Embryology of Koeberlinia (Koeberliniaceae): Evidence for Core-Brassicalean Affinities Author(S): Hiroshi Tobe and Peter H
Total Page:16
File Type:pdf, Size:1020Kb
Embryology of Koeberlinia (Koeberliniaceae): Evidence for core-Brassicalean affinities Author(s): Hiroshi Tobe and Peter H. Raven Source: American Journal of Botany, Vol. 95, No. 11 (November 2008), pp. 1475-1486 Published by: Wiley Stable URL: https://www.jstor.org/stable/41922997 Accessed: 06-07-2021 08:55 UTC REFERENCES Linked references are available on JSTOR for this article: https://www.jstor.org/stable/41922997?seq=1&cid=pdf-reference#references_tab_contents You may need to log in to JSTOR to access the linked references. JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact [email protected]. Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at https://about.jstor.org/terms Wiley is collaborating with JSTOR to digitize, preserve and extend access to American Journal of Botany This content downloaded from 86.59.13.237 on Tue, 06 Jul 2021 08:55:17 UTC All use subject to https://about.jstor.org/terms American Journal of Botany 95(1 1): 1475-1486. 2008. Embryology of Koeberlinia (Koeberliniaceae): Evidence for core-Brassic alean affinities1 Hiroshi Tobe24 and Peter H. Raven3 department of Botany, Graduate School of Science, Kyoto University, Kyoto 606-8502, Japan; and 3Missouri Botanical Garden, P. O. Box 299, St. Louis, Missouri 63166 USA Koeberlinia , comprising a single xerophytic species K spinosa , had previously been placed in various families, mainly Cap- paraceae. Current molecular evidence now places it in its own family Koeberliniaceae, thought to be related to the Bataceae/Sal- vadoraceae among the 17 other families of the Brassicales. We investigated 55 embryological characters of the genus, most of which are not understood yet, and thereby assessed its systematic relationships. Koeberlinia has many embryological features in common with the Capparaceae and seven other core-Brassicalean families (i.e., Brassicaceae, Cleomaceae, Emblingiaceae, Gy- rostemonaceae, Pentadiplandraceae, Resedaceae, and Tovariaceae), specifically by possessing a campylotropous ovule with a nonmultiplicative (two-cell-layered) outer integument, reniform seeds with a curved embryo, and a fibrous exotegmen in the ma- ture seed coat. However, Koeberlinia is clearly distinguished from them by a tenuinucellate rather than crassinucellate ovule as previously reported, markedly enlarged apical nucellar epidermal cells, and an "exotestal" seed coat. Embryologically, Koeber- linia resembles neither the Bataceae nor the Salvadoraceae, although only limited embryological data are available for these two families. Embryological evidence thus favors its joining the core Brassicales, but additional molecular analyses and embryological studies on the missing data of the Bataceae and Salvadoraceae are needed for final confirmation of its phylogenetic position. Key words: Brassicales; embryology; Koeberlinia ; Koeberliniaceae; ovule; seed. Koeberlinia , a genus consisting of the single leafless sificationxero- (Haston et al., 2007; see also APG website [Stevens, phytic species K. spinosa Zucc., occurs in the southwestern 2001 onward]). United States and adjacent Mexico (Airy Shaw, 1973). In The the purpose of this paper is to embryologically character- past, it had been assigned to various families: e.g., Pittospo- ize Koeberlinia by examining the development of the anther, raceae, Rutaceae, Simaroubaceae, and Capparaceae (for review, ovule, and seed. These characters will be used to demonstrate see Mehta and Moseley, 1981). Importantly, Koeberlinia the, like embryological similarities or dissimilarities compared to the Capparaceae and related families, was shown to have the my- other families of the Brassicales, particularly the Cappar- rosin cells (Gibson, 1979), a feature formerly thought to aceae, occur Bataceae, and Salvadoraceae, within an overall frame- only in taxa producing glucosinolates. In the 1980s, despitework of phylogenetic relationships. Several embryological Koeberlinia not producing glucosinolates unlike other mem- features of Koeberlinia have been reported by Mehta and bers of the Capparaceae, a general agreement held that the Moseley two (1981: table 2), including a multinucleate anther ta- were closely related; alternatively, Koeberlinia was regarded petum, as a fibrous endothecium, two-celled pollen grains at the a rather isolated or aberrant genus (often a distinct subfamily time or of shedding, anatropous and crassinucellate ovules with tribe) in that family (Cronquist, 1981, 1988; Dahlgren, 1983;a zigzag micropyle formed by both integuments, and a linear Thorne 1983, 1992; Takhtajan, 1986, 1987) or in its own familytetrad of megaspores. However, many other features, particu- Koeberliniaceae near the Capparaceae (Mehta and Moseley, larly of the seeds and seed coats that are of critical importance 1981). On the basis of morphological or molecular analyses, in considering relationships with other Brassicalean families Rodman (1991) and Rodman et al. (1993, 1996, 1998) accepted (see Tobe and Raven, 1991) are reported here for the first Koeberlinia as belonging to the distinct family Koeberliniaceae time. Additionally, several corrections to the previous data are in the order Brassicales ("Capparales sensu lato"), suggesting reported here including the following: the number of nuclei in its sister-group relationship with Bataceae + Salvadoraceae a tapetai in a cell is two rather than multinucleate, the orientation position just outside of the core Brassicales (Brassicaceae, of the mature ovule is campylotropous rather than anatropous, Capparaceae, Cleomaceae, Emblingiaceae, Gyrostemonaceae, and the structure of the nucellus is tenuinucellate rather than Pentadiplandraceae, Resedaceae, and Tovariaceae) (see crassinucellate. also Hall et al., 2004; Ronse de Craene and Haston, 2006; Fig. 1). The family Koeberliniaceae has now been adopted by the APG MATERIALS AND METHODS II (Angiosperm Phylogeny Group, 2003) and post APG II clas- Flower buds and fruits of Koeberlinia spinosa in various stages of devel- opment were collected from several localities in Arizona and Texas in the 1 Manuscript received 28 June 2008; revision accepted 25 August 2008. United States (vouchers: Baker 7745 at the herbarium of Arizona State Uni- The authors thank M. Baker, E. Joyal, L. R. Landrum, B. Parffit, versityand A. [= ASU], Landrum and Parfitt 6228 [ASU], Powell 5571 at the her- M. Powell for collecting the fixed materials used in this study. The barium study of Sul Ross State University). They were fixed in FAA (five parts was supported by a Grant-in-Aid for Scientific Research from the stock Japan formalin; five parts glacial acetic acid; 90 parts 50% ethanol). Flower Society for the Promotion of Science (No. 18370036) and by the buds Global and seeds were dehydrated through a /-butyl alcohol series, and then COE Program A06 to Kyoto University. embedded in Paraplast with a melting point (mp) of 57-58°C for sectioning. 4 Author for correspondence (e-mail: [email protected]) Serial sections cut to a thickness of 8-10 |im were stained with Heidenhain' s hematoxylin, safranin, and fast-green FCF, and mounted in Entellan (Merck, doi:10.3732/ajb.0800218 Darmstadt, Germany). Paraffin-embedded seeds were softened by a mixture 1475 This content downloaded from 86.59.13.237 on Tue, 06 Jul 2021 08:55:17 UTC All use subject to https://about.jstor.org/terms 1476 American Journal of Botany [Vol. 95 Meiosis in a microspore mother cell is accompanied by si- multaneous cytokinesis (Fig. 2E), and the resultant microspore tetrads are predominantly tetrahedral. Pollen grains are two- celled at the time they are shed (Fig. 2G) as reported by Mehta and Moseley (1981: table 2). Integuments - The ovule is bitegmic (Figs. 3A, I, 4A, B), as reported by Mehta and Moseley (1981: Table 2). Both the inner and the outer integument are derived from dermal cells of the nucellus (Fig. 3A) and initially have two cell layers. The inner cell layer of the inner integument undergoes periclinal division, resulting in a three-cell-layered inner integument. The outer in- tegument remains two-cell-layered throughout ovule develop- ment. The micropyle is formed by both the inner and the outer integument, as shown by Mehta and Moseley (1981: table 2). Neither the inner integument nor the outer integument has vas- cular bundles. Megagametophyte and nucellus - The archesporium is three- to six-celled, differentiating beneath the apical dermal layer of the nucellus (Fig. 3A). The archesporial cells enlarge and dif- ferentiate directly into megaspore mother cells (Fig. 3B). The ovule is therefore tenuinucellate, not crassinucellate as reported by Mehta and Moseley (1981: table 2). It is always the case, however, that only one megaspore mother cell undergoes mei- otic division and gives rise to a dyad (Fig. 3C) and finally a linear tetrad (rarely triad) of megaspores. In the megaspore tetrad, the chalazal megaspore is functional, while the three upper me- gaspores degenerate (Fig. 3D). A functional megaspore develops successively into a two- (Fig. 3E), four- (Fig. 3F), and eight-nu- cleate embryo sac (Fig. 3G, H). Therefore, the mode of embryo Fig. 1 . Phylogenetic tree sac formationof the is of the Brassicales, Polygonum type. Three antipodal indicating cells are a position