Solid-State Biology and Seed Longevity: A Mechanical Analysis of Glasses in Pea and Soybean Embryonic Axes

Abstract

The cytoplasm of anhydrobiotes (organisms that persist in the absence of water) solidifies during drying. Despite this stabilization, anhydrobiotes vary in how long they persist while dry. In this paper, we call upon concepts currently used to explain stability of amorphous solids (also known as glasses) in synthetic polymers, foods, and pharmaceuticals to the understand variation in longevity of biological systems. We use embryonic axes of pea (Pisum sativum) and soybean (Glycine max) seeds as test systems that have relatively long and short survival times, respectively, but similar geometries and water sorption behaviors. We used dynamic mechanical analysis to gain insights on structural and mobility properties that relate to stability of other organic systems with controlled composition. Changes of elastic and loss moduli, and the dissipation function, tan δ, in response to moisture and temperature were compared in pea and soybean axes containing less than 0.2 g H₂O g^-1 dry mass. The work shows high complexity of structure-regulated molecular mobility within dried seed matrices. As was previously observed for pea cotyledons, there were multiple relaxations of structural constraints to molecular movement, which demonstrate substantial localized, "fast" motions within solidified cytoplasm. There was detected variation in the coordination among long-range slow, diffusive and short-range fast, vibrational motions in glasses of pea compared to soybean, which suggest higher constraints to motion in pea and greater "fragility" in soybean. We are suggesting that differences in fragility contribute to variation of seed longevity. Indeed, fragility and coordination of short and long range motions are linked to stability and physical aging of synthetic polymers.

Keywords: aging; anhydrobiosis; glass fragility; glass stability; longevity; mechanical analysis; seed; storage.

FIGURE 1

Longevity of soybean (open triangles) and pea (solid circles) seeds stored at 35°C and RH ranging from 1 to 92%. The water contents indicated are for embryonic axes treated at the specific RH/temperature combination, which was determined using water sorption isotherms similar to those given by Vertucci and Leopold (1987). The P50 values (time for viability to decrease to 50% was calculated from deterioration time courses (not shown) that were fitted to an Avrami equation (Walters et al., 2005a,b). The vertical lines indicate key properties of the solid at 35°C that were characterized by DMA measurements: water content for the peak and onset of α relaxations (Figure 6) and breaks in the pattern of separation of different relaxation events (Figure 7).

FIGURE 2

Experimental set-up for DMA experiments on pea (A) and soybean (B) embryonic axes. The 1 mm diameter probe was placed on the convex surface of axes between the plumule and the radicle to deliver force evenly across surface.

FIGURE 3

The relationship between temperature and storage modulus (A), loss modulus (B), and tan δ (C) functions in pea (black symbols) and soybean (gray symbols) embryonic axes containing 0.05 g H₂O g^–1 dry mass. Relaxation events are labeled as α, β, and γ, according to convention in mechanical analyses. The slope of the storage modulus with temperature $\left(\frac{\partial E^{\prime }}{\partial T}\right)$ was calculated from the points between –80 and +20°C and the relationship of this slope to changes in moisture is summarized in Table 1. The size of α relaxation (Δ tan δ) was determined from the baseline to the peak of the tan δ curve, as indicated and changes in this parameter with moisture are summarized in Figure 8. Scans are representative of replicate treatments.

FIGURE 4

The relationship between temperature and storage modulus (A), loss modulus (B), and tan δ (C) functions in pea embryonic axes containing 0.09 (black symbols) and 0.02 (gray symbols) g H₂O g^–1 dry mass. Relaxation events are as labeled in Figure 3. Additional high temperature transitions, typical of starchy materials, are also labeled.

FIGURE 5

The relationship between temperature and storage modulus (A), loss modulus (B), and tan δ (C) functions in soybean embryonic axes containing 0.09 (black symbols), 0.07 (black, open symbols), and 0.02 (gray symbols) g H₂O g^–1 dry mass. Relaxation events are as labeled in Figure 3.

FIGURE 6

Plasticization curves for glasses of pea (A) and soybean (B) embryonic axes showing the relationships between water content and temperature of α (circles), β (triangles), and γ (squares) relaxations. Data are provided for the onset of the relaxation (filled symbols) and peak (open symbols). Lines are the calculated regression relationships (Table 2). Two lines were calculated for the γ relaxation data for soybean using data ≥ and ≤0.05 g H₂O g^–1 dry mass.

FIGURE 7

Temperature differences separating α and β (A) and β and γ (B) relaxation events in pea (solid circles) and soybean (open triangles) embryonic axes. Data for pea cotyledons (Ballesteros and Walters, 2011; blue Xs) are provided for comparative purposes. In (A), the temperature for the onset of α relaxation is subtracted from the peak of the β relaxation and in (B), the temperature for the onset of the β relaxation is subtracted from the peak of the γ relaxation. The lines represent regression relationships calculated for tissues having water contents ≥ and ≤0.05 g H₂O g^–1 dw. In (A), relationships were significant (P < 0.05) for T_α – T_β in soybean axes (all wc), pea cotyledons (all wc) and pea axes (wc ≥ 0.05 g H₂O g^–1 dw). The slopes for pea and soybean axes were similar (P > 0.05) and significantly greater (P < 0.05) than pea cotyledons. The x intercepts were calculated as 0.09 g H₂O g^–1 dw for pea and soybean axes and 0.17 g H₂O g^–1 dry mass for pea cotyledons. In (B), relationships were significant (P < 0.05) for T_β – T_γ in only soybean axes and pea cotyledons at water contents ≤0.05 g H₂O g^–1 dw.

FIGURE 8

The size of α (A) and β (B) relaxation signals in pea (solid circles) and soybean (open triangles) embryonic axes. Data for pea cotyledons (Ballesteros and Walters, 2011; blue Xs) are provided for comparative purposes. Signal size was calculated from Δ tan δ functions as indicated in Figure 3C. The lines represent regression relationships calculated for all water contents except for soybean axes in (A), where regressions were calculated separately for water contents ≥ and ≤0.05 g H₂O g^–1 dw. All regressions were significant (P > 0.05). Slopes for the size of α relaxation were similar in pea and soybean embryonic axes at water contents ≥0.05 g H₂O g^–1 dw (P > 0.05) and for β relaxation were similar for all tissues.