Structure of the zygotic embryos and seedlings of Butia capitata (Arecaceae)
- First Online:
- Cite this article as:
- Magalhães, H.M., Lopes, P.S.N., Ribeiro, L.M. et al. Trees (2013) 27: 273. doi:10.1007/s00468-012-0797-1
- 345 Views
Butia capitata, an endemic palm of the Brazilian savanna threatened by deforestation, demonstrates low germinability due to seed dormancy. The present study characterizes the structure of the zygotic embryo and describes germination and seedling development. Pyrenes were sown into sandy soil substrates to germinate, and their embryos were also cultivated in vitro in MS medium; structural evaluations were made during their development. Seedling growth through the endocarp germ pore culminates in the protrusion of the cotyledonary petiole, with the root and leaf sheaths subsequently being emitted laterally from its extremity. The embryos are composed of the cotyledon (whose proximal third has a haustorial function) and a diminutive embryo axis that is contained within the cotyledonary petiole. The protoderm, ground meristem, and procambium can be observed in their typical positions in the embryo axis and cotyledon. The development of the vegetative axis could be observed on the second day of in vitro cultivation, with elongation of the embryo axis and the beginning of the differentiation of the first eophyll. Elongation of the cotyledonary petiole and the differentiation of the parenchyma and tracheary elements were observed during the second to fifth day. Although the hypocotyl–radicle axis is less differentiated than the plumule, root protrusion occurs on the eighth day, and the leaf sheaths are only emitted between the twelfth and the sixteenth days; the haustorium atrophied during this stage. The embryonic structure of B. capitata does not impose limitations on seed germination as dormancy is of the non-profound physiological type, and the 50 % elongation of the cotyledonary petiole serves as a morphological indicator of germination.
KeywordsEmbryo cultureMorphological dormancyGerminationPalms
Butia capitata Mart. (Becc.), also known as “coquinho-azedo”, is an endemic palm tree of the savanna (cerrado) regions of central Brazil (Lorenzi et al. 2010). This palm has significant ecological value and economic potential as its fruits are eaten by the local fauna and consumed by humans in natura or used to prepare juices, liquors, and ice-cream (Mercadante-Simões et al. 2006; Moura et al. 2010). The species has a potential for ornamental use because a closely related species, Butia odorata (Barb. Rodr.) Noblick & Pirani (until recently considered synonymous with B. capitata) is commonly used in landscape projects in southern Brazil (Lorenzi et al. 2010) and in the southern regions of the United States (Broschat 1998). B. capitata is threatened by deforestation as well as by predation of its seedlings and inflorescences by domestic cattle and horses (Mercadante-Simões et al. 2006). Commercial plantings of this species, its utilization in recuperating degraded areas, as well as the conservation of established natural populations have been limited by difficulties encountered in seedling production due to seed dormancy resulting in uneven and slow germination at low rates (Lopes et al. 2011).
The structure of the diaspores and the germination processes in palm trees are quite peculiar (Orozco-Segovia et al. 2003; Henderson 2006). After the abscission of the fruits, the exocarp and mesocarp are normally consumed by animal dispersers while the seed remains contained within the endocarp, forming a pyrene (Orozco-Segovia et al. 2003). The seed is albuminous, with a linear embryo (Baskin and Baskin 1998) whose distal portion assumes a haustorial function in the mobilization of the endosperm reserves (DeMason 1985; DeMason et al. 1985). Germination is hypogeal (Aguiar and Mendonça 2002), and the protrusion of the seedling occurs as the cotyledonary petiole grows through the germ pore of the endocarp (Orozco-Segovia et al. 2003). There are wide variations in the morphologies of palm tree seedlings (Henderson 2006) and, in spite of the size and importance of this family, this subject has been relatively poorly studied (Aguiar and Mendonça 2003; Orozco-Segovia et al. 2003; Panza et al. 2004; Silva et al. 2006). Structural studies are relevant to seedling production efforts because they aid our comprehension of the posterior stages of germination, including seedling growth and establishment (Henderson 2006). Additionally, these studies aid in interpreting germination and tetrazolium tests and can contribute to standardizing the terminology used in propagation studies (Silva et al. 2006; Mendonça et al. 2008; Ribeiro et al. 2010).
Dormancy in palm trees has been classified as morphological (Baskin and Baskin 1998) and is associated with the immaturity of the embryo, among other factors. This immaturity retards the initiation of germination due to the time required to achieve full embryo differentiation (Orozco-Segovia et al. 2003). Although Arecaceae comprises more than 2,200 species, studies focusing on embryo and seedling structures in this family have so far been restricted to just a few taxa (Orozco-Segovia et al. 2003; Panza et al. 2004). Studies of the embryos of Phoenix dactylifera (DeMason and Thomson 1981), Washingtonia filifera (DeMason 1988), Euterpe precatoria (Aguiar and Mendonça 2003), and E. edulis (Panza et al. 2004) showed the embryo axes to have variable degrees of differentiation, although the plumule and the hypocotyl–radicle axes were relatively well-developed in these species. Haccius and Philip (1979) suggested the term “long-term differentiation of the embryo” because differentiation of the radicle in Cocos nucifera embryos occurs relatively slowly and becomes recognizable with the growth of the first eophyll primordium. This condition is not a general rule in Arecaceae because in Acrocomia aculeata Ribeiro et al. (2012) identified a partially differentiated hypocotyl–radicle axis, but without a distinct protoderm.
Morpho-anatomical studies of the germination and development of B. capitata seedlings have encountered difficulties due to the low and erratic germination rates. Lopes et al. (2011) reported that the germination of seeds still enveloped by the endocarp (pyrenes) occurred only irregularly during the year and at levels below 1 %. In vitro embryo culture may favor the study of germination and propagation in species demonstrating dormancy by making seedling production more rapid and uniform, and by allowing the production of large numbers of seedlings in short periods of time (Tzec-Sima et al. 2006; Pech y Aké et al. 2007). Ribeiro et al. (2011) observed germination rates of 72 % in 30 days in in vitro cultures of isolated embryos of B. capitata, which indicated that this cultivation technique could aid in more detailed examination of this process. Additionally, embryo culture can aid viability and vigor testing of dormant seeds such as those of B. capitata (Bhojwani and Razdan 1996; Ribeiro et al. 2010). Anatomical studies of this type are important for defining criteria of embryonic axis elongation [which is considered indicative of successful germination (Bewley and Black 1994)], as this structure is microscopic and internal to the cotyledonary petiole in palms (Orozco-Segovia et al. 2003; Panza et al. 2004; Ribeiro et al. 2012) and seedling morphology is very variable in that group (Henderson 2006).
The present work characterizes the morphology of seedlings obtained from the germination of pyrenes and examines the zygotic embryos and seedlings of B. capitata cultured in vitro from a morphological and anatomical point of view. We aimed at understanding possible causes of dormancy in this species and to define the morphological indicators of germination.
Materials and methods
The fruits of B. capitata were obtained from a savanna (cerrado sensu stricto area) where these plants naturally occur, in the municipality Montes Claros, Minas Gerais State, Brazil. Three to four bunches of ripe fruits having yellow exocarps were harvested from three different plants. One hundred fruits were removed from each bunch and their pulp was manually removed in the laboratory. The pyrenes were washed in distilled water and subsequently dried in the shade for 15 days.
Four hundred pyrenes were scarified by removing the protective tissues from the germination pore (Lopes et al. 2011) and then were sown in a greenhouse in trays containing a sandy soil substrate gathered from cerrado sites, at depths of 5 cm, with daily watering. The pyrenes were removed from the substrate weekly and evaluated for cotyledonary petiole protrusion, during 360 days. The morphology of the developing seedlings was described and illustrated using the terminology suggested by Henderson (2006).
The seeds used for in vitro culture were obtained by breaking the endocarp; seeds were then immersed in a solution of undiluted commercial sodium hypochlorite for 20 min, followed by three rinses in sterile distilled water. Three hundred embryos were excised and disinfected in a solution of 0.5 % commercial sodium hypochlorite for 10 min and then rinsed three times in sterile distilled water in a laminar flow chamber. The embryos were inoculated into test tubes containing 10 mL of MS culture medium (Murashige and Skoog 1962) at 75 % concentration supplemented with 30 g L−1 sucrose, 0.5 mg L−1 thiamine, 1 mg L−1 pyridoxine, 0.5 mg L−1 nicotinic acid, 3 g L−1 activated charcoal, and 6 g L−1 agar; the pH was adjusted to 5.7 (Ribeiro et al. 2011). The medium was sterilized in an autoclave at 120 °C for 15 min. The cultures were maintained in a germination chamber at 25 °C in darkness.
Ten embryos and ten seedlings were removed after 2, 5, 8, 12, 16, 26, 30, and 37 days of culture for measurements of the lengths and diameters of the cotyledonary petioles, sheaths, haustoria, and roots. The embryos and seedlings from each of these harvests were illustrated using a Cannon A-650/S digital camera coupled to a Primo Star Zeiss stereomicroscope. Anatomical evaluations of ten embryos were first made shortly after collecting the fruits, and of 10 seedlings after 2, 5, 8, 12 and 16 days of culture. The samples were fixed under vacuum in FAA50 (Johansen 1940) for 24 h, dehydrated in an ethanol series, and subsequently stored in 70 % ethanol (Jensen 1962). Some of the samples were embedded in Leica™ 2-hydroxyethyl-methacrylate according to Paiva et al. (2011). Transversal and longitudinal sections (6–7 μm thick) were obtained using a rotary microtome, stained with 0.05 % toluidine blue at pH 6.8 (O’Brien et al. 1964), and then mounted on slides with Entellan™. Photomicrographs were taken using a Zeiss Stemi 2000-C stereomicroscope and a Primo Star Zeiss optical microscope, both coupled with a Cannon A-650/S digital camera. Anatomical descriptions were based on the terminology of Haccius and Philip (1979), DeMason (1988), and Panza et al. (2004).
The plumule is located within a proximal cavity in the cotyledon that is formed by the ligule, and it is composed of two leaf sheath primordia and by the semicircular promeristem with small, dense, isodiametric cells with large nuclei (Fig. 2c, d). The cavity and the first sheath are covered by a unistratified protoderm composed of cubical cells. The leaf sheaths have a ground meristem and distinct procambial strands. The hypocotyl–radicle axis appears to be less differentiated than the plumule as it has no visible protoderm and only consists of the radicle promeristem formed by just a few cells, a ground meristem, and procambium (Fig. 2c).
Protoderm, ground meristem, and the procambium can be distinguished in the cotyledon. Protoderm cells have radially elongated rectangular outlines in transversal section, principally in the haustorium (Fig. 2e, f). A groove perpendicular to the embryo axis can be seen in the proximal portion of the cotyledon (Fig. 2d). The cells of the ground meristem are larger than the others, and tend to be isodiametric; these same cells are larger in the central and distal regions of the cotyledon than in the proximal and peripheral regions (Fig. 2d–f). The procambial cells are narrow, longitudinally elongated, and have thin walls; their nucleus is not very evident. The vascular system appears as a set of procambial strands that form from traces emitted by the embryo axis in the region of the cotyledon node (Fig. 2b, d–f). Traces emitted towards the proximal extremity of the embryo form 9–10 procambial strands that vascularize the ligule surrounding the plumule (Fig. 2d). Traces emitted towards the distal region of the embryo give rise to approximately 10–12 procambial strands (Fig. 2e) that ramify and produce 16–20 procambial strands in the distal portion of the haustorium (Fig. 2f).
The cells of the ground meristem along the hypocotyl–radicle axis increased in volume, which characterizes the initiation of parenchyma differentiation and the elongation of the embryo axis. The volume of the leaf sheaths in the plumule also increased slightly, and cellular alterations were similar to those taking place in the hypocotyl–radicle axis; the initiation of differentiation of the first eophyll could also be seen (Fig. 6d). Increases in cell volume and anticlinal cell divisions were observed in the median region of the cotyledonary petiole (Fig. 6e); in the distal region, the peripheral zone of the haustorium had intense mitotic activity, with periclinal divisions similar to those occurring in the proximal region leading to formation of new cell layers (Fig. 6f). Longitudinally, elongated procambial strands and cells with thin walls and less evident nuclei were distinguished.
The initiation of differentiation of the root cap was observed on the eighth day of cultivation. A cell layer with accumulated phenolic compounds adjacent to the root favored the rupture of the cell layers of the extremity near the cotyledon. These layers were flattened by the expansion of the root and appeared to be bursting and sloughing off, thus facilitating the emergence of the radicle (Fig. 7e). The stem apex also became more evident at this stage, the leaf sheaths began to elongate, and the first eophyll was differentiating (Fig. 7f). The distal region of the haustorium did not show any apparent modifications in relation to the previous phase.
After 16 days of cultivation, the initiation of differentiation of the second eophyll was observed (Fig. 8f) and the radicle cortex and root cap were distinct. The cotyledonary petiole continued growing and the haustorial region atrophied; no peripheral meristematic activity was observed (Fig. 8g).
The germination of B. capitata seeds is of the remote tubular type, according to Henderson (2006). The ligule does not expand, as oppose to E. precatoria (Aguiar and Mendonça 2002), but the cotyledonary petiole elongates favoring seedling development at some distance from the seed itself. According to Pinheiro (2002) and Orozco-Segovia et al. (2003), the adaptation of some palms to dry habitats has been associated with a strategy involving the pronounced elongation of the cotyledonary petiole. The maintenance of the embryo axis below the dry soil surface, in deeper layers where the humidity is higher, confers an ecological advantage that favors seedling development while protecting its growing point from perturbations at the soil surface caused by fire, physical damage, or drying. The observations made here with B. capitata corroborate this interpretation, as this species is typical of savanna environments where such adverse conditions are recurrent. We observed a pronounced dormancy in this species because only 9.5 % seeds contained within the endocarp germinated after 1 year, even though they were scarified. Erratic germination at low frequency constitutes an adaptive advantage in highly seasonal environments (Bewley and Black 1994; Baskin and Baskin 1998, 2004), but makes detailed evaluations of the biometric and anatomical development of those seedlings more difficult.
Panza et al. (2004) reported that the embryos of palm trees are not very anatomically distinct and that differences among them are principally related to the disposition of the embryo axis. According to the classification of DeMason (1988), the embryo axis of B. capitata is oblique in relation to the cotyledon like in C. nucifera (Haccius and Philip 1979), W. filifera (DeMason 1988), and A. aculeata (Ribeiro et al. 2010, 2012). In P. dactylifera, however, the embryo axis is parallel to the cotyledon (DeMason and Thomson 1981). The embryo axes of palm trees are composed of the plumule, which may have two or three leaf primordia and the hypocotyl–radicle axis (Panza et al. 2004). In B. capitata, as in other species, the plumule is encased in the cotyledon cavity and has two leaf sheaths. The proximal region of the cotyledon in some palm trees, such as Phoenix canariensis, Trachycarpus excelsa, Livistona australis, and Licuala horrida, is characterized by the presence of a layer of flattened cells called the “M zone” that delimits the radicle region into the cotyledonary petiole (Gatin 1906). The “M zone” in B. capitata is quite well-defined and is composed of a group of flattened cells that are pressured by the emerging root. Martius (1823–1850) described this same region, referring to it as the “hialina zone”. The “M zone” was also noted by Haccius and Philip (1979) who described it as a superficial arc of cells discerned in longitudinal sections of C. nucifera embryos.
Similarly to other palms, a haustorium composes the distal region of the embryo of B. capitata (DeMason and Thomson 1981; DeMason 1988; Panza et al. 2004), although difficulties have been encountered in establishing the precise extension of this portion at the time of dispersal. In E. precatoria, the whitish-yellow portion of the embryo corresponds to the proximal region, and the pale-white portion to the haustorium (Aguiar and Mendonça 2003); the haustorium in A. aculeata can be identified by its whitish color and by the occurrence of invaginations (Ribeiro et al. 2012). There are no apparent macroscopic differences between these two structural regions in B. capitata and they can be distinguished principally by the peripheral distribution of the procambial strands visible in the distal third of the cotyledon. The presence of peripheral procambial strands in the haustorium is recurrent in different species of palms although the numbers of strands can vary (see DeMason and Thomson 1981; Ribeiro et al. 2012).
According to Bewley and Black (1994), germination initiates with seed imbibition and ends with the elongation of the embryo axis. On the second day of cultivation of B. capitata embryos, a number of indications of germination were observed even before the phase of significant cotyledonary petiole elongation, which normally occurs after 5 days. On the second day, cell volumes increased in the ground meristem of the plumule and in the hypocotyl–radicle axis, differentiation of the first eophyll started, and the plumule showing slight signs of development that was characterized by the beginning of embryo axis elongation and its transition to a vegetative axis. These microscopic observations coincided with an increase of approximately 50 % in the length of the cotyledonary petiole (which could be considered a morphological indication of germination of B. capitata) that occurred after 8 days of culture. The elongation of the parenchyma cells in the proximal region of the cotyledonary petiole is responsible for the protrusion of the cotyledonary petiole, which is commonly considered indicative of palm seed germination (DeMason and Thomson 1981; Aguiar and Mendonça 2002; Oliveira et al. 2010). Additionally, the pronounced and continuous elongation of the parenchyma cells of the cotyledonary petiole in B. capitata also contributes to seedling development distant from the seed itself.
When grown in culture medium, the haustorium of B. capitata did not expand as much as if it were still encased within the endosperm. DeMason et al. (1985) hypothesized that digestive enzymes were synthesized by the haustorium and secreted into the endosperm, or synthesized in the endosperm itself upon receiving inductive signals from the haustorium. During the in vitro cultivation of A. aculeata embryos (Ribeiro et al. 2012), the haustorium showed signs of atrophy on the seventh day after inoculation, which may have been related to a lack of interaction with the endosperm, as was also observed in the present work. The expansion of the haustorial normally aids in maintaining the embryo in constant contact with the nutrient source (the endosperm, in the case of the intact seed), but by furnishing an adequate nutrient medium the haustorium ceases to be the single source of nutrient transport to the embryo and the plant might invest less in the haustorium expansion, resulting in its progressive atrophy.
Morphological dormancy in palm trees has been frequently mentioned as being associated with embryo immaturity at the time of seed dispersal (Corner 1966; Baskin and Baskin 1998; Orozco-Segovia et al. 2003). However, the characteristics that define this condition and their role in limiting or slowing germination are not yet very clear. Primary meristematic tissues were observed in the cotyledon and in the embryo axis of B. capitata, although the hypocotyl–radicle axis was less differentiated than the plumule and did not have a distinct protoderm (as it is not inserted in a cavity). DeMason (1988) noted that the radicle pole of the embryo of W. filifera was poorly developed and that it was difficult to determine if the embryo axis was curved or straight; Haccius and Philip (1979) likewise reported that the differentiation of the radicle pole of C. nucifera embryos occurred very slowly and only became recognizable when the first plumule primordium developed. In E. edulis, a species with recalcitrant seeds, the hypocotyl–radicle axis was more developed than in the cited species, having a complete root cap and a very evident radicle (Panza et al. 2004). The differentiation of the hypocotyl–radicle axis in B. capitata was not observed to restrict its development, which was more precocious than that of the plumule because roots were emitted before the emission of the leaf sheaths.
The effectiveness of in vitro culture of isolated embryos in overcoming dormancy (most of the seedlings demonstrated sheaths and roots after 30 days of culture) is typical in non-profound physiological dormancy. In cases of morphological dormancy, embryos are known to demonstrate a lag period of approximately 30 days before germinating (Baskin and Baskin 1998, 2004). Recent work with B. capitata seeds has also shown that the tegument does not significantly restrict water absorption by the embryo (unpublished data) and that the removal of the operculum (which is composed of tissues that restrict embryo growth: the opercular tegument and the micropylar endosperm) generates the same effect as in vitro culture of isolated embryos (Fior et al. 2011; Oliveira 2012). This confirms the classification of dormancy in the species as non-profound physiological. The same type of dormancy has been reported in other Cerrado palms such as A. aculeata and Attalea vitrivir Zona (Ribeiro et al. 2011; Neves 2012). The results of this and other recent studies therefore indicate that the concept of morphological dormancy being common in palm trees needs to be further investigated.
In spite of the fact that the structures of the embryos and seedlings of B. capitata share many characteristics with other palms, there are certain peculiarities that confirm the observations of Henderson (2006) of significant variability in the germination patterns within the family. Seed dormancy in B. capitata and the developmental patterns of seedlings make it well adapted to the seasonal climate of the Cerrado biome. A 50 % elongation of the cotyledonary petiole is a morphological indication of germination and is related to the elongation of the parenchymatous cells of the petiole and the embryonic axis, and the development of the “M Zone”. Precocious root emission does occur indicating a lack of structural limitations to germination in spite of the fact that the hypocotyl–radicle axis is less differentiated than the plumule. As such, dormancy in this species can be classified as non-profound physiological and the view that morphological dormancy is common in palms should be re-examined.
The authors would like to thank the Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES/PROCAD 213/207), the Conselho de Desenvolvimento Científico e Tecnológico (CNPq 27/2008) for their financial support and for the study grant awarded to the first author. We also thank professor Santos D’Angelo Neto for the illustration of seed germination. Denise M. T. Oliveira thanks CNPq for the research productivity grant (process number 304716/2008-1), Leonardo M. Ribeiro thanks the Fundação de Amparo à Pesquisa de Minas Gerais for the research productivity grant (FAPEMIG process number CRA-BIP-00137-11), and Paulo S. N. Lopes thanks CNPq for the technological development and innovative extension grant (process number 313116/2009-1).