Plant neighbor detection through touching leaf tips precedes phytochrome signals

Abstract

Plants in dense vegetation compete for resources, including light, and optimize their growth based on neighbor detection cues. The best studied of such behaviors is the shade-avoidance syndrome that positions leaves in optimally lit zones of a vegetation. Although proximate vegetation is known to be sensed through a reduced ratio between red and far-red light, we show here through computational modeling and manipulative experiments that leaves of the rosette species Arabidopsis thaliana first need to move upward to generate sufficient light reflection potential for subsequent occurrence and perception of a reduced red to far-red ratio. This early hyponastic leaf growth response is not induced by known neighbor detection cues under both climate chamber and natural sunlight conditions, and we identify a unique way for plants to detect future competitors through touching of leaf tips. This signal occurs before light signals and appears to be the earliest means of above-ground plant-plant signaling in horizontally growing rosette plants.

Fig. 1.

Shade-avoidance traits in a developing Arabidopsis canopy. (A) Representative photographs of the inner nine plants of a canopy at different LAIs with above each canopy photo two individual rosettes of the same age, of which the left is always a single-grown, and the right is a canopy-grown plant. (B) R:FR of reflected light inside the canopy, petiole angles (left y axis, black lines) and lengths (right y axis, blue lines) of the third-youngest leaves of canopy-grown (filled symbols) and single-grown (open symbols) plants (n = 27 from three individual canopies for canopy plants, n = 9 for single-grown plants). Petiole angles from canopy plants are significantly different from single plants from LAI = 0.65 onward; petiole lengths from LAI = 1.27 onward (Student’s t test, P < 0.05). (C) XTH15 (left y axis, black lines) and ATHB2 (right y axis, blue lines) relative expression (quantitative RT-PCR) in canopy-grown (filled symbols) and single-grown (open symbols) plants (n = 4). Expression from both genes is significantly induced in canopy plants at LAI = 2.9 (Student’s t test, P < 0.05). Data represent means ± SE. Data for single plants are plotted against canopy LAI and represent plants of the same age as their equivalent in the canopy at a specific LAI.

Fig. 2.

FR- and neighbor-induced hyponasty. (A) Petiole angles after 6 h of R:FR 1.5, R:FR 1.2, and R:FR 0.9 (n = 10). (B) Percentage from total leaves that are hyponastic in a Col-0 canopy and in a wei8-1 canopy. (C) Representative photograph of hyponastic wei8-1 canopy and (D) of a canopy with supplemental R LEDs. (E) Percentage from total leaves that are hyponastic in a canopy without LEDs, with green supplemental LEDs, and with red supplemental LEDs. Data represent means ± SE. Asterisk indicates significant difference; NS, not significant (one-way ANOVA, P < 0.05, n = 18 plants from two individual canopies).

Fig. 3.

Touch-induced hyponasty. Representative photographs of leaves hyponastic through touch in (A) canopy-grown plants, (B) two single-grown plants with touching leaf tips, and (C) a single-grown plant touching a transparent tag. (D) Percentage of total leaves in canopy that is hyponastic, and percentage of hyponastic leaves that is being touched by other leaves. (E) Petiole angles after 24 h and 48 h from leaves of plants touching a transparent tag (touch) and untouched leaves from the same plants (systemic) or control plants (n = 10). Data represent means ± SE. Asterisk represents significant difference (one-way ANOVA, P < 0.05).

Fig. 4.

Simulations of a developing Arabidopsis canopy. (A) Virtual representation of a canopy at LAI 0.65. The brightness of the color corresponds to the R:FR on the leaves. (B) R:FR output of virtual canopy model for lamina of leaves 6–9 from canopy-grown and single-grown plants. These are the most responsive leaves and include the leaf measured in Fig.1 (leaf 8). Data for single plants are plotted against canopy LAI and represent plants of the same age as their equivalent in the canopy at a specific LAI. Model output for all leaves is given in Fig. S6. (C) Average R:FR output for all lamina of a virtual canopy plant at LAI = 0.65 with manipulated leaf angles in background light with R:FR 2.1 or 1.2. Data represent means from nine simulated inner canopy or single-grown plants.