Allozyme and DNA sequence comparisons of nine species of Encephalartos (Zamiaceae)

https://doi.org/10.1016/S0305-1978(00)00064-8Get rights and content

Abstract

Phylogenetic relationships between Encephalartos altensteinii Lehmann, E. arenarius R.A. Dyer, E. horridus (Jacquin) Lehmann, E. latifrons Lehmann, E. lehmannii Lehmann, E. longifolius (Jacquin) Lehmann, E. princeps R.A. Dyer and E. trispinosus (Hooker) R.A. Dyer were studied, using E. ferox Bertoloni f. as outgroup. Three continuous and one discontinuous buffer systems were used and gene products of 14 enzyme coding loci were examined by horizontal starch gel-electrophoresis. Genetic variation was studied in a cultivated population of E. lehmannii and the average heterozygosity value for this population is 13.5%, which falls within the range reported for other cycad species. Fixed allele differences between the species studied was not found at any of the loci studied, which suggest that these species are closely related. DNA sequence analysis of rbcL and ITS 1 & 2 genes (1428 and 895 basepairs, respectively) confirmed the close genetic relationships between these taxa. According to ITS and rbcL sequences E. altensteinii and E. princeps are sibling taxa which form a sister group to E. arenarius, E. horridus, E. latifrons, E. lehmannii, E. longifolius, and E. trispinosus. The genetic distances between both groups were 0.12–0.47% for ITS and 0.08–0.16% for rbcL DNA. The results indicate recent (probably pleistocenic) speciation for this group of cycads, and the relationships are discussed with reference to affinities based on morphology and distribution.

Introduction

The taxonomy of the Cycadales is problematic because really distinctive morphological characteristics are lacking in most groups, and species are usually defined in terms of sets of characteristics. Furthermore, the characteristics used for identification may not be of phylogenetic significance due to a high degree of convergence or divergence. It also seems reasonable to assume that at least some of the species in a given geographical area are closely related (i.e. derived from a common ancestor) due to slow reproduction rates (Tang, 1990) and absence of long-distance dispersal mechanisms for seeds. In the Eastern Cape Province of South Africa eight of the 13 known Encephalartos species are thought to be closely related. These are Encephalartos arenarius R.A. Dyer, E. horridus (Jacquin) Lehmann, E. latifrons Lehmann, E. lehmannii Lehmann, E. longifolius (Jacquin) Lehmann, E. princeps R.A. Dyer, and E. trispinosus (Hooker) R.A. Dyer, with E. altensteinii Lehmann thought to be somewhat more distantly related. These are the species used for this study. The remaining species, E. caffer (Thunberg) Lehmann, E. cycadifolius (Jacquin) Lehmann, E. friderici-guilielmi Lehmann, E. ghellinckii Lemaire, and E. villosus Lemaire, most certainly belong to other, more distantly related groups within the genus. The studied species have either or both very strongly dentate (almost dentate–lobed) leaflets and very glaucous leaves (Dyer (1965a), Dyer (1965b)). Such dentate leaflets are not known anywhere else in the distribution area of the genus, and glaucous leaves are uncommon elsewhere (Vorster (1993), Vorster (1999)). Apart from these two characteristics, these eight species vary greatly in respect of growth form (development of an aerial trunk) and cone morphology. The development of an aerial trunk may well be a rather plastic (albeit at present genetically fixed) reaction to local environmental conditions, as the habitat becomes markedly drier further inland from the coast. The cones, being the reproductive structures, could be expected to provide a more reliable indication of phylogenetic relationships than vegetative characteristics. However, cone morphology varies considerably, and cannot be satisfactorally correlated with vegetative morphological features (Vorster, 1993).

The purpose of this study is to compare the eight species of Encephalartos listed above in respect of DNA sequence (ITS 1 & 2 and rbcL) and allozyme data to determine whether they form a phylogenetically coherent group and to determine the relationships between the species. Encephalartos ferox Bertoloni f. is included as outgroup species. We have also analysed a cultivated population of E. lehmannii to compare within- and between species allozymic variation.

Section snippets

Allozyme study

Electrophoretic data for 126 individuals, from nine species (Table 1), were compared. We used cultivated plants, firstly because of the remote locations of natural populations, and secondly because plants in nature have become very scarce due to illicit collecting for the ornamental plant trade. We have, however, used only plants of undisputed identity, and with collection data where possible. When sampling different specimens of the same species, we took care not to sample a clone twice.

Allozyme study

Locus abbreviations, enzyme commission numbers, and monomorphic loci are listed in Table 2. Allele products at the following loci were monomorphic: GPI-1, MDH, PGDH, PGM-1, -2, PEPD and PEP-LT1, and those of ACP and PER migrated cathodally. Allele frequencies for polymorphic loci are presented in Table 3, and allozyme phenotypes of putative heterozygotes were congruent with those expected on the basis of the quaternary structure of the enzyme (Ward, 1977).

Conclusions

Encephalartos ferox, selected as outgroup, is not closely related to any of the other species investigated. However, the allozyme and DNA sequence data implicate close and recent evolutionary relationships between the ingroup taxa, and E. altensteinii is separated from the other Eastern Cape species studied by both allozyme and DNA data, as shown by its basal position in all the trees (Fig. 2, Fig. 3).

Encephalartos arenarius, in spite of its very strong vegetative resemblance to E. latifrons,

Acknowledgements

We would like to thank Susan Myburg from the University of Pretoria and the curator of Kirstenbosch National Botanical Gardens for cycad samples.

References (41)

  • J.S Farris

    Estimating phylogenetic trees from distance matrices

    Am. Nat.

    (1972)
  • G.G Goncharenko et al.

    Population structure, gene diversity and differentiation in a natural population of Ceder pines (Pinus subsect. Cembae, Pinaceae) in SSR

    Pl. Syst. Evol.

    (1992)
  • L.D Gottlieb

    Electrophoretic evidence and plant populations

  • Hamrick, J.L., Mitton, J.B., Linhart, Y.B., 1979. Level of genetic variation in trees: influence of life history...
  • S.R Kephart

    Starch gel electrophoresis of plant isozymesa comparative analysis of techniques

    Am. J. Bot.

    (1990)
  • C.L Markert et al.

    Lactate dehydrogenase isozyme patterns of fish

    J. Exp. Zool.

    (1965)
  • M Nei

    Genetic distance between populations

    Am. Nat.

    (1972)
  • M Nei

    Estimation of average heterozygosity and genetic distance from a small number of individuals

    Genetics

    (1978)
  • M Nei

    Definition and estimation of fixation indices

    Evolution

    (1986)
  • M Nei et al.

    Accuracy of estimated phylogenetic trees from molecular data

    J. Mol. Evol.

    (1983)
  • Cited by (23)

    • Species relationships and phylogenetic diversity of the African genus Encephalartos Lehm. (Zamiaceae)

      2023, South African Journal of Botany
      Citation Excerpt :

      The results of the phylogenetic tree also confirm that Encephalartos is a monophyletic group (PP=1 / UFB=100) – agreeing with previous work (van der Bank et al., 2001; Vorster et al., 2004; Treutlein et al., 2005; Rousseau, 2012; Mankga et al., 2020). The earliest phylogenies produced by van der Bank et al. (2001) and Vorster et al. (2004) included nine and 10 taxa, respectively and were able to show a clear distinction between Encephalartos ferox (clade five) and species from the eastern Cape (clade seven). The limited number of species used within the matrix did not provide enough resolution to differentiate between species or to determine species groups.

    View all citing articles on Scopus
    View full text