Fig 1.
Arabidopsis accessions display natural variation in ovule number per flower.
(A) Boxplot of ovule number in 189 Arabidopsis accessions. The detailed names of these accessions are in S1 Table. (B) Dissected siliques of low, medium, and high ovule number accessions (Bar = 2.0mm). (C) Geographic distribution of accessions. Groups 1 to 6 correspond to ovule number ranking (from low to high), with the colors corresponding to panel A. The pie chart indicates the percentage of accessions in each country in each ovule number group, while the size of the pie chart corresponds to the total number of accessions per country. (D) Cladogram of accessions used in this study (generated in MEGA7). The 15 accessions labeled in blue had the lowest ovule numbers, and the 15 accessions labeled with red had the highest ovule numbers. The arrow points to a clade with clustered low ovule number accessions.
Fig 2.
GWAS identifies candidate loci associated with ovule number per flower.
(A) Manhattan plot for the SNPs associated with ovule number per flower. Chromosomes are depicted in different colors. The horizontal blue dashed line corresponds to a–log10(p values) ≥ 6.2 and the red dashed line corresponds to–log10(p values) ≥ 7.5 after Benjamini Hochberg (False Discovery Rate) correction. (B) The genomic region surrounding the two most significant GWA peaks on Chromosome 3. (C) Genes in the genomic region surrounding the two significant GWA peaks in (B). The red boxes are AT3G51050 (NERD1) and AT3G60660 (ONA2).
Fig 3.
nerd1 mutants have reduced ovule number and fertility.
(A) Scatter plot of ovule number per flower in Col-0, nerd1-2, nerd1-4 and their complementation lines. “***” indicates statistical significance (p value<0.001 determined by Student’s t-test). (B-C) nerd1-2/nerd1-2 displays lower ovule number, infertile ovules, and short siliques compared to Col-0 and complemented lines. (D) Quantification of normal vs abnormal embryo sacs from developmental stages FG2 to mature ovule in Col-0 vs nerd1-2/nerd1-2. (E) During female gametophyte development, instead of the functional megaspore developing into female gametophyte by sequential mitotic divisions as seen in Col-0, nerd1-2/nerd1-2 displays defective embryo sacs from stages FG2 to mature ovule. The central vacuole (V) is absent in FG3 and FG4 nerd1 embryo sacs and mature ovules often have only 2 nuclei (red arrows) (F-G) DAPI staining of early stages of pollen development. (F) At tetrad stage in pollen development, nerd1-2/nerd1-2 only has one developed haploid microspores instead of four microspores in Col-0 (white arrowheads indicate normal microspores and red arrowheads indicate aborted microspores). (G) After separation of the microspores, Col-0 has microspores with decondensed chromosomes, while most nerd1-2/nerd1-2 microspores appear empty (white arrow heads indicate some normal chromosomes and red arrow head indicate empty microspores). (H) Alexander-stained Col-0 anthers have viable pollen grains (red indicates viable pollen while the green is non-viable). Alexander-stained nerd1-2/nerd1-2 anthers have no viable pollen. Bars = 2.5mm (B-C), 30μm (E), 10μm (F-G), 100μm (H).
Table 1.
nerd1 segregation in F2 populations.
Table 2.
Transmission efficiency of the nerd1-2 allele determined by reciprocal crosses with wild-type Col-0.
Fig 4.
NERD1 is conserved across plant lineages.
(A) Phylogenetic tree from amino acid alignment of NERD1 generated using the Neighbor-joining method in MEGA7. (B) NERD1 alignment showing sequence conservation among species. The green boxes correspond to conserved regions of NERD1 and the lines represent gaps in the alignment. The red box indicates the FG-GAP domain annotated in NERD1 by Langhans, et al., 2017.
Fig 5.
NERD1 co-localizes with a Golgi marker in N. benthamiana epidermal cells.
(A) NERD1 protein domains determined by TMHMM 2.0. (B) NERD1-GFP (green signal) co-localizes with Golgi-mCherry (magenta signal) in N. benthamiana epidermal cells. (C) NERD1-GFP (green signal) does not co-localize with ER-mCherry (magenta signal). (D) NERD1-GFP (green signal) partially overlaps with PM-mCherry (magenta signal). Bars = 15 μm.
Fig 6.
NERD1 expression in plant development.
The NERD1pro::NERD1-GUS reporter (blue signal) is detected in inflorescence (A-B), the flower in stage 9 (C), the pistil in stage 8 (D), the pistil in stage 9 (E), mature pistil (F), mature flower (G), mature anther (H), and SAM and RAM of seedlings (I). Bars = 25 μm.
Fig 7.
NERD1 expression is reduced in developing flowers of some specific low ovule number accessions.
(A) qRT-PCR of NERD1 in Col-0, Altai-5 and Kas-2 inflorescences. (B) qRT-PCR of NERD1 in Col-0, Altai-5 and Kas-2 leaves. (C) qRT-PCR of NERD1 in Col-0, Gre-0, Hh-0, Gifu-0 and Yo-0 inflorescences. (D) qRT-PCR of NERD1 in Col-0, Gre-0, Hh-0, Gifu-0 and Yo-0 leaves. “***” indicates statistical significance at p value<0.001, “**” indicates statistical significance at p value<0.01, “*” indicates statistical significance at p value<0.05 determined by Student’s t-test.
Fig 8.
Overexpression of NERD1 affects plant architecture and fertility.
(A-B) 35S::NERD1-GFP expressed in Col-0 and Altai-5 backgrounds. Bars = 5cm. Scatter plots of total ovule number per flower (C), total branch number (D), total flower number per plant (E), number of flowers within 15 cm of the main shoot (F), and distance between adjacent flowers (G-H) in 35S::NERD1-GFP transformants in the Col-0 and Altai-5 backgrounds compared to untransformed controls.
Fig 9.
Sequence variation in the NERD1 locus correlates with ovule number.
(A) SNPs in and around the NERD1 gene. The blue line indicates the 16 lowest ovule number accessions and the red line indicates the 16 highest ovule number accessions. SNPs are identified based on comparison to the Col-0 reference genome. The red boxes highlight low-ovule number associated SNPs (the turquoise bar indicates two SNPs in adjacent nucleotides, a non-synonymous C-A SNP and a synonymous C-G SNP). Only genomes that were available on the SALK 1,001 genomes browser (http://signal.salk.edu/atg1001/3.0/gebrowser.php) were considered. (B) One of the ovule number-associated SNPs in NERD1 causes a serine-tyrosine substitution in low ovule number accessions. (C) Average ovule number in NERD1 “A” and “C” allele-containing accessions. The “A” allele is significantly associated with lower ovule numbers. “***” indicates statistical significance at p value<0.001 determined by Student’s t-test.