| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Reproductive Biology |
Department of Ecology and Evolutionary Biology, University of Toronto, Toronto, Ontario, Canada M5S 3B2; Laboratory of Systematic and Evolutionary Botany, Institute of Botany, Chinese Academy of Sciences, Xiangshan, Beijing, People's Republic of China 100093
Received for publication November 20, 2006. Accepted for publication May 24, 2007.
ABSTRACT
Spatial features of pollen tube growth and the composition of the extracellular matrix (ECM) of transmitting tissue in carpels of Kadsura longipedunculata, a member of the basal angiosperm taxon Schisandraceae, were characterized to identify features of transmitting tissue that might have been important for pollen–carpel interactions during the early history of angiosperms. In addition to growing extracellularly along epidermal cells that make up stigmatic crests of individual carpels, pollen tubes grow on abaxial carpel epidermal cells between unfused carpels along an extragynoecial compitum to subsequently enter an adjacent carpel, a feature important for enhancing seed set in apocarpous species. Histo- and immunochemical data indicated that transmitting tissue ECM is not freely flowing as previously hypothesized. Rather, the ECM is similar to that of a dry-type stigma whereby a cuticular boundary with associated esterase activity confines a matrix containing methyl-esterified homogalacturonans. The Schisandraceae joins an increasing number of basal angiosperm taxa that have a transmitting tissue ECM similar to a dry-type stigma, thereby challenging traditional views that the ancestral pollen tube pathway was similar to a wet-type stigma covered with a freely flowing exudate. Dry-type stigmas are posited to provide tighter control over pollen capture, retention, and germination than wet-type stigmas.
Key Words: dry-type stigma • extracellular matrix • extragynoecial compitum • Schisandraceae • transmitting tissue
The evolution of the carpel and carpel closure resulted in the elimination of the ovule as the site of pollen capture. As a result, the need arose for a specialized carpel tissue that functioned to maintain the evolutionary trend that originated with the ovule habit to provide improved capture, retention, and germination of pollen and the transmission of the male gametophyte to the female gametophyte (Thomas, 1934
; Bailey and Swamy, 1951
; Lloyd and Wells, 1992
; Endress, 1994
; Sage et al., 1994
). This tissue in angiosperms, termed transmitting tissue, is most commonly composed of a stigma, style, the adaxial carpel epidermis within the ovary locule, and frequently the funiculus, which can be modified into an obturator (Vasil and Johri, 1964
; Tilton and Horner, 1980
). In comparison to other seed plants, the angiosperm transmitting tissue of the carpel has a longer pathway for pollen tube growth with a greater diversity of sites for prezygotic interactions between the male gametophyte and female sporophytic tissues. These interactions are important for male gametophytic competition and selection (Mulcahy, 1975
; Willson and Burley, 1983
) and recognition of self through mechanisms of self-incompatibility (deNettancourt, 1977
, 1997
, 2001
).
The origin and ancestral features of angiosperm transmitting tissue have been long-standing topics of interest. Using the well-resolved picture of basal angiosperm relationships, comparative studies on reproductive traits of basal groups have led to the conclusion that pollen tube growth in the archetype carpel occurred on transmitting tissue comprising adaxial epidermal secretory cells that are positioned in extant species, in part, at the site of carpel closure (Endress and Igersheim, 2000
; Endress, 2001
). This idea is similar to earlier views that the first stigma consisted of glandular hairs diffusely situated on the adaxial surface of early carpellary structures (Thomas, 1934
; Bailey and Swamy, 1951
; Lloyd and Wells, 1992
). Lloyd and Wells (1992)
suggested that the diffuse stigma originated when the site of secretion of the pollination droplet shifted from the ovule to the adaxial "carpel" surface. This ancestral adaxial secretory transmitting tissue has been compared to a wet- vs. dry-type stigma (Lloyd and Wells, 1992
; Endress and Igersheim, 2000
). The extracellular matrix (ECM) of both wet- and dry-type stigmas are characterized by the presence of a protein layer with esterase activity (pellicle; Heslop-Harrison and Heslop-Harrison, 1970
; Mattsson et al., 1974
) that is associated with a cuticle (Heslop-Harrison and Shivanna, 1977
). However, in a dry-type stigma, the pellicle and cuticle remain intact in the functional phase, whereas they become discontinuous in the wet-type thereby resulting in a freely flowing secretion (Heslop-Harrison and Shivanna, 1977
; Shivanna and Sastri, 1981
). The dry-type stigma has been hypothesized to have evolved later in angiosperm history and is viewed as providing tighter control over pollen capture, adhesion, recognition, hydration, and germination than the wet-type because of the presence of the cuticle and associated pellicle (Dickinson, 1995
). The functional roles of some ECM components important for pollen capture, retention, recognition, hydration, and germination have been hypothesized to have been transferred from a wet-type stigma ECM to the pollen ECM during the transition from a wet- to dry-type stigma (Dickinson, 1995
). Fossil evidence led Friis and Crept (1987)
to the currently held view that the style appeared later in the history of angiosperms.
The Schisandraceae consists of two genera of scandent and twining woody vines, Kadsura (16 species; Saunders, 1998
) and Schisandra (23 species; Saunders, 2000
). The family is widely distributed throughout east and southeastern Asia and in a more limited region in southeastern USA and Mexico (Smith, 1947
; Saunders, 1998
, 2000
). The Schisandraceae, which has a fossil record extending back to the Early Cretaceous (Chmura, 1973
; Guo, 1984
; Saunders, 1998
), is a member of the paraphyletic group of eight families situated at the base of the angiosperm phylogeny (Mathews and Donoghue, 1999
; Parkinson et al., 1999
; Qiu et al., 1999
, 2001
; Barkman et al., 2000
; Graham and Olmstead, 2000
; Soltis et al., 2000
; Zanis et al., 2002
; Angiosperm Phylogeny Group, 2003
; Saarela et al., 2007
). The taxon figured prominently in the formation of early hypotheses regarding the evolution of angiosperm reproductive structures because of the occurrence of primitive floral features in the unisexual flowers with spirally arranged organs (Tucker and Bourland, 1994
; Igersheim and Endress, 1997
; Saunders, 1998
).
The gynoecium of Schisandraceae has many features in common with other members of angiosperms currently situated at the base of the phylogeny. The gynoecium is apocarpous and the ascidate carpels lack a style. Unicellular, epidermal cells line the margins of individual carpels at the site of carpel closure to form a stigmatic crest that is continuous with similar epidermal cells of the ovary locule and funiculus (Igersheim and Endress, 1997
). A portion of the stigmatic crest protrudes above the tightly packed carpels (Igersheim and Endress, 1997
; Saunders, 1998
, 2000
) where it is exposed for reception of pollen grains. The remaining region of the stigmatic crest is hidden between tightly packed adjacent carpels (Igersheim and Endress, 1997
; Saunders, 1998
, 2000
) and thus isolated from pollen grain capture. This hidden portion of the stigmatic crest has been referred to as a pseudostyle (Smith, 1947
; Leinfellner, 1966
; Saunders, 1998
; Endress and Igersheim, 2000
; Endress, 2001
). A copious ECM extending along the stigmatic crest and ovary locule of individual carpels appears to be continuous with an ECM observed between the adjacent unfused carpels in the Schisandraceae (Igersheim and Endress, 1997
; Endress and Igersheim, 2000
). The gynoecial ECM associated with the stigmatic crest is viewed as freely flowing (wet-type stigma) and has been proposed to function in pollen tube transmission and competition between male gametophytes (Igersheim and Endress, 1997
; Saunders, 1998
; Endress and Igersheim, 2000
). The freely flowing exudates between the adjacent carpels have been posited to transmit pollen tubes from one carpel to another, thereby operating as a cryptic extragynoecial compitum (Endress and Igersheim, 2000
). In syncarpous angiosperms, a compitum is a region of shared pollen tube transmitting tissue allowing pollen tubes full access to all ovules within a compound ovary (Carr and Carr, 1961
; Armbruster et al., 2002
). Although structural studies have provided insight into the distribution of gynoecial exudates, data confirming the spatial relationships of pollen tube transmission within and between carpels with respect to the ECM distribution are absent for the Schisandraceae as are data confirming the histo- and immunochemical nature of the ECM. It is not apparent whether the ECM is similar to that of a wet- vs. dry-type stigma and whether any of the molecules that have been deemed important for nutrition and guidance of compatible pollen tube growth in the angiosperm transmitting tissue ECM including methyl-esterified homogalacturonans (Lenartowska et al., 2001
, Khosravi et al., 2003
), lipids (Lush et al., 1998
; Zinkl et al., 1999
), arabinogalactan-proteins (AGPs, Wu et al., 2000
), and other proteins (Kim et al., 2003
; Marton et al., 2005
) are present in the transmitting tissue ECM of species in the Schisandraceae.
In this study, the spatial features of pollen tube growth, and hence transmitting tissue distribution, as well as the histo- and immunochemical composition of the transmitting tissue ECM of a monoecious species within the Schisandraceae, Kadsura longipedunculata Finet and Gagnep, were assessed. These studies were undertaken to provide comparative information to assist in characterizing spatial features of transmitting tissue and chemical makeup of transmitting tissue ECM that might have been present and important for pollen–carpel interactions during the early history of angiosperms. Three questions were addressed: (1) Does a freely flowing ECM associated with epidermal cells extending along the stigmatic crest form a continuous pathway for pollen tube growth within a carpel? (2) Is an extragynoecial compitum present in association with a freely flowing ECM? (3) And are any molecules that are regarded as important for pollen tube growth in angiosperms (methyl-esterified homogalacturonans, lipids, and AGPs) present in the transmitting tissue ECM?
MATERIALS AND METHODS
Plant material
Plants from three populations of K. longipedunculata used in this study were located in Langshan National Geological Park, Province of Hunan in southern China (26.3866° N, 110.7751° E). At this site, K. longipedunculata grows as an understory woody vine. The flowers are monoecious with staminate and carpellate flowers in different inflorescences (Fig. 1A–C). Carpels are green with white stigmas (Fig. 1A), and stamens are whitish cream (Fig. 1B, C). Tepals in both staminate and carpellate flowers tend to be green, and the androecium of staminate flowers is red (Fig. 1B) or yellow (Fig. 1C). Cross- and self-pollinations were conducted in the natural populations on flowers at the time of floral opening by tapping dehiscent anthers onto receptive stigmas. Pollination by wind or insects was avoided by enclosing floral buds in pollination bags constructed from bridal veil prior to anthesis as described by Bernhardt et al. (2003)
. Bags were replaced on flowers following pollination until harvesting for fluorescence microscopy as described later.
|
. Pollinated gynoecia that were viewed with aniline blue fluorescence microscopy were subsequently prepared for scanning electron microscopy (SEM) as described by Sage and Williams (1995)
. In addition, pollinated flowers were fixed in acetic acid : 100% ethanol (1 : 3) and prepared for SEM.
Transmitting tissue ECM composition and structure
Histochemical features of unpollinated gynoecial ECM were characterized on whole mounts and freehand sections of fresh and fixed carpels (N = 3–5 flowers/plant/population/stain) at stigma receptivity using the following stains: (a) nile red in 75% glycerol for neutral lipids (Greenspan et al., 1985
); (b) 0.01% auramine-O in 0.05 M TRIS/HCl buffer at pH 7.2 observed under UV light for lipids and cutin (Heslop-Harrison and Shivanna, 1977
); (c) 0.001% 8-anilinonapthalene-1-sulfonic acid (8-ANS) observed under UV light to localize proteins (Mattsson et al., 1974
; Fulcher and Wong, 1980
); and (d) indoxyl acetate in 0.1 M Tris/HCl buffer at pH 7.0, in 0.1 M potassium ferrocyanide and potassium ferricyanide for detection of nonspecific esterase activity of the stigma pellicle (Dejong et al., 1967
). Controls were conducted by omitting the stain or omitting the substrate for the detection of nonspecific esterase activity.
The presence or absence and spatial distribution of homogalacturonans and AGPs were detected by using monoclonal antibodies that recognize epitopes of high methyl-esterified homogalacuronans (JIM7), low methyl-esterified homogalacturonans (JIM5), and AGPs (JIM13). Unpollinated flowers in the female phase of ontogeny at the time of stigma receptivity were cryofixed, freeze-substituted, and embedded as described by Lam et al. (2001)
. Serial sections (70 nm thick) of 3–6 carpels per plant were collected on formvar-coated grids and incubated with JIM7, JIM5, or JIM13 (PlantProbes, Leeds, UK). Primary and secondary (anti-rat IgG-gold conjugate 18 nm for JIMs; Jackson Immunoresearch, West Grove, Pennsylvania, USA) antibody dilutions were 1 : 50 and 1 : 20, respectively. Incubation times in 1° and 2° antibodies were 2 and 1 h, respectively. Controls were run by omitting 1° antibody. To determine the density of JIM5, 7, and 13 epitopes, the number of gold particles per square micrometer were quantified using Image-Pro Plus (Media Cybernetiks, Silver Springs, Maryland, USA). Epitope density per carpel was contrasted using one-way ANOVA (Sigma Stat 2.03). No differences were observed between carpels, and all values per epitope per carpel were pooled. Pairwise comparisons between epitope densities at different regions of the transmitting tissue were made with a Mann–Whitney rank sum test. Images were captured on the Phillips 201 TEM equipped with an Advantage HR Camera System (Advanced Microscopy Techniques Corp., Danvers, Massachusetts, USA).
RESULTS
Spatial features of the pollen tube pathway
Tubes from germinating cross- and self-pollen grew extracellularly downward from the terminal stigma toward and along the pseudostyle solely upon the surface of papillate epidermal cells that make up the stigmatic crest (Figs. 2A–C, 3A–C). After extending approximately three-quarters the length of the pseudostyle, pollen tubes followed one of two routes. Tubes turned abruptly and grew either toward the pseudostyle of a neighboring carpel on abaxial epidermal cells of the adjacent carpels (Figs. 2C and 3C) or into the locule of the carpel that originally received the pollen grains (Fig. 2C, D; Fig. 3C) to subsequently grow along papillate epidermal cells of an obturator (Fig. 3E, F). Pollen tubes grew toward the pseudostyle of a neighboring carpel in a region of the carpel where the cuticle of the abaxial epidermal cells changed from rugose to smooth (Fig. 3D). Tubes that traversed the carpel epidermal cells with the smooth cuticle subsequently entered the ovary locule and ovule (Fig. 2E) of an adjacent carpel only. Tubes were not observed extending beyond an adjacent carpel. Pollen tubes had penetrated the ovule micropyle by 24 h postpollination.
|
|
|
|
|
|
The present investigation on the pollen tube pathway in the monoecious species K. longipedunculata provides novel data on spatial features of pollen tube growth as well as the fine structure and histo- and immunochemical makeup of the transmitting tissue ECM. We demonstrated intercarpellary growth of pollen tubes on the surface of abaxial epidermal cells of adjacent, appressed carpels, thereby confirming the hypothesis that an extragynoecial compitum is present in the Schisandraceae (Igersheim and Endress, 1997
; Endress and Igersheim, 2000
). However, ultrastructural and histochemical data showed that at stigma receptivity, the ECM associated with pollen tube growth is not freely flowing. Rather, the outermost region of the entire transmitting tissue ECM within a carpel at the time of stigma receptivity is composed of a continuous cuticle with associated esterase activity. This cuticle confines the transmitting tissue ECM present in a carpel solely to that carpel. Similarly, no evidence was found for any freely flowing secretion associated with epidermal cells at the site of intercarpellary growth of pollen tubes. The transmitting tissue ECM within a carpel contained molecules such as high and low methyl-esterified homogalacturonans and lipids that are regarded as important for pollen tube growth. The ECM composition differed noticeably among the stigma, pseudostyle and region of intercarpellary growth at floral receptivity. Most notably, nile red-positive droplike aggregates indicating the presence of neutral lipids were found only at the stigma. The presence of lipids at the stigma and not the pseudostyle or extragynoecial compitum is noteworthy given that stigmatic lipids are known to play an important role in pollen hydration, adhesion, and subsequent pollen tube growth (Lush et al., 1998
; Zinkl et al., 1999
; Wolters-Arts et al., 2002
).
We will address the results of this study first by placing the structural and histo/immunochemical data indicating the presence of a cuticle at the outermost boundary of the transmitting tissue ECM in the context of similar information available for other angiosperms and second by addressing the significance of these results for pollen-carpel interactions in K. longipedunculata. The presence of an extragynoecial compitum has important implications for increasing seed production in K. longipedunculata. Therefore, these spatial aspects of pollen tube growth are also compared with similar features in other species where the extragynoecial compitum is known to have a positive impact on reproduction. Pollen tube growth from the pseudostyle to the extragynoecial compitum results from an abrupt change in direction of tube growth. Hence, these observations are discussed in the context of the role of the transmitting tissue ECM in control of pollen tube growth. Finally, because the Schisandraceae has many features that may have been common to ancestral angiosperms, and the taxon is a member of the grade of eight families situated at the base of the angiosperm phylogeny, we conclude by placing the results from the present study in the context of hypotheses regarding the early evolution of angiosperm transmitting tissue and pollen–carpel interactions therein.
Structural and functional significance of transmitting tissue ECM within carpels of K. longipedunculata
The present study provides evidence that the transmitting tissue ECM within a carpel that extends along the stigma and pseudostyle (stigmatic crest) of K. longipedunculata is structurally and histo- and immunochemically similar to the ECM of a dry- vs. wet-type stigma in angiosperms (Heslop-Harrison and Shivanna, 1977
). The most notable similarity between a dry-type stigma of other angiosperms and the transmitting tissue ECM within a carpel of K. longipedunculata is the presence of an intact outermost region that is auramine-O- and 8-ANS-positive with associated esterase activity. The auramine-O-, 8-ANS-, and esterase-positive layers are indicative of an intact cuticle and associated protein pellicle present in dry- but not wet-type stigmas at receptivity (Heslop-Harrison and Heslop-Harrison, 1970
; Mattsson et al., 1974
; Heslop-Harrison and Shivanna, 1977
). At the ultrastructural level, the thin outermost layer of the ECM of stigmatic epidermal cells of K. longipedunculata corresponds to a nonlamellate cuticle proper present on the epidermal cells of dry-type stigmas and leaves of many species (Heslop-Harrison and Shivanna, 1977
; Jeffree, 1996
). A striking feature of the dry-type transmitting tissue ECM found in the pseudostyle but not the stigma of K. longipedunculata is the presence of a thick globular reticulate layer below the cuticle proper that extends up to the primary wall and contains high and low methyl-esterified homogalacturonans. These regions of the ECM correspond to a thick cuticular layer composed of cutin and polysaccharides in models of the plant cuticle (Jeffree, 1996
) and also occur in the dry-type stigmas of monocots and eudicots (Heslop-Harrison and Heslop-Harrison, 1982
) as well as leaves of other seed plants (Jeffree, 1996
).
Results indicating that the ECM of the stigma and pseudostyle of K. longipedunculata are structurally similar to those of a dry-type stigma in other angiosperms are surprising given that the ECM of pollen tube transmitting tissue of the Schisandraceae has been considered to be freely flowing and thus similar to that of a wet-type stigma (Saunders, 1998
, 2000
; Endress and Igersheim, 2000
). The main differences in conclusions regarding the nature of the ECM of K. longipedunculata between the present study and others likely arise from dissimilarities between tissue fixation techniques. Histochemical observations on fresh flowers combined with cryofixation/freeze-substitution techniques that were utilized in the present study provide optimal preservation and retention of the transmitting tissue ECM (Koehl, 2002
; Thien et al., 2003
; Hristova et al., 2005
; Hristova-Sarkovski, 2006
). In contrast, fixation techniques that utilize ethanol or formalin:acetic acid:alcohol (Endress and Igersheim, 2000
; SEM data from present study) remove some ECM components and disperse others, giving the appearance of a freely flowing ECM. Optimal preservation and retention of the transmitting tissue ECM is essential for identification of ECM components, their spatial relationships to one another and hence, interpretation of the functional significance of each component for pollen tube growth in K. longipedunculata as well as other angiosperms. Documentation of a cuticle and pellicle at the site of pollen capture and retention in K. longipedunculata is noteworthy because the cuticle and pellicle of the dry-type stigma of angiosperms is posited to exert a higher degree of control over pollen adhesion, recognition, and hydration than a wet-type stigma (Dickinson, 1995
; Zinkl et al., 1999
).
Extragynoecial compitum of K. longipedunculata
The Schisandraceae represents the third angiosperm taxon with apocarpy whereby pollen tubes exit a carpel and grow from one ovary to subsequently enter a second. An extragynoecial compitum has also been demonstrated in the Illiciaceae (Williams et al., 1993
) and Alismataceae (Wang et al., 2002
, 2006
). Intercarpellary growth of pollen tubes in apocarpous species has been posited to provide a site of sexual competition between male gametophytes as well as a mechanism of increasing efficiency of seed set (Williams et al., 1993
) in a fashion similar to what has been proposed and observed for a compitum in angiosperms with syncarpy (Carr and Carr, 1961
; Endress, 1982
; Armbruster et al., 2002
). Enhanced seed set via pollen tube growth in the extragynoecial compitum occurs because of the ability of pollen tubes from a pollinated carpel to enter the ovary of an unpollinated carpel (Williams et al., 1993
; Wang et al., 2006
). Interesting differences are apparent between patterns of intercarpellary growth of pollen tubes in K. longipedunculata and Illicium floridanum. Pollen tubes growing in the extragynoecial compitum of I. floridanum enter ovaries situated one or more than one carpel away from the original pollinated carpel (Williams et al., 1993
). In contrast, observations from the present study indicate that pollen tubes of K. longipedunculata only enter the ovary of an adjacent carpel. Hence, the extragynoecial compitum of K. longipedunculata may be more limited in the extent to which it may function to increase seed set in comparison to I. floridanum. These differences in pollen tube growth within the extragynoecial compitum between the two species likely result from dissimilar spatial arrays of carpels and transmitting tissue within flowers of each species. All carpels in I. floridanum appear in a series around a centrally positioned apical residuum (Robertson and Tucker, 1979
). The epidermis of the apical residuum as opposed to the abaxial carpel epidermis functions as a conduit for pollen tubes to grow around, unimpeded en route from ovary-to-ovary-to-ovary. Notably, although an apical residuum is present in the Schisandraceae (Tucker and Bourland, 1994
), only a very limited number of carpels are situated adjacent to the apical residuum because carpels are spirally arranged on an elongated axis. Interestingly, neither an apical residuum nor the abaxial epidermis of carpels arranged in whorls on a receptacle (Igersheim et al., 2001
) appears to be involved in pollen tube growth in Alismataceae species (Ranalisma rostratum, Sagittaria guyanensis, and S. potamogetifolia; Wang et al., 2002
, 2006
). The extragynoecial compitum in the Alismataceae is apparently composed of subepidermal layers of the receptacle. And, while pollen tube growth in the receptacle functions to increase seed set in the apocarpous condition (Wang et al., 2006
), it is not clear whether pollen tubes can grow to the ovary of carpels situated more than one carpel away as observed in I. floridanum. More detailed studies are needed to confirm the presence or absence of an extragynoecial compitum in other species of the Schisandraceae and the impact of the extragynoecial compitum on seed set in K. longipedunculata and other taxa where an extragynoecial compitum has been demonstrated or is suspected to occur.
The directionality of pollen tube growth in seed plants is a well-known phenomenon, and identification of the components of the transmitting tissue ECM that control pollen tube growth has been the subject of intense research (Wu et al., 2000
; Lord and Russell, 2002
; Khosravi et al., 2003
; Kim et al., 2003
). Components likely important for the control of pollen tube growth, including methyl-esterified homogalacturonans (Lenartowska et al., 2001
; Khosravi et al., 2003
), lipids, AGPs (Wu et al., 2000
), and select proteins ( Kim et al., 2003
; Marton et al., 2005
), may be present in transmitting tissue ECM before pollination, or their presence may be induced by pollination (Sedgley and Scholefield, 1980
; Arbeloa and Herrero, 1987
; Pontieri and Sage, 1999
; Lenartowska et al., 2001
; Koehl, 2002
).
The present study on K. longipedunculata provides novel information on the histo- and immunochemical and ultrastructural features of an external compitum where pollen tubes will grow in an angiosperm. Notably, pollen tubes track the surface of smooth, thick-cuticled epidermal cells along the external compitum and do not penetrate the thick cuticular layer as appears to occur when tubes grow down the pseudostyle. Significantly, discontinuous regions of low methyl-esterified homogalacturonans are situated on the surface of the cuticle of epidermal cells along the external compitum in K. longipedunculata. Low methyl-esterified homogalacturonans strongly bind Ca2+, and it has been posited that low methyl-esterified homogalacturonans within the transmitting tissue may act as a Ca2+ reservoir to be released after pollination-induced degradation of the homogalacturonans. Calcium, which is essential for controlling directionality of pollen tube growth (Taylor and Hepler, 1997
), is then available. Low methyl-esterified homogalacturonans may also act in combination with a cysteine-rich protein with sequence similarity to plant lipid transfer proteins for proper adhesion of pollen tubes to the cell wall of transmitting tissue cells (Mollet et al., 2000
; Lord, 2003
). It is noteworthy that pollen tubes adhere tightly to epidermal cells of the extragynoecial compitum of K. longipedunculata. The role of low methyl-esterified homogalacturonans in growth of pollen tubes along the extragynoecial compitum in K. longipedunculata remains to be determined, particularly because similar deposits of low methyl-esterified homogalacturonans are located on all other carpellary epidermal cells in the species. Because the cuticle morphology is so distinct along the extragynoecial compitum, it also remains to be clarified whether the smooth vs. rugose cuticle is functionally significant for the control of pollen tube growth. Finally, although this study did not note the presence of any pollination-induced ECM components along the extragynoecial compitum, their presence cannot be discounted in the absence of observations on cryofixed/freeze-substituted pollinated carpels. The accessibility of the extragynoecial ECM in K. longipedunculata should enhance the ease with which such data can be obtained.
Conclusion
What does the structure and histo/immunochemical composition of extant basal angiosperm stigmas allow us to infer about the ancestral stigma?
The Schisandraceae is the sixth basal angiosperm taxon hypothesized to have species with a wet-type stigma that has been subsequently demonstrated to have a dry-type stigma with high and low methyl-esterified homogalacturonans and a limited accumulation of lipids contained under the cuticle proper at stigma receptivity. Stigmas with similar characteristics have also been reported for the Saururaceae (Pontieri and Sage, 1999
), Illiciaceae (Koehl, 2002
), Trimeniaceae (Bernhardt et al., 2003
), Amborellaceae (Thien et al., 2003
; Hristova et al., 2005
; Hristova-Sarkovski, 2006
), and Chloranthaceae (Hristova et al., 2005
). Taken as a whole, results from the present study and others on the nature of the stigma in extant basal angiosperms call into question previous views that regard the ancestral stigma as a site of freely flowing secretions with limited control of pollen capture, adhesion, hydration, and germination (Pontieri and Sage, 1999
; Bernhardt et al., 2003
; Thien et al., 2003
; Hristova et al., 2005
). Rather, the results provide the intriguing possibility that the dry-type stigma, which tightly regulates pollen capture, adhesion, hydration, and germination, was the ancestral state.
Was transmitting tissue, and thus the sole site of male gametophyte growth prior to ovule entry, confined to the adaxial epidermal carpel cells of the ancestral carpel?
Collectively, results from the present study on spatial patterns of pollen tube growth within a carpel in K. longipedunculata and similar studies on other species within taxa rooted at the base of the angiosperm phylogeny indicate that transmitting tissue is not always confined to the adaxial carpel epidermis. Notably, the Schisandraceae is only the second basal angiosperm taxon in which pollen tube growth in a species is confined to the epidermis of the carpel en route to the ovule following stigmatic germination. Pollen tubes also track the epidermis of the carpel in I. floridanum (Illiciaceae; Williams et al., 1993
; Koehl, 2002
; Koehl et al., 2004
). In contrast, pollen tubes from other basal angiosperms grow intercellularly following germination within ground tissue that is differentiated into transmitting tissue (Nymphaeaceae: Orban and Bouharmount, 1995
; Winteraceae: Sage et al., 1998
; Saururaceae: Pontieri and Sage, 1999
; Sage and Sampson, 2003
; Amborellaceae: Thien et al., 2003
; Trimeniaceae: Bernhardt et al., 2003
; Chloranthaceae: Hristova et al., 2005
). It has been proposed that carpel closure during the early history of angiosperms would have protected pollen tubes growing on the adaxial carpel surface from foraging insects (Lloyd and Wells, 1992
). Alternatively, intercellular growth of pollen tubes following germination would have protected tubes prior to carpel closure (Hristova et al., 2005
). Intercellular growth of pollen tubes within ground tissue that is differentiated into transmitting tissue likely requires a different suite of physiological parameters vs. those needed to grow extracellularly along the epidermal surface (Bell, 1995
; Dickinson, 1995
).
Although pollen tubes grow along the epidermis of carpel cells in K. longipedunculata and I. floridanum, each species has a unique spatial pattern of transmitting tissue distribution and hence pollen tube growth. The present study demonstrates that prior to growth along the extragynoecial compitum, pollen tubes of K. longipedunculata grow solely on the epidermal cells of the stigmatic crest. In contrast, pollen tubes of I. floridanum plunge down following germination on the epidermal cells of the stigmatic crest to grow between the epidermal cells of the appressed adaxial carpel surfaces (Williams et al., 1993
; Koehl, 2002
; Koehl et al., 2004
). The spatial differences in pollen tube growth between K. longipedunculata and I. floridanum provide alternative modes of protection for pollen tubes from abiotic factors and may be associated with differences in the arrangement of carpels within a flower. Pollen tubes growing on epidermal cells along the stigmatic crest in K. longipedunculata are protected because the majority of the stigmatic crest, the pseudostyle, is shielded by adjacent carpels (Smith, 1947
; Leinfellner, 1966
; Igersheim and Endress, 1997
; Saunders, 1998
; Endress and Igersheim, 2000
). In contrast, the stigmatic crest where the pollen germinates is entirely exposed to abiotic factors in I. floridanum.
An extragynoecial compitum has been proposed to be a prominent feature of the basal angiosperms with more than one free carpel per flower (Endress and Igersheim, 2000
). To date, the Schisandraceae and Illiciaceae are the only two basal angiosperm taxa with a species in which such a feature has been demonstrated to be present. This shared trait of pollen tube growth between K. longipedunculata and I. floridanum is intriguing because the Schisandraceae and Illiciaceae are sister taxa within a clade that also includes Austrobaileyaceae (Angiosperm Phylogeny Group, 2003
). Although extragynoecial exudates have been reported to occur between carpels in Austrobaileya scandens (Endress and Igersheim, 2000
), spatial patterns of pollen tube growth in the species have not yet been reported. Confirmation of spatial patterns of pollen tube growth in other basal angiosperms and the composition of the transmitting tissue ECM will be essential to provide a complete data set for a reconstruction of the evolution of the pollen tube pathway and hence interactions therein that were important for pollen tubes to complete the journey from the site of capture to the site of fertilization in the ovule during the early history of angiosperms.
FOOTNOTES
1 Research was funded by a Connaught New Faculty Award, University of Toronto start-up funds, and a Natural Sciences and Engineering Research Council of Canada grant to T.L.S. 
4 Author for correspondence (tsage{at}eeb.utoronto.ca
)
LITERATURE CITED
Angiosperm Phylogeny Group II.. 2003. An update of the angiosperm phylogeny group classification for the orders and families of flowering plants. Botanical Journal of the Linnean Society 141: 399-436.[CrossRef][Web of Science]
Arbeloa A. Herrero M.. 1987. The significance of the obturator in the control of pollen-tube entry into the ovary in peach (Prunus persica). Annals of Botany 60: 681-685.
Armbruster W. S. Debevec E. M. Wilson M. F.. 2002. Evolution of syncarpy in angiosperms: theoretical and phylogenetic analyses of the effects of carpel fusion on offspring quantity and quality. Journal of Evolutionary Biology 15: 657-672.[CrossRef][Web of Science]
Bailey I. W. Swamy B. G. L.. 1951. The conduplicate carpel of dicotyledons and its initial trends of specialization. American Journal of Botany 38: 373-378.[CrossRef][Web of Science]
Barkman T. L. Chenery G. McNeal J. R. Lyons-Weiler J. Ellisens W. Moore G. Wolfe A. D. dePamphilis C. W.. 2000. Independent and combined analyses of sequences from all three genomic compartments converge on the root of flowering plant phylogeny. Proceedings of the National Academy of Sciences, USA 97: 13166-13171.
Bell P. R.. 1995. Incompatibility in flowering plants: adaptation of an ancient response. Plant Cell 7: 5-16.[CrossRef][Web of Science][Medline]
Bernhardt P. Sage T. L. Weston P. Azuma H. Lam M. Thien L. B. Bruhl J.. 2003. The pollination of Trimenia moorei (Trimeniaceae): floral volatiles, insect/wind pollen vectors, and stigmatic self-incompatibility in a basal angiosperm. Annals of Botany 92: 1-14.
Carr S. G. M. Carr D. J.. 1961. The functional significance of syncarpy. Phytomorphology 11: 249-256.
Chmura C. A.. 1973. Upper Cretaceous (Campanian–Maastrichtian) angiosperm pollen from the western San Joanchin Valley, California, USA. Palaeontographica Abteilung B 141: 89-171.
De Nettancourt D.. 1977. Incompatibility in angiosperms Springer-Verlag, Berlin, Germany.
De Nettancourt D.. 1997. Incompatibility in angiosperms. Sexual Plant Reproduction 10: 185-199.[CrossRef][Web of Science]
De Nettancourt D.. 2001. Incompatibility and incongruity in wild and cultivated plants Springer-Verlag, New York, New York, USA.
Dejong D. W. Jansen E. F. Olsen A. C.. 1967. Oxidoreductive and hydrolytic enzyme patterns in plant suspension culture cells: local and time relationships. Experimental Cell Research 47: 139-156.[CrossRef][Web of Science][Medline]
Dickinson H. G.. 1995. Dry stigmas, water and self-incompatibility in Brassica. Sexual Plant Reproduction 8: 1-10.[Web of Science]
Endress P. K.. 1982. Syncarpy and alternative modes of escaping disadvantages of apocarpy in primitive angiosperms. Taxon 31: 48-52.[CrossRef][Web of Science]
Endress P. K.. 1994. Diversity and evolutionary biology of tropical flowers Cambridge University Press, Cambridge, UK.
Endress P. K.. 2001. The flowers in basal extant angiosperms and inferences on ancestral flowers. International Journal of Plant Sciences 162: 1111-1140.[CrossRef][Web of Science]
Endress P. K. Igersheim A.. 2000. Gynoecium structure and evolution in basal angiosperms. International Journal of Plant Sciences 161: S211-S223.[CrossRef][Web of Science]
Friis E. M. Crepet W. L.. 1987. Time of appearance of floral features. In E. M. Friis, W. G. Chaloner, and P. R. Crane [eds.], The origins of angiosperms and their biological consequences 145-179 Cambridge University Press, Cambridge, UK.
Fulcher R. G. Wong S. I.. 1980. Inside cereals—a fluorescence microchemical view. In G. E. Inglett and L. Munck [eds.], Cereals for food and beverages 1-25 Academic Press, New York, New York, USA.
Graham S. W. Olmstead R. G.. 2000. Utility of 17 chloroplast genes for inferring the phylogeny of the basal angiosperms. American Journal of Botany 87: 1712-1730.
Greenspan P. Meyer E. P. Flower S. D.. 1985. Nile red: a selective fluorescent stain for intracellular lipid droplets. Journal of Cell Biology 100: 965-973.
Guo S.. 1984. The tertiary climates of China, as indicated by plant megafossils. In R. O. Whyte [ed.], The evolution of the east Asian environment, vol. 2 467-469 University of Hong Kong, Hong Kong, China.
Heslop-Harrison J. Heslop-Harrison Y.. 1970. Evaluation of pollen viability by enzymatically induced fluorescence: intracellular hydrolysis of fluorescein diacetate. Stain Technology 45: 115-120.[Web of Science][Medline]
Heslop-Harrison J. Heslop-Harrison Y.. 1982. The specialized cuticles of the receptive surfaces of angiosperm stigmas. In D. F. Cutler, K. L. Alvin, and C. E. Price [eds.], The plant cuticle, Linnean Society Symposium No. 10 99-120 Academic Press, London, UK.
Heslop-Harrison Y. Shivanna K. R.. 1977. The receptive surface of the angiosperm stigma. Annals of Botany 41: 1233-1258.
Hristova K. Lam M. Field T. Sage T. L.. 2005. Transmitting tissue ECM distribution and composition, and pollen germinability in Sarcandra glabra and Chloranthus japonicus (Chloranthaceae). Annals of Botany 96: 779-791.
Hristova-Sarkovski K.. 2006. Functional reproductive biology of the basal angiosperms Sarcandra glabra and Chloranthus japonicus (Chloranthaceae) and Amborella trichopoda (Amborellaceae). M.S. thesis University of Toronto, Toronto, Canada.
Igersheim A. Buzgo M. Endress P. K.. 2001. Gynoecium diversity and systematics in basal monocots. Botanical Journal of the Linnean Society 136: 1-6.[CrossRef][Web of Science]
Igersheim A. Endress P. K.. 1997. Gynoecium diversity and systematics of the Magnoliales and winteroids. Botanical Journal of the Linnean Society 124: 213-271.[CrossRef][Web of Science]
Jeffree C. E.. 1996. Structure and ontogeny of plant cuticles. In G. Kersteins [ed.], Plant cuticles, an integrated and functional approach Bios Scientific Publishers, Oxford, UK.
Khosravi D. Joulaie R. Shore J. S.. 2003. Immunocytochemical distribution of polygalacturonase and pectins in styles of distylous and homostylous Turneraceae. Sexual Plant Reproduction 16: 179-190.[CrossRef][Web of Science]
Kim S. Mollet J. Dong J. Zhang K. Park S. Lord E. M.. 2003. Chemocyanin, a small basic protein from the lily stigma, induces pollen tube chemotropism. Proceedings of the National Academy of Sciences, USA 100: 16125-16130.
Koehl V.. 2002. Functional reproductive biology of Illicium floridanum (Illiciaceae) M.S. thesis, University of Toronto, Toronto, Canada.
Koehl V. Thien L. B. Heij E. G. Sage T. L.. 2004. The causes of self-sterility in natural populations of the relictual angiosperm, Illicium floridanum (Illiciaceae). Annals of Botany 94: 43-50.
Lam B. C. H. Sage T. L. Bianchi F. Blumwald E.. 2001. Role of SH3 domain-containing proteins in clathrin-mediated vesicle trafficking in Arabidopsis. Plant Cell 13: 2499-2512.
Leinfellner W.. 1966. Über die Karpelle verschiedener Magnoliales. III. Schisandra (Schisandraceae). Österreichische Botanische Zeitschrift 113: 563-569.[CrossRef]
Lenartowska M. Rodriguez M. I. Bednarska E.. 2001. Immunocytochemical localization of esterified and unesterified pectins in unpollinated and pollinated styles of Petunia hybrida Hort. Planta 213: 182-191.[CrossRef][Web of Science][Medline]
Lloyd D. G. Wells M. S.. 1992. Reproductive biology of a primitive angiosperm, Pseudowintera colorata (Winteraceae), and the evolution of pollination systems in the Anthophyta. Plant Systematics and Evolution 181: 77-95.[CrossRef][Web of Science]
Lord E. M.. 2003. Adhesion and guidance in compatible pollination. Journal of Experimental Botany 54: 47-54.
Lord E. M. Russell S. D.. 2002. The mechanisms of pollination and fertilization in plants. Annual Reviews of Cell Developmental Biology 18: 81-105.[CrossRef][Web of Science][Medline]
Lush W. M. Grieser F. Wolter-Arts M.. 1998. Directional guidance of Nicotiana alata pollen tubes in vitro and on the stigma. Plant Physiology 118: 733-741.
Martin F. M.. 1959. Staining and observing pollen tubes in the style by means of fluorescence. Stain Technology 34: 436-437.
Marton M. L. Cordts S. Broadhvest J. Dresselhaus T.. 2005. Micropylar pollen tube guidance by egg apparatus 1 of maize. Science 307: 573-576.
Mathews S. Donoghue M. J.. 1999. The root of angiosperm phylogeny inferred from duplicate phytochrome genes. Science 286: 947-950.
Mattsson O. Knox R. B. Heslop-Harrison J. Heslop-Harrison Y.. 1974. Protein pellicle of stigmatic papillae as a probable recognition site in incompatibility reactions. Nature 247: 298-300.[CrossRef]
Mollet J. C. Park S. Y. Nothnagel E. A. Lord E. M.. 2000. A lily stylar pectin is necessary for pollen tube adhesion to an in vitro stylar matrix. Plant Cell 12: 1737-1749.
Mulcahy D. L.. 1975. Gamete competition in plants and animals North-Holland, Amsterdam, Netherlands.
Orban I. Bouharmont J.. 1995. Reproductive biology of Nymphaea capensis Thunb. var. zanzibarensis (Casp.) Verdc. (Nymphaeaceae). Botanical Journal of the Linnean Society 119: 35-43.[CrossRef][Web of Science]
Parkinson C. L. Adams K. L. Palmer J. D.. 1999. Multi-gene analyses identify the three earliest lineages of extant flowering plants. Current Biology 9: 1485-1488.[CrossRef][Web of Science][Medline]
Pontieri V. Sage T. L.. 1999. Evidence for stigmatic self-incompatibility, pollination induced ovule enlargement and transmitting tissue exudates in the paleoherb, Saururus cernuus L. (Saururuaceae). Annals of Botany 84: 507-519.
Qiu Y. L. Lee J. Bernasconi-Quadroni F. Soltis D. E. Soltis P. S. Zanis M. Zimmer E. A. Chen Z. D. Savolainen V. Chase M. W.. 1999. The earliest angiosperms: evidence from mitochondrial, plastid and nuclear genomes. Nature 405: 101-120.
Qiu Y. L. Lee J. Whitlock B. A. Bernasconi-Quadroni F. Dombrovska O.. 2001. Was the ANITA rooting of the angiosperm phylogeny affected by long-branch attraction?. Molecular Biology and Evolution 18: 1745-1753.
Robertson R. E. Tucker S. C.. 1979. Floral ontogeny of Illicium floridanum, with emphasis on stamen and carpel development. American Journal of Botany 66: 605-617.[CrossRef][Web of Science]
Saarela J. M. Rai H. S. Doyle J. A. Endress P. K. Mathews S. Marchant A. D. Briggs B. G. Graham S. W.. 2007. Hydatellaceae identified as a new branch near the base of the angiosperm phylogenetic tree. Nature 446: 312-315.[CrossRef][Medline]
Sage T. L. Bertin R. Williams E. G.. 1994. Ovarian and other late-acting self-incompatibility. In E. G. Williams, R. B. Knox, and A. E. Clarke [eds.], Genetic control of self-incompatibility and reproductive development in flowering plants 116-140 Kluwer, Dordrecht, Netherlands.
Sage T. L. Sampson F. B.. 2003. Evidence for ovarian self-incompatibility as a cause of self-sterility in the primitive woody angiosperm, Pseudowintera axillaris (Winteraceae). Annals of Botany 91: 1-10.
Sage T. L. Sampson F. B. Bayliss P. Gordon M. G. Heij E. G.. 1998. Self-sterility in the Winteraceae. In S. J. Owens and P. J. Rudall [eds.], Reproductive biology in systematics, conservation and economic botany 317-328 Royal Botanic Gardens, England, UK.
Sage T. L. Williams E. G.. 1995. Structure, ultrastructure, and histochemistry of the pollen tube pathway in the milkweed Asclepias exaltata. L. Sexual Plant Reproduction 8: 257-265.[Web of Science]
Saunders R. M. K.. 1998. Monographs of Kadsura (Schisandraceae). Systematic Botany Monographs 54: 1-106.
Saunders R. M. K.. 2000. Monographs of Schisandra (Schisandraceae). Systematic Botany Monographs 58: 1-146.
Sedgley M. Scholefield P. B.. 1980. Stigma secretion in the watermelon before and after pollination. Botanical Gazette 141: 428-434.
Shivanna K. R. Sastri D. C.. 1981. Stigma–surface esterase activity and stigma receptivity in some taxa characterized by wet stigmas. Annals of Botany 47: 53-64.
Smith A. C.. 1947. The families Illiciaceae and Schisandraceae. Sargentia 7: 1-224.
Soltis D. E. Soltis P. S. Chase M. W. Mort M. E. Albach D. C. Zanis M. Savolainen V.. 2000. Angiosperm phylogeny inferred from a combined data set of 18S rDNA, rbcL, and atpB sequences. Botanical Journal of the Linnean Society 133: 381-461.[CrossRef][Web of Science]
Taylor L. P. Hepler P. K.. 1997. Pollen germination and tube growth. Annual Review of Plant Physiology and Plant Molecular Biology 48: 461-481.[CrossRef][Web of Science][Medline]
Thien L. B. Sage T. L. Jaffre T. Bernhardt P. Pontieri V. Weston P. Malloch D. Azuma H. Graham S. W. McPherson M. A. Rai H. S. Sage R. F. Dupre J. L.. 2003. The population structure and floral biology of Amborella trichopoda Baillon (Amborellaceae). Annals of the Missouri Botanical Garden 90: 466-490.[CrossRef][Web of Science]
Thomas H. H.. 1934. The nature and origin of the stigma. New Phytologist 33: 173-198.[CrossRef]
Tilton V. R. Horner H. T.. 1980. Stigma, style, and obturator of Ornithogalum caudatum (Liliaceae) and their function in the reproductive process. American Journal of Botany 67: 1113-1131.[CrossRef][Web of Science]
Tucker S. C. Bourland J. A.. 1994. Ontogeny of staminate and carpellate flowers of Schisandra glabra (Schisandra). Plant Systematics and Evolution 8: 137-158.
Vasil I. K. Johri M. M.. 1964. The style, stigma and pollen tube. I. Phytomorphology 14: 352-369.
Wang X. F. Tan Y. Y. Chen J. H. Lu Y. T.. 2006. Pollen tube reallocation in two preanthesis cleistogamous species Ranalisma rostratum and Sagittaria guyanesis ssp. lappula (Alismataceae). Aquatic Botany 85: 233-240.[CrossRef][Web of Science]
Wang X. F. Tao Y. B. Lu Y. T.. 2002. Pollen tubes enter neighbouring ovules by way of receptacle tissue, resulting in increased fruit set in Sagittaria potamogetifolia Merr. Annals of Botany 89: 791-796.
Williams E. G. Sage T. L. Thien L. B.. 1993. Functional syncarpy by intercarpellary growth of pollen tubes in a primitive apocarpous angiosperm, Illicium floridanum (Illiciaceae). American Journal of Botany 80: 137-142.[CrossRef][Web of Science]
Willson M. F. Burley N.. 1983. Mate choice in plants Princeton University Press, Princeton, New Jersey, USA.
Wolters-Arts M. Van Der Weerd L. Aelst Van A. C. Van Der Weerd J. Van As H. Mariani C.. 2002. Water-conducting properties of lipids during pollen hydration. Plant, Cell and Environment 25: 513-519.[CrossRef]
Wu H. M. Wong E. Ogdahl J. Cheung A. Y.. 2000. A pollen tube growth-promoting arabinogalactan protein from Nicotiana alata is similar to the tobacco TTS protein. Plant Journal 22: 165-176.[CrossRef][Web of Science][Medline]
Zanis M. J. Soltis D. E. Soltis P. S. Mathews S. Donoghue M. J.. 2002. The root of the angiosperms revisited. Proceedings of the National Academy of Sciences, USA 99: 6848-6853.
Zinkl G. M. Zwiebel B. I. Grier D. G. Preuss D.. 1999. Pollen–stigma adhesion in Arabidopsis: a species-specific interaction mediated by lipophilic molecules in the pollen exine. Development 126: 5431-5440.[Abstract]
This article has been cited by other articles:
|
|
 |
|
  P. K. Endress and J. A. Doyle Reconstructing the ancestral angiosperm flower and its initial specializations Am. J. Botany, January 1, 2009; 96(1): 22 - 66. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. L. Sage, K. Hristova-Sarkovski, V. Koehl, J. Lyew, V. Pontieri, P. Bernhardt, P. Weston, S. Bagha, and G. Chiu Transmitting tissue architecture in basal-relictual angiosperms: Implications for transmitting tissue origins Am. J. Botany, January 1, 2009; 96(1): 183 - 206. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. B. Thien, P. Bernhardt, M. S. Devall, Z.-d. Chen, Y.-b. Luo, J.-H. Fan, L.-C. Yuan, and J. H. Williams Pollination biology of basal angiosperms (ANITA grade) Am. J. Botany, January 1, 2009; 96(1): 166 - 182. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. H. Williams Amborella trichopoda (Amborellaceae) and the evolutionary developmental origins of the angiosperm progamic phase Am. J. Botany, January 1, 2009; 96(1): 144 - 165. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
0.05). Error bars represent 95% confidence interval