Several studies have indicated that the value of LET affects DNA lesion formation in several ways, including the efficiency and the density of double‐stranded break induction along the particle path. We assumed that the mutation type can be altered by selecting an appropriate LET value. Here, we quantitatively demonstrate differences in the mutation type induced by irradiation with two representative ions, Ar ions (LET: 290 keV μm⁻¹) and C ions (LET: 30.0 keV μm⁻¹), by whole‐genome resequencing of the Arabidopsis mutants produced by these irradiations. Ar ions caused chromosomal rearrangements or large deletions (≥100 bp) more frequently than C ions, with 10.2 and 2.3 per mutant genome under Ar‐ and C‐ion irradiation, respectively. Conversely, C ions induced more single‐base substitutions and small indels (<100 bp) than Ar ions, with 28.1 and 56.9 per mutant genome under Ar‐ and C‐ion irradiation, respectively. Moreover, the rearrangements induced by Ar‐ion irradiation were more complex than those induced by C‐ion irradiation, and tended to accompany single base substitutions or small indels located close by. In conjunction with the detection of causative genes through high‐throughput sequencing, selective irradiation by beams with different effects will be a powerful tool for forward genetics as well as studies on chromosomal rearrangements.

How do separate sexes originate and evolve? Plants provide many opportunities to address this question as they have diverse mating systems and separate sexes (dioecy) that evolved many times independently. The classic “two-factor” model for evolution of separate sexes proposes that males and females can evolve from hermaphrodites via the spread of male and female sterility mutations that turn hermaphrodites into females and males, respectively. This widely accepted model was inspired by early genetic work in dioecious white campion (Silene latifolia) that revealed the presence of two sex-determining factors on the Y-chromosome, though the actual genes remained unknown. Here, we report identification and functional analysis of the putative sex-determining gene in S. latifolia, corresponding to the gynoecium suppression factor (GSF). We demonstrate that GSF likely corresponds to a Y-linked CLV3-like gene that is specifically expressed in early male flower buds and encodes the protein that suppresses gynoecium development in S. latifolia. Interestingly, GSFY has a dysfunctional X-linked homolog (GSFX) and their synonymous divergence (dS = 17.9%) is consistent with the age of sex chromosomes in this species. We propose that female development in S. latifolia is controlled via the WUSCHEL-CLAVATA feedback loop, with the X-linked WUSCHEL-like and Y-linked CLV3-like genes, respectively. Evolution of dioecy in the S. latifolia ancestor likely involved inclusion of ancestral GSFY into the nonrecombining region on the nascent Y-chromosome and GSFX loss of function, which resulted in disbalance of the WUSCHEL-CLAVATA feedback loop between the sexes and ensured gynoecium suppression in males.

The loss of functional genes from non-recombining sex-specific chromosomes [1, 2], such as the Y chromosomes in mammals [3] or W chromosomes in birds [4], should result in an imbalance of gene products for sex-linked genes [5]. Different chromosome-wide systems that rebalance gene expression are known to operate in organisms with relatively old sex chromosomes [6]; e.g., Drosophila overexpress X-linked genes in males [7], while mammals shut down one of the X chromosomes in females [8]. It is not known how long it takes for a chromosome-wide dosage compensation system to evolve. To shed light on the early evolution of dosage compensation, we constructed a high-density Y-deletion map and used deletion mutants to manipulate gene dose and analyze gene expression in white campion (Silene latifolia), which evolved dioecy and sex chromosomes only 11 million years ago [9]. We demonstrate that immediate dosage compensation can be triggered by deletions in a large portion of the p arm of the Y chromosome. Our results indicate that dosage compensation in S. latifolia does not have to evolve gene by gene because a system to upregulate gene expression is already operating on part of the X chromosome, which likely represents an intermediate step in the evolution of a chromosome-wide dosage compensation system in this species.

Recently, a recessive Arabidopsis thaliana mutant with abundant stromules in leaf epidermal pavement cells was visually screened and isolated. The gene responsible for this mutant phenotype was identified as PARC6, a chloroplast division site regulator gene. The mutant allele parc6-5 carried two point mutations (G62R and W700stop) at the N- and C-terminal ends of the coding sequence, respectively. Here, we further characterized parc6-5 and other parc6 mutant alleles, and showed that PARC6 plays a critical role in plastid morphogenesis in all cell types of the leaf epidermis: pavement cells, trichome cells, and guard cells. Transient expression of PARC6 transit peptide (TP) fused to the green fluorescent protein (GFP) in plant cells showed that the G62R mutation has no or little effect on the TP activity of the PARC6 N-terminal region. Then, plastid morphology was microscopically analyzed in the leaf epidermis of wild-type (WT) and parc6 mutants (parc6-1, parc6-3, parc6-4 and parc6-5) with the aid of stroma-targeted fluorescent proteins. In parc6 pavement cells, plastids often assumed aberrant grape-like morphology, similar to those in severe plastid division mutants, atminE1, and arc6. In parc6 trichome cells, plastids exhibited extreme grape-like aggregations, without the production of giant plastids (>6 µm diameter), as a general phenotype. In parc6 guard cells, plastids exhibited a variety of abnormal phenotypes, including reduced number, enlarged size, and activated stromules, similar to those in atminE1 and arc6 guard cells. Nevertheless, unlike atminE1 and arc6, parc6 exhibited a low number of mini-chloroplasts (< 2 µm diameter) and rarely produced chloroplast-deficient guard cells. Importantly, unlike parc6, the chloroplast division site mutant arc11 exhibited WT-like plastid phenotypes in trichome and guard cells. Finally, observation of parc6 complementation lines expressing a functional PARC6-GFP protein indicated that PARC6-GFP formed a ring-like structure in both constricting and non-constricting chloroplasts, and that PARC6 dynamically changes its configuration during the process of chloroplast division.