Dispersal biophysics and adaptive significance of dimorphic diaspores in the annual Aethionema arabicum (Brassicaceae)

Abstract

Heteromorphic diaspores (fruits and seeds) are an adaptive bet-hedging strategy to cope with spatiotemporally variable environments, particularly fluctuations in favourable temperatures and unpredictable precipitation regimes in arid climates. We conducted comparative analyses of the biophysical and ecophysiological properties of the two distinct diaspores (mucilaginous seed (M⁺ ) vs indehiscent (IND) fruit) in the dimorphic annual Aethionema arabicum (Brassicaceae), linking fruit biomechanics, dispersal aerodynamics, pericarp-imposed dormancy, diaspore abscisic acid (ABA) concentration, and phenotypic plasticity of dimorphic diaspore production to its natural habitat and climate. Two very contrasting dispersal mechanisms of the A. arabicum dimorphic diaspores were revealed. Dehiscence of large fruits leads to the release of M⁺ seed diaspores, which adhere to substrata via seed coat mucilage, thereby preventing dispersal (antitelechory). IND fruit diaspores (containing nonmucilaginous seeds) disperse by wind or water currents, promoting dispersal (telechory) over a longer range. The pericarp properties confer enhanced dispersal ability and degree of dormancy on the IND fruit morph to support telechory, while the M⁺ seed morph supports antitelechory. Combined with the phenotypic plasticity to produce more IND fruit diaspores in colder temperatures, this constitutes a bet-hedging survival strategy to magnify the prevalence in response to selection pressures acting over hilly terrain.

Keywords: Aethionema arabicum; abscisic acid (ABA); bet-hedging; dimorphic diaspores; dispersal by wind and water; environmental adaptations; fruit biomechanics; pericarp-imposed properties.

Figure 1

Dispersal, germination, habitat and climate of the dimorphic fruits and seeds of Aethionema arabicum. (a) Aethionema arabicum infructescence showing dehiscent (DEH) and indehiscent (IND) fruit morphs. Mature DEH fruits do not abscise from the mature infructescence, but instead disperse multiple mucilaginous (M⁺) seed diaspores. Mature IND fruit diaspores, containing a single, nonmucilaginous (M⁻) seed, disperse in their entirety from the mature infructescence. (b) Upon imbibition, the M⁺ seed diaspore produces mucilage from the outer cell walls of the seed coat epidermal layers, forming conical papillae of up to 200 µm. (c) By contrast, germination of the IND fruit diaspore is a slower process, in part mediated by pericarp‐imposed dormancy. A single M⁻ seed remains enclosed within the winged IND fruit, unless they are released by external mechanical means, and the radicle protrudes from within the pericarp. Bars, 1 mm. (d) Characteristic stony‐slope and steppe habitat of A. arabicum, taken in the Kırşehir Province in the central Anatolian region of Turkey. (e) Climate diagram of sites in Turkey from which A. arabicum has been collected. The maximum (solid red) and minimum (solid blue) temperatures of an average day for every month are shown. The averages of the hottest day (dashed red) and coldest night (dashed blue) of each month from the last 30 yr are also shown. The dashed magenta line indicates the mean monthly temperature. Mean monthly precipitation (bars) is shown. All climate data were obtained from Meteoblue (developed at the University of Basel, Switzerland, based on weather forecast models of the National Oceanic and Atmospheric Administration and National Centers for Environmental Prediction) from a total of 25 specimens archived at the Global Biodiversity Information Facility (GBIF.org, https://doi.org/10.15468/dl.w63b7k; accessed 31 January 2018). Data are means ± 1 SE. Numbers above the climate diagram indicate typical life cycle stages of diaspore persistence in seed bank (1,5), germination and seedling establishment (2), vegetative, flowering, and reproductive growth (3), and diaspore dispersal and plant senescence (4). (f) Seasonal variation of mean maximum (red) and mean minimum (blue) temperatures at low (800 m) and high (2200 m) elevations that support A. arabicum growth. Data are means ± 1 SE.

Figure 2

Biomechanics of dehiscent (DEH) and indehiscent (IND) Aethionema arabicum fruits. (a, b) Comparative maximum force (F _max) (a) and slope (approx. modulus of elasticity) (b) required to separate fruit valves from fruits under dry conditions. (c) Comparisons of F _max and slope in 17%, 65% and 100% relative humidities show a trend towards gradual decrease in stiffness in both fruits, while F _max also decreases in the IND fruit but remains unchanged in the DEH fruit. (d, e) Characteristic force–displacement curves of mechanical tests in which fruit dehiscence occurred in A. arabicum, revealing distinct fracture biomechanical properties of slow, gradual failure of the DEH fruit (d), and sudden, complete failure for the IND fruit (e). n = 30. Error bars ± 1 SEM.

Figure 3

Comparative fall rates of Aethionema arabicum diaspores during descent. Mean rate of descent (m s⁻¹) of multiple mucilaginous (M⁺) seed diaspores and intact indehiscent (IND) fruit diaspores when released from a height of 1.08 m in still air. IND fruits required greater time to fall, confirming that the pericarp can be regarded as an adaptation for wind dispersal. The fall rates of nonmucilaginous (M⁻) seeds (mechanically removed from IND fruits) are shown for comparison. Differences between M⁺ seed and IND fruit diaspores are significant (P < 0.001). n = 100. Error bars ± 1 SEM.

Figure 4

Frequency distribution patterns (‘seed dispersal shadows’) of the dispersibility of Aethionema arabicum diaspores from the mother plant. (a, b) The dispersibilities of dry multiple mucilaginous (M⁺) seed (a) and dry indehiscent (IND) fruit diaspores (b) differ significantly (F _2,313 = 437.2, P < 0.001) in a 4 m s⁻¹ continuous current of air. M⁺ seeds achieve only short‐distance dispersal from the mother plant, while the longer‐distance dispersibility of IND fruit diaspores may influence colonization patterns in new steppe habitats. (c) Comparisons are made to rare cases where M⁺ seeds imbibe while still attached to the replum, and are subsequently redried and then dispersed. n = 100. Dotted lines indicate log‐normal curves fitted to individual frequency distributions.

Figure 5

Comparative behaviour of Aethionema arabicum diaspores on fine‐grained sandy soil substrata. (a, b) Relative changes in mass of dry and imbibed multiple mucilaginous (M⁺) seed and indehiscent (IND) fruit diaspores upon exposure to dry (a) and water‐saturated (b) sand. By comparing the initial and final weights of the diaspores, a percentage was obtained by which the mass of the dispersal unit had been increased by adherent sand particles, thus illustrating the effectiveness of mucilage production in M⁺ seed diaspores as a means of antitelechory. The unchanged mass of IND fruits suggests the fruit diaspores retain high dispersibility. n = 3, each with 25 replicates. Error bars ± 1 SEM.

Figure 6

Surface water runoff displacement and buoyancy of Aethionema arabicum fruit and seed diaspores. (a) Distance displaced by surface water runoff across a sloped sandpaper plate of dry, redried and imbibed multiple mucilaginous (M⁺) seeds, in comparison to indehiscent (IND) fruits. Nonmucilaginous (M⁻) seeds (not shown), when manually excised from the IND fruit, were displaced by 32.3 ± 1.9 cm. n = 100. Error bars ± 1 SEM. Post hoc pairwise comparisons confirmed that the dimorphic diaspores were statistically different from each other (P < 0.001). (b) Buoyancy of A. arabicum seed and fruit diaspores. Dry M⁺ seeds, redried M⁺ seeds, and imbibed M⁺ seeds show progressive sinking as a result of mucilage extrusion. IND fruit diaspores, by contrast, start to sink after 5 d of shaking, while all M⁻ seeds (not shown) remained floating at the end of the experimental treatment. Symbols are offset on the x‐axis for clarity. n = 3, each with 25 replicates. Error bars ± 1 SEM.