Elsevier

Mycological Research

Volume 110, Issue 4, April 2006, Pages 423-430
Mycological Research

Genetic diversity in the blackberry rust pathogen, Phragmidium violaceum, in Europe and Australasia as revealed by analysis of SAMPL

https://doi.org/10.1016/j.mycres.2005.11.014Get rights and content

Abstract

Indigenous to Europe, the blackberry rust fungus Phragmidium violaceum was introduced to Australia and subsequently appeared in New Zealand, with the most recent authorised introductions to Australia specifically for the biological control of European blackberry. Markers for ‘selective amplification of microsatellite polymorphic loci’ (SAMPL) were developed for studying the population genetics of P. violaceum. Modification of one of the two SAMPL primers with a HaeIII adapter (H) revealed significantly greater levels of genetic variation than primers used to generate AFLPs, the latter revealing little or no variation among 25 Australasian and 19 European isolates of P. violaceum. SAMPL was used to describe genetic variation among these 44 isolates of P. violaceum from 51 loci generated using primer pairs (GACA)4 + H–G and R1 + H–G. The European isolates were more diverse than Australasian isolates, with 37 and 22 % polymorphic loci, respectively. Cluster analysis revealed geographic clades, with Australasian isolates forming one cluster separated from two clusters comprising the European isolates. However, low bootstrap support at these clades suggested that Australian isolates had not differentiated significantly from European isolates since the first record of P. violaceum in Australia in 1984. In general, the results support two hypotheses. First, that the population of P. violaceum in Australia was founded from a subset of individuals originating from Europe. Second, that P. violaceum in New Zealand originated from the Australian population of P. violaceum, probably by wind dispersal of urediniospores across the Tasman Sea. The application of SAMPL markers to the current biological control programme for European blackberry is discussed.

Introduction

Population genetics is being applied increasingly to understand the dynamics of plant diseases in time and space (Milgroom & Peever 2003). The selection and deliberate release of exotic plant pathogens, especially rust fungi (Uredinales), for successful biological control of introduced weeds relies on a sound knowledge of epidemiology and host–pathogen interactions. Molecular techniques aid investigation of relationships between weed and pathogen diversity, and can be applied for identifying and monitoring the fate of the released pathogen genotypes with certainty (Evans & Gomez 2004). In the case of European blackberry (Rubus fruticosus aggr.) in Australia, efficient DNA markers will allow the study of evolutionary processes in Phragmidium violaceum, before and after the release of additional strains of this biological control agent.

At least 15 species of European blackberry are naturalised in Australia, where it is considered a ‘weed of national significance’ (Thorp and Lynch, 2000, Evans et al., 2004). It is also considered an important weed in New Zealand, South Africa, North America and Chile (Amor et al. 1998). Phragmidium violaceum, the cause of leaf rust of European blackberry, was first reported in Australia in 1984 and subsequently in New Zealand in 1990, following an unauthorised introduction first identified in Victoria, Australia (Marks et al., 1984, Pennycook, 1998). Subsequently, there were two authorised introductions of P. violaceum for the biological control of European blackberry in Australia. First, strain F15 from France was released throughout south eastern and south western Australia in 1991 and 1992 (Bruzzese & Lane 1996). Second, eight strains of P. violaceum were released in New South Wales, Western Australia and Victoria in 2004 after their collection in France in 2000 (Scott et al. 2002). All spore states of this autoecious, macrocyclic rust fungus can form under Australian conditions (Washington 1985).

Before authorised releases of additional strains of P. violaceum in Australia in 2004, Evans et al. (2000) identified RFLPs using M13 DNA to probe genomic DNA of Australian isolates of P. violaceum digested with the restriction enzyme HaeIII. The robustness of this technique was demonstrated when 13 DNA phenotypes were identified among 18 single-uredinium isolates of P. violaceum from various locations in mainland Australia between 1997 and 1999 (Evans et al. 2000). This DNA marker also identified strain F15 of P. violaceum as being genetically distinct from the 18 isolates screened in the same study. The disadvantage of the RFLP protocol is that it uses relatively large quantities of DNA when compared with a PCR-based marker system. Furthermore, the M13 RFLP technique of Evans et al. (2000) was limited to analysis of restriction fragments in the range 4–9 kbp and additional markers were deemed necessary to expand the proportion of the genome analysed for genetic diversity.

AFLPs are used frequently in studies of genetic diversity in plants and fungi due to their ease of development, efficiency, high multiplex ratio and reproducibility (Zabeau and Vos, 1993, Majer et al., 1996). AFLPs have been demonstrated in at least seven species of Uredinales (Pei and Ruiz, 2000, Samils et al., 2001, Hovmøller et al., 2002, Keiper et al., 2003, Yourman and Luster, 2004). A modification of the AFLP procedure called ‘selective amplification of microsatellite polymorphic loci’ (SAMPL, Morgante & Vogel 1994) combines the advantages of AFLP with the analysis of highly variable microsatellite regions of eukaryotic genomes. SAMPL has been reported to be more powerful than AFLPs in discriminating between closely related individuals in several plant complexes (Paglia and Morgante, 1998, Roy et al., 2002, Singh et al., 2002, Tosti and Negri, 2002) and in the rust pathogen, Puccinia striiformis f. sp. tritici, which is strictly clonal (Stubbs, 1985, Keiper et al., 2003).

We report the development of a modified SAMPL technique using adapters linked to the blunt end restriction enzyme, HaeIII, following on from the useful resolution identified by RFLPs using this same enzyme (Evans et al. 2000). The utility of SAMPL, relative to AFLP, in describing genetic differentiation within and among populations of P. violaceum from Europe and Australasia, is demonstrated.

Section snippets

Origin of isolates

Throughout this paper, individual collections of Phragmidium violaceum will be called isolates, whereas isolates that have been characterised for the purpose of biological control will be called strains. A total of 44 single uredinium-derived isolates of P. violaceum were analysed through the course of developing markers for the fungus (Table 1). Eighteen single uredinium-derived isolates of P. violaceum from Australia originated from collections made by Evans et al. (2000). For the purpose of

AFLP and modified AFLP

No polymorphic loci were observed among DNA preparations of Phragmidium violaceum isolates SA1, V1 and V2 using 20 MseI/EcoRI AFLP primer pairs, each of which generated at least 25 scoreable loci. Loci were also monomorphic among DNA from the seven reference isolates of P. violaceum (F15, SA1, V1, V2, WA9, WA12 and WP69) using primer pairs E–AC + M–C, E–0 + M–0 or E–0 + M–T. The use of primer pair E–0 + M–0 to detect AFLPs among 44 European and Australasian isolates revealed a single 300 bp polymorphic

Discussion

This study highlights the importance not only of screening AFLP primers for polymorphisms but also of choosing restriction enzymes that are polymorphic for the species being studied, as practised during development of RFLP markers. In contrast to the application of AFLP for genetic analysis of other Uredinales by means of commonly-used linkers and primers (e.g. Hovmøller et al. 2002), genetic diversity was not detected in Phragmidium violaceum using AFLP. This result was inconsistent with our

Acknowledgements

The Cooperative Research Centre for Australian Weed Management supported this research and the PhD scholarship of D.R.G. Research in New Zealand (NZ) was funded by the NZ Forest Health Research Collaborative and the Bay of Plenty Regional Council, NZ. Special thanks go to staff of Landcare Research, NZ, including Paula Wilkie for technical assistance, Duckchul Park for extraction of DNA and David Glenny for identification of blackberry species in NZ. We also thank John K. Scott and Andy

References (35)

  • T.K. Crosby et al.

    Area codes for recording specimen localities in the New Zealand subregion

    New Zealand Journal of Zoology

    (1998)
  • K.J. Evans et al.

    Genetic markers in rust fungi and their application to weed biocontrol

  • K.J. Evans et al.

    DNA phenotypes of the blackberry biological control agent, Phragmidium violaceum, in Australia

    Australasian Plant Pathology

    (2000)
  • K.J. Evans et al.

    Taxonomic update and Lucid key for introduced blackberry in Australia

  • K.J. Evans et al.

    Rubus anglocandicans (Rosaceae) is the most widespread taxon of blackberry in Australia

    Australian Systematic Botany

    (2003)
  • A. Gustafsson

    The genesis of the European blackberry flora

    Acta Universitas Lundensis

    (1943)
  • M.S. Hovmøller et al.

    Clonality and long-distance migration of Puccinia striiformis f. sp. tritici in north-west Europe

    Plant Pathology

    (2002)
  • Cited by (13)

    • Applying molecular-based approaches to classical biological control of weeds

      2011, Biological Control
      Citation Excerpt :

      A “trap garden” was established in France, containing a representative selection of the blackberry genotypes present in Australia, to source new strains of the rust (Scott et al., 2002). Molecular analysis using SAMPL showed that the strains recovered from the trap garden were genetically different from those already released and established in Australia and New Zealand (Gomez et al., 2006). A suitable molecular technique was required to follow the fate of the new strains released, since they could not easily be distinguished from those already present in the field.

    View all citing articles on Scopus
    View full text