| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
0 Botanical Institute, Evolutionary Botany, University of Göteborg, Box 461, SE-405 30 Göteborg, Sweden
Received for publication May 11, 1999. Accepted for publication August 24, 1999.
ABSTRACT
An investigation of the embryology of Potentilla nivea was carried out using a confocal laser scanning microscope. The crassinucellate nature of the ovule as well as the presence of a multicellular archesporium was confirmed. A "nucellar cap" builds up as the result of mitotic divisions in the parietal cell tissue and the nucellar epidermis. Earlier reports that several mature megagametophytes are present in each ovule was confirmed. The processes of megasporogenesis and megagametogenesis in Potentilla turned out to differ from descriptions in previous reports, and we propose a reinterpretation. Most importantly, the "archesporium" of previous authors is here considered to be parietal cells, whereas the "chalazal cells" are here interpreted as the archesporium. Meioses commonly occur, giving rise to tetrahedral tetrads. Cytologically unreduced megagametophytes arise from generative tissue (archesporial cells or megasporocytes) only and are diplosporous. Apospory does not occur. The megagametophyte is monosporic and five-nucleate; the three antipodal nuclei normally present in eight-nucleate megagametophytes are missing.
Key Words: apospory • crassinucellate • diplospory • embryology • five-nucleate megagametophytes • laser scanning microscopy • multicellular archesporium • Potentilla • Rosaceae • tetrahedral tetrads
What is the sexual potential in the facultative agamosperm Potentilla nivea? Or in other words, what is the proportion of cytologically reduced megagametophytes in an ovule or a flower relative to the number of unreduced megagametophytes? This was the question that set off an investigation on the embryology of Potentilla. However, megasporogenesis as well as megagametogenesis turned out to follow patterns different from those previously described, and, before the question about sexual vs. asexual reproduction could be answered, we found it necessary to reinterpret megaspore and megagametophyte development in Potentilla.
The facultatively agamospermous and pseudogamous nature of Potentilla was established by Müntzing (1928, 1931)
, and the embryology of the genus was described some years later by Rutishauser (1943—P. canescens, P. verna, and P. arenaria; 1945—P. heptaphylla), Håkansson (1946—P. opaca, P. crantzii, P. tabernaemontani, P. collina, and P. argentea), and Hunziker (1954—P. argentea, P. verna, and P. canescens). The interpretation of the sporogenous tissue in apomictic members of Rosoideae began in 1901 with Murbeck's work on Alchemilla and was elaborated by Strasburger (1904)
and Böös (1917)
. Since the nucelli of Alchemilla and Potentilla are very similar in structure, the description of the megagametophyte development in Potentilla rests, to a large extent, on the earlier works on Alchemilla. Characteristic features for both genera include crassinucellate ovules with a multicellular archesporium (Maheshwari, 1950
). In a crassinucellate ovule many layers of cells build up between the archesporium and the nucellus epidermis, and in a multicellular archesporium more than one cell may enter meiosis and contribute to megagametophyte production. The great number of cells involved complicates the interpretation of structures observed in sectioned material. The many drawings and published results indicate that previous authors invested a lot of time and effort in their embryological studies. It is also obvious that the sectioning technique available in the first half of this century placed limits on what could be observed.
In this study we have made use of a Confocal Laser Scanning Microscope (CLSM), which was first used for plant embryological studies by Fredrikson, Carlsson, and Franksson (1988)
. The instrument produces optical sections by using a laser beam and relies on autofluorescence in cleared tissue. The optical sectioning technique (1) allows screening of thousands of ovules for identification of different developmental stages, (2) provides very thin sections (0.5–1 µm compared with ~10 µm for microtome sections) that makes it possible to follow the curved outlines of cells, (3) provides video sequences, which make it easier to follow the outline of cells in three dimensions, and (4) makes possible optical sectioning of each ovule over and over again to find better cutting angles or to show nuclei more clearly. In 1950 Maheshwari described embryological work by studying microtome sectioned material this way: "several workers have confessed with a feeling of disappointment that, in spite of repeated efforts and the study of hundreds of ovules they failed to find many of the critical stages in the process." The superiority of the optical sectioning technique over these older methods is clear.
Megasporocytes failing to undergo meiosis are usually considered to be the cause of agamospermy in Potentilla (Rutishauser, 1945
; Håkansson, 1946
; Hunziker, 1954
; Smith, 1963a
—P. tabernaemontani; this study). Cytologically unreduced megagametophytes develop and the embryos are derived from the egg or some other cell of the gametophyte. This type of agamospermy is classified as recurrent apomixis by Maheshwari (1950)
. Recurrent apomixis can be divided into two subclasses: (1) generative apospory, implying that the megagametophyte is derived from the archesporium and (2) somatic apospory, implying that the megagametophyte is derived from some other cell of the nucellus. Nogler (1984)
and Asker and Jerling (1992)
advocate the use of the terms diplospory instead of generative apospory or gonial apospory (Battaglia, 1963
) and apospory instead of somatic apospory. That terminology is accepted here. Both diplospory and apospory (sensu Asker and Jerling, 1992
) are reported to exist in Potentilla, sometimes within a single ovule (Rutishauser, 1945
; Håkansson, 1946
; Hunziker, 1954
; Smith, 1963a, b
—P. crantzii). Maheshwari (1950)
states that "the distinction between the two forms, known as generative and somatic apospory, is somewhat artificial, as it is frequently difficult to say whether the cell in question belongs to the archesporial tissue or to the "somatic" tissues of the nucellus," and we have emphasized this view in our attempt to disentangle the confusion that has arisen.
Facultatively agamospermous plants are capable of reproducing sexually as well as asexually. In the case of sexual reproduction, meiosis takes place in the megasporocyte. In plants having a multicellular archesporium, several megaspore tetrads are formed, and a number of megaspores may begin to enlarge (Maheshwari, 1950
). Since the present study was originally initiated to specify the potential degree of sexuality in a population of Potentilla nivea, traces of meiotic events were looked for in particular. Reports of linear tetrads with the chalazal megaspore functional in several species of Potentilla (Rutishauser, 1945
; Håkansson, 1946
; Hunziker, 1954
) have been generally accepted. In this study megaspore tetrads were located, and the spatial arrangement as well as the developmental history of all megaspores was studied.
Megagametophytes in Potentilla are classified as monosporic and eight-nucleate (Håkansson, 1946
; Smith, 1963a
). The antipodal cells, however, are according to Czapik (1961—P. crantzii) difficult to discern during the development of a megagametophyte. In this study we followed nuclear divisions from the megaspore to the mature megagametophyte stage.
MATERIALS AND METHODS
Flowering and fruiting material was collected from four individuals of Potentilla nivea L., nom. cons. prop. (Rosaceae; Eriksen, Jonsell, and Nilsson, 1999
) in their natural habitat in the Latnjajaure Valley, northern Swedish Lapland (68°22' N, 18°13' E) at 1000 m altitude. Potentilla nivea inhabits scree slopes and other open soil on calcareous substrates. It is a perennial and, in this particular population, a polyploid (2n = 6x and 2n = 7x), consisting of one to many rosettes of leaves united by a common rootstock. Inflorescences are lateral and bear 1–7 yellow flowers each. The gynoecium of Potentilla is apocarpous with a mean of 35 ± 6 ovules (mean ± 1 SD, N = 77). During a period of time from mid-June to mid-July 1997 buds, flowers, or seed heads were continuously harvested in order to enhance the likelihood of getting a complete series of stages in the development of the female gametophyte. The plant material was preserved in FPA50 (formalin: propionic acid: 50% alcohol, 5:5:90) for 24 h and subsequently stored in 70% ethanol. A total of ~1500 ovaries were studied. Each ovary was dissected, and, at least in older stages, the integument was removed from the ovule. The nucellus tissue with megagametophytes or their initials was put into Herr's clearing fluid (Herr, 1971
) on a slide and covered with a coverglass. The material was processed immediately or within a few hours. Optical sections of <1 µm in thickness were obtained by using a Molecular DynamicsTM (California, USA) confocal laser scanning microscope. The procedure is described in detail in Fredrikson, Carlsson, and Franksson (1988)
. Images were processed by a Kodak color video printer or electronically on a computer. In contrast to mechanically sectioned and stained tissue, which appears dark on a pale background in the light microscope, the optical sectioning technique provides an image showing the autofluorescence from the tissue as white on a black background. The advantage with the latter type of image is the logic equivalence between brightness and cellular activity.
RESULTS
The archesporium in Potentilla nivea consists of a few cells, i.e., is multicellular. In the initial stage, the archesporial cells are covered by a single-layered nucellus-epidermis. We did not find such an early stage of development in the material used for this study. The archesporial cells divide once mitotically and the lowermost of the resulting pair of cells, the primary sporogenous cells (Maheshwari, 1950
), enter a resting stage. The uppermost cells of all pairs, the primary parietal cells (Maheshwari, 1950
) continue to divide mitotically and constantly add to the number of cell layers between the primary sporogenous cells and the epidermis (Figs. 1, 2). The epidermal cells divide periclinally (Fig. 3), and together with the parietal cells contribute to a cap-like structure. The term "nucellar cap" is usually applied to the layers derived from the epidermis only, but since we cannot distinguish between cells derived from the parietal cells and cells derived from the epidermis in the cap, we name the entire structure the nucellar cap. The nucellus is, thus, crassinucellate.
|
Each megaspore, whether reduced or unreduced, has the potential to develop into a megagametophyte, and often all four megaspores from a meiotic event and the undivided megaspores originating from the apomictic megasporocytes will initiate megagametophyte development (Figs. 7, 8). In each ovule, typically between two and six megagametophytes will reach maturity. Most often an ovule will produce either reduced or unreduced megagametophytes, but sometimes both types seem to occur. Due to the fact that all megaspores have what appear to be equal opportunities to reach maturity, the final product, the mature ovule, has the potential to contain a mixture of megagametophytes with reduced and unreduced chromosome numbers.
|
|
|
|
|
Redefinition of tissue in the nucellus
The nucellar tissue interpreted by Rutishauser (1943, 1945
), Håkansson (1946)
, and later authors (Hunziker, 1954
; Czapik, 1961
; Smith, 1963a, b
) as the archesporium (embryo mother cells) is by us considered to be parietal cells and, together with the epidermis, responsible for the growth of the nucellar cap (Fig. 14). According to our observations the cells divide mitotically and give rise to continuous rows of cells throughout the nucellus (Fig. 4). The tissue consists of cells that in most sections appear to differ in size and shape from most other cells of the nucellus. Because of their central position, superior size, and level of activity in the early stages of ovule development, the primary parietal cells were previously interpreted as the sporogenous tissue. Rutishauser (1945)
wrote: "... of the generative cells the chalazal ones are usually longer than the others and are considered to be the true megasporocytes." As has been shown in a study on orchids (Fredrikson, Carlsson, and Franksson, 1988
), cell outlines in sectioned tissue depend to a large extent on the cutting angle relative to the direction of the cells. A cell may appear comparatively large in a section cut at a specific angle but equivalent to or even smaller than surrounding cells when cut at a less oblique angle. It is very important to translate the two-dimensional sections obtained by cutting into a three-dimensional view and not to ascribe special properties to cells on incomplete observations.
|
). In most plants megaspores appear in linear tetrads, which is also the configuration reported from Potentilla by all previous workers. According to our observations, however, we do not find linear tetrads in Potentilla but rather tetrahedral tetrads. Maheshwari (1950)
pointed out that a tetrahedral arrangement of megaspores is very rare and has been reported only as an abnormality. We would like to challenge this view and claim that tetrahedral tetrads form as the normal condition in Potentilla. The nucellar tissue interpreted by Rutishauser (1943)
, Håkansson (1946)
, Hunziker (1954)
, and later authors (Czapik, 1961
; Smith, 1963a, b
) as "chalazal" is here considered to be sporogenous and ultimately may or may not undergo meiosis, but in any case, give rise to megagametophytes. Our study shows that meioses take place during a very short period of time in the young bud. Often in our study of meiosis, the ovules examined were either too young or too old. Only by screening more than a thousand ovules were we able to find a few in which meiosis had just occurred. In Figs. 5 and 6 two tetrahedral tetrads are seen in the nucellus. In Fig. 6 the tetrad to the right has just started to enlarge. In Fig. 5 the development of the right-hand tetrad was arrested after meiosis, whereas the left-hand tetrad continued to develop and has enlarged considerably. The parietal cells have disintegrated.
In previous studies, illustrations show rows of cells (often many in line), of which four are interpreted as belonging to the tetrad. This picture is typical in quite young ovules in which the parietal cells are actively dividing and still forming the nucellar cap. Sometimes the "tetrads" are shown as one functional megaspore and three degenerating ones. This situation is often found in somewhat older ovules, either when the parietal cells begin to disintegrate or as an artifact resulting from cutting. The interpretation of the parietal cell rows as linear tetrads comes from the work of Murbeck (1901)
on Alchemilla. According to his studies Alchemilla was obligately agamospermous and the presence of tetrads puzzled him. He wrote: "despite the tetrad development the nuclear division of the megasporocyte is not connected with a reduction in chromosome number." By definition a megaspore tetrad will not form without being preceded by meiosis and therefore the "tetrads" observed by Murbeck (1901)
were most certainly parietal cell rows rather than megaspores. Unfortunately the misinterpretation was carried on to later studies on Potentilla.
By the time megasporocytes have gone through meiosis, the parietal cells begin to disintegrate and are seen as veils among the much larger megaspores. Linear tetrads were previously reported to be seen among what we interpret as parietal cells, and the meiotic events preceding the formation of megaspores were expected to be found among these cells as well. Different stages of meiosis were reported to occur in this region of the nucellus, but already at an early point there were some doubts as to whether the divisions seen were mitotic rather than meiotic. Gustafsson (1946)
citing Rutishauser (1943)
writes that "the EMCs at first show early stages of meiosis (?) ... " and "the EMCs remain in the meiosis-like stage and ultimately degenerate" (EMC is embryosac mother cell, or in modern terms megasporocyte). Another example demonstrates that meioses, in what we interpret as megasporocytes, have been observed by other authors, but had not been taken seriously. Czapik (1961)
notes: "occasionally some of the chalazal cells protruding inside the archesporium seemed to attain the premeiotic stage, prophase I and even some further stages of division (dyad; metaphase)" and she continues: "it should be added that Rutishauser (1948)
has reported the occurrence of premeiotic stages in the chalazal region of ovules of the hybrid P. canescens x argentea. This author, however, considered the resp. phenomena as due to a displacement of archesporial cells."
The autofluorescence intensity of a tissue is a measure of its nuclear activity. In photographs it is best illustrated by applying pseudocolors to the section (Fig. 15). Orange and red colors indicate a high nuclear activity. In young buds the activity lies in the cap-producing cells. When the cap is complete and meiosis is about to occur, activity moves down to the megasporocytes at the base of the ovule. Just before the flower opens, activity is seen in the nucellar cap as megagametophytes enlarge towards the micropyle.
|
and Battaglia (1963)
, but we refrain from commenting on the actual disturbance mechanism at this point. We believe it is important to look through the entire process once again in order to straighten out the concepts in light of the radically changed description of the nucellus tissue presented here.
Megagametophyte development
In most angiosperms it is the chalazal megaspore of the tetrad that gives rise to the megagametophyte, while the remaining three megaspores degenerate and disappear. Rarely all or any one of the four megaspores can enlarge and divide (Maheshwari, 1950
). Again, megagametophyte development in Potentilla is of a rare type. The general rule is that more than one and often all four megaspores will initiate megagametophyte development. In Fig. 6 the tetrad to the right has started to enlarge and in Figs. 7 and 9 the megaspores begin to send out the tubes that will develop into megagametophytes.
The function of the tubes was not understood properly at first, as they were interpreted mainly as haustoria. "A peculiar condition [megaspore haustoria] occurs in ... Potentilla (Rutishauser, 1945
) ... in which the megaspores give out lateral tubes which subsequently begin to grow upward and are in the state of competition with one another" (Maheshwari, 1950
). The enlarging reduced or unreduced cells and their tubes soon fill the entire space available as the parietal cells degenerate (Fig. 16).
|
explained the difference as follows: "In some species of Potentilla there is a multicellular archesporium without any well-marked distinction between the axial and lateral cells of the sporogenous tissue. Embryo sacs [megagametophytes] may arise either from such cells by ordinary mitotic divisions as in P. verna (generative apospory), or from the chalazal cells of the nucellus as in P. canescens, P. praecox, and P. argentea (somatic apospory)". Diplospory and apospory were reported to work within a single ovule in Potentilla (Rutishauser, 1945
; Håkansson, 1946
; Hunziker, 1954
; Smith, 1963a
). We propose that all megagametophytes originate from sporogenous chalazal cells, either by or without meiotic divisions. The high and low positions of enlarging megagametophytes in the nucellus are determined by the initial position of the reduced or unreduced megaspores rather than by "somatic" or sporogenous origin. Unreduced megagametophytes are usually confined to the chalazal part of the nucellus because of the fact that no reduction division has taken place. Cytologically reduced megaspores are found in large tetrads, which fill up the entire space in the center of the ovule, and the megagametophytes may originate from any megaspore positioned from the base of the nucellus to the lower edge of the nucellar cap. The reduction level of the megagametophyte cannot be deduced from the position of the megagametophyte alone. The important issue is whether or not if the terms diplospory and apospory are to be considered for cytologically unreduced megagametophytes in Potentilla nivea, the correct term for the situation is diplospory. The diplosporous megagametophyte arise from megasporocytes that fail to undergo meiosis and, thus, originate from potentially sporogenous tissue. Apospory does not exist at all in Potentilla nivea according to our observations.
The number of megaspore tetrads or unreduced megaspores in each ovule is commonly 1–3, which potentially may result in 1–12 megagametophytes. The number of mature megagametophytes found in each ovule is typically 2–6, indicating that some of the megagametophytes become arrested in their development at a comparatively early stage, as also noted by Maheshwari (1950)
. In Fig. 5, the nucleus of each megaspore in the tetrad on the right is located in the inactive megaspore, whereas all megaspores in the tetrad to the left have enlarged, and the nuclei have moved out into the growing tubes. Figure 17 shows an entire ovule with a total of six megagametophytes. Two of the megagametophytes are seen in their full length.
|
; Smith, 1963a
). The monosporic nature of the megagametophyte is confirmed, but we were unable to verify that the mature megagametophyte is eight-nucleate. Rather it is five-nucleate and cannot easily be assigned to the megagametophyte typology proposed by Johri (1963)
. Megagametophyte development approaches the Polygonum-type (Maheshwari, 1950
; Johri, 1963
) but differs by the lack of nuclear divisions leading to the formation of antipodal cells. A first division of the megasporocyte nucleus into two nuclei, of which the micropylar nucleus undergoes the second division (Fig. 18), initiates the megagametophyte development. The third division is simultaneous and involves the two micropylar nuclei resulting from the second division. By this mode of division no antipodal cells are formed. This might explain why Czapik (1961)
found antipodals difficult to discern during the development of a megagametophyte and Smith (1963a)
claims that the three antipodal cells show signs of degeneration at an early stage and that they were never observed in the mature megagametophyte.
|
has shown that, when sexual, the egg cell and central nucleus in one of the embryo sacs (megagametophytes) are fertilized. In diplosporous megagametophytes the egg cell is expected to remain unfertilized. The pseudogamous endosperm formation takes place by free nuclear division.
FOOTNOTES
1 The authors thank Dr. Ulf Molau, Prof. Gunnar Harling, Göteborg University, Prof. Inger Nordal, Oslo University, and Prof. David Murray, University of Alaska, Fairbanks, for reading and commenting on earlier versions of this paper; and Prof. Emeritus J. M. Herr, Jr. and an anonymous reviewer for very helpful comments. Financial support was received from The Swedish Natural Science Research Council (NFR B-AA/BU 10480–308, Eriksen) and Enanderska Fonden of the Royal Academy of Sciences for which we are grateful. 
LITERATURE CITED
Asker, S., and L. Jerling. 1992 Apomixis in plants. CRC Press, Boca Raton, Florida, USA.
Battaglia, E. 1963 Apomixis. In P. Maheshwari [ed.], Recent advances in the embryology of angiosperms, Chapter 8. Catholic Press, Ranchi, India.
Böös, G. 1917 Parthenogenesis in der Gruppe Aphanes der Gattung Alchemilla. Lunds Universitets Årsskrift, Naturvetenskapliga Fakulteten, avdelning 2, 13: 1–37.
Czapik, R. 1961 Embryological studies in the genus Potentilla L. P. crantzii. Acta Biologica Cracoviensia, Series Botanica 4: 97–119.
Eriksen, B., B. Jonsell, and Ö. Nilsson. 1999 (1394) Proposal to conserve the species name Potentilla nivea L. (Rosaceae) with a conserved type. Taxon 48: 165–166.[CrossRef]
Fredrikson, M., K. Carlsson, and O. Franksson. 1988 Confocal scanning laser microscopy, a new technique used in an embryological study of Dactylorhiza maculata (Orchidaceae). Nordic Journal of Botany 8: 369–374.[ISI]
Gustafsson, Å. 1946 Apomixis in higher plants. Part. I. The mechanism of apomixis. Lunds Universitets Årsskrift. Naturvetenskapliga Fakulteten, avdelning 2, 42: 1–67.
Håkansson, A. 1946 Untersuchungen über die Embryologie einiger Potentilla-formen. Lunds Universitets Årsskrift. Naturvetenskapliga Fakulteten, avdelning 2, 42: 1–70.
Herr, J. M. Jr., 1971 A new clearing squash technique for the study of ovule development in angiosperms. American Journal of Botany 58: 785–790.[CrossRef][ISI]
Hunziker, R. 1954 Beitrag zur Aposporie und ihrer Genetik bei Potentilla. Archiv der Julius Klaus-Stiftung 29: 135–222.
Johri, B. M. 1963 Female gametophyte. In P. Maheshwari [ed.], Recent advances in the embryology of angiosperms, Chapter 4. Catholic Press, Ranchi, India.
Maheshwari, P. 1950 An introduction to the embryology of angiosperms. McGraw-Hill, New York, New York, USA.
Müntzing, A. 1928 Pseudogamie in der Gattung Potentilla. Hereditas 11: 267–283.
———. 1931 Note on the cytology of some apomictic Potentilla-species. Hereditas 15: 166–178.[ISI]
Murbeck, S. 1901 Parthenogenetische Embyobildung in der Gattung Alchemilla. Lunds Universitets Årsskrift, avd. 2, 38: 1–41.
Nogler, G. A. 1984 Gametophytic apomixis. In B. M. Johri [ed.], Embryology of angiosperms, 475–518. Springer-Verlag, Berlin.
Rutishauser, A. 1943 Untersuchungen über die Fortpflanzung und Bastardbildung apomiktischer Potentillen. Berichte der Schweizerischen Botanischen Gesellschaft 53: 1–83.
———. 1945 Zur Embryologie amphimiktischer Potentillen. Berichte der Schweizerischen Botanischen Gesellschaft 55: 19–32.
———. 1948 Pseudogamie und Polymorphie in der Gattung Potentilla. Archiv der Julius Klaus-Stiftung 23: 247–424.
Smith, G. L. 1963a Studies in Potentilla L. I. Embryological investigations into the mechanism of agamospermy in British P. tabernaemontani Aschers. New Phytologist 62: 264–282.[CrossRef]
———. 1963b Studies in Potentilla L. II. Cytological aspects of apomixis in P. crantzii (Cr.) Beck ex Fritsch. New Phytologist 62: 283–300.[CrossRef]
Strasburger, E. 1904 Die Apogamie der Eualchimillen. Jahrbüchern für wissenschaftliche Botanik 41: 88–164.
This article has been cited by other articles:
|
|
 |
|
  K. T. Hansen, R. Elven, and C. Brochmann Molecules and morphology in concert: tests of some hypotheses in arctic Potentilla (Rosaceae) Am. J. Botany, October 1, 2000; 87(10): 1466 - 1479. [Abstract] [Full Text]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
