| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
|
First published online January 9, 2009; doi:10.3732/ajb.0800142 American Journal of Botany 96: 391-408 (2009) © 2009 Botanical Society of America, Inc. |
What's this? |
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Anatomy and Morphology |
2 University of Missouri–St. Louis, 1 University Boulevard, St. Louis, Missouri 63121 USA 3 Missouri Botanical Garden, 4500 Shaw Boulevard, St. Louis, Missouri 63110 USA 4 Department of Ecology and Evolutionary Biology, Brown University, 80 Waterman St., Box G-W, Providence, Rhode Island 02912 USA
Received for publication 21 April 2008. Accepted for publication 6 October 2008.
ABSTRACT
The cacti have undergone extensive specialization in their evolutionary history, providing an excellent system in which to address large-scale questions of morphological and physiological adaptation. Recent molecular phylogenetic studies suggest that (1) Pereskia, the leafy genus long interpreted as the sister group of all other cacti, is likely paraphyletic, and (2) Cactaceae are nested within a paraphyletic Portulacaceae as a member of the "ACPT" clade (Anacampseroteae, Cactaceae, Portulaca, and Talinum). We collected new data on the vegetative anatomy of the ACPT clade and relatives to evaluate whether patterns in the distributions of traits may provide insight into early events in the evolutionary transition to the cactus life form. Many traits had high levels of homoplasy and were mostly equivocal with regard to infraclade relationships of ACPT, although several characters do lend further support to a paraphyletic Pereskia. These include a thick stem cuticle, prominent stem mucilage cells, and hypodermal calcium oxalate druses, all of which are likely to be important traits for stem water storage and photosynthesis. We hypothesize that high lability of many putative "precursor" traits may have been critical in generating the organismal context necessary for the evolution of an efficient and integrated photosynthetic stem.
Key Words: adaptation • Cactaceae • character evolution • homoplasy • Portulacaceae • Portulacineae • stem photosynthesis • vegetative anatomy
Cactaceae are one of the most distinctive and immediately recognizable plant groups due to their unusual growth forms, which reflect adaptations that allow them to persist and flourish in extremely arid environments. Characteristics such as stem-based photosynthesis and a concomitant reduction of leaves, stem succulence, spines, and crassulacean acid metabolism (CAM) photosynthesis are all considered to be adaptations of cacti to water-limited environments (Gibson and Nobel, 1986
), a supposition that is supported by the independent acquisition of different combinations of these traits in unrelated lineages that are also drought adapted (e.g., Euphorbiaceae, Agavaceae, Aizoaceae).
Cacti thus are an extremely useful system in which to examine questions surrounding the evolution of specialized morphologies and physiologies in plants, especially considering that within the group there is a range of variation in these traits. Such questions are best approached within an explicitly phylogenetic framework. In the case of Cactaceae, phylogenetic relationships both within and outside of the group have historically proven difficult to resolve by traditional morphology-based methods, but two recent breakthroughs in cactus molecular systematics now enable us to ask cogent evolutionary questions about the group. The first is that the genus Pereskia Mill., generally interpreted as the sister group to the rest of Cactaceae, appears to be paraphyletic (Nyffeler, 2002; Butterworth and Wallace, 2005; Edwards et al., 2005). The second is the finding that Cactaceae are nested within a paraphyletic Portulacaceae (Hershkovitz and Zimmer, 1997; Applequist and Wallace, 2001; Edwards et al., 2005; Nyffeler, 2007).
Phylogenetic context
Within Cactaceae, Pereskia has traditionally been viewed as the sister group to the remaining cacti, retaining many ancestral traits that have been lost in other species. Such traits include fully expanded photosynthetic leaves, a relatively nonsucculent stem, unspecialized wood, and a tree or shrub habit (Gibson and Nobel, 1986; Leuenberger, 1986; Mauseth and Landrum, 1997). Recent studies confirm this phylogenetic placement but further suggest that Pereskia as traditionally circumscribed is likely paraphyletic, with a mainly Caribbean/Central American/northern South American clade (hereafter, the northern clade) sister to a mainly Andean/southern South American clade (hereafter, the Andean/SSA clade) plus the remainder of Cactaceae (Edwards et al., 2005) (Fig. 1). Close examination of this Pereskia sensu lato reveals variation in some characters that may be taxonomically useful. Although no character perfectly distinguishes the two Pereskia clades, all northern clade species have precocious stem periderm onset and lack stem stomata, while most Andean/SSA clade species have delayed stem periderm (except P. aculeata) and possess stem stomata (except P. nemorosa). No Pereskia species have been reported to engage in significant photosynthetic carbon uptake through the stem (Nobel and Hartsock, 1986; Martin and Wallace, 2000; Edwards and Donoghue, 2006), but the presence of stem stomata and delayed periderm unites the Andean/SSA species with the most speciose cactus lineages, Opuntioideae and Cactoideae ("core cacti" of Edwards et al., 2005). The changes in these characters subsequent to the divergence of the northern Pereskia lineage from other cacti may have therefore been important preconditions for the evolution of the functionally leafless, stem-photosynthetic life form of the core cacti. However, assigning the polarity of transitions in periderm timing and stem stomata, as well as the evolutionary importance of changes in these characters, requires knowledge about these character states in outgroup taxa (Fig.1, arrow).
|
) (Hershkovitz and Zimmer, 1997; Applequist and Wallace, 2001; Cuénod et al., 2002; Applequist et al., 2006; Nyffeler, 2007). Moreover, these molecular studies have clearly indicated that Portulacaceae are paraphyletic with respect to Cactaceae, Didiereaceae, and Hectorellaceae (Fig. 2), a result that was hinted at previously in some morphological studies (Chorinsky, 1931; Hershkovitz, 1993). Thus the nearest relatives of the cactus family appear to be three lineages of predominantly African, eastern American, and Australian taxa within Portulacaceae: Talinum Adans. (including Talinella Baillon), Portulaca L., and Anacampseroteae Nyananyo ex. G. D. Rowley (Nyffeler, 2007) (Fig. 2). Because a formal nomenclatural classification for this group has not yet been proposed, the clade consisting of these three portulacaceous lineages plus Cactaceae will hereafter be referred to by the working title ACPT (Anacampseroteae, Cactaceae, Portulaca, and Talinum).
|
; matK and ndhF: Nyffeler, 2007), mitochondrial (nad1: Nyffeler, 2007), and nuclear markers (ITS : Hershkovitz and Zimmer, 1997; phyC: Edwards et al., 2005), the exact relationships among the four major clades of ACPT and within suborder Portulacineae as a whole have proven difficult to resolve. In ACPT, there is generally strong support across genes and methods of analysis for monophyly of the four major component clades of Cactaceae, Anacampseroteae, Portulaca, and Talinum, as well as for the ACPT clade itself. The nodes indicating infraclade relationships in ACPT, however, are less well resolved. In particular, the relationships among Portulaca, Anacampseroteae, and Cactaceae have yielded conflicting results (Applequist and Wallace, 2001; Edwards et al., 2005; Nyffeler, 2007).
Using this improved phylogenetic framework, we examined vegetative anatomical variation in Cactaceae and their nearest relatives. Anatomical characters are a particularly appropriate focus for investigations of the relationship between form, function, and adaptation at a macroevolutionary scale. Although vegetative anatomy in Cactaceae has been fairly well studied (see Mauseth, 2006, and references therein), it is less well characterized in Portulacaceae s.l. (but see Carlquist, 1998, for wood anatomy; Landrum, 2002). Furthermore, an explicit link between Cactaceae and their nearest portulacaceous relatives has generally not been considered in previous anatomical studies. We therefore examined vegetative anatomical characters within ACPT and related Portulacineae to (1) provide a general characterization of the anatomical variation within ACPT, (2) identify potential synapomorphic characters that would support any one of the alternative hypotheses for relationships within ACPT, and (3) examine trait patterns in the ACPT clade as they relate to the evolutionary transition to the cactus life form.
MATERIALS AND METHODS
Anatomical investigations
We investigated 34 taxa sampled broadly from all of the portulacaceous taxa within ACPT, as well as representatives of both Pereskia clades and Pereskiopsis gatesii, an opuntioid species (see Appendix 1 for taxa examined, nomenclatural authorities, and voucher numbers). Among outgroups of ACPT, we sampled Portulacaria afra and Ceraria fruticulosa from Didiereaceae and Claytonia virginica, Phemeranthus teretifolius, and Parakeelya pickeringii from Montiaceae. Some taxa were field-collected, but the majority originated from cultivated material, whether grown in commercial or institutional greenhouses (see Appendix 1 for details). The focus of this study was on anatomical characters of stems and leaves in all taxa examined. Roots could not be reliably obtained for all taxa and so were not observed as extensively. Hereafter, mention of characters relating to xylem, cortex, etc. will apply to stems only unless specifically stated otherwise.
Vegetative anatomical characters were investigated primarily through freehand sectioning of fixed material. Material was fixed in formalin-acetic acid-alcohol (FAA) (Ruzin, 1999) or 70% ethanol for at least 48 h prior to sectioning. Sections were stained with cresyl violet acetate (CVA) in a 15% ethanol solution before mounting in calcium chloride (Herr, 1992; Keating, 1996). CVA is a metachromatic dye that produces a striking and visually pleasing color contrast between cellulosic (staining pink to red) and lignified tissues (staining blue), although its use has not been widely adopted by botanists (Keating, 1996). For sections to examine calcium oxalate crystal structure, separate unstained sections were mounted in glycerin due to the tendency of calcium oxalate crystals and druses to erode over time in calcium chloride.
For stems, transverse, longitudinal radial, and longitudinal tangential sections were cut at nodes and internodes. To address ontogenetic variation within a single plant, whenever possible we obtained samples from stems in a state of primary growth, early secondary growth (i.e., shortly after the initiation of the vascular cambium) and mature growth from near the base of the stem. Stem epidermal peels were obtained at various ages as well. Mature leaves were sectioned transversely and leaf epidermal peels were obtained.
Image capture and processing
All photographs were taken with a Nikon Coolpix 8700 (Nikon, Tokyo, Japan) mounted on a Martin Microscopes MM99 adapter (Martin Microscopes, Easley, South Carolina, USA). Most slides were viewed on an Olympus BX40 light microscope (Olympus, Tokyo, Japan). A Zeiss Universal microscope (Carl Zeiss, Oberkochen, Germany) was used to visualize calcium oxalate crystals and druses using polarized light filters. Images were processed using Photoshop 8.0 (Adobe Systems, San Jose, California). Processing was limited to reduction of image size, application of an unsharp mask filter, and adjustment of color levels. Original image files are available upon request from the first author.
Character evolution analyses
We generated measures of homoplasy for 14 stem characters that could be coded into discrete categories. Briefly, we calculated consistency indices (CI = minimum number of possible changes/observed number of changes) for each character using the current best hypothesis of relationships within Portulacineae (e.g., based on Hershkovitz and Zimmer, 1997; Nyffeler, 2002; Butterworth and Wallace, 2005; Edwards et al., 2005; Nyffeler, 2007). Our phylogeny included several unresolved nodes, specifically (1) the relationships between Montiaceae, Didiereaceae, and the ACPT clade; (2) relationships between Portulaca, Anacampseroteae, and cacti; (3) relationships within Anacampseros; and (4) relationships within Portulaca. From the CI (measured using the program MacClade version 4.06, Sinauer, Sunderland, Massachusetts, USA), we calculated a homoplasy index (HI = 1 – CI). The nexus file used to generate CI values is available from the authors upon request.
We also estimated the degree of correlated evolution between a subset of stem characters we considered to be potentially important "developmental enablers" (sensu Donoghue, 2005) for the further evolution of stem photosynthesis and water storage in cacti. We employed a Bayesian reversible-jump Markov chain Monte Carlo (MCMC) approach (Pagel and Meade, 2006) as implemented in the program BayesTraits version 1.0 (available at website http://www.evolution.rdg.ac.uk). To account for topological uncertainties, we generated a set of 1000 trees with randomly resolved polytomies (using the program Mesquite version 2.01, Maddison and Maddison, 2006) and used this sample of trees for the Bayesian analyses. Briefly, for each pair of traits, we performed MCMC runs employing both a dependent (i.e., correlated) and independent (i.e., uncorrelated) model of trait evolution. For each analysis, we ran 5 050 000 iterations, discarding the first 50 000 as burn-in, and sampling every 100 iterations. We seeded the mean and variance of the gamma distribution of rate coefficents drawing from a uniform (0–10) hyperprior distribution. For each trait pair, we calculated a log BayesFactor statistic from the harmonic means of the likelihoods of both dependent and independent runs, estimated as logBayes factor = 2 x [harmonic mean (dependent model) – harmonic mean (independent model)]. In general, values of logBayes factor greater than 5 provide strong support for the dependent (correlated) model, and values of greater than 10 provide very strong support (Pagel and Meade, 2006).
RESULTS
Anatomical observations are summarized in Tables 1 and 2. General patterns are described next.
|
|
|
Periderm onset is delayed in the Andean/SSA Pereskia group, except P. aculeata, while all northern Pereskia species observed have precocious periderm formation. Anacampseroteae are characterized by precocious periderm onset, with the exception of Anacampseros sect. Avonia, in which stem periderm was not observed. Similarly, the herbaceous Portulaca generally lacks periderm in all but the oldest stem bases, while both precocious and delayed periderm initiation occur with no clear phylogenetic pattern in the Talinum clade.
Location of periderm onset also shows variation. In most Portulacineae observed, periderm initiates in the epidermis (Fig. 3A, B, green arrows). Exceptions occur in Pereskia guamacho, P. portulacifolia, Talinopsis frutescens, Grahamia bracteata, and Talinum portulacifolium, in which it initiates in the outer cortex, eventually cutting off the epidermis, hypodermis, and outermost cortical layers as rhytidome (Fig. 3C, D). Also noteworthy is the occurrence in most Pereskia species of stratified concentric bands of sclereids in the phellem layer of periderm (Fig. 3E). Lignified bands consisting of thinner-walled, flattened cells occur in the phellem of Talinopsis frutescens (Fig. 3F), Grahamia bracteata, and Anacampseros kurtzii, as well as in Portulacaria afra and Ceraria fruticulosa.
|
Leaves of most taxa observed here, including those of Didiereaceae and Montiaceae, are amphistomatic, with essentially equal densities of stomata on their abaxial and adaxial surfaces, or equally around the entire leaf surface in the case of terete leaves. An exception to this was Portulaca oligosperma, in which stomata occurred, atypically for most leafy plants, on the adaxial leaf surface only. All observed taxa of ACPT and Didiereaceae have parallelocytic stomata, in which the stomate is laterally invested by a series of at least three alternating subsidiary cells that become increasingly larger progressing out from the guard cells (Fig. 4A). In contrast, the species of Montiaceae observed all have brachyparacytic stomata, in which two subsidiary cells surround but do not completely enclose the stomatal apparatus (Fig. 4B). Most Anacampseroteae and Portulaca sampled have transversely oriented leaf stomata, while in other taxa sampled here leaf stomatal orientation is more or less random.
|
Aggregated isodiametric brachysclereids ("stone cells") occur in the outer cortex of P. guamacho and are also associated with the primary phloem of P. guamacho (Figs. 3E, 5A) and P. portulacifolia, both northern clade species. P. diaz-romeroana and P. weberiana, representatives of the Andean/SSA clade, lack cortical sclereids but have fusiform sclereids occurring singly or occasionally in aggregate in the secondary phloem (Fig. 5B). No sclereids were found in P. quisqueyana.
|
Thin-walled, elongated cells with lignified walls are associated with extraxylary fiber caps in Talinum portulacifolium, T. triangulare, and Talinella pachypoda, as well as Portulacaria afra and Ceraria fruticulosa (Fig. 5D). When viewed in longitudinal section, these sclereids are typically not fusiform but are squared-off and similar in shape to surrounding parenchyma cells. They do not appear to develop thicker secondary walls as do the sclereids of Anacampseroteae and Pereskia. Subelliptic brachysclereids with more or less thin walls occur in root cortex of Portulaca amilis, P. pilosa, and P. grandiflora (online Appendix S1D).
Mucilage cells and cavities in leaves and stems
Mucilage-containing cells in leaves and stems are common throughout Portulacineae. These cells are often larger than surrounding parenchyma and contain mucilage that stains red in CVA. Of the taxa sampled here, the only ones lacking mucilage cells in leaf mesophyll are Claytonia virginica and Parakeelya pickeringii, both members of Montiaceae (Phemeranthus, the other sampled Montiaceae, does have leaf mucilage cells). Within ACPT, even taxa with relatively thin, nonsucculent leaves (e.g., Talinum paniculatum, Pereskia quisqueyana) have abundant mucilage cells.
In stems, mucilage cells were lacking in most Montiaceae but observed in most ACPT and Didiereaceae (Table 1
), commonly in the outer cortex (Fig. 3A, white arrows). In most cases, however, they were not highly conspicuous components of stem ground tissues, occurring sparsely and/or ephemerally in young shoots only. However, in older stems of Portulaca amilis, P. oligosperma, the Andean/SSA Pereskia clade, and Pereskiopsis, large mucilage cells are retained and are a prominent feature of cortical and pith tissues (Fig. 6A, B).
|
; Gibson, 1994; Landrum, 2002), although specific tests to determine their content were not performed. These cells were common in the cortex, rays, and pith of all Didiereaceae and Talinum species sampled except T. paniculatum and Talinella pachypoda (Figs. 3A, 6C). Viewed in longitudinal section, they occur in extensive columns of contiguous cells (online Appendix S1E, green arrow). Tannin cells were also found in the leaf epidermis of P. afra (Fig. 6D) and early-formed phellogen of C. fruticulosa. In Talinella pachypoda, tanniniferous cell contents occur in axillary prophyll scales (see description in next section) (online Appendix S1F), in scarious leaf tips, and in stem epidermal papillae. Tannin cells were not observed in any taxa of Portulaca, Anacampseroteae, Cactaceae, or Montiaceae that were examined.
Nodal hairs/bristles
Cactaceae, Anacampseroteae, and Portulaca are characterized by the presence of distinctive hair- or bristle-like outgrowths, typically at nodes in the leaf axil, but in some taxa borne elsewhere on the stem or the leaf lamina. These structures are here interpreted as trichomes, as opposed to stipules, in that they grow from the epidermal layer, are not vascularized, and are noncaducous.
The hair-like trichomes range from one (uniseriate) to three or more (multiseriate) cells in width. In contrast, "bristles" are extremely wide, flattened multiseriate trichomes that are up to 20 or more cells in width. All Portulaca, Anacampseroteae, and Cactaceae observed have hairs at nodes (thus in the areoles of cacti), although they are highly reduced (i.e., not visible macroscopically) in Portulaca oligosperma, P. oleracea, and Xenia vulcanensis. The occurrence of bristles is more limited, being observed here in Grahamia bracteata and Anacampseros sect. Anacampseros. Anacampseros sect. Avonia has broad axillary scales that completely enclose the leaf distal to the subtending leaf and are lignified at the tip.
Talinum species have scarious scale-like structures in leaf axils, often paired, which can superficially appear similar to the hairs of other ACPT. However, closer examination reveals these structures to be the tips of prophylls, which may subsequently expand as full leaves (onlinen Appendix S1F and S1G). Indeed, the scale-like structures are typically still visible at the tips of fully expanded leaves in most Talinum (online Appendix S1H). We also observed vascularization leading to these axillary prophylls in Talinum paniculatum, which never occurs in the hairs or bristles of ACP.
Druses and crystals
Calcium oxalate druses (i.e., spherical crystal aggregations) and single crystals are widespread in all major subclades of ACPT as well as in Didiereaceae and Parakeelya pickeringii. Position of druses and crystals in stem and leaf tissues is highly variable even within individual plants and does not appear to carry a strong phylogenetic signal. An exception is found in Pereskiopsis and Pereskia diaz-romeroana, P. weberiana, P. aculeata, and P. guamacho, which are distinctive in consistently having druses immediately internal to the epidermal layer (Fig. 7A). In P. guamacho these druses are lost early in development, when they are cut off by the activity of periderm in the outer cortex (Fig. 3D).
|
Wide-band tracheids/wood characters
Short, barrel-shaped tracheids with wide helical or annular thickenings projecting into the cell lumen occurred in stems and leaves of all species of Anacampseroteae except Talinopsis frutescens, Grahamia bracteata, and Xenia vulcanensis. These wide-band tracheids (WBTs) (Bailey, 1964; Mauseth et al., 1995) always occur in xylem rays or in the ground tissue of the pith and are never associated with vessel elements in vascular bundles. Despite occurring in rays, they are usually oriented longitudinally in the stem. No WBTs were observed in Portulaca, Talinum, either Pereskia clade, or in any of the outgroups of ACPT that were examined. In Anacampseroteae, WBTs do not occur in Talinopsis frutescens and Grahamia bracteata, the only two species that have libriform xylem fibers. Xenia appears to have neither WBTs nor libriform fibers.
Epidermis
Papillae arising from the stem epidermis are a common feature in Portulacaria afra (Didiereaceae) and most Talinum species observed here and occur in a few other ACPT. These papillae are often outgrowths of a single epidermal cell, although in Talinum triangulare, T. caffrum, and less commonly Portulacaria afra, they are made up of two to four cells (Fig. 6C). They vary extensively among species in their lifespan, with some persisting in spite of extensive periderm (Portulacaria afra, Portulaca oligosperma, and some Talinum species), while others are ephemeral (other sampled Portulaca species). Papillae are generally lacking in the Anacampseroteae species observed here, occurring only on very young stems and bases of the leaf lamina in Xenia vulcanensis (online Appendix S1B). Although stem epidermal papillae were not found in Anacampseros sect. Anacampseros, elongate epidermal papillae were observed at the leaf tip in A. rufescens and over the entire lamina in A. filamentosa. Papillae also were observed on leaf margins of numerous Talinum species. Epidermal papillae were not observed on stems of either Pereskia group, although they did occur on leaf margins of P. weberiana and P. diaz-romeroana.
Pereskiopsis gatesii and three of the Andean/SSA Pereskia species, P. diaz-romeroana, P. nemorosa, and P. weberiana, are unique among the taxa observed here in possessing a distinctly thickened stem cuticle that stains purple in CVA (Fig. 7A, B). In P. diaz-romeroana, P. nemorosa, and P. weberiana the cuticle is intercalated between the lateral walls of the epidermal cells.
Leaves
Leaves in taxa observed here are typically pinnately veined, i.e., with a more or less prominent midvein joined along its length by secondary veins. Pereskiopsis, Ceraria, and Portulacaria have palmately veined leaves.
Leaves of all taxa observed in this study have little differentiation between palisade and spongy layers of mesophyll tissues, with the exception of Claytonia virginica. In general, taxa with fleshy, bifacial leaves (Talinum, Pereskia, some Anacampseroteae) have greater mesophyll differentiation than those with more succulent, subterete to terete leaves.
Character evolution analyses
In general, many traits exhibited very high levels of homoplasy (Table 3). The two most labile traits were the presence of stem stomata and the timing of periderm formation, each having arisen multiple times across all of Portulacineae. Several characters were less labile. Axillary hairs, for example, appear to have arisen once, along the stem to the ACP clade, and were subsequently lost within Anacampseros sect. Avonia. Similarly, a thickened stem cuticle is likely to have arisen once in the caulocactus clade, and to have been subsequently lost in Pereskia aculeata.
|
) and negatively correlated with the number of states to which characters may be assigned (Donoghue and Ree, 2000). The first issue is not a concern because all HI values were generated using the same tree. While some of our stem characters had two states and others three, the three-state characters (stem sclereids and mucilage cells) had among the highest HI values; therefore, we do not think their homoplasy was significantly underestimated. Our Bayesian analyses also provided little evidence for significantly correlated evolution between many of these labile traits (Table 4). There were four notable exceptions: there is strong evidence for a correlation between delayed periderm and a thickened cuticle; between hypodermal druses and a thickened cuticle; between hypodermal druses and stem stomata; and very strong evidence for a correlation between delayed periderm and enlarged, persistent mucilage cells. The two most labile stem traits, delayed periderm and stem stomata, had nearly the lowest support for any correlation between them, with a logBayes factor of –1.00.
|
Morphological evidence for relationships in ACPT and greater Portulacineae
No clear morphological synapomorphies for the ACPT clade were identified in this study. However, the shared presence of parallelocytic leaf stomata, tannin cells, and pericyclic sclereids in both ACPT and Didiereaceae suggests a close relationship between these two clades, as hinted in at least one study (Nyffeler, 2007). These characters have been reported from other Didiereaceae that were not sampled here (Gibson, 1994), and, as far as is known, are lacking in Montiaceae and Basellaceae (Metcalfe and Chalk, 1950; Sperling, 1987; Carlquist, 1999) (Table 5). If ACPT and Didiereaceae are ultimately shown to be sister taxa, these characters may be considered potential synapomorphies of such a clade.
|
; Nyffeler, 2007). The morphological characters studied here similarly fail to provide satisfactory evidence to support one phylogenetic hypothesis over another for relationships within the ACP grouping. In common with the northern Pereskia clade, and in contrast with Portulaca, the ancestral condition for stems in Anacampseroteae is inferred as fibrous-woody and nonsucculent, with precocious periderm initiating in the outer cortex (see Table 1, Talinopsis frutescens and Grahamia bracteata), which could be interpreted as supporting an affinity between Cactaceae and Anacampseroteae. However, stem woodiness and position and timing of periderm onset vary considerably across all of Portulacineae, and within well-defined clades, suggesting that these characters may not be phylogenetically informative. Wide-band tracheids occur in both Anacampseroteae and Cactaceae, although their phylogenetic distribution indicates that they are derived independently in both clades and likely evolved in parallel (Mauseth, 2004; Landrum, 2006).
Characters supporting a possible phylogenetic association between Anacampseroteae and Portulaca include transversely oriented leaf stomata, a lack of extraxylary fibers, and a base chromosome number of x = 9 (but also x = 8 in some Portulaca) (Goldblatt and Johnson, 1979). Again, however, all these characters are labile in the larger phylogenetic context (Table 5). The reduced, herbaceous habit in Portulaca and Anacampseros appears to be convergent and is not evidence of a close relationship of these two clades.
Calcium oxalate druses distributed in the hypodermis have previously been reported from many Cactaceae, including both Pereskia lineages (Metcalfe and Chalk, 1950; Mauseth and Landrum, 1997; Mauseth, 2005). Their absence in the portulacaceous clades of ACPT observed here indicates that they may be a synapomorphy of Cactaceae.
Evolution in Cactaceae
The core cacti represent a relative extreme in plant morphological evolution; leaves are highly reduced, and stems have been modified for the dual purposes of water storage and photosynthetic carbon uptake. Reconstructing the evolution of the cactus life form will therefore have general implications for our understanding of major morphological transitions in plants. A paraphyletic Pereskia provides a starting point for analyzing the evolution of the core cacti from leafy, nonsucculent ancestors (Fig. 1). The presence of delayed periderm and stem stomata unites the Andean/SSA Pereskia clade with core cacti. While stems of many plants may photosynthetically recycle respiratory carbon dioxide under periderm and in the absence of stem stomata (Pfanz et al., 2002), these coincident character changes were clearly necessary precursors for the evolution of the stem as the major photosynthetic organ. Edwards et al. (2005) tentatively named the clade comprising the core cacti + Andean/SSA Pereskia clades the "caulo cacti" to highlight these early and potentially important stem developments (Fig. 1).
The current study, however, demonstrates that both of these characters are highly variable in Portulacineae (Tables 3, 5). Stem stomata of various densities occur at least ephemerally in most Talinum species observed and also appear in Portulaca, Anacampseroteae, Didiereaceae, and Montiaceae. Many Portulacineae also have delayed or absent stem periderm, especially in Portulaca and Montiaceae. Furthermore, the specific combination of delayed or no periderm and fairly dense stem stomata occurs in many taxa (e.g., Talinum caffrum, Claytonia virginica; Table 1; although they are not significantly correlated across the tree, Table 4). The apparent homoplasy of these traits across Portulacineae makes it currently difficult to infer whether they were lost in the northern Pereskia clade or gained in the caulocactus clade. However, neither the uncertainty about the polarity of these changes nor their observed lability should exclude these traits from a discussion of their importance in early cactus evolution. The activity and significance of a given character needs to be considered in its context, both in terms of the other characters occurring within the organism, as well as the ecological setting in which that organism is found (de Queiroz, 2002; Donoghue, 2005). For example, almost all examples of stem stomata and delayed periderm observed here outside of the caulocactus clade are found in reduced, herbaceous plants. It is likely that in the cacti, it was the organismal context of the change in these stem traits, i.e., their occurrence in a perennial, woody plant, that was critical in the evolution of the stem as a long-lived, photosynthetic organ.
Along these lines, a further question would be whether other stem characters can be identified that may have interacted with delayed periderm and stem stomata in the evolution of cacti. These characters may also have changed between the two Pereskia lineages, or they may be invariant, their importance being based in their interaction with changes in the previously identified epidermal characters. If such characters can be found and are lacking in outgroup taxa, then they may be candidates for important components of complex morphological and physiological innovation in cacti and would be worthy of further investigation.
Three characters identified here are potential candidates for such a role, and each carries physiological implications for stem water storage and/or photosynthesis. They have been mentioned by previous authors, but not in the context of the hypothesis of a paraphyletic Pereskia.
The first of these is the development of a thick stem cuticle seen in the Andean/SSA Pereskia clade and the core cacti (Fig. 7A, B). Such a cuticle is qualitatively different from those seen in either leaves or stems of other taxa here, including young preperiderm shoots of the northern Pereskia lineage, and has been shown in core cacti to be an important factor in the ability of the stem to store water (Gibson and Nobel, 1986). Stems of the Andean/SSA Pereskia clade possess an only slightly succulent cortex when compared with core cacti, and therefore the evolution of a thickened cuticle appears to have preceded that of major stem succulence. It is noteworthy that the only Andean/SSA Pereskia species seen here to lack a thick cuticle, P. aculeata, is also the only species in this group with precocious periderm onset. If the cuticle is secreted by the epidermis (Esau, 1977), then it follows that precocious bark formation would disrupt the epidermis before a cuticle could form, making delayed periderm essentially a requirement for the development of a thickened cuticle. Indeed, the strong statistical support for their correlated evolution (Table 4) further suggests that these two traits may be mechanistically linked. This relationship illustrates the importance of delayed periderm as a potentially key developmental enabler (Donoghue, 2005) for subsequent stem evolution in the cacti.
A second character of interest is the development of greatly enlarged, persistent mucilage cells or cavities in the cortex and pith of all Andean/SSA Pereskia species observed in this study (Fig. 6B, Table 5). In contrast, stem mucilage cells in the northern Pereskia species are generally lacking or are poorly developed (Table 5). Similarly, although mucilage cells are common in leaves of Portulacineae, they are typically ephemeral and/or poorly developed in stems and are often no larger than surrounding parenchyma cells (Fig. 3A). Outside of the Andean/SSA Pereskia group, the largest stem mucilage cells seen in this study are in Portulaca amilis (Fig. 6A), but these are still a qualitatively less prominent component of stem tissues than in the former taxa. Chemically, the mucilage found in cacti consists of complex fibrous polysaccharides that may contain up to 30 000 sugar subunits in a single molecule (Gibson and Nobel, 1986). Mucilage cells have been demonstrated to serve as effective reservoirs for water molecules due to the chemical affinities of these polysaccharides (Gibson and Nobel, 1986). The distribution of mucilage cells in leaves of most Portulacineae correlates well with the role of leaves as the major water-storing organs in these taxa, and it follows that the gain of enlarged mucilage cells in stems of the Andean/SSA Pereskia group may also have been important in the assumption of water storage by stems. Interestingly, the correlation between mucilage cells and delayed periderm is the strongest of our data set (Table 4), suggesting a possible functional link between these two traits. Because suberized cork tissue mitigates water loss in perennating stems, a delay in periderm onset may require the development of some small degree of cortical water storage to buffer evaporative loss from the stem surface, which could be provided by the gain of cells with copious mucilage. This functional analogy between periderm and mucilage is supported by the presence of ephemeral mucilage cells in young stems of many taxa with early periderm onset, which are subsequently lost as the stem matures (Fig. 3A).
A third character that may have played a role in the evolution of the cactus stem as the major photosynthetic organ is the distribution of calcium oxalate druses in the hypodermal layer. It has been demonstrated that an excess of calcium ions in the vicinity of guard cells results in stomatal closure (de Silva et al., 1985) and that formation of druses serves to prevent calcium from interfering with stomatal function (Ruiz and Mansfield, 1994). The ability to sequester calcium ions in this manner may be relevant for the correct functioning of stem stomata. It is interesting to consider the strong support for the correlated evolution of hypodermal druses and stem stomata in our data set (Table 4); currently, however, there is not much evidence that these traits are functionally linked, because stem stomata in the Andean/SSA Pereskia lineage do not appear to be heavily involved in gas exchange (Martin and Wallace, 2000; Edwards and Donoghue, 2006). An alternative explanation for a functional role of hypodermal druses may be as a deterrent to herbivory, which would be important in plants that are lacking the protective functions of an early periderm. However, the correlation between periderm timing and hypodermal druses is weaker in our data set (Table 4). It also bears noting that hypodermal druses were found in both Pereskia lineages, although the onset of periderm in the cortex in P. guamacho means that they are soon lost with the outer cortex and epidermis, emphasizing that the interaction between characters may be most useful from a functional standpoint.
The results obtained here underscore the importance of sampling multiple ontogenetic stages when examining anatomical characters. Many important characters, such as mucilage cells, epidermal papillae, and calcium oxalate druses occurred ephemerally in younger tissues and were often subsequently either lost or greatly altered during ontogeny. Failure to sample a wide range of ontogenetic states would result in an erroneous interpretation of character distributions. Ontogenetic variability in the occurrence of many characters also raises the potential of examining heterochronic shifts in the timing of developmental events as a mechanism of morphological evolution in the ACPT clade.
A link between trait lability and evolutionary innovation?
In summary, morphological and anatomical characters show complex patterns of variation both within ACPT and Portulacineae as a whole. Characters such as leaf succulence, stem woodiness, and periderm timing, to name a few, are highly variable both within and between clades (Tables 3, 5), making interpretation of character evolution and assessment of morphological support for phylogenetic hypotheses both challenging endeavors. However, the distribution of a number of stem traits supports a paraphyletic Pereskia, and furthermore, all of these characters are potentially significant for the development of the stem as a photosynthetic, water-storing organ. The apparent lability of these characters underscores the importance of considering whole organismal context when assigning functional or evolutionary significance to a trait: although all of these traits appear in other lineages across the Portulacineae, it is only in the caulo cacti that we find the combination of a perennial habit, delayed periderm, stem stomata, enlarged stem mucilage cells, a thickened cuticle, and hypodermal calcium oxalate druses. We hypothesize that it is this particular combination of traits, rather than any one in isolation, that provided the necessary organismal setting for the further evolution of the cactus stem.
Many of these traits have unknown adaptive significance when considered in isolation; it is not immediately obvious, for example, what sort of competitive advantage delayed periderm formation brings to a plant with no photosynthetic stem tissue. Similarly, it is not clear what sort of significant costs delayed periderm would present. If many of these traits are more or less adaptively neutral in isolation and are also evolving independently from one another, then it follows that higher evolutionary lability in these traits will increase the likelihood that one lineage will evolve the "right" character states in all of them, thus providing the organismal context that would promote further innovation and eventual functional integration. It is important to note that while these traits are labile across our sample of Portulacineae, they are all highly conserved within the leafless, stem succulent cacti (Gibson and Nobel, 1986; Mauseth and Landrum, 1997; Mauseth, 2005, 2006), indicating that they have become integrated components of the stem photosynthetic and water storage systems. Similar patterns of variation in the tempo and mode of trait evolution during a major morphological transition have been previously reported: for example, the great variation in floral phyllotaxis in ANITA grade and magnoliid angiosperms as compared to the highly conserved whorl found in monocots and eudicots (Endress, 1987; see also Donoghue, 1989, and references therein), or the variation in embryo sac type among the deepest lineages of angiosperms (Friedman, 2006). High trait lability at the base of evolutionary radiations has been noted in the metazoa as well (Simpson, 1944; Gould, 1989; Foote, 1992). These examples highlight Riedl’s notion of burden (Riedl, 1978; Donoghue, 1989) and that the likelihood of character change is dynamic and highly dependent upon that character’s integration (and resulting burden) with other traits.
Homoplasy has been extensively discussed within the context of key innovations, but primarily as an indicator of adaptation (i.e., the repeated evolution of a certain structure in response to a certain habitat illustrates the adaptive nature of that structure) (Armbruster, 1996), or as outlined earlier, as a simple observation that homoplasy of traits may decrease over time as they become fundamentally integrated into a novel structure or pathway. The notion that homoplasy may itself serve as an engine of morphological novelty has only been discussed in a very general way (see Sanderson and Hufford, 1996). Here we propose a potential mechanism: in the sense that any "key innovation" is in reality a sum of many traits, assembling the right combination of those traits in one organism is an essential first step in their integration into a novel structure or function. We suggest that there may be an optimal rate of evolution for these traits that will maximize the likelihood of this initial assembling of characters. In the case of cacti, two prominent developmental enabler traits, stem stomata and delayed bark formation, had the highest lability of any traits in our data set. While we do not currently know the mechanisms that generated high lability in these traits, we hypothesize that initially elevated rates of trait evolution may promote the accumulation of greater combinations of new characters in lineages, which in turn may be a primary driver of evolutionary innovation.
Appendix 1. Taxon names, nomenclatural authorities, and voucher numbers used in this study.
|
FOOTNOTES
1 The authors thank B. Leuenberger (Berlin Botanical Garden), H. Forbes (University of California Botanical Garden at Berkeley), P. Forster (Queensland Herbarium), R. Abbott, P. Schnebelen, and the Royal Botanic Gardens, Kew, for generously providing plant material for this study. They also thank P. Stevens, E. Kellogg, R. Keating, A. Doust, and M. Richardson for guidance in developing the project, comments on the manuscript, and assistance with technical aspects of the work. 
5 Author for correspondence (e-mail address: mogburn{at}brown.edu); present address: Department of Ecology and Evolutionary Biology, Brown University, 80 Waterman St., Box G-W, Providence, Rhode Island 02912 USA
LITERATURE CITED
Applequist, W. L., W. L. Wagner, E. A. Zimmer, AND M. Nepokroeff. 2006. Molecular evidence resolving the systematic position of Hectorella (Portulacaceae). Systematic Botany 31: 310–319.[CrossRef][Web of Science]
Applequist, W. L., AND R. S. Wallace. 2001. Phylogeny of the portulacaceous cohort based on ndhF sequence data. Systematic Botany 26: 406–419.[Web of Science]
Armbruster, W. S. 1996. Exaptation, adaptation, and homoplasy: Evolution of ecological traits in Dalechampia vines. In M. J. Sanderson, and L. Hufford [eds.], Homoplasy: The recurrence of similarity in evolution, 227–243. Academic Press, San Diego, California, USA.
Bailey, I. W. 1964. Comparative anatomy of the leaf-bearing Cactaceae, XI: The xylem of Pereskiopsis and Quiabentia. Journal of the Arnold Arboretum 45: 140–157.
Butterworth, C. A., AND R. S. Wallace. 2005. Molecular phylogenetics of the leafy cactus genus Pereskia (Cactaceae). Systematic Botany 30: 800–808.[CrossRef][Web of Science]
Carlquist, S. 1998. Wood anatomy of Portulacaceae and Hectorellaceae: Ecological, habital, and systematic implications. Aliso 16: 137–153.
Carlquist, S. 1999. Wood, stem, and root anatomy of Basellaceae with relation to habit, systematics, and cambial variants. Flora 194: 1–12.
Chorinsky, F. 1931. Vergleichend-anatomische Untersuchung der Haargebilde bei Portulacaceen und Cactaceen. Österreichische Botanische Zeitschrift 80: 308–327.[CrossRef]
Cuénod, P., V. Savolainen, L. W. Chatrou, M. Powell, R. J. Grayer, AND M. W. Chase. 2002. Molecular phylogenetics of Caryophyllales based on nuclear 18s rDNA and plastid rbcLatpB, and matK DNA sequences. American Journal of Botany 89: 132–144.
De Queiroz, A. 2002. Contingent predictability in evolution: Key traits and diversification. Systematic Biology 51: 917–929.
de Silva, D. L. R., A. M. Hetherington, AND T. A. Mansfield. 1985. Synergism between calcium ions and abscisic acid in preventing stomatal opening. New Phytologist 100: 473–482.[CrossRef][Web of Science]
Donoghue, M. J. 1989. Phylogenies and the analysis of evolutionary sequences, with examples from the seed plants. Evolution 43: 1137–1156.[CrossRef][Web of Science]
Donoghue, M. J. 2005. Key innovations, convergence, and success: Macroevolutionary lessons from plant phylogeny. Paleobiology 31: 77–93.
Donoghue, M. J., AND R. H. Ree. 2000. Homoplasy and developmental constraint: A model and an example from plants. American Zoologist 40: 759–769.[CrossRef][Web of Science]
Edwards, E. J., AND M. J. Donoghue. 2006. Pereskia and the origin of the cactus life-form. American Naturalist 167: 777–793.[CrossRef][Web of Science]
Edwards, E. J., R. Nyffeler, AND M. J. Donoghue. 2005. Basal cactus phylogeny: Implications of Pereskia paraphyly for the transition to the cactus life form. American Journal of Botany 92: 1177–1188.
Endress, P. K. 1987. Floral phyllotaxis and floral evolution. Botanische Jahrbücher für Systematik 108: 417–438.
Esau, K. 1977. Anatomy of seed plants, 2nd ed. Wiley, New York, New York, USA.
Foote, M. 1992. Paleozoic record of morphological diversity in blastozoan echinoderms. Proceedings of the National Academy of Sciences, USA 89: 7325–7329.
Friedman, W. E. 2006. Evidence for developmental lability during early angiosperm evolution. Nature 441: 337–340.[CrossRef][Medline]
Gibson, A. C. 1994. Vascular tissues. In H.-D. Behnke, and T. J. Mabry [eds.], Caryophyllales: Evolution and systematics, 45–74. Springer-Verlag, Berlin, Germany.
Gibson, A. C., AND P. S. Nobel. 1986. The cactus primer. Harvard University Press, Cambridge, Massachusetts, USA.
Goldblatt, P., AND D. E. Johnson. [eds.]. 1979 [onward]. Index to plant chromosome numbers data base. Missouri Botanical Garden, St. Louis, Missouri, USA. Website http://mobot.mobot.org/W3T/Search/ipcn.html [accessed 15 May 2007].
Gould, S. J. 1989. Wonderful life: The Burgess Shale and the nature of history. W.W. Norton, New York, New York, USA.
Herr, J. M. 1992. New uses for calcium chloride solution as a mounting medium. Biotechnic & Histochemistry 67: 9–13.[CrossRef][Web of Science][Medline]
Hershkovitz, M. A. 1993. Revised circumscriptions and subgeneric taxonomies of Calandrinia and Montiopsis (Portulacaceae) with notes on phylogeny of the portulacaceous alliance. Annals of the Missouri Botanical Garden 80: 333–365.[CrossRef][Web of Science]
Hershkovitz, M. A., AND E. A. Zimmer. 1997. On the evolutionary origins of the cacti. Taxon 46: 217–232.[CrossRef][Web of Science]
Keating, R. C. 1996. Anther investigations: A review of methods. In W. G. D’Arcy, and R. C. Keating [eds.], The anther: Form, function and phylogeny, 255–271. Cambridge University Press, New York, New York, USA.
Landrum, J. V. 2002. Four succulent families and 40 million years of evolution and adaptation to xeric environments: What can stem and leaf anatomical characters tell us about their phylogeny? Taxon 51: 463–473.[CrossRef][Web of Science]
Landrum, J. V. 2006. Wide band tracheids in genera of Portulacaceae: Novel, non-xylary tracheids possibly evolved as an adaptation to water stress. Journal of Plant Research 119: 497–504.[CrossRef][Web of Science][Medline]
Leuenberger, B. E. 1986. Pereskia (Cactaceae). Memoirs of the New York Botanical Garden 41: 1–141.
Maddison, W. P., AND D. R. Maddison. 2006. Mesquite: A modular system for evolutionary analysis, version 2.01. Computer program and documentation distributed by the authors, website http://mesquiteproject.org. [accessed 3 March 2008].
Martin, C. E., AND R. S. Wallace. 2000. Photosynthetic pathway variation in leafy members of two subfamilies of the Cactaceae. International Journal of Plant Sciences 161: 639–650.[CrossRef][Web of Science]
Mauseth, J. D. 2004. Wide-band tracheids are present in almost all species of Cactaceae. Journal of Plant Research 117: 69–76.[CrossRef][Web of Science][Medline]
Mauseth, J. D. 2005. Anatomical characters, other than wood, in subfamily Opuntioideae (Cactaceae). Haseltonia 11: 113–125.[CrossRef]
Mauseth, J. D. 2006. Structure–function relationships in highly modified shoots of Cactaceae. Annals of Botany 98: 901–926.
Mauseth, J. D., AND J. V. Landrum. 1997. Relictual vegetative anatomical characters in Cactaceae: The genus Pereskia. Journal of Plant Research 110: 55–64.[CrossRef][Web of Science]
Mauseth, J. D., Y. Uozumi, B. J. Plemons, AND J. V. Landrum. 1995. Structural and systematic study of an unusual tracheid type in cacti. Journal of Plant Research 108: 517–526.[CrossRef][Web of Science]
Metcalfe, C. R., AND L. Chalk. 1950. Anatomy of the dicotyledons. Oxford University Press, London, UK.
Nobel, P. S., AND T. L. Hartsock. 1986. Leaf and stem CO2 uptake in the three subfamilies of the Cactaceae. Plant Physiology 80: 913–917.
Nyffeler, R. 2002. Phylogenetic relationships in the cactus family (Cactaceae) based on evidence from trnK/matK and trnL-trnF sequences. American Journal of Botany 89: 312–326.
Nyffeler, R. 2007. The closest relatives of cacti: Insights from phylogenetic analyses of chloroplast and mitochondrial sequences with special emphasis on the tribe Anacampseroteae. American Journal of Botany 94: 89–101.
Pagel, M., AND A. Meade. 2006. Bayesian analysis of correlated evolution of discrete characters by reversible-jump Markov chain Monte Carlo. American Naturalist 167: 808–825.[CrossRef][Web of Science]
Pfanz, H., G. Aschan, R. Langenfeld-Heyser, C. Wittmann, AND M. Loose. 2002. Ecology and ecophysiology of tree stems: Corticular and wood photosynthesis. Die Naturwissenschaften 89: 147–162.[CrossRef][Web of Science][Medline]
Riedl, R. 1978. Order in living organisms. Wiley, New York, New York, USA.
Ruiz, L. P., AND T. A. Mansfield. 1994. A postulated role for calcium oxalate in the regulation of calcium ions in the vicinity of stomatal guard cells. New Phytologist 127: 473–481.[CrossRef][Web of Science]
Ruzin, S. E. 1999. Plant microtechnique and microscopy. Oxford University Press, New York, New York, USA.
Sanderson, M. J., AND M. J. Donoghue. 1989. Patterns of variation in levels of homoplasy. Evolution 43: 1781–1795.[CrossRef][Web of Science]
Sanderson, M. J., AND L. Hufford. 1996. Homoplasy and the evolutionary process: An afterword. In M. J. Sanderson, and L. Hufford [eds.], Homoplasy: The recurrence of similarity in evolution, 327–330. Academic Press, San Diego, California, USA.
Simpson, G. G. 1944. Tempo and mode in evolution. Columbia University Press, New York, New York, USA.
Sperling, C. 1987. Systematics of the Basellaceae. Ph.D. dissertation. Harvard University, Cambridge, Massachusetts, USA.
Thorne, R. F. 1976. A phylogenetic classification of the Angiospermae. Evolutionary Biology 9: 35–106.
CiteULike Complore Connotea Del.icio.us Digg Facebook Reddit Technorati Twitter What's this?
This article has been cited by other articles:
|
|
 |
|
  D. J. Hearn Developmental patterns in anatomy are shared among separate evolutionary origins of stem succulent and storage root-bearing growth habits in Adenia (Passifloraceae) Am. J. Botany, November 1, 2009; 96(11): 1941 - 1956. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
