MicroRNA319a-targeted Brassica rapa ssp. pekinensis TCP genes modulate head shape in chinese cabbage by differential cell division arrest in leaf regions

Abstract

Leafy heads of cabbage (Brassica oleracea), Chinese cabbage (Brassica rapa), and lettuce (Lactuca sativa) are composed of extremely incurved leaves. The shape of these heads often dictates the quality, and thus the commercial value, of these crops. Using quantitative trait locus mapping of head traits within a population of 150 recombinant inbred lines of Chinese cabbage, we investigated the relationship between expression levels of microRNA-targeted Brassica rapa ssp. pekinensis TEOSINTE BRANCHED1, cycloidea, and PCF transcription factor4 (BrpTCP4) genes and head shape. Here, we demonstrate that a cylindrical head shape is associated with relatively low BrpTCP4-1 expression, whereas a round head shape is associated with high BrpTCP4-1 expression. In the round-type Chinese cabbage, microRNA319 (miR319) accumulation and BrpTCP4-1 expression decrease from the apical to central regions of leaves. Overexpression of BrpMIR319a2 reduced the expression levels of BrpTCP4 and resulted in an even distribution of BrpTCP4 transcripts within all leaf regions. Changes in temporal and spatial patterns of BrpTCP4 expression appear to be associated with excess growth of both apical and interveinal regions, straightened leaf tips, and a transition from the round to the cylindrical head shape. These results suggest that the miR319a-targeted BrpTCP gene regulates the round shape of leafy heads via differential cell division arrest in leaf regions. Therefore, the manipulation of miR319a and BrpTCP4 genes is a potentially important tool for use in the genetic improvement of head shape in these crops.

Figure 1.

The rosettes and leafy heads of the representative RILs and the transgenic plants overexpressing the BrpMIR319a gene. A to D, The plants of RILs with round (A), oblong (B), cylindrical (C), and cone-like (D) heads at the rosette stage. E to H, The round (E), oblong (F), cylindrical (G), and cone-like (H) heads. I and J, The plant (I) and head (J) of cv Bre. K and L, The plant (K) and head (L) of 319a2-2. M and N, The plants of 319a2-1 (M) and miR319a2-3 (N). O and P, Cross sections of shoot apex of the wild type (O) and 319a2-2 (P) at the heading stage. hl, Head leaf; rl, rosette leaf; s, stalk; yhl, young head leaf. Bars = 5 cm in A to L and 0.5 mm in M and N. [See online article for color version of this figure.]

Figure 2.

Phylogenetic trees of BrpMIR319 and miR319a-targeted BrpTCP genes. A, Multiple sequence alignment of BrpMIR319 genes with Arabidopsis and Chiifu-401-42 homologous genes. B, Multiple sequence alignment of BrpTCP genes with Arabidopsis and Chiifu-401-42 homologs. C, Multiple sequence alignment and complementation of mature miRNAs in BrpMIR319a and the complementary sequences in BrpTCP genes. The phylogenetic trees are constructed using the maximum likelihood method based on the Tamura-Nei model by MEGA5. The sequences of AtMIR159a and AtTCP1 are designated as the out group.

Figure 3.

Box plots and Kruskal-Wallis test for differences between the RIL subpopulations and BrpTCP4-1 expression. The data are from Supplemental Table S2. A, Composition of the RIL population and subpopulations. Sub1 and Sub2 represent the compact and loose subpopulations, respectively. Sub1-C, Sub1-N, Sub1-O, and Sub1-R represent the cylindrical, cone-like, oblong, and round subpopulations, respectively. The numbers of RILs are shown in parentheses. B and C, Box plots for the compact and loose subpopulations (B) and for the cylindrical, cone-like, oblong, and round subpopulations (C). On each box, the central mark is the median, the edges of the box are the 25th and 75th percentiles, and the whiskers extend to the most extreme data points. Asterisks indicate the relative expression levels of BrpTCP4-1. D, P values using the Kruskal-Wallis test.

Figure 4.

Overexpression of BrpMIR319a2 and silencing of BrpTCP genes in the plants transgenic for BrpMIR319a2. A, Schematic diagram showing the structure of T-DNA containing p35S::BrpMIR319a2. B, Southern hybridization of the three transgenic lines (F3 generation) with p35S::BrpMIR319a2 using CaMV 35S probes. C, Northern blotting showing the expression levels of BrpMIR319a2 in young rosette leaves (less than 1 cm long) of the three transgenic lines. D, Real-time PCR showing the expression levels of three BrpTCP genes in young rosette leaves of the three transgenic lines (n = 3; 1 cm long). E, Phenotypes of cv Bre and 319a2-2 transgenic plants at the cotyledon, seedling, and heading stages. EV, Transgenic plants with empty vectors; LB, left border of the T-DNA; M, molecular size markers; PAT, phosphinothricin acetyl transferase; p35S, promoter of CaMV 35S; RB, right border of the T-DNA; Tnos, terminator of the nopaline synthase gene; T35S, terminator of CaMV 35S; WT, wild type. [See online article for color version of this figure.]

Figure 5.

Altered leaf phenotypes of the plants transgenic for BrpMIR319a2. A, The fifth seedling leaves of the wild type (left) and transgenic 319a2-2 (right) at the seedling stage (eight leaves in total). B, The fifth rosette leaves of the wild type (left) and 319a2-2 (right) at the rosette stage (21 leaves in total). C, The last rosette leaves of the wild type (left) and transgenic 319a2-2 (right) at the heading stage (43 leaves in total). D, The third head leaves (counted from outside to inside) of the wild type (left) and transgenic 319a2-2 (right) at the heading stage (43 leaves in total). E, Leaf areas of the first six rosette leaves of wild-type and 319a2-2 plants at the heading stage (43 leaves in total). The number of leaves for each measurement is more than 20. Error bars indicate sd. F, Local bulges in head leaf of cv Bre and 319a2-2. G, Scanning electron microscopy results showing the hydathodes on leaf margins of the fifth heading leaves of the wild type (left) and 319a2-2 (right) at the heading stage. Arrowheads indicate the hydathodes. H, Vein patterns in the local leaf margins of the wild type (left) and 319a2-2 (right) at the heading stage (43 leaves in total). Bars = 1 cm in A, 5 cm in B to D, and 0.5 cm in F to H. [See online article for color version of this figure.]

Figure 6.

Temporal and spatial expression of BrpMIR319a2 and BrpTCP4 genes in three leaf regions. A, Northern blotting showing the expression levels of BrpMIR319a2 in the tip, middle (mid), and base of the fifth rosette leaves and the third heading leaves of the wild type (left) and 319a2-2 (right) at the heading stage. B, In situ hybridization showing the localized expression of BrpMIR319a and BrpTCP4 genes in the fifth rosette leaves and the third heading leaves of the wild type (left) and 319a2-2 (right) at the heading stage. Bars = 100 μm. C, Real-time PCR showing the relative expression of BrpTCP4-1 in the fifth rosette leaves and the third heading leaves of the wild type (left) and 319a2-2 (right) at the heading stage. ACTIN was used as the loading control. The error bars indicate three biological replicates. D, Longitudinal sections through the middle of the first head leaf at the heading stage showing H4 expression (only the top half of the section is shown). Bars = 100 μm.

Figure 7.

Size of the epidermal cells in three leaf regions of rosette and head leaves. A, Scanning electron microscopy images showing the relative cell size in the tip, middle (Mid), and base of the fifth rosette leaves and third head leaves of the wild type and 319a2-2 at the heading stage. B, Epidermal cell sizes of rosette and head leaves of the fifth rosette leaves and third head leaves of the wild type and 319a2-2. Bars = 50 μm. C, Flow cytometry assay showing the distributions of mean leaf cell ploidy (n = 5). The different patterns represent the percentage of cells with nuclei of 2C, 4C, and 8C (from bottom to top). [See online article for color version of this figure.]