Skip to main content
NCBI home page
Annals of Botany logoLink to Annals of Botany
. 2018 Dec 8;123(5):815–829. doi: 10.1093/aob/mcy213

The role of the testa during the establishment of physical dormancy in the pea seed

Anna Janská 1, Eva Pecková 1, Bogna Sczepaniak 1, Petr Smýkal 2, Aleš Soukup 1,✉
PMCID: PMC6526324  PMID: 30534972

Abstract

Background

A water-impermeable testa acts as a barrier to a seed’s imbibition, thereby imposing dormancy. The physical and functional properties of the macrosclereids are thought to be critical determinants of dormancy; however, the mechanisms underlying the maintenance of and release from dormancy in pea are not well understood.

Methods

Seeds of six pea accessions of contrasting dormancy type were tested for their ability to imbibe and the permeability of their testa was evaluated. Release from dormancy was monitored following temperature oscillation, lipid removal and drying. Histochemical and microscopic approaches were used to characterize the structure of the testa.

Key results

The strophiole was identified as representing the major site for the entry of water into non-dormant seeds, while water entry into dormant seeds was distributed rather than localized. The major barrier for water uptake in dormant seeds was the upper section of the macrosclereids, referred to as the ‘light line’. Dormancy could be released by thermocycling, dehydration or chloroform treatment. Assays based on either periodic acid or ruthenium red were used to visualize penetration through the testa. Lipids were detected within a subcuticular waxy layer in both dormant and non-dormant seeds. The waxy layer and the light line both formed at the same time as the establishment of secondary cell walls at the tip of the macrosclereids.

Conclusions

The light line was identified as the major barrier to water penetration in dormant seeds. Its outer border abuts a waxy subcuticular layer, which is consistent with the suggestion that the light line represents the interface between two distinct environments – the waxy subcuticular layer and the cellulose-rich secondary cell wall. The mechanistic basis of dormancy break includes changes in the testa’s lipid layer, along with the mechanical disruption induced by oscillation in temperature and by a decreased moisture content of the embryo.

Keywords: Hardseedness, light line, macrosclereid, physical dormancy, Pisum sativum seed, seed coat, subcuticular lipids, testa, water permeability

INTRODUCTION

The timing of germination is critical for the successful completion of a plant’s life cycle. Some species have evolved the means to delay germination, a phenomenon referred to as dormancy (Baskin and Baskin, 2014). A major form of dormancy involves the seed developing a water-impermeable coat or testa (de Souza et al., 2012), thereby inhibiting the uptake of water (imbibition) required to initiate the growth of the embryo (de Souza and Marcos-Filho, 2001). This physical dormancy is particularly widespread among legume species, although it is thought to have evolved independently several times (Smýkal et al., 2014; Willis et al. 2014). The testa gradually becomes impermeable during the later stages of seed maturation (Gillikin and Graham, 1991); the biochemical basis of this hardening varies among families and species. Testa thickness is influential (Miao et al., 2001), but is not the sole basis of impermeability (de Souza and Marcos-Filho, 2001), which is also affected by the testa’s composition (Smýkal et al., 2014; Cechová et al., 2017; Hradilová et al., 2017). Cell wall thickness in the outer integuments varies throughout seed development (Verdier et al., 2013), along with the accumulation of polyphenolic compounds; oxidation products of the latter are thought to contribute to impermeability in both barrel medic (Medicago truncatula) and pea (Pisum sativum) (Marbach and Mayer, 1974; Werker et al., 1979; Moïse et al., 2005), but also modifies permeability in Arabidopsis (Debeaujon et al., 2000). Comparisons among a set of black-seeded soybean (Glycine max) accessions have suggested that the abundance of epicatechin correlates positively with hardseededness (Zhang et al., 2011). Callose, deposited either in the macrosclereid layer (Finch-Savage and Leubner-Metzger, 2006; de Souza et al., 2012) or the parenchyma (Bhalla and Slattery, 1984), has been implicated as forming a barrier to water uptake in some dormant seeds. However, based on probing with callose-specific antibodies, this appears not to be the case in the pea seed (Hradilová et al., 2017).

Mapping progeny from a cross between wild (Glycine soja) and cultivated soybean led Sun et al. (2015) to the conclusion that impermeability is controlled by the product of a gene that encodes a calcineurin-like transmembrane protein, which is deposited most abundantly in the testa’s macrosclereid layer. A second gene, encoding endo-1,4-β-glucanase, has also been identified (Jang et al., 2015); allelic variants of this gene are associated with variation in the accumulation of β-1,4- and β-(1,3)(1,4)-glucans, which reinforce testa impermeability. In both Arabidopsis thaliana and the legume species, seed permeability is altered in mutants compromised for either the synthesis (Verdier et al., 2013) or the transport (Fedi, 2017) of extracellular lipids. According to Molina et al. (2008), waxy polymers are present in the A. thaliana and Brassica napus testa, but their contribution to testa permeability has not been established.

Physical dormancy can be broken by manipulating the seeds’ thermal and/or humidity environment, as well as by mechanical abrasion, microbial action or passage through an animal gut (Baskin and Baskin, 2000; de Souza and Marcos-Filho, 2001; Long et al., 2015). The primary pathway of water entry into the pea seed (as well as that of some other legumes) remains unclear, despite much discussion in the literature (Baskin et al., 2000; Meyer et al., 2007; Ranathunge et al., 2010; Baskin and Baskin, 2014; Smýkal et al., 2014, Hradilová et al., 2017). While the consensus is that one or more of the strophiole, hilum and micropyle are the most likely candidates, Ma et al. (2004) have suggested that in soybean small dispersed fissures in the cuticle, rather than any particular specialized structure, are the entry point. The testa’s surface is paved by macrosclereids and covered with cuticle (Smýkal et al., 2014; Hradilová et al., 2017), which has a fatty acid composition rather different from that of the shoot cuticle (Shao et al., 2007). The terminal cap of the macrosclereid is separated from the rest of the cell by a so-called ‘light line’ (LL), the prominence of which has been implicated as a major determinant of testa impermeability (Stevenson, 1937; Harris, 1987); the nature of the LL, however, remains unresolved (Hradilová et al., 2017). Here, the anatomical structure and chemical composition of non-dormant and dormant pea seeds have been compared, with a particular focus on identifying the primary site of water entry into the testa and the mechanisms of dormancy release.

MATERIALS AND METHODS

Plant material

Six pea accessions, three producing dormant seeds and three non-dormant seeds, were compared. The three accessions producing dormant seeds, all belonging to ssp. elatius, were JI64 from Turkey, L100 from Israel and VIR320 from Palestine; the VIR320 testa, unlike that of the other two ssp. elatius accessions, has a smooth surface (Bogdanova et al., 2012). The three producing non-dormant seeds were landrace JI92 from Afghanistan and the commercial cultivars Cameor from France and Terno from the Czech Republic. Seeds of these six accessions were obtained from the John Innes Pisum Collection (Norwich, UK; JI92, JI64), INRA (urgi.versailles.inra.fr/siregal/siregal/grc.do; Cameor), Vavilov Institute Research of Plant Industry (St Petersburg, Russia; VIR320), Russian Academy of Sciences (Novosibirsk, Russia; L100) and the Czech National GeneBank (Prague; Terno). Plants were grown in a glasshouse during February–April of both 2016 and 2017. Mature seeds were air-dried and stored for between 1 and 6 months at room temperature in the dark. Dormancy was quantified following the methods given by Hradilová et al. (2017). Immature seeds were harvested at defined developmental stages, fixed in 0.1 m phosphate buffer (pH 7.2) containing 1 % v/v formaldehyde, then stored at 4 °C.

Imbibition

A sample of ten seeds per accession per treatment was imbibed on moistened cellulose wadding at room temperature. Before imbibition, the hilum, the strophiole or the hilum plus strophiole plus micropyle was covered with warm lanolin, while control seeds were left untreated. The swelling of the seeds was captured at 1-h intervals using a camera, and the time-related changes in size were analysed using NIS Elements software (www.nisoftware.net).

Treatments to influence testa properties

Dry seeds were subjected to a number of treatments to break dormancy. A thermocycling treatment (25 °C, ~30 % RH for 24 h/70 °C, ~10 % RH for 24 h, cycled over 2 weeks) was applied by holding dormant seeds in a Vacucell programmable incubator (BMT, Brno, Czech Republic). Seeds of JI92 and JI64 were freeze-dried in a Finn-Aqua Lyovac GT2 device (www.sterislifesciences.com) for 24 h. In order to define the role of the embryo in the dormancy release process, comparisons were made between seeds containing the embryo and testas without an embryo. Intact seeds of JI64 were immersed in chloroform for 3 d, then washed in distilled water to remove surface lipids. Cross-sections of JI92 and JI64 seeds were immersed in either chloroform or hexane at 60 °C for 48 h to solubilize specific classes of lipid. Both treated and non-treated sections were stained with 0.01 % w/v Sudan red 7B (Sigma-Aldrich, St Louis, MO, USA) to assay for waxy compounds. The sections were observed by bright-field microscopy.

Seed anatomy

The testa was peeled from mature seeds of each accession (from at least five seeds per accession) and held in 2 % w/v sucrose under vacuum for 1 h, after which an equal volume of Shandon™ Cryomatrix™ embedding medium (www.thermofisher.com) was added and the mixture was shaken overnight. The testa samples were frozen to −25 °C and cut with a cryotome (Shandon SME, Astmoor, UK) into 12-µm transverse sections (Soukup and Tylová, 2014), which were stained with 0.01 % w/v aqueous toluidine blue (Sigma–Aldrich), 0.01 % w/v (in 0.1 m K2HPO4, pH 9) aniline blue fluorochrome (Anilinblau, Dr G. Grubler & Co., Leipzig, Germany) or 0.01 % w/v Sudan red 7B (Sigma–Aldrich), following Soukup (2014). The presence of callose was assayed with an immunogenicity-based assay, using an antibody that recognized (1,3)-β-glucan (Soukup, 2014). For scanning electron microscopy, intact mature seeds were vacuum-dried and gold-coated using an SCD 050 sputter coater (Bal-Tec) before being imaged using a JSM-6380LV device (JEOL, Tokyo, Japan). The structure of the macrosclereids was observed using confocal microscopy: intact testas were observed using two-photon excitation (720 nm), while the subcuticular wax was detected from its autofluorescence. Individual macrosclereids were released from the testa by exposure to 0.1 % w/v pectinase in 0.1 m citrate buffer (pH 5.1), and were then stained with aqueous 0.01 % w/v calcofluor white.

Permeability

The surface permeability of the seeds was assayed by immersing intact seeds in either aqueous 0.1 % w/v periodic acid (H5IO6) (Sigma–Aldrich) followed by reducing solution (2 h) according to Soukup et al. (2007) or in aqueous 0.01 % w/v ruthenium red (Sigma–Aldrich) for 2 h. Modifications to the cell wall induced by the incursion of periodic acid were detected using Schiff’s solution (Pearse, 1968). For materials stained with ruthenium red, excess dye was removed by rinsing in distilled water. The permeability of the testa was quantified on the basis of the penetration depth of the two tracer compounds. Periodic acid solution is not a specific apoplastic tracer and its oxidative properties might affect structures such as membranes. In spite of known limitations, it is highly suitable for the demonstration of penetration of surrounding solution across the seed-coat surface. For this purpose the periodic acid–Schiff procedure is more reliable than tracers, which might be washed out of the cell wall during subsequent processing (Peckova et al., 2016).

Statistics

All statistical evaluations used routines implemented in the NCSS 9 package (www.ncss.com). The effects of entry and treatment on the various traits were tested using conventional analysis of variance and the Tukey–Kramer multiple comparison test.

RESULTS

The primary site of water entry differs between dormant and non-dormant seeds

The seed of the three ssp. elatius entries were significantly more dormant than those of the cultivated types; there were also distinct differences in the pigmentation and thickness of the testa. Imbibition for 24 h hardly altered the appearance of dormant seeds, whereas non-dormant seeds exhibited extensive changes, particularly in the region of the strophiole (Supplementary Data Fig. S1). After a 17-d imbibition period, half of the VIR320 seeds had germinated, whereas none of the L100 and JI64 seeds had done so. Exploration of the site of water entry in the cultivated type entries (germination in the dormant types was too slow and too unsynchronized for this experiment), performed by sealing the hilum, strophiole and/or micropyle, identified the strophiole as the key structure (Fig. 1). Sealing had no effect on the germination of VIR320 seeds. The outer surface of both JI64 and L100 seeds was uneven (‘gritty’), resulting from protrusions emanating from the tips of the macrosclereids, which extended above the LL (Figs 2A–D and 3A, D, G); in contrast the surface of all three non-dormant type seeds, as well as seeds of the partially dormant type VIR320, was quite smooth (Figs 2E, F and 3B, C; Supplementary Data Fig. S1).

Fig. 1.

Fig. 1.

Open in a new tab

Control of water entry during the germination of non-dormant seeds of Terno, Cameor and J192. The hilum (H), strophiole (S) and hilum + strophiole + micropyle (HSM) were sealed with warm lanolin; control seeds (C) were not treated with lanolin. Statistically significant within-treatment differences are indicated by different letters at the head of each column.

Fig. 2.

Fig. 2.

Open in a new tab

Testa surface of dormant seeds of JI64, L100 and VIR320 before (A, C, E) and after (B, D, F) thermocycling treatment applied to (A, B) JI64, (C, D) L100 and (E, F) VIR320 seeds. Fissures (arrowed) developed on the seed surface of JI64 and L100, but not on the surface of the other seeds. Arrowheads indicate gritty structures on the surface of JI64 and L100 seeds. Scale bars = 0.5 mm (50 μm in the insets).

Fig. 3.

Fig. 3.

Open in a new tab

Variation in testa permeability between dormant (JI64: A, D, G, J, M), non-dormant (JI92: B, E, H, K, N) and Terno (C, F, I, L, O) seeds. Permeability was assayed either using periodic acid–Schiff solution (violet staining) or ruthenium red (red staining). (A–C) Schiff solution staining without periodic acid treatment. Scale bars = 50 μm. (D–F) Periodic acid treatment. Scale bars = 50 μm (12.5 μm in the insets). (G–I) Ruthenium red staining. Scale bars = 50 μm. (J–L) Ruthenium red staining around the hilum. Scale bars = 100 μm. (M–O) Ruthenium red penetrating the testa. Arrows in (A–L) indicate the LL.

Variation in testa permeability

The permeability of the testa was assessed by monitoring the uptake of either periodic acid or ruthenium red. In non-dormant seeds, both reagents were able to enter the seed through the seed side, hilum and strophiole (Fig. 3E, F, H, I, K, L), and both penetrated the cuticle and the terminal caps, passing through the LL. Some parts of the testa were more permeable than others, particularly where the testa was wrinkled (Fig. 3N, O). The testa of non-imbibed, non-dormant seeds was free of any significant cracks, presumably reflecting the greater elasticity of their testa compared with that formed by dormant seeds. In dormant seeds, by contrast, the reagents passed through the cuticle and terminal caps, but were blocked at the LL (Fig. 3D, G, J). Paradermal sections revealed that ruthenium red reached the pectin-rich middle lamella of the macrosclereids (Fig. 4A–C). The cuticle of dormant seeds featured a number of fine cracks, some of which developed in suitable environmental conditions into large fissures, thereby enabling release from dormancy (Fig. 2B, D). Overall, the conclusion was that, once the cuticle had been disrupted by local fissures after exposure to proper environmental conditions, the site of water uptake into dormant seeds was dispersed over the whole testa surface in dormant seeds.

Fig. 4.

Fig. 4.

Open in a new tab

The presence of pectin in the testa cell wall, as shown by ruthenium red staining of paradermal sections. (A) In JI92, the macrosclereid cavity (arrow) branches into fine channels and the cavity is filled with extracellular waxy material. (B) An oblique paradermal section of Terno demonstrates the arrangement of the central cavity of the macrosclereids at various distances from the cuticle; fine channels near the cuticle are indicated by arrows, and larger ones at a greater distance from the cuticle by asterisks. (C) At the surface of a JI64 seed, ruthenium red stains only the tops of the gritty surface. (A–C) Macrosclereid cavities are indicated by arrows. Compound middle lamellae are indicated by arrowheads. Scale bars = 25 μm.

Composition and structure of the testa surface

Light microscopy was unable to differentiate between the cuticles of the six entries. Both Cameor and Terno (non-dormant entries) contained a high level of polyanionic cell wall components (pectins) in the compound middle lamella of their macrosclereids. The non-dormant (although pigmented) entry JI92 contained less of these polyanionic compounds in the pigmented area of its testa. Metachromasy in the dormant seed was mostly limited to the macrosclereid tips. The LL was stainable only by the aniline blue fluorochrome, indicative of a predominance of callose or similar compounds. Nevertheless, probing with an anti-callose antibody failed to confirm callose as a major component of the LL (data not shown), as also suggested by Hradilová et al. (2017). A detailed inspection demonstrated that the LL was associated with the presence of compact cell wall, which reduced the space for protoplast (Fig. 5B), just below the region enriched for waxy compounds. The presence of non-cuticular waxy extracellular material in the tips of the macrosclereids just above the LL was revealed by both Sudan red staining (Fig. 6A, B) and its autofluorescence under both UV and two-photon 720-nm excitation (Fig. 5D). The waxy layer was present in the seeds of both dormant and non-dormant entries. However, its patterning in JI92 was different from that in JI64 (Fig. 6A, B). Images acquired by scanning electron microscopy also identified the structure of the macrosclereid layer, where a waxy layer covered the upper surface of the macrosclereids (Fig. 5A).

Fig. 5.

Fig. 5.

Open in a new tab

Micrographs showing the detailed structure of the macrosclereids. (A) Longitudinal section of the JI64 testa, showing the outer part of macrosclereids, including their terminal caps, composed of compact extracellular material, and the subcuticular waxy layer. Arrows indicate the lower surface of the LL. Scale bar = 10 μm. (B) Individual macrosclereids of JI64 released from the testa after pectinase treatment and stained with calcofluor white. Arrows indicate the lower surface of the LL and asterisks the macrosclereids’ internal channels. The inset box indicates enlarged portion of the macrosclereids presented on the left. Scale bar = 10 μm. (C) Three-dimensional projection illustrating deposition of waxy material within the branched internal macrosclereid cavity. (D) Autofluorescence (two-photon 720-nm excitation) due to the waxy content of the branched internal macrosclereid cavity, as seen from the testa surface. Scale bar = 10 μm. (E) Surface paradermal section showing differences in the macrosclereid cavity at various distances from the cuticle. The macrosclereid cavity is branched at the terminal cap; the branches are fine in the vicinity of the LL and more extensive in the basal part of the macrosclereid. Scale bar = 25 μm.

Fig. 6.

Fig. 6.

Open in a new tab

Lipids in the testa visualized in cross-sections of JI64 (A, C, E, F) and JI92 (B, D). The testas were stained with Sudan red for 2 h (A, B), treated with hexane before staining with Sudan red (C, D) or treated with sulphuric acid (E, F). The sections were photographed either in bright-field illumination (A–E) or with UV excitation (F) to generate autofluorescence. The LL is indicated by arrows, the subcuticular waxy layer by stars and the position of cuticle by arrowheads. Scale bars = 25 μm.

Dormancy release

Once physical dormancy had been broken by thermocycling (Fig. 2B, D, F), it was clear that the LL barrier was breached at the hilum, strophiole and seed side in each of the dormant entries (Fig. 7A–U). The implication was that the waxy layer was modified by the thermocycling treatment. When intact seeds of JI64 (dormant type) were immersed in chloroform for 3 d then imbibed in the presence of either ruthenium red or periodic acid, it was apparent that the macrosclereids had become permeable to periodic acid over the entire testa surface (Fig. 8A–C), whereas ruthenium red staining was limited mostly to side areas of the seed (Fig. 8D–F). The explanation for this apparent inconsistency was that ruthenium red preferentially stains pectins, and this makes it a far less sensitive and reliable assay, particularly in pectin-poor dormant types, than is periodic acid. Yet the strong induction of increased permeability is obvious also in this case. The development of large fissures in the testa of dormant seeds following thermocycling (Fig. 2B, D) implied the induction of mechanical forces. Major cracks formed in the testa of the dormant JI64 after drying, particularly after rehydration. The testas of JI92 (whether or not the embryo was present) and JI64 (without an embryo) did not change in response to dehydration, but those of JI64 containing an embryo became extensively ruptured (Fig. 9D, E). The ruptures appeared similar to those induced by both thermocycling (Figs 2B, D and 9G, H) and chloroform treatment (Fig. 9F). The implication was that the fissuring of the testa was strongly promoted by the reduction and expansion of the embryo as it dehydrated and rehydrated.

Fig. 7.

Fig. 7.

Open in a new tab

Effect of thermocycling on testa permeability. Surface permeability of VIR320 (A–C, M–O), L100 (D–F, P–R) and JI64 (G–I, S–U) seeds after 2 weeks of thermocycling as assayed with periodic acid (A–I), with Schiff’s solution without periodic acid (J–L) or with ruthenium red (M–U). The arrows point to the LL. (A, D, G, J, M, P, S) Seed side. Scale bars = 50 μm. (B, E, H, K, N, Q, T) Hilum. Scale bars (B, E, H) = 200 μm, (K, Q) = 100 μm, (N, T) = 50 μm. (C, F, I, L, O, R, U) Strophiole. Scale bars (C, I) = 200 μm, (F, L, O, R, U) = 100 μm.

Fig. 8.

Fig. 8.

Open in a new tab

Changes in the permeability of the testa of the dormant entry JI64 following chloroform treatment. Surface permeability was assayed using periodic acid (A–C), ruthenium red (D–F) or Schiff’s solution without periodic acid (G–I). The LL is indicated by an arrow. (A, D, G) Seed side. Scale bar = 50 μm. (B, E, H) Hilum. Scale bar = 100 μm. (C, F, H) Strophiole. Scale bar = 100 μm. (J–L) Macrographs of intact JI64 seeds following ruthenium red staining of chloroform-treated material. (J) Seed side, (K) hilum, (L) strophiole.

Fig. 9.

Fig. 9.

Open in a new tab

Cracking of the testa after freeze-drying and chloroform treatment (A, B). Control (no treatment) (C, D). Immediately after freeze-drying treatment. Isolated testas are shown on the left and testas with embryo on the right (E, F). Rehydration following freeze-drying treatment. Isolated testas are shown on the left and testas with embryo on the right (G, H). Thermocycling treatment (A, C) J192, (B, D, F, G, H) J164.

Composition and development of the subcuticular waxy layer above the LL

The organic solvents (chloroform and hexane) used to remove subcuticular wax from seed cross-sections left only the cuticle stainable with Sudan red, which implied that the waxy material was mostly non-polymeric (Fig. 6C, D). This conclusion was supported by the outcome of the sulphuric acid treatment, which dissolved everything except for cutin in seed cross-sections of both JI64 and JI92 (Fig. 6E, F). Cross-sections of immature seeds (JI92 sampled at 10, 14 and 17 d post-anthesis and JI64 sampled at 9, 14, 15 and 17 d post-anthesis) stained with Sudan red and toluidine blue showed that the waxy layer became detectable during day 15 in JI64, and within a further 2 d the macrosclereids appeared as they did in mature seeds (Fig. 10I, J). The development of the waxy layer was similar in JI92 (Fig. 10B, C). The LL formed just below the Sudan red-stained waxy layer, even in JI64, where Sudan red staining extended through the whole macrosclereid cap, including the part above the LL. A large quantity of phenolic compounds (most likely tannins) was responsible for the negative metachromatic toluidine blue staining in the protoplasts of non-dormant seed types and the first two developmental stages of dormant ones. Following the development of secondary walls in the tips of the macrosclereids, tannins also became detectable in the secondary walls present in JI64 (Fig. 10M) but not in those of JI92, and the volume of protoplasts was reduced. In both entries, the development of secondary cell walls revealed the LL. Positive metachromasy generated by the presence of polyanionic compounds was notable in the upper parts of the JI92 macrosclereids, but faded as the testa developed, remaining only between the cuticle and the non-cuticular waxy layer (Fig. 10F). During development, the upper portion of the JI64 macrosclereids stained even more intensely blue, especially just below the LL (Fig. 10M, N), most likely reflecting the accumulation of cellulose and possibly other uncharged polysaccharides. Positive metachromasy was observed in the macrosclereids’ compound middle lamella and in the region between the cuticle and the non-cuticular waxy layer.

Fig. 10.

Fig. 10.

Open in a new tab

Development of the cuticle, subcuticular waxy layer and the LL. Cross-sections of seeds of JI92 harvested at 10, 14 and 17 d post-anthesis (DPA) (A–F) and JI64 harvested at 9, 14, 15 and 17 DPA (G–N) stained with Sudan red (A–C, G–J) or toluidine blue (D–F, K–N). The LL is indicated by arrows, the subcuticular waxy layer by stars and the cuticle by arrowheads. (A, D) 10 DPA, (B, E) 14 DPA, (C, F) 17 DPA; (G, K) 9 DPA, (H, L) 14 DPA, (I, M) 15 DPA, (J, N) 17 DPA. Scale bars = 50 μm (20 μm in inset).

The layer of waxy material formed within the fine channels of the terminal caps was detectable by bright-field (Figs 4, 5E and 6), confocal (Fig. 5B) and scanning electron (Fig. 5A) microscopy. The cavity (cell lumen) of macrosclereids appeared branched at the terminal caps (where it appeared as fine channels) and was filled with waxy material (Figs 4 and 5C, D). However, in the vicinity of the LL the cavity’s size was small in comparison with that of the lower part of macrosclereids (Fig. 5B, E). Underneath the LL, the cavity opened to channel filled with the remains of degenerating protoplast, which in dormant seeds was enriched for tannins (Figs 4, 5B, E and 10E, F, M, N). The volume of the palisade and counter-palisade layers in the hilum of both dormant and non-dormant seeds responded readily to changes in the seed’s moisture status; the effect of differential swelling was to alter the extent of the hilum’s aperture (Fig. 11).

Fig. 11.

Fig. 11.

Open in a new tab

The hilum aperture responds to water entry. Cross-sections of JI64 (A, C) and JI92 (B, D) seeds taken before (A, B) and after (C, D) imbibition. Arrows indicate the aperture. Scale bar = 200 μm.

DISCUSSION

During the late stages of the pea embryo’s development, it is the hilum that regulates the drying process, opening when humidity is low and closing when it rises. The mechanical basis for this process involves differences in swelling and shrinkage of palisade and counter-palisade cells, a process that was observable in cross-sections of both JI92 and JI64 seeds. When its moisture content reaches ~10 %, the seed enters dormancy, a state in which its level of metabolic activity is minimal (Hyde, 1954; Lush and Evans, 1980). The requirements for the release of dormancy have been widely discussed (see review by Baskin and Baskin, 2014). One possible scenario is that the primary rehydration pathway operates via specific structures (Baskin, 2003; Gama-Arachchige et al., 2013; Jaganathan et al., 2017), while an alternative holds that minor fissures form in the cuticle across the whole seed surface (Chachalis and Smith, 2000; Ma et al., 2004). Here, the finding was that the principal site of water entry differed between dormant ssp. eliatus seeds and non-dormant commercial pea cultivars. A contribution of the strophiole, as similarly suggested by both Karaki et al. (2012) and Gama-Arachchige et al. (2013), was implied by the observation that its blockage significantly decreased the rate of imbibition (Fig. 1), while fissures appeared to be rather frequently formed in its vicinity. In certain Phaseoleae species, the strophiole (and also the hilum and micropyle) have been identified as important for water entry (Korban et al., 1981; Agbo et al., 1987).

Water uptake may not depend on a localized disruption of the cuticle of dormant seeds. The subcuticular macrosclereids can impose a strong barrier; while water entry is not inhibited by the LL in non-dormant seeds, it does function as an effective barrier in dormant ones (Serrato-Valenti et al., 1993; de Souza et al., 2012, Hradilová et al., 2017). The nature of the LL remains unresolved. Various suggestions have been made to account for its distinct optical appearance. Its acceptance of aniline blue fluorochrome has been taken to imply that it contains abundant callose (de Souza et al., 2012), but immunogenicity-based experiments in pea using an anti-callose antibody do not support this hypothesis (Hradilová et al., 2017). According to Harris (1983), the pea LL represents the junction between the cellulosic tips of the macrosclereids and a subcuticular waxy layer; in white clover, Martens et al. (1995) have proposed that it forms as a result of an altered orientation of cellulose microfibrils in palisade cell walls; Ma et al. (2004) have suggested that in soybean the LL is formed as a result of its tight appression to secondary cell walls; finally, Manning and van Staden’s (1985) suggestion was that the LL in Erythrina sp. is an optical phenomenon linked to the formation of a structural discontinuity at the boundary between the lignified and non-lignified parts of the cell walls. No evidence was found here regarding any lignification of the macrosclereids, in either dormant or non-dormant seeds, although the LL was invariably visible (data not shown). The conclusion is that the origin of the LL is more nuanced. It features a compacted secondary cell wall, which diffracts polarized light differently from the rest of the macrosclereids, indicating a different orientation of its cellulosic microfibrils. The presence of a subcuticular waxy layer just above it likely contributes to its acting as a barrier to the passage of water as well as to its optical visibility. In both dormant and non-dormant seeds, the LL developed during the course of the macrosclereids’ differentiation, simultaneously with the subcuticular waxy layer. A similar lipid-rich layer has also been identified in the A. thaliana seed (Molina et al., 2008). The promotion of seed hardness through endo-1,4-β-glucanase activity, as suggested by Jang et al. (2015), is consistent with the known requirement for hydrolase activity in coordinating both the development of the cell wall (Nicol et al., 1998; Shani et al., 2006) and the deposition of secondary cell walls. It is expected that mechanical properties of the seed coat associated with cell wall development are connected with formation of fissures by volume changes of desiccating embryo of dormant genotypes.

The temperature-cycling regime employed to break dormancy was designed to reproduce conditions occurring in nature. It may have altered the nature of the subcuticular waxy layer, thereby allowing the testa to become more permeable, consistent with the observation that permeability in barrel medic seeds can be affected both in mutants in which extracellular lipid synthesis is compromised (Verdier et al., 2013) and in the KNOX4 mutant, which is both non-dormant and features an altered composition of lipids (Chai et al., 2016). Similarly, in A. thaliana, permeability is enhanced by mutations that affect the synthesis of waxy polymers (Beisson et al., 2007). Here, when the testa of the highly dormant seeds produced by entry JI64 was chemically dewaxed, the major barrier to water entry had clearly been removed; the chloroform treatment not only dissolved the surface wax, but also induced cracking of the surface, similar to the effect of thermocycling (Fig. 8). In effect, therefore, both the establishment and the breaking of testa impermeability are complex processes that depend on a number of mechanisms. Seeds can be softened by temperature-induced changes in the structure of the waxy components of the macrosclereid cell wall (Zeng et al., 2005), so it is of relevance that the testa of dormant pea seeds contains a number of unique hydroxylated fatty acids (Cechová et al., 2017).

The testa of non-dormant seeds tends to be thinner, softer and more elastic (as a result of a higher content of pectin, lower contents of cellulose and tannin and less extensive cell wall deposition) than that of dormant ones (Miao et al., 2001; Hradilová et al., 2017). This generalization ties in with the present observation that the testa of non-dormant seeds, following their dehydration and subsequent rehydration, developed areas showing a higher permeability, and that these areas coincided with sites of testa flections where cuticle continuity is disrupted but more extensive fissures are not generally present. As suggested by Harris (1983), the LL lies at the interface between two distinct environments: the waxy subcuticular layer and the cellulosic secondary wall. Dewaxing treatments revealed that the wax localized in the subcuticular layer was non-polymeric, while treatment with sulphuric acid dissolved everything except lignin, suberin and cutin (Hohnel, 1878; Zimmermann, 1892). According to Shao et al. (2007), the cuticle in water-impermeable soybean seeds contains a higher proportion of hydroxylated fatty acids than is present in permeable seeds. While dormancy in white clover has been associated with a more pronounced cuticle (Galussi and Moya, 2017), in the dormant pea entries it was not the cuticle that represented the major barrier to water entry. As noted by Cechová et al. (2017), hydroxylated fatty acids are an important determining factor in the dormancy expressed by JI92 and JI64 seeds, which agrees with the present conclusion, based on anatomical and histochemical analyses, that waxy compounds are among the more important features that distinguish dormant from non-dormant pea seed. As yet it has not been possible to determine whether the hydroxylated fatty acids are located within the cuticle or whether they are associated with the subcuticular wax layer.

Conclusions

The mechanistic basis of both the establishment and the breaking of physical dormancy is both rather complex and genotype-specific, so that it is unlikely that the products of just one or two genes will be implicated as the major factors underlying what is a key domestication trait in pea. While the intactness of the cuticle is important in order to limit the passage of water through the testa in non-dormant seeds, it is the LL that represents the main barrier in dormant ones. The LL develops in concert with the waxy subcuticular layer, which supports the notion that the LL represents an interface between the waxy subcuticular layer and the compact cellulosic secondary cell walls. Along with the influence of environmental factors (chiefly temperature and moisture), breaking dormancy depends on changes in the composition of the subcuticular wax and the mechanical disruption of the testa related to the cell wall properties of macrosclereids. While the former may be achieved by temperature variation, the latter is dependent on the swelling of the embryo, which generates a sufficiently large mechanical force to induce the formation of cracks.

SUPPLEMENTARY DATA

Supplementary data are available online at https://academic.oup.com/aob and consist of Figure S1: morphology of the testa surface, as documented by scanning electron microscopy.

mcy213_suppl_Supplementary_Data_Figure_S1
Click here for additional data file. (6.9MB, jpeg)

ACKNOWLEDGEMENTS

This research was financially supported by the Czech Republic Ministry of Education Youth and Sports (NPUI-LO1417) and by the Grant Agency of the Czech Republic (14-11782S and 16-21053S).

LITERATURE CITED

  1. Agbo GN, Hosfield MA, Uebersax MA, Klomparens K. 1987. Seed microstructure and its relationship to water uptake in isogenic lines and a cultivar of dry beans (Phaseolus vulgaris L.). Food Microstructure 6: 91–102. [Google Scholar]
  2. Baskin CC. 2003. Breaking physical dormancy in seeds: focussing on the lens. New Phytologist 158: 229–232. [Google Scholar]
  3. Baskin JM, Baskin CC. 2000. Evolutionary considerations of claims for physical dormancy-break by microbial action and abrasion by soil particles. Seed Science Research 10: 409–413. [Google Scholar]
  4. Baskin CC, Baskin JM. 2014. Seeds: ecology, biogeography and evolution of dormancy and germination, 2nd edn. New York: Academic Press. [Google Scholar]
  5. Baskin JM, Baskin CC, Li X. 2000. Taxonomy, anatomy and evolution of physical dormancy in seeds. Plant Species Biology 15: 139–152. [Google Scholar]
  6. Beisson F, Li Y, Bonaventure G, Pollard M, Ohlrogge JB. 2007. The acyltransferase GPAT5 is required for the synthesis of suberin in seed coat and root of Arabidopsis. Plant Cell 19: 351–368. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Bhalla PL, Slattery HD. 1984. Callose deposits make clover seeds impermeable to water. Annals of Botany 53: 125–128. [Google Scholar]
  8. Bogdanova VS, Galieva ER, Yadrikhinskiy AK, Kosterin OE. 2012. Inheritance and genetic mapping of two nuclear genes involved in nuclear-cytoplasmic incompatibility in peas (Pisum sativum L.). Theoretical and Applied Genetics 124: 1503–1512. [DOI] [PubMed] [Google Scholar]
  9. Cechová M, Válková M, Hradilová I, et al. 2017. Towards better understanding of pea seed dormancy using laser desorption/ionization mass spectrometry. International Journal of Molecular Sciences 18: 2196. [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Chachalis D, Smith ML. 2000. Imbibition behavior of soybean (Glycine max (L.) Merrill) accessions with different testa characteristics. Seed Science and Technology 28: 321–331. [Google Scholar]
  11. Chai M, Zhou C, Molina I, et al. 2016. A class II KNOX gene, KNOX4, controls seed physical dormancy. Proceedings of the National Academy of Sciences of the USA 113: 6997–7002. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Debeaujon I, Léon-Kloosterziel KM, Koornneef M. 2000. Influence of the testa on seed dormancy, germination, and longevity in Arabidopsis. Plant Physiology 122: 403–414. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Fedi F, O’Neil C, Menard G, et al. 2017. Awake1, an ABC-type transporter, reveals an essential role for suberin in the control of seed dormancy. Plant Physiology 174: 276–283. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Finch-Savage WE, Leubner-Metzger G. 2006. Seed dormancy and the control of germination. New Phytologist 171: 501–523. [DOI] [PubMed] [Google Scholar]
  15. Galussi AA, Moya ME. 2017. Anatomical and chemical insights into the white clover (Trifolium repens L.) seed coat associated to water permeability. In: Jimenez-Lopez JC, ed. Advances in Seed Biology. IntechOpen. doi: 10.5772/intechopen.70313. [DOI] [Google Scholar]
  16. Gama-Arachchige NS, Baskin JM, Geneve RL, Baskin CC. 2013. Identification and characterization of ten new water gaps in seeds and fruits with physical dormancy and classification of water-gap complexes. Annals of Botany 112: 69–84. [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Gillikin J, Graham JS. 1991. Purification and developmental analysis of the major anionic peroxidase from the seed coat of Glycine max. Plant Physiology 96: 214–220. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Harris WM. 1983. On the development of macrosclereids in seed coats of Pisum sativum L. American Journal of Botany 70: 1528–1535. [Google Scholar]
  19. Harris WM. 1987. Comparative ultrastructure of developing seed coats of ‘hard-seeded’ and ‘soft-seeded’ varieties of soybean, Glycine max (L.) Merr. Botanical Gazette 148: 324–331. [Google Scholar]
  20. Höhnel F, Ritter von. 1878. Über den Kork und verkorkte Gewebe Überhaupt. Mathematisch-naturwissenschaftliche classe lxxvi: 507–662. [Google Scholar]
  21. Hradilová I, Trněný O, Válková M, et al. 2017. A combined comparative transcriptomic, metabolomic, and anatomical analyses of two key domestication traits: pod dehiscence and seed dormancy in pea (Pisum sp.). Frontiers in Plant Science 8: 542. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Hyde EOC. 1954. The function of the hilum in some Papilionaceae in relation to the ripening of the seed and the permeability of the testa. Annals of Botany 18: 241–256. [Google Scholar]
  23. Jaganathan GK, Wu G-R, Han Y-Y, Liu BL. 2017. Role of the lens in controlling physical dormancy break and germination of Delonix regia (Fabaceae: Caesalpinioideae). Plant Biology 19: 53–60. [DOI] [PubMed] [Google Scholar]
  24. Jang S-J, Sato M, Sato K, et al. 2015. A single-nucleotide polymorphism in an endo-1,4-β-glucanase gene controls seed coat permeability in soybean. PLoS ONE 10: e0128527. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Karaki T, Watanabe Y, Kondo T, Koike T. 2012. Strophiole of seeds of the black locust acts as a water gap. Plant Species Biology 27: 226–232. [Google Scholar]
  26. Korban SS, Coyne DP, Weihing JL. 1981. Evaluation, variation, and genetic control of seed coat whiteness in dry beans (Phaseolus vulgaris L.). Journal of the American Society for Horticultural Science 166: 575–579. [Google Scholar]
  27. Long RL, Gorecki MJ, Renton M, et al. 2015. The ecophysiology of seed persistence: a mechanistic view of the journey to germination or demise. Biological Reviews 90: 31–59. [DOI] [PubMed] [Google Scholar]
  28. Lush WM, Evans LT. 1980. The seed coats of cowpeas and other grain legumes: structure in relation to function. Field Crops Research 3: 267–286. [Google Scholar]
  29. Ma F, Cholewa E, Mohamed T, Peterson CA, Gijzen M. 2004. Cracks in the palisade cuticle of soybean seed coats correlate with their permeability to water. Annals of Botany 94: 213–228. [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Manning JP, Van Staden J. 1985. The development and ultrastructure of the testa and tracheid bar in Erythrina lysistemon Hutch. (Leguminosae: Papilionoideae). Protoplasma 129: 157–167. [Google Scholar]
  31. Marbach I, Mayer AM. 1974. Permeability of seed coats to water as related to drying conditions and metabolism phenolics. Plant Physiology 54: 817–820. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Martens H, Jakobsen HB, Lyshede OB. 1995. Development of the strophiole in seeds of white clover (Trifolium repens L.). Seed Science Research 5: 171–176. [Google Scholar]
  33. Meyer CJ, Steudle E, Peterson CA. 2007. Patterns and kinetics of water uptake by soybean seeds. Journal of Experimental Botany 58: 717–732. [DOI] [PubMed] [Google Scholar]
  34. Miao ZH, Fortune JA, Gallagher J. 2001. Anatomical structure and nutritive value of lupin seed coats. Australian Journal of Agricultural Research 52: 985–993. [Google Scholar]
  35. Moïse JA, Han S, Gudynaite-Savitch L, Johnson DA, Miki BLA. 2005. Seed coats: structure, development, composition, and biotechnology. In vitro Cellular and Developmental Biology – Plant 41: 620–644. [Google Scholar]
  36. Molina I, Ohlrogge JB, Pollard M. 2008. Deposition and localization of lipid polyester in developing seeds of Brassica napus and Arabidopsis thaliana. Plant Journal 53: 437–449. [DOI] [PubMed] [Google Scholar]
  37. Nicol F, His I, Jauneau A, Vernhettes S, Canut H, Höfte H. 1998. A plasma membrane-bound putative endo-1,4-beta-D-glucanase is required for normal wall assembly and cell elongation in Arabidopsis. EMBO Journal 17: 5563–5576. [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Pearse A. 1968. Histochemistry, theoretical and applied. London: Churchill. [Google Scholar]
  39. Peckova E, Tylova E, Soukup A. 2016. Tracing root permeability: comparison of tracer methods. Biologia Plantarum 60: 695–705. [Google Scholar]
  40. Ranathunge K, Shao S, Qutob D, Gijzen M, Peterson CA, Bernards MA. 2010. Properties of the soybean seed coat cuticle change during development. Planta 231: 1171–1188. [DOI] [PubMed] [Google Scholar]
  41. Serrato-Valenti G, Cornara L, Ferrando M, Modenesi P. 1993. Structural and histochemical features of Stylosanthes scabra (Leguminosae; Papilionoideae) seed coat as related to water entry. Canadian Journal of Botany 71: 834–840. [Google Scholar]
  42. Shani Z, Dekel M, Roiz L, et al. 2006. Expression of endo-1,4-β-glucanase (cel1) in Arabidopsis thaliana is associated with plant growth, xylem development and cell wall thickening. Plant Cell Reports 25: 1067–1074. [DOI] [PubMed] [Google Scholar]
  43. Shao S, Meyer CJ, Ma F, Peterson CA, Bernards MA. 2007. The outermost cuticle of soybean seeds: chemical composition and function during imbibition. Journal of Experimental Botany 58: 1071–1082. [DOI] [PubMed] [Google Scholar]
  44. Smýkal P, Vernoud V, Blair MW, Soukup A, Thompson RD. 2014. The role of the testa during development and in establishment of dormancy of the legume seed. Frontiers in Plant Science 5: 351. [DOI] [PMC free article] [PubMed] [Google Scholar]
  45. Soukup A. 2014. Selected simple methods of plant cell wall histochemistry and staining for light microscopy. In: Žársky V, Cvrčková F, eds. Methods in molecular biology, Vol. 1080. Plant cell morphogenesis: methods and protocols. New York: Springer: 25–40. [DOI] [PubMed] [Google Scholar]
  46. Soukup A, Tylová E. 2014. Essential methods of plant sample preparation for light microscopy. In: Žárský V, Cvrčková F, eds. Plant Cell Morphogenesis Totowa: Humana Press. [Google Scholar]
  47. Soukup A, Armstrong W, Schreiber L, Franke R, Votrubová O. 2007. Apoplastic barriers to radial oxygen loss and solute penetration: a chemical and functional comparison of the exodermis of two wetland species, Phragmites australis and Glyceria maxima. New Phytologist 173: 264–278. [DOI] [PubMed] [Google Scholar]
  48. de Souza FHD, Marcos-Filho J. 2001. The seed coat as a modulator of seed-environment relationship in Fabaceae. Brazilian Journal of Botany 24: 365–375. [Google Scholar]
  49. de Souza TV, Voltolini CH, Santos M, Silveira Paulilo MT. 2012. Water absorption and dormancy-breaking requirements of physically dormant seeds of Schizolobium parahyba (Fabaceae – Caesalpinioideae). Seed Science Research 22: 169–176. [Google Scholar]
  50. Stevenson TM. 1937. Sweet clover studies on habit of growth, seed pigmentation and permeability of the seed coat. Scientific Agricola 17: 627–654. [Google Scholar]
  51. Sun L, Miao Z, Cai C, et al. 2015. GmHs1-1, encoding a calcineurin-like protein, controls hard-seededness in soybean. Nature Genetics 47: 939–943. [DOI] [PubMed] [Google Scholar]
  52. Verdier J, Dessaint F, Schneider C, Abirached-Darmency M. 2013. A combined histology and transcriptome analysis unravels novel questions on Medicago truncatula seed coat. Journal of Experimental Botany 64: 459–470. [DOI] [PMC free article] [PubMed] [Google Scholar]
  53. Werker E, Marbach I, Mayer AM. 1979. Relation between the anatomy of the testa, water permeability and the presence of phenolics in the genus Pisum. Annals of Botany 43: 765–771. [Google Scholar]
  54. Willis CG, Baskin CC, Baskin JM, et al. 2014. The evolution of seed dormancy: environmental cues, evolutionary hubs, and diversification of the seed plants. New Phytologist 203: 300–309. [DOI] [PubMed] [Google Scholar]
  55. Zeng LW, Cocks PS, Kailis SG, Kuo J. 2005. The role of fractures and lipids in the seed coat in the loss of hardseededness of six Mediterranean legume species. Journal of Agricultural Science 143: 43–55. [Google Scholar]
  56. Zhang RF, Zhang FX, Zhang MW, et al. 2011. Phenolic composition and antioxidant activity in seed coats of 60 Chinese black soybean (Glycine max L. Merr.) varieties. Journal of Agricultural and Food Chemistry 59: 5935–5944. [DOI] [PubMed] [Google Scholar]
  57. Zimmermann A. 1892. Die botanische Mikrotechnik: ein Handbuch der mikroskopischen Präparations-, Reaktions- und Tinktionsmethoden. Tübingen: H. Laupp’sche Buchhandlung. [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

mcy213_suppl_Supplementary_Data_Figure_S1
Click here for additional data file. (6.9MB, jpeg)

Articles from Annals of Botany are provided here courtesy of Oxford University Press

RESOURCES