AOBPreview published online on February 13, 2007
Annals of Botany, doi:10.1093/aob/mcm002
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Ultrastructural Evidence for a Dual Function of the Phloem and Programmed Cell Death in the Floral Nectary of Digitalis purpurea
1 Institut für Biologie, Lehrstuhl für Pharmazeutische Biologie der Friedrich-Alexander-Universität Erlangen-Nürnberg, Staudtstraße 5, D-91058 Erlangen, Germany
2 Dermatologische Klinik, Krankenhaus Bethesda, Euelbuchstraße 39, D-57258 Freudenberg, Germany
* For correspondence. E-mail kgaffal{at}biologie.uni-erlangen.de
Received: 10 July 2006 Returned for revision: 12 October 2006 Accepted: 8 December 2006
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONACKNOWLEDGEMENTLITERATURE CITED |
|---|
BACKGROUND AND AIMS: The floral nectary of Digitalis purpurea is a transitory organ with stomatal exudation of nectar. In this type of nectary, the nectar is thought to be transported to the exterior via intercellular ducts that traverse the nectariferous tissue. The latter is also traversed by a ramified system of phloem strands from which prenectar sugar is most probably unloaded. The aims of this study were to provide some of the basic information needed to evaluate the possible mechanism involved in nectar secretion and to discover the fate of the nectary.
METHODS: The ultrastructure of the nectary was investigated at different stages of development by analysis of a series of ultrathin (7 x 10–8 m) sections 7 x 10–7 m apart from one another. Proportions of the cells typical of the nectary were documented by 3D-reconstruction and morphometry.
KEY RESULTS: The phloem consisted of variably shaped sieve elements and companion cells which, as a rule, were more voluminous than the sieve elements. Direct contact between the phloem strands and intercellular ducts was observed. In contrast to the phloem, which remained structurally intact beyond the secretory phase, the nectariferous tissue exhibited degenerative changes reminiscent of programmed cell death (PCD), which started as early as the onset of secretion and progressed in a cascade-like fashion until final cell death occurred in the exhausted nectary. Hallmarks of PCD were: increased vacuolation; increase in electron opacity of individual cells; progressive incorporation of plasmatic components into the vacuole reminiscent of autophagy; degradation of plastids starting with hydrolysis of starch; deformation of the nucleus and gradual disappearance of chromatin; loss of tonoplast integrity and subsequent autolysis of the rest of cellular debris. Degeneration of the cells occurred against a background of increasing cell size.
CONCLUSIONS: The cytological and anatomical evidence presented here, and calculations of the solute fluxes necessary for accumulation of starch and for the production of nectar support the view that: (a) in the foxgloves' nectary, apoplastic phloem unloading dominates, at least during exudation of nectar; (b) the obsolete nectary may be dismantled by PCD; and (c) at least the products of late nectary degradation are loaded via the apoplast into the unchanged phloem and exported to sinks elsewhere in the plant for reallocation.
Key words: Floral nectary, Digitalis purpurea, 3D-reconstruction, morphometry, fluid-filled intercellular space, phloem innervation, programmed cell death, nectariferous tissue
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONACKNOWLEDGEMENTLITERATURE CITED |
|---|
Although it is generally accepted that nectar originates from phloem sap (Frey-Wyssling and Agthe, 1950; Fahn, 2000), the ultrastructure of the phloem innervating nectaries has received less attention than the ultrastructure of the nectariferous tissue (secretory cells; nectary parenchyma). The sugar-secreting nectaries are undisputedly sink organs dependent on source tissue assimilates. Most of them are vascularized exclusively by phloem strands (Frei, 1955). Therefore they typify model organs for the study of unloading (release) phloem. The efflux of carbohydrates may occur across plasma membranes of the sieve element–companion cell complexes (apoplastic pathway) or through plasmodesmata interlinking these complexes with adjacent cells (symplastic pathway; Patrick et al., 2001).
According to Thompson and Holbrook (2003), phloem transport depends strongly on sieve tube radius, sieve tube length and sieve plate geometry. They stressed the need for satisfactory anatomical measurements in any phloem study. The aims of the present study were (a) to make 3D-reconstructions and morphometry of phloem elements from series of ultrathin sections, (b) to look for ultrastructural criteria indicating which unloading pathway is taken in the floral nectary of Digitalis purpurea during nectar secretion, and (c) to discover the fate of the nectary after it has fulfilled its temporary function. Deletion of cells that are no longer needed is one of the functions of programmed cell death (PCD) (Pennell and Lamb, 1997; Ranganath and Nagashree, 2001). In view of the importance attached by Gray and Johal (1998) to extensive microscopic examination of the temporal sequence of subcellular events that occur during PCD, this was tried with the nectary.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONACKNOWLEDGEMENTLITERATURE CITED |
|---|
Plant material
Ovaries of Digitalis purpurea with nectaries at their base were collected from plants growing in a garden near Erlangen (Germany) on 1 August 1996 for investigation by transmission electron microscopy (TEM) and on 25 July 2005 for light microscopical examination. Three different stages of nectary development were chosen: (1) the onset of secretion, i.e. just before the first pair of anthers dehisced; (2) the peak period of secretion, i.e. during stigma receptivity; and (3) the cessation of secretion, i.e. after the abscission of the corolla.
Preparation and analysis
For TEM, small pieces (max. 10–9 m3) of the nectaries were treated for 1 h in collidine-buffered (50 mM) 2 % glutaraldehyde containing 5 mM CaCl2 at room temperature. After three 10-min rinses in collidine buffer plus 5 mM CaCl2, the samples were post-fixed overnight in 2 % OsO4 plus 0·8 % K3Fe3(CN)6 at 4 °C. After three 10-min rinses the samples were stained en block for 2 h with 2 % aqueous uranyl acetate and then rinsed again three times. After standard dehydration and embedding (Spurr, 1969) a series of thin sections (7 x 10–8 m thick) were cut on a Reichert-Jung Ultracut E microtome with a diamond knife (45° Diatome, Balzers Union). From this series only every tenth section was collected and analysed. Accordingly, the distance between adjacent sections was 7 x 10–7 m.
True-to-scale models were produced either by conventional or by computer-aided 3D-reconstruction. In both cases the profiles of the structures were first traced onto transparent paper. For conventional reconstruction, the profiles were projected and traced onto polystyrene sheets of appropriate thickness. The cut-out polystyrene profiles were then well-matched and glued together. The stepped surface area was smoothed by trimming the edges of the steps and by filling the remaining corners of the steps with a paste of gypsum. After the gypsum was hard and dry the models were painted. For computer-aided 3D-reconstruction, all data processing took place using the programs ANAT3D, EDIT3D and PIC SUM (Genims International, Luxembourg) on an Atari Mega ST4 computer (Atari Corp., Sunnyvale, CA, USA). The x/y co-ordinates of the series of points placed along the profiles were digitized using a cross-hair cursor of the digitizing tablet (1st CRP Koruk, Germany). By this way the profiles of the structures were represented by polygons. The distance between adjacent profiles was 7 x 10–7 m. The reconstructions were rotated/scaled on the monitor screen until the most instructive perspective was found by eye. Stereopairs were created by rotating the reconstructed object along the vertical y-axis (3º to the left and 3º to the right). The three-dimensional details of stereopictures can be perceived by a stereoviewer or by naked-eye stereopsis. An easily learned technique for achieving unaided stereoviewing was described by McKeon and Gaffield (1990). Polychromy enhanced the discrimination of different structure elements. The same software was also used for the morphometry.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONACKNOWLEDGEMENTLITERATURE CITED |
|---|
Ultrastructure of the nectary at the onset of secretion
A region of nectariferous tissue (NT) approx. 40–50 cells wide was located below the epidermis (Gaffal et al., 1998). The latter was interspersed with permanently open stomata (Fig. 1). At the onset of secretion its cells were isodiametric and relatively small, making the nucleus appear relatively large. The cytoplasm stained homogeneously. Apart from tiny septa the intercellular spaces were devoid of structures (Fig. 2). The nuclei were more or less rotund with a smooth surface. In contrast to the more internally located nucleolus, condensed chromatin was preferentially located at the periphery, tightly attached to the nuclear envelope (Fig. 2). Several cells of the NT had a spongy appearance due to the presence of numerous smaller vacuoles that tended to be empty and to aggregate. The septa between the vacuoles differed in thickness; the thinner ones consisted of membranes only, the thicker ones were filled with cytoplasm (=plasmatic septa). Other cells had more voluminous vacuoles with inclusions that varied considerably in size, shape and structure from one another. Frequently these inclusions were membrane-bound. Occasionally irregularly shaped evaginations of the cytoplasm into the vacuoles were observed. Sporadically occurring transverse walls that were significantly thinner than mature cell walls indicated previous cell divisions. The nectariferous cells (NCs) possessed plastids (amyloplasts) containing relatively densely stained starch grains (Fig. 2). This was also the case with the epidermis including its guard cells (not shown).
|
|
Ultrastructure of the optimal secreting nectary
During secretion both the epidermal cells and the NCs differed markedly in their staining properties (Fig. 3). Consequently, the alterations of the cells were non-synchronous. This made the chronological ordering of the cytological events difficult. Although transition stages existed, the ultrastructural characteristics were classified into three categories. (1) Cytoplasm, nucleoplasm and the endomembrane system were so electron opaque that the compartmentation of the cell was hardly discernible. The existence of vesicles, for example, was indirectly indicated by their electron-transparent lumina (Fig. 3A, cells E3, E4 and N1). (2) Staining of the cytoplasm and nucleoplasm was moderately strong; except for vesicles, the lumina of the rest of endomembrane system were heavily stained (Fig. 3B, cells N3 and N4). (3) Staining of cytoplasm, nucleoplasm and endomembrane system was weak by way of comparison (Fig. 3B, cells N5 and N6).
|
Generally, more voluminous vacuoles predominated over aggregates of smaller ones. Analysis of serial sections revealed that most of the vacuoles found apparently separated in one section (Fig. 3A) were joined in adjacent sections. It was repeatedly found that only a single large vacuole existed. In addition to vacuolar inclusions resembling those found at the onset of secretion, other types of inclusions occurred: (a) more or less spherical structures with variably arranged electron opaque deposits (Figs 3 and 4F), (b) more or less dispersed electron-dense grana (Fig. 3A); (c) flocculent material (Fig. 4A, asterisk); (d) bizarre membranous structures (Fig. 3A, asterisks); and (e) scalloped lumps with an enigmatic internal structure that had diameters of up to 3 x 10–6 m (Figs 4Q–V).
|
In particular, the nuclei of the NCs tended to become flattened with a wavy outline. While the nucleoli remained unchanged, a gradual weakening of the contrast between chromatin and nucleoplasm occurred until the chromatin vanished completely. The chromatin when still visible was less often tightly attached to the nuclear envelope (Fig. 3).
The amyloplasts tended to approach the vacuole. At the sites of contact, either membrane-bound evaginations (Fig. 4A–F) or small openings [42 ± 20 x 10–8 m (± s.d.); n = 8] were repeatedly found. Via these openings at least the proximate starch grain appeared to be exposed to the vacuole sap (Fig. 4G–J). Generally, the starch grains were less electron dense than at the onset of secretion. In addition to the plastid-associated evaginations, which were electron translucent or filled with amorphous material (Fig. 4A–F), a great variety of other plasmatic evaginations into the vacuole were found, the larger of which were usually bounded by a double membrane (Fig. 4D, N).
In the majority of epidermal, nectariferous and companion cells, the protoplast had retracted from the cell wall leaving gaps between the wall and the plasma membrane. These gaps occurred mainly where the plasmatic coat of the vacuole was thin. They contained membranous connections between the wall and the plasma membrane (Figs. 4K–M), and/or membrane-bound vesicles (not shown), and/or irregularly shaped membrane aggregates (Figs 3 and 5A), and/or heavy deposits of electron-opaque material (Fig. 4E, F, Q–V).
|
The vascular supply entering the NT consisted of reticulating phloem strands that terminated approx. 8–12 cell layers below the epidermis (Gaffal et al., 1998). Most probably it is here that unloading occurs. Diagnostic features of the sieve elements (SEs) are either the sieve plates by which they are joined to each other or peculiar plastids (Behnke, 1991), which in this case are of the SS-type (Fig. 5A, E).
At the magnification usually used in the examination of the secreting nectary, it was very difficult to identify with certainty the cells adjacent to the SEs. Two types of cells were present: those that were similar in internal structure to NCs and those that usually displayed maximum staining of the cytoplasm and lacked starch grains. The latter proved to be companion cells (CCs) not only by the absence of starch but also by being linked with their SEs via pore–plasmodesma complexes (Fig. 5E). There were relatively few plasmodesmata connecting the sieve element–companion cell–complexes (SE-CC-Cs) to surrounding cells; i.e. the phloem was somewhat isolated symplastically.
Intercellular spaces were not only found between the epidermal cells (inclusive guard cells) and the NT, and between the NCs (Figs 1 and 3) but they also abutted the phloem strands (Figs 5A, E, 6 and 8A). These contacts support the view that the phloem termini lack typical bundle sheath cells. Most of the intercellular spaces contained small electron-opaque particles (Figs 3 and 4Q–V) that were similar to those particles found at the periphery of the protoplast of several cells (Fig. 4K–M), and in some of the gaps between the cell wall and the plasma membrane (Fig. 4F, Q–V).
Ultrastructure of the exhausted nectary
After cessation of nectar exudation, heterogeneity in cellular features of the NT still existed. The epidermal cells (exclusive of guard cells) and the NT were devoid of plastids, indicating that not only the starch but also other components of the plastids had been hydrolysed during the secretory stage. The NCs, now appearing more or less empty, varied greatly in volume (1330–4480 x 10–9 m3), shape and contents from one another. The mean value (2250 ± 1140 x 10–9 m3, n = 6) was almost double that of the NCs in the active nectary (1199 ± 319 x 10–9 m3, Gaffal et al., 1998). Frequently the interfaces of neighbouring NCs undulated (Fig. 5E). Analysis of serial sections revealed that such cells, by lacking a nucleus and a structurally intact cytoplasm, were nothing more than cell corpses. Their lumina contained variable aggregates of granules and lipid-like globules, which were obviously connected by remnants of the plasmalemma (Fig. 5B–E). However, some NCs still contained a nucleus and a few mitochondria, with a single central vacuole occupying most of their volume and with the cytoplasm being confined to a narrow peripheral layer surrounding the vacuole. Gaps between the cell wall and plasma membrane were no longer found (Fig. 5B). The cytoplasm contained lipid-like globules that were devoid of internal structure and did not appear to be membrane-bound. Although they tended to distend the tonoplast they were not pinched off into the vacuole (Fig. 5B, C), but remained attached to cytoplasmic remnants until very late in cytoplasmic degeneration. Cytoplasmic collapse was preceded by loss of the tonoplast, detachment of the plasma membrane from the cell wall and spread of the vacuolar sap across the whole cell lumen (Fig. 5D). Mitochondria and the usually flattened nucleus seemed to be degraded along with or immediately after the tonoplast. The majority of intercellular spaces were again devoid of internal structures. Occasionally, however, lipid-like globules were found (Fig. 5B).
In contrast to the NT, the phloem innervation remained almost unchanged. Therefore, the CCs still rich in mitochondria could be easily identified (Fig. 5B, E). It could be shown that there was, on average, one CC for each SE. Ultrastructural differences between the SE-CC-Cs of the active and the exhausted nectary are restricted to an apparent numerical increase of vacuolar inclusions in the CCs of the exhausted nectary. The SEs, however, displayed an identical ultrastructure. The sieve plate pore diameter was 40 ± 4 x 10–8 m (n = 19) in the active nectary and 41 ± 5 x 10–8 m (n = 29) in the exhausted nectary with a plate thickness of 56 ± 6 x 10–8 m (n = 16) and 58 ± 7 x 10–8 m (n = 18), respectively, and the total pore area was approx. 50 % of the plate area. The pores appeared to be filled with two differently stained components: an electron-translucent callose collar surrounded a darkly stained and irregularly shaped core (Fig. 5A, E). The SEs not only displayed various shapes but they also did not form straight files, i.e. a zigzag path of sieve tubes existed (Figs 6 and 8B). The SEs differed from one another by the number of interfaces that they shared with other SEs. Terminal SEs were connected by only one sieve plate with subterminal SEs, whereas the latter and other interstitial SEs shared at least two sieve plates with their neighbours. Individual SEs were shown to be in contact with up to five other SEs (Fig. 7A).
|
|
|
The intercellular contact between SEs and other types of neighbouring cells was also variable. As a rule, SEs were in contact with more than one CC (Fig. 7B). These interfaces, however, differed markedly in size. The largest interface indisputably resulted from division of the mother cell into the SE-CC sisters. These sisters sometimes had almost equal volumes, but on average CCs were more voluminous (V = 650 ± 134 x 10–9 m3; n = 6; range: 489–809 x 10–9 m3) than SEs (V = 489 ± 103 x 10–9 m3; n = 5; range: 384 – 603 x 10–9 m3). Whereas SEs inside the phloem strands lacked contact both with NCs and with intercellular spaces, peripheral and, in particular, terminal SEs shared up to 68 % of their surface area with NCs, and up to 6 % with intercellular spaces (Fig. 9).
|
The fate of the post-secretory nectary
Light-microscopical examination of hand sections through fresh ovaries revealed that (a) the diameter of the active nectary increased by approx. 25 % during secretion, (b) while the maximum diameter of the ovary increased about 2-fold, the peripheral diameter of the exhausted nectary increased only slightly by about 10–20 % after cessation of secretion, and (c) while the horziontal height of the corpse of the nectary bulge did not change markedly, the vertical width was reduced significantly to about 30 % of that in the active nectary (Table 1). The basal nectary bulge became inconspicuous and a brownish stained layer of crushed cell corpses remained.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONACKNOWLEDGEMENTLITERATURE CITED |
|---|
Sink to source transition
According to the classification of sinks into the categories (i) consuming, (ii) accumulating, (iii) secreting (Zamski, 1996), the nectary of D. purpurea changes from both a consuming and accumulating body during growth and starch deposition to a secreting one. During secretion the nectary is not only a sink but it also becomes a source, because a number of hallmark features that indicated the onset of PCD were observed (see ‘Hallmarks of PCD’). Nutrient element mobilization and recycling in surviving parts of the plant are central features of PCD (Rubinstein, 2000; Huelskamp and Schnittinger, 2004). There is reason to believe the SEs and CCs remain functional until very late, perhaps dying only after the surrounding NCs have collapsed. The surviving phloem appears to be involved in the uptake and export of products of cellular catabolism from the dying nectary to sinks elsewhere in the plant. In addition to senescing leaves and corollas (Matile and Winkenbach, 1971; Noodén and Leopold, 1988; Bieleski, 1995), the nectary is another example of the marked persistence of phloem activity in a background of dying cells and the rapid reversal in sink–source behaviour.
Solute flux necessary for nectar production
The nectar of D. purpurea was composed of 78·4 % sucrose, 10·5 % glucose and 11·1 % fructose (Lichius et al., 1990). On average, the daily amount of nectar produced by one flower of D. purpurea was 11·6 mg containing 16–27 % sugar (Percival and Morgan, 1965), 13·7 mg containing 24–45 % sugar (Halmágyi and Gulyás, 1970), and 11·1 mg containing 23·6 % sugar (Assmann, 1986). This is equivalent to about 470–1300 mM sucrose. The lower concentration resulted from daily exploitation, thus indicating that the sugar concentration of nascent nectar was near this value. In nectar that is exposed to the environment for longer times, sugars accumulate to higher concentrations due to evaporation of water. In the phloem sap, sugar (in most species almost exclusively sucrose) also reached very high levels, usually ranging from 200 to 1600 mM (Winter and Huber, 2000). Assuming that (a) sugar concentration in nascent nectar is 440 mM, one flower may secrete 15–30 x 10–9 m3 nectar a day, (b) the nascent nectar and the phloem sap coincide in sucrose concentration, and (c) that only the 78·4% share of sucrose in the nectar originates from the phloem sap, then approx. 118–236 x 10–10 m3 phloem sap has to be translocated per day via the phloem vasculature of the nectary. The volume of the nectary is approx. 2 x 10–9 m3 (Gaffal et al., 1998), but only a small fraction is allotted to the sieve tubes. Estimating 1%, then nectar secretion loads new sap to the extent that the entire volume of the sieve tubes must be refilled 590–1180 times per day.
Solute flux necessary for starch production
In order to get an idea of the sucrose flux necessary for storage of starch, it was estimated that at the onset of nectar secretion, 10–20 % of the volume of the nectary (about 2 x 10–9 m3) of D. purpurea is occupied by starch, i.e. a maximum of 4 x 10–10 m3. By taking a median density of starch (1·34 mg mm–3), it is converted to a maximum of 0·54 mg starch. Because the amount of energy per mass of starch (4·15 kcal g–1 from maize) and of sucrose (4·1 kcal g–1) are almost equivalent to one another (Neumüller, 1979), it can be roughly estimated that 0·54 mg starch originates from about 0·54 mg of sucrose. In this balance of energy the expenditure of energy for the synthesis of starch from sucrose is not taken into consideration. Therefore this value is underestimated to an unknown degree. However, the error is reduced because the calculation is with the upper limit of starch content. Because 3·6 x 10–9 m3 of a 15 % solution contain 0·54 mg sucrose, the phloem vasculature of the nectary (2 x 10–11 m3) has to be refilled 180 times. It appears that, similar to starch accumulation in other nectaries (Razem and Davis, 1999; Peng et al., 2004), it takes > 1 d to deposit the starch. But even if this 1-d interval is used, the rate of sucrose flux during the accumulation of starch would be at least three to seven times smaller than during the secretion of nectar. Hence, the sink strength of the nectary seems to increase during secretion. The temporarily enhanced sucrose unloading capacity may result either from intensification of the unloading system or from turning on a specific efflux mechanism.
Origin of nectar
The coincidence of nectar secretion and degradation of starch in the NT has been a well-known phenomenon for more than a century (Behrens, 1879). Recently, NEC1, a gene predominantly expressed in the nectaries of Petunia hybrida, was cloned (Ge et al., 2000). The pattern of NEC1 expression appeared to follow the temporal events of starch hydrolysis, thus supporting the hypothesis that NEC1 is involved in the process of nectar secretion. However, as early as 1886, Stadler questioned the origin of nectar in nectaries that lacked adequate quantities of starch (Stadler, 1886). He, and more recently Horner et al. (2003), suggested that such nectaries, after being fed with fluid nectar-producing precursors from sources elsewhere in the plant, might secrete it directly without an intermediate storage form of reserves. The calculations above suggest that sugar originating from starch degradation plays only a minor role in nectar production of D. purpurea. The bulk of nectar seems to be directly unloaded from the phloem sap. The decrease in nectar production after reduction of the assimilate transport by girdling flowering shoots (von Czarnowski, 1952; Wykes, 1952) or by darkening and defoliation experiments (von Czarnowski, 1952; recently Pleasants and Chaplin, 1983; Nepi et al., 2005) strongly favours the view that a good deal of nectar is supplied from the current assimilate flux.
Morphotype of the nectary phloem
Morphometry established that the CCs of the phloem strands innervating the floral nectaries of D. purpurea and Isoplexis canariensis (Gaffal and El-Gammal, 2003) are typically more voluminous than the SEs. A similar ratio in size is not only evident from studies on other nectaries (fig. 3 in Figier, 1971; Findlay and Mercer, 1971; Durkee et al., 1981; Durkee, 1983; Davis et al., 1986, 1988; Razem and Davis, 1999; fig. 3 in Zhu and Hu, 2002; Wist and Davis, 2006) but also from studies on the vasculature of other sink tissues. CCs at least as wide as the SEs were found in the unloading phloem of the rice pericarp (Oparka and Gates, 1981), in the minor bundle of the developing apple fruit (fig. 7a, c in Peng et al., 2003), and in the phloem feeding the pericarp and the cotyledons (named ‘seed pericarp’) of developing walnut fruit (fig. 2 in Wu et al., 2004). The large size of the CCs relative to the SEs in special regions of phloem unloading is the reverse of that found in the transport phloem, but quite similar to that of leaf minor veins (for instance, Haritatos et al., 2000), where phloem loading occurs. The large size of the CCs is thought to reflect their active role not only in phloem loading (Barth et al., 2003), but also in the unloading process (Oparka and Gates, 1981; Findlay, 1988).
Ramification of the phloem innervation causing bifurcation of the solute flux necessitates at least tripolar SEs. Multipolar (more than tripolar) SEs are junctions where solute inflow from more than one SE and solute outflow into more than one SE can meet. The amount of inflow from and outflow into adjacent SEs seems to depend on the number of sieve pores per interface. Solute distribution may be controlled in this way. Multipolar SEs interlinked via sieve plates with more than two SEs are known from leaf veins (Koch, 1884) and from wound phloem (Eschrich, 1953, Schulz, 1986). The sieve tubes of the latter, obviously an unloading phloem, are zigzag shaped, similar to those of the nectary phloem. Between two given points a straight sieve tube is not only shorter than a zigzag-shaped one, but will also exert less frictional resistance to the solute flux. If future studies corroborate that such anatomical features are typical of unloading phloem, models of solute flux should incorporate the effects introduced by shape.
Open access of phloem to intercellular space
Phloem elements in contact with intercellular spaces were found not only in this and more recent studies on nectaries of (a) Helianthus annuus var. macrocarpus (fig. 5 in Sammataro et al., 1985); (b) Vicia faba (Davis et al., 1988); (c) Pisum sativum (Razem and Davis, 1999); (d) Isoplexis canariensis (Gaffal and El-Gammal, 2003); (e) Echinacea purpurea (Wist and Davis, 2006), but also more than a century ago. Intercellular ducts abutting the vasculature of the nectary of Serratula lycophila were interpreted to be involved in the secretion of nectar (von Wettstein, 1889). In the nectaries of Paeoniaceae (Zimmermann, 1932; Frey-Wyssling and Häusermann, 1960; Hiepko, 1966) cells of the so-called ‘conducting tissue’ [Leitparenchym (obviously phloem elements)] even adjoined the substomatal spaces of those ‘stomata’ (Discusspalten) through which the nectar escaped to the outside. This structural specialization, which most probably also exists in Helianthus (fig. 5 in Sammataro et al., 1985), indicated the possibility of a direct unloading of assimilates from the phloem to the exterior of the plant (Zimmermann, 1932; Hiepko, 1966; Schnepf, 1973), and might also explain why exclusively sucrose was found in the nectar of Paeonia (Frey-Wyssling and Häusermann, 1960). A sucrose-dominant nectar is also secreted from the petal tip of Medinilla magnifica (Tobe et al., 1989). There any NT (nectary parenchyma) is missing, but the associated vascular bundle consisting of phloem only is thickened. This anatomical evidence again indicates that the nectary is the phloem-rich median petal bundle itself. Although it remains to be proven that the reticulum of intercellular spaces terminating at the permanently open stomata of the nectary (Gaffal et al., 1998) is connected with the reticulum of intercellular spaces arriving at the phloem strands, there is a good chance for the existence of continuity. Two fundamental differences exist between the nectaries of Paeonia and Medinilla on the one hand, and D. purpurea on the other hand. (1) The length of the path the nectar has to pass through the intercellular space of the NT until it is discharged to the surface of the nectary via the modified stomata is much longer in the nectary of D. purpurea. The chemical composition of the nectar may be altered along this passage. For instance, invertases located in the cell wall boundary of the intercellular spaces may form glucose and fructose from the raw material of the nascent nectar (sucrose). (2) Starch in the nectary parenchyma is hydrolysed during secretion, and sugars originating from this decomposition may enter the intercellular spaces and influence the chemical composition of the nectar.
Intercellular spaces that abut against phloem elements are not limited to nectaries. They were clearly seen (a) between unloading phloem cells of the developing apple fruit (fig. 7a, c in Peng et al., 2003), (b) in minor veins of developing tobacco leaves (fig. 4 in Wright et al., 2003), (c) in minor veins of the leaves of transgenic potato plants (figs 18 and 19 in Schulz et al., 1998) and (d) in adaxial minor veins of Cucurbita pepo leaves (fig. 3 in Turgeon and Webb, 1976), which were suggested to constitute the preferential pathway for the import of assimilates.
The apoplast of the secreting nectary
During secretion the intercellular space of the nectary is most probably filled with nascent nectar. Positive staining for carbohydrates indicated their presence in the intercellular spaces of the extrafloral nectaries of Australian acacias (Marginson et al., 1985). Because the prenectar has to pass the cell wall, we think that the sugar concentrations of the fluid in the wall apoplast and in the intercellular space are identical. Accordingly, the sink end of the phloem pathway may be surrounded by an apoplast solution with a concentration of osmotically active solutes not drastically different from that of the phloem sap. Therefore, unloading of sucrose from the phloem pathway that is driven by large transmembrane concentration differences seems to play a minor role. High sucrose concentration in the cell wall apoplast favours an energy-dependent and possibly carrier-mediated process as postulated for strong sinks (Wolswinkel, 1985), such as ripening grape berries (Wang et al., 2003). In theory, the unloading of sucrose into the apoplast could occur by specific effluxers that either use a proton antiport mechanism or are directly energized by ATP (Lalonde et al., 2004). ATP-ase activity was detected by cytochemistry in the phloem tissue adjacent to the nectaries of Gossypium (Eleftheriou and Hall, 1983) and Hibiscus (Sawidis, 1991).
The mechanisms of nectar secretion proposed by Vassilyev (2003) and recently Koteyeva et al. (2005) were based on the idea that the apoplast of the NT forms the main route taken by the nectar sugars on their way to the stomatal opening. Although the actual direction faced by the wall ingrowths of the nectary CCs (which are transfer cells) of Vicia faba (Davis et al., 1988) clearly suggested an avenue for unloading of pre-nectar constituents directly to intercellular spaces (i.e. apoplast), Vassilyev (2003) and recently Koteyeva et al. (2005) either ignored the CCs of the nectary phloem or assigned them only a passive role in assimilate transport. Except for the substomatal cavity, Vassilyev's fig. 1 also lacked intercellular spaces. In contradiction to this model, the sites of open access of the SE-CC-Cs to the intercellular system were considered to be the shortest pathways for an apoplastic efflux of assimilates (Gaffal and Heimler, 2000).
Hallmarks of PCD
During secretion and the final degradation phase of the nectary, a number of hallmark features arise that are reminescent of autophagy and autolysis. These symptoms are common in plants and appeared during normal pericarp development and may therefore indicate developmental PCD (Dangl et al., 2000; van Doorn and Woltering, 2005). The phenotype of PCD in the foxglove's nectary is definitively very similar to that in the nectary of Glycine max (Horner et al., 2003), and to the phenotype of senescence in the petals of Ipomoea (Matile and Winkenbach, 1971), Dianthus (Smith et al., 1992) and Hemerocallis (Stead and van Doorn, 1994). Senescence is considered to be a type of PCD (Noodén, 2004). The sequence of cell degeneration discussed here is an interpretation without independent confirmation that one stage neccessarily leads to another.
(a) Increased vacuolation
One of the earliest events observed is the fusion of smaller vacuoles, which culminates in the formation of a single central vacuole. This obviously occurs by rupture or degradation of the tonoplast-coated septa between adjacent vacuolar spaces (Figs 3A and 4A, E, F). Some of the vacuolar inclusions seem to originate from these events. Further enlargement of the central vacuole is associated with the extension growth of the dying but still expanding NCs. The NCs of Eccremocarpus scaber also increased 2-fold in volume during their development (Belmonte et al., 1994). Significant changes in the vacuome, usually in the form of increased vacuolation, were widely reported in plant PCD (Rogers, 2005).
(b) Autophagy
Vacuolar inclusions also seem to originate from the pinching off of plasmatic evaginations into the vacuole. With progression through the PCD process, incorporation of plasmatic components into the vacuole increases in number. Degradation of vacuolar inclusions seems to occur through autophagy. In dying plant cells there is ample evidence for vacuolar autophagy (Krishnamurthy et al., 2000). Autophagic PCD is normally used for recycling cellular components (Lam, 2004), since progressive elimination of cytoplasm by hydrolysis in vacuoles extends the survival of the dying cell and efficient withdrawal of nutrients from the cells can be achieved (Jones, 2000; Mittler and Cheung, 2004).
(c) Degradation of plastids
The decrease in electron opacity of the starch grains is the first obvious symptom of starch degradation. The coincident occurrence of small gaps between the vacuole and the interior of the amyloplast may allow leakage of vacuolar sap into this organelle. Evidence that the vacuole represents the lytic compartment of plant cells that contains hydrolytic enzymes has been largely established by Matile (1975). The products of hydrolase activity can, on the other hand, be released into the vacuole via these gaps. As long as they are small, digestion may be under control and homeostasis maintained. If this idea and the widely accepted idea that the sugars produced by starch lysis enter the nectar are true, then these sugars must pass at least two membranes – the tonoplast and the plasma membrane (plasmalemma). In contrast to the open access of the vacuolar sap to the interior of the amyloplasts, indicated by the present study, engulfing of the plastids by the vacuole and subsequent hydrolysis of starch in the vacuole were described in the floral nectaries of Eccremocarpus saber (Belmonte et al., 1994) and Cucumis sativus (Peng et al., 2004), and in senescing French bean leaves (Minamikawa et al., 2001).
Accumulation of the products of starch degradation in the vacuole may serve one more transient function. If the sugar concentration in the apoplast of the nectary is at least 440 mM (see ‘Solute flux necessary for nectar production’), a significant osmotic stress exists. These apoplastic solutes will lower cell turgor and cause plasmolysis. Transient local detachment of the protoplasts from the cell wall indicates such an event. As the toxic effect of prolonged plasmolysis will cause premature death, an increase in the solute concentration in the vacuole will counterbalance too much loss of water in order to facilitate extension growth of the NCs by enlargement of the vacuole, and provide adequate turgor for effective PCD, which is a slowly proceeding process (van Doorn and Woltering, 2005).
(d) Nuclear changes
Deformation of the nuclei and gradual disappearance of chromatin were found in this study to be relatively early symptoms of nuclear degradation. Changes in nuclear shape were also observed in dying cells of the tapetum (Papini et al., 1999) and the nucellus (Greenwood et al., 2005). However, the ultimate breakdown is a relatively late event, just as in most senescing cells (Noodén and Leopold, 1988).
(e) Intensification of electron opacity
The increase in the osmiophilic properties of the cytoplasm, nucleoplasm and contents of the endomembrane system correlates with the occurrence of osmiophilic particles at the periphery of the protoplast, in the gaps between the cell wall and the plasma membrane, and in the intercellular spaces. Conclusive evidence for possible ontogenetical relationships between these events is missing, however. Characteristic increases in cytoplasmic density were also observed in the so-called dense cells of the Arabidopsis thaliana nectary (Zhu and Hu, 2002) and during early degeneration of the tobacco synergid (Huang et al., 1993).
(f) Disintegration of the tonoplast
As degradation of the NT advances to its end point, tonoplast integrity is lost. Dissolution and subsequent release of hydrolytic enzymes into the vestiges of cytoplasma, which promote final cell death, are common features of plant PCD (Jones, 2000; Moriyasu and Klionsky, 2004; Noodén, 2004). Eventually total autolysis of the remaining cytoplasmic constituents ensues, ending with leakage of the catabolites into the apoplast (Lam, 2004), from which they may be loaded into the structurally intact phloem and exported to sinks elsewhere in the plant. The breakdown of the remaining mitochondria seems to coincide with the breakdown of the tonoplast. The lipid-like globules, whose origin and function are completely unknown, and the plasma membrane, are among the most persistent cellular components.
The fate of the cell walls of the NT
The breakdown of cellular contents is paralleled by the occurrence of wavy cell walls. This buckling is followed by further collapse of the NC corpses as a result of forces exerted by the swelling fruit. Ensuing nectar secretion degeneration and the collapse of NCs was also observed in Glycine max (Horner et al., 2003) and Hexisea imbricata (Stpiczy
ska et al., 2005). Cell walls are typically not eliminated during PCD in plants (Mittler, 1998), and crushing, tearing or overgrowth of dead cells by their expanding neighbours is one of their fates (Beers, 1997). The fact that cell walls are not degraded may reflect an energetic budget that is too high (Gray and Johal, 1998) or the cell walls perform specialized functions such as mechanical support or water transport (Krishnamurthy et al., 2000). We think that once the mission of the nectary is accomplished, it alters to a mode in which not only resource mobilization but also defence should be the most important goals for preservation of the next generation. The production of a physical barrier composed of compressed cell walls and collapsed intercellular spaces may inhibit the invasion of pathogens via the permanently open stomata into living parts of the ovary. Nectar clearly represents a preferred medium for bacterial and fungal growth (Fig. 3B), and the bacterial pathogen Erwinia amylovora is known to follow this route during infection (Buban et al., 2003).
Future directions
PCD is usually regarded as a genetically encoded active process that performs a detectable function in the life of the plant (Gunawardena et al., 2005). If nectary death is correctly interpreted as programmed, a genetic control can be expected. In this case nectaries could be excellent model systems for the study of fundamental cell death processes, because the whole natural period from maturity to death is relatively short. In particular the nectaries of species that lend themselves readily to genetic manipulation, mutagenesis or transformation, such as Arabidopsis thaliana, may be employed. Enlargement of the vacuoles in the post-secretory nectary of Arabidopsis thaliana (Davis, 1994) and the ultrastructural changes observed during degeneration of the NT (Zhu and Hu, 2002) seem to be hallmarks of PCD. Genetic analysis using arabidopsis mutants that show abnormalities in nectary morphology – as for instance the petal loss-1 (ptl-1) mutation (Bowman, 1994) – may provide clues to the molecular mechanisms underlying PCD in the nectary.
|
ACKNOWLEDGEMENT |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONACKNOWLEDGEMENTLITERATURE CITED |
|---|
We are indebted to Dr A. R. Davis for an excellent revision of the manuscript.
|
LITERATURE CITED |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONACKNOWLEDGEMENTLITERATURE CITED |
|---|
-
Assmann U. (1986) Nektaruntersuchungen an Digitalis purpurea Beständen. PhD Thesis, FB Biology,(University of Marburg, Germany).
Barth I, Meyer S, Sauer N. (2003) PmSUC3: characterization of a SUT2/SUC3-type sucrose transporter from Plantago major. The Plant Cell 15:1375–1385.
Beers EP. (1997) Programmed cell death during plant growth and development. Cell Death and Differentiation 4:649–661.[CrossRef][Web of Science][Medline]
Behnke H-D. (1991) Distribution and evolution of forms and types of sieve-element plastids in the dicotyledons. Aliso 13:167–182.
Behrens WJ. (1879) Die Nektarien der Blüten. Flora 62:433–457.
Belmonte E, Cardemil L, Kalin-Arroyo MJ. (1994) Floral nectary structure and nectar composition in Eccremocarpus scaber (Bignoniaceae), a hummingbird-pollinated plant of central Chile. American Journal of Botany 81:493–503.[CrossRef]
Bieleski RL. (1995) Onset of phloem export from senescent petals of daylily. Plant Physiology 109:557–565.[Abstract]
Bowman J. (1994) Arabidopsis – an atlas of morphology and development(Springer-Verlag, New York, NY).
Buban T, Orosz-Kovacs Z, Farkas A. (2003) The nectary is the primary site of infection by Erwinia amylovora (Burr.) Winslow et al.: a mini review. Plant Systematics and Evolution 238:183–194.
von Czarnowski C. (1952) Untersuchungen zur Frage der Nektarabsonderung. Archiv für Geflügelzucht und Kleintierkunde 1:23–44.
Dangl JL, Dietrich RA, Thomas H. (2000) Senescence and programmed cell death. In Buchanan B, Gruissem W, Jones R (Eds.). Biochemistry and molecular biology of plants(American Society of Plant Biologists, Rockville, MD) pp. 1044–1100.
Davis AR. (1994) Nectary morphology and anatomy. In Bowman J (Ed.). Arabidopsis – an atlas of morphology and development(Springer-Verlag, New York, NY) pp. 172–177.
Davis AR, Peterson RL, Shuel RW. (1986) Anatomy and vasculature of the floral nectaries of Brassica napus (Brassicaceae). Canadian Journal of Botany 64:2508–2516.
Davis AR, Peterson RL, Shuel RW. (1988) Vasculature and ultrastructure of the floral and stipular nectaries of Vicia faba (Leguminosae). Canadian Journal of Botany 66:1435–1448.
van Doorn WG and Woltering EJ. (2005) Many ways to exit? Cell death categories in plants. Trends in Plant Science 10:117–122.[Web of Science][Medline]
Durkee LT. (1983) The ultrastructure of floral and extra-floral nectaries. In Bentley B and Elias T (Eds.). The biology of nectaries(Columbia University Press, New York, NY) pp. 1–26.
Durkee LT, Gaal DJ, Reisner WH. (1981) The floral and extrafloral nectaries of Passiflora. I. The floral nectary. American Journal of Botany 68:453–462.[CrossRef]
Eleftheriou EP and Hall JL. (1983) The extrafloral nectaries of cotton. II. Cytochemical localization of ATPase activity and Ca2+ binding sites, and selective osmium impregnation. Journal of Experimental Botany 34:1066–1079.
Eschrich W. (1953) Beiträge zur Kenntnis der Wundsiebröhrenentwicklung bei Impatiens holstii. Planta 43:37–74.[CrossRef]
Fahn A. (2000) Structure and function of secretory cells. Advances in Botanical Research 31:37–75.[CrossRef]
Figier J. (1971) Etude infrastructurale de la stipule de Vicia faba L. au niveau du nectaire. Planta 98:31–49.[CrossRef]
Findlay N. (1988) Nectaries and other glands. In Baker D and Hall J (Eds.). Solute transport in plant cells(Longman Scientific and Technical, New York, NY) pp. 538–560.
Findlay N and Mercer FV. (1971) Nectar production in Abutilon. II. Submicroscopic structure of the nectary. Australian Journal of Biological Sciences 24:657–664.
Frei E. (1955) Die Innervierung der floralen Nektarien dikotyler Pflanzenfamilien. Berichte der Schweizerischen Botanischen Gesellschaft 65:60–114.
Frey-Wyssling A and Agthe C. (1950) Nektar ist ausgeschiedener Phloemsaft. Verhandlungen der Schweizerischen Naturforschenden Gesellschaft 130:175–176.
Frey-Wyssling A and Häusermann E. (1960) Deutung der gestaltlosen Nektarien. Berichte der Schweizerischen Botanischen Gesellschaft 70:150–162.
Gaffal KP and Heimler W. (2000) Die Nektarien von herzglycosidhaltigen Rachenblütlern – eine Quelle der Speise für Götter mit Herzinsuffizienz? Mikrokosmos 89:129–138.
Gaffal KP and El-Gammal S. (2003) Nectar and honey analyses for prognosis of phloem transport of natural plant toxins? Drogenreport 16:9–13 [in German with Summary and Conclusions in English].
Gaffal KP, Heimler W, El-Gammal S. (1998) The floral nectary of Digitalis purpurea L, structure and nectar secretion. Annals of Botany 81:251–262.
Ge YX, Angenent GC, Wittich PE, Peters J, Franken J, Busscher M, et al. (2000) NEC1, a novel gene, highly expressed in nectary tissue of Petunia hybrida. The Plant Journal 24:725–734.[CrossRef][Web of Science][Medline]
Gray J and Johal GS. (1998) Programmed cell death in plants. In Anderson M and Roberts J (Eds.). Arabidopsis(CRC Press, Boca Raton) pp. 360–394 Annual Plant Reviews.
Greenwood JS, Helm M, Gietl C. (2005) Ricinosomes and endosperm transfer cell structure in programmed cell death of the nucellus during Ricinus seed development. Proceedings of the National Academy of Sciences of the USA 102:2238–2243.
Gunawardena AHLAN, Sault K, Donnelly P, Greenwood JS, Dengler NG. (2005) Programmed cell death and leaf morphogenesis in Monstera obliqua (Araceae). Planta 221:607–618.[CrossRef][Web of Science][Medline]
Halmágyi L and Gulyás S. (1970) Nektarium und Nektarproduktion der Digitalis-Arten. Acta Biologica Szeged 16:43–50.
Haritatos E, Medville R, Turgeon R. (2000) Minor vein structure and sugar transport in Arabidopsis thaliana. Planta 211:105–111.[CrossRef][Web of Science][Medline]
Hiepko P. (1966) Zur Morphologie, Anatomie und Funktion des Diskus der Paeoniaceae. Berichte der Deutschen Botanischen Gesellschaft 79:233–245.
Horner HT, Healy RA, Cervantes-Martinez T, Palmer RG. (2003) Floral nectary fine structure and development in Glycine max L. (Fabaceae). International Journal of Plant Sciences 164:675–690.[CrossRef]
Huang BQ, Strout WG, Russell SD. (1993) Fertilization in Nicotiana tabacum – ultrastructural organization of propane-jet-frozen embryo sacs in-vivo. Planta 191:256–264.
Huelskamp M and Schnittinger A. (2004) Programmed cell death in development of plant vegetative tissue (leaf and root). In Gray J (Ed.). Programmed cell death in plants(Blackwell Publishing, Oxford) pp. 106–130.
Jones AM. (2000) Does the plant mitochondrion integrate cellular stress and regulate programmed cell death? Trends in Plant Science 5:225–230.[CrossRef][Web of Science][Medline]
Koch A. (1884) Über den Verlauf und die Endigungen der Siebröhren in den Blättern. Botanische Zeitung 42:401–411.
Koteyeva NK, Vassilyev AE, Tarlyn N, Franceschi VR. (2005) On mechanisms of nectar secretion. XVII International Botanical Congress (Austria, Vienna) Abstract 10·3·4.
Krishnamurthy KV, Krishnaraj R, Chozhavendan R, Christopher FS. (2000) The programme of cell death in plants and animals – a comparison. Current Science 79:1169–1181.
Lalonde S, Wipf D, Frommer WB. (2004) Transport mechanisms for organic forms of carbon and nitrogen between source and sink. Annual Review of Plant Biology 55:341–372.[CrossRef][Medline]
Lam E. (2004) Controlled cell death, plant survival and development. Nature Reviews Molecular Cell Biology 5:305–315.[CrossRef][Web of Science][Medline]
Lichius JJ, Daniel M, Fingerhut T, Wärtgen T. (1990) Der Nektar von Digitalis. Deutsche Apotheker Zeitung 130:2191–2193.
McKeon TA and Gaffield W. (1990) Viewing stereopictures in three dimensions with naked eyes. Trends in Biochemical Sciences 15:412–413.[CrossRef][Web of Science][Medline]
Marginson R, Sedgley M, Douglas TJ, Knox RB. (1985) Structure and secretion of the extrafloral nectaries of Australian acacias. Israel Journal of Botany 34:91–102.
Matile P. (1975) The lytic compartment of plant cells. Cell biology monographs(Springer-Verlag, Berlin) 1:.
Matile P and Winkenbach F. (1971) Function of lysosomes and lysosomal enzymes in the senescing corolla of the morning glory (Ipomoea purpurea). Journal of Experimental Botany 22:759–771.
Minamikawa T, Toyooka K, Okamoto T, Hara-Nishimura I. (2001) Degradation of ribulose 1,5 biphosphate carboxylase oxygenase by vascular enzymes of senescing French bean leaves: immunocytochemical and ultrastructural observations. Protoplasma 218:144–153.[CrossRef][Web of Science][Medline]
Mittler R. (1998) Cell death in plants. In Locksin RA (Ed.), et al. When cells die(Wiley-Liss, New York, NY) pp. 147–174.
Mittler R and Cheung AY. (2004) Cell death in plant development and defense. In Lockshin RA and Zakeri Z (Eds.). When cells die II(John Wiley & Sons, Hoboken, NJ) pp. 99–121.
Moriyasu Y and Klionsky DJ. (2004) Autophagy in plants. In Klionsky DJ (Ed.). Autophagy(Landes Bioscience, Georgetown, TX) pp. 208–215.
Nepi M, Pacini E, Cresti L, Guarnieri M, Artese D. (2005) Nectar carbohydrate sources: facts and hypothesis. XVII. International Botanical Congress (Austria, Vienna) Abstract 10·3·6.
Neumüller O-A. (1979) Römpps Chemie-Lexikon(Franckh'sche Verlagshandlung, Stuttgart) 8. Aufl.
Noodén LD. (2004) Plant cell death processes(Elsevier, Amsterdam).
Noodén LD and Leopold AC. (1988) Senescence and aging in plants(Academic Press, San Diego, CA).
Oparka KJ and Gates P. (1981) Transport of assimilates in the developing caryopsis of rice (Oryza sativa L.): ultrastructure of the pericarp vascular bundle and its connection with the aleurone layer. Planta 151:561–573.[CrossRef]
Papini A, Mosti S, Brighigna L. (1999) Programmed-cell-death events during tapetum development of angiosperms. Protoplasma 207:213–221.[CrossRef]
Patrick JW, Zhang WH, Tyerman SD, Offler CE, Walker NA. (2001) Role of membrane transport in phloem translocation of assimilates and water. Australian Journal of Plant Physiology 28:695–707.[Web of Science]
Peng Y-B, Lu YF, Zhang DP. (2003) Abscisic acid activates ATPase in developing apple fruit especially in fruit phloem cells. Plant, Cell and Environment 26:1329–1342.[CrossRef]
Peng Y-B, Li Y-Q, Hao Y-J, Xu Z-H, Bai S-N. (2004) Nectar production and transportation in the nectaries of the female Cucumis sativus L. flower during anthesis. Protoplasma 224:71–78.[Web of Science][Medline]
Pennell RI and Lamb C. (1997) Programmed cell death in plants. The Plant Cell 9:1157–1168.[CrossRef][Web of Science][Medline]
Percival M and Morgan P. (1965) Observations on the floral biology of Digitalis species. New Phytologist 64:1–22.[Medline]
Pleasants JM and Chaplin SJ. (1983) Nectar production rates of Asclepias quadrifolia: causes and consequences of individual variation. Oecologia 59:232–238.[CrossRef][Web of Science]
Ranganath RM and Nagashree NR. (2001) Role of programmed cell death in development. International Reviews of Cytology 202:159–242.
Razem FA and Davis AR. (1999) Anatomical and ultrastructural changes of the floral nectary of Pisum sativum L. during flower development. Protoplasma 206:57–72.
Rogers HJ. (2005) Cell death and organ development in plants. Current Topics in Developmental Biology 71:225–261.[Web of Science][Medline]
Rubinstein B. (2000) Regulation of cell death in flower petals. Plant Molecular Biology 44:303–318.[CrossRef][Web of Science][Medline]
Sammataro D, Erickson EH, Garment MB. (1985) Ultrastructure of the sunflower nectary. Journal of Apicultural Research 24:150–160.
Sawidis T. (1991) A histochemical study of nectaries of Hibiscus rosa-sinensis. Journal of Experimental Botany 42:1477–1487.
Schnepf E. (1973) Sezernierende und exzernierende Zellen bei Pflanzen. In Hirsch GC, Ruska H, Sitte P (Eds.). Grundlagen der Cytologie(Fischer, Jena) pp. 461–477.
Schulz A. (1986) Wound phloem in transition to bundle phloem in primary roots of Pisum sativum L. I. Development of bundle-leaving wound-sieve tubes. Protoplasma 130:12–26.[CrossRef]
Schulz A, Kühn C, Riesmeier JW, Frommer WR. (1998) Ultrastructural effects in potato leaves due to antisense-inhibition of the sucrose transporter indicate an apoplasmic mode of phloem loading. Planta 206:533–543.[CrossRef]
Smith MT, Saks Y, van Staden J. (1992) Ultrastructural changes in the petals of senescing flowers of Dianthus caryophyllus L. Annals of Botany 69:277–285.
Spurr AR. (1969) A low-viscosity epoxy resin embedding medium for electron microscopy. Journal of Ultrastructural Research 26:31–43.
Stadler S. (1886) Beiträge zur Kenntnis der Nektarien und Biologie der Blüten(Friedländer, Berlin).
Stead AD and van Doorn WG. (1994) Strategies of flower senescence – a review. In Scott RJ and Stead AD (Eds.). Molecular and cellular aspects of plant reproduction(Cambridge University Press, Cambridge) pp. 215–237.
Stpiczyska M, Davies KL, Gregg A. (2005) Comparative account of nectary structure in Hexisea imbricata (Lindl.) Rchb. f. (Orchidaceae). Annals of Botany 95:749–756.
Thompson MV and Holbrook NM. (2003) Scaling phloem transport: water potential equilibrium and osmoregulatory flow. Plant, Cell and Environment 26:1561–1577.[CrossRef]
Tobe H, Hakki MI, Langhammer L. (1989) Floral nectary in Medinilla magnifica, an Old World Melastomataceae. Botanische Jahrbücher für Systematik, Pflanzengeschichte und Pflanzengeographie 111:57–62.
Turgeon R and Webb JA. (1976) Leaf development and phloem transport in Cucurbita pepo: maturation of the minor vein. Planta 129:265–269.[CrossRef]
Vassilyev AE. (2003) Pocemu vydielajetsia nektar? O mechanizmie nectarovydielenija. Botaniceskij Zurnal 88:1–8.
Wang ZP, Deloire A, Carbonneau A, Federspiel B, Lopez F. (2003) An in vivo experimental system to study sugar phloem unloading in ripening grape berries during water deficiency stress. Annals of Botany 92:523–528.
von Wettstein R. (1889) Über die Compositen der österreichisch-ungarischen Flora mit zuckerabscheidenden Hüllschuppen. Sitzungsberichte der Kaiserlichen Akademie der Wissenschaften in Wien. Mathematisch-naturwissenschaftliche Classe 97:570–589.
Winter H and Huber SC. (2000) Regulation of sucrose metabolism in higher plants: localization and regulation of activity of key enzymes. Critical Reviews in Plant Sciences 19:31–67.[CrossRef]
Wist J and Davis AR. (2006) Floral nectar production and nectary anatomy and ultrastructure of Echinacea purpurea (Asteraceae). Annals of Botany 97:177–193.
Wolswinkel P. (1985) Phloem unloading and turgor-sensitive transport: factors involved in sink control of assimilate partitioning. Physiologia Plantarum 65:331–339.[CrossRef]
Wright KM, Roberts AG, Martens HJ, Sauer N, Oparka KJ. (2003) Structural and functional vein maturation in developing tobacco leaves in relation to AtSUC2 promotor activity. Plant Physiology 131:1555–1565.
Wu G-L, Zhang X-Y, Zhang L-Y, Pan Q-H, Shen Y-Y, Zhang D-P. (2004) Phloem unloading in developing walnut fruit is symplasmic in the seed pericarp, and apoplasmic in the fleshy pericarp. Plant and Cell Physiology 45:1461–1470.
Wykes GR. (1952) The influence of variations in the supply of carbohydrate on the process of nectar secretion. New Phytologist 51:294–300.[CrossRef]
Zamski E. (1996) Anatomical and physiological characteristics of sink cells. In Zamski E and Schaffer AA (Eds.). Photoassimilate distribution in plants and crops: source–sink relationships(Marcel Dekker, New York, NY) pp. 283–310.
Zhu J and Hu Z-H. (2002) Cytological studies on the development of sieve element and floral nectary tissue in Arabidopsis thaliana. Acta Botanica Sinensis 44:9–14.
Zimmermann JG. (1932) Über die extrafloralen Nektarien der Angiospermen. Botanisches Centralblatt/Beihefte 1. Abteilung 49:99–196.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  E. A. S. Paiva and S. R. Machado The Floral Nectary of Hymenaea stigonocarpa (Fabaceae, Caesalpinioideae): Structural Aspects During Floral Development Ann. Bot., January 1, 2008; 101(1): 125 - 133. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||