Core Mechanisms Regulating Developmentally Timed and Environmentally Triggered Abscission

Abstract

Drought-triggered abscission is a strategy used by plants to avoid the full consequences of drought; however, it is poorly understood at the molecular genetic level. Here, we show that Arabidopsis (Arabidopsis thaliana) can be used to elucidate the pathway controlling drought-triggered leaf shedding. We further show that much of the pathway regulating developmentally timed floral organ abscission is conserved in regulating drought-triggered leaf abscission. Gene expression of HAESA (HAE) and INFLORESCENCE DEFICIENT IN ABSCISSION (IDA) is induced in cauline leaf abscission zones when the leaves become wilted in response to limited water and HAE continues to accumulate in the leaf abscission zones through the abscission process. The genes that encode HAE/HAESA-LIKE2, IDA, NEVERSHED, and MAPK KINASE4 and 5 are all necessary for drought-induced leaf abscission. Our findings offer a molecular mechanism explaining drought-triggered leaf abscission. Furthermore, the ability to study leaf abscission in Arabidopsis opens up a new avenue to tease apart mechanisms involved in abscission that have been difficult to separate from flower development as well as for understanding the mechanistic role of water and turgor pressure in abscission.

Figure 1.

Arabidopsis abscises its cauline leaves when plants wilted by drought (water deficit) are rewatered. A, Well-watered Arabidopsis does not abscise its cauline leaves. The first three cauline leaves to develop are numbered. B and C, A plant that abscised cauline leaves after being wilted by drought and rewatered. The same plant is shown in B and C where B is before gently touching the leaves and C is after touching. The plants shown in A to C are laid on their side for the photo. The image shown in C has abscised leaves placed near where they abscised from. Red tape equals 9.5 mm in length (A–C). D, Number of leaves abscised from well-watered or drought treatments on all inflorescences (primary or secondary). E, Breakstrength required to remove leaves from the plant. F, Breakstrength measured 5 d after plants reached 40% SWC (not rewatered). Thirty percent of plants never develop a third cauline leaf on their primary inflorescence. In these plants the third cauline leaf to develop is on the first secondary inflorescence to develop. Therefore, data for leaves 1 and 2 are exclusively from the primary inflorescence, while data for leaf 3 (E and F) are a mix of leaves from both primary and secondary inflorescences. Data are mean ± se; n = 10 biological replicates (one plant each; D), n = 6 (E and F); t test versus well-watered (D and E); *P < 0.05, **P < 0.01.

Figure 2.

HAE is preferentially expressed in cauline leaf abscission zones from plants exposed to drought stress. A, Well-watered plant and drought/rewatered plant with cauline leaves 1 to 3 used for microscopy. B to G, Cauline leaf abscission zones from a well-watered plant. H to M, Cauline leaf abscission zones from a drought/rewatered plant. Images (B–M) were taken at the same magnification; scale bar = 0.5 mm. Images are representative from eight plants (n = 8).

Figure 3.

HAE is expressed prior to cell separation when leaves become wilted. A to E, A representative soil drying and rewatering time course of cauline leaf 2 including well-watered/88% SWC (A), 80% SWC (B), 40% SWC (C), 1 d after rewatering (D), and 2 d after rewatering (E). Top shows cauline leaf 2, middle is a close-up image of the abscission zone with reflected white light, and bottom is the same as the middle where YFP fluorescence is being imaged. The time course was repeated four times (n = 4) with similar results. Red tape is 9.5 mm in length (A–E, top) and scale bar is 0.5 mm (A–E, middle and bottom) where images were taken under the same magnification. F, Leaf relative water content at different SWC or soil water potentials. Visible wilting occurs at 50% SWC, and plants were rewatered once SWC reached 40%. Relative water content of cauline leaves is shown for wilted cauline leaves as in C and after rewatering as in D. Data are mean ± se; n = 4 biological replicates (one plant each). G, Both HAE and IDA transcripts increase in abscission zones from wilted plants as in C compared with well-watered controls and continue to increase after rewatering. Data are mean ± se; n = 6 biological replicates (one plant each) for well-watered and wilted, n = 3 for rewatered; different letters indicate statistically different quantities within a gene target, t test P < 0.05.

Figure 4.

Mutants defective in floral organ abscission are also defective in drought-induced leaf abscission. A, Wild type (WT; Col-0) plants abscise after drought/rewatering, while hae-3 hsl2-3, ida-2, nev-3, mkk4/5 RNAi, and bop1 bop2 leaves yellow but do not abscise. bop1 bop2 lacks an abscission zone altogether, while hae-3 hsl2-3, ida-2, nev-3, and mkk4/5 RNAi all have a clear band (abscission zone) at the leaf-stem boundary. Images in top panels are wilted leaves (leaf 2) right before rewatering. Images in middle panels are leaves 2 d after rewatering. Red tape is 9.5 mm in length. Images in bottom panels are magnified versions of middle panels showing the abscission zone. Scale bar = 0.5 mm. B, Number of leaves abscised per plant for various genotypes. C, Breakstrength required to remove leaves from the plant for various genotypes. Data are mean ± se; n = 5 biological replicates (one plant each; B and C); t test versus the wild type; *P < 0.1, **P < 0.05, ***P < 0.01. D and E, Number of abscised cauline leaves in hae mutant plants (D) and hsl2 mutant plants (E) after a drought and rewatering treatment. Data are mean ± se; n = 6 biological replicates (one plant each; D and E); t test versus the wild type; ***P < 0.005.

Figure 5.

HAE-YFP expression correlates with partial fruit abscission at the vestigial pedicel abscission zone. A, pHAE:HAE-YFP inflorescence with flowers/siliques numbered by floral position. B, Same inflorescence as in A after flowers or siliques were manually pulled off. After position 7 all siliques tear off at the base of the pedicel, indicating partial abscission has occurred. Before position 8 tearing occurs in the middle of the pedicel. C, Close-up view of B. D, Scar left after pulling off a full-length silique. The top half tears off relatively smoothly, while the bottom half has a rough surface due to the abscission zone only being present at the upper part of the base of the pedicel (Cho and Cosgrove, 2000). E, Close-up view of vestigial abscission zones from inflorescence in A with floral position and stage indicated. Top panels are extended depth of field bright-field images, and bottom panels show HAE-YFP fluorescence. F and G, Pedicel abscission is not drought-inducible. Bright-field (F) or HAE-YFP florescence (G) of a position 11 pedicel abscission zone from plants treated with drought followed by rewatering. All siliques break at the base of the pedicel when pulled after position 8 (H). Red tape is 9.5 mm in length (A–C, H). Scale bar = 100 µm (D–G). Experiments were repeated three times (n = 3) with similar results.

Figure 6.

Mutations in HAE/HSL2 or IDA have minimal seed yield penalty compared with other abscission-defective mutants. A, Seed yield grown under nonstressed conditions. Data are mean ± se; n = 3 biological replicates (one plant each); t test versus the wild type; *P < 0.05. B, A model of drought-induced cauline leaf abscission in Arabidopsis. BOP1 and BOP2 are required for cauline leaf abscission zone development. IDA, HAE/HSL2, MKK4/5, and NEV are all required for drought-induced leaf abscission as well as floral organ abscission.