OUP user menu

Multiple gene genealogies reveal important relationships between species of Phaeophleospora infecting Eucalyptus leaves

Vera Andjic, Giles E. StJ. Hardy, Maria Noel Cortinas, Michael J. Wingfield, Treena I. Burgess
DOI: http://dx.doi.org/10.1111/j.1574-6968.2007.00637.x 22-33 First published online: 1 March 2007


The majority of Eucalyptus species are native to Australia, but worldwide there are over 3 million ha of exotic plantations, especially in the tropics and subtropics. Of the numerous known leaf diseases, three species of Phaeophleospora can cause severe defoliation of young Eucalyptus; Phaeophleospora destructans, Phaeophleospora eucalypti and Phaeophleospora epicoccoides. Phaeophleospora destructans has a major impact on seedling survival in Asia and has not, as yet, been found in Australia where it is considered a serious threat to the biosecurity of native eucalypts. It can be difficult to distinguish Phaeophleospora species based on symptoms and micromorphology and an unequivocal diagnostic tool for quarantine purposes would be useful. In this study, a multiple gene genealogy of these Phaeophleospora species and designed specific primers has been constructed to detect their presence from leaf samples. The phylogenetic position of these Phaeophleospora species within Mycosphaerella was established. They are closely related to each other and to other important Eucalyptus pathogens, Mycosphaerella nubilosa, Mycosphaerella cryptica and Colletogloeopsis zuluensis. The specific primers developed can now be used for diagnostic and screening purposes within Australia.

  • species specific primers
  • diagnostics
  • biosecurity
  • quarantine
  • native forests
  • Eucalyptus plantations


Eucalyptus species are highly favoured for the establishment of plantations. This is due to their rapid growth, ease of cultivation and their adaptation to a wide variety of different growing conditions (Turnbull, 2000). The timber of these trees is an important source of fibre for the international paper and pulp industry (Turnbull, 2000). In Australia, plantation forestry is rapidly increasing in size (National Forestry Inventory, 2004) and a number of fungal foliar pathogens have been reported to impact negatively on yields of these plantations. Among the most important of these pathogens are Mycosphaerella spp. (Carnegie et al., 1997; Park et al., 2000; Barber et al., 2003; Maxwell et al., 2003) and their incidence and severity is increasing as the areas under cultivation expand (Park et al., 2000; Maxwell et al., 2003).

Phaeophleospora Rangel is an anamorph genus assigned to some species of Mycosphaerella (Crous, 1998; Crous et al., 2001, 2004; Maxwell et al., 2003). Six species are known to cause disease on leaves of Eucalyptus species These are Phaeophleospora epicoccoides (Cooke & Massee) Crous, Ferreira & Sutton, Phaeophleospora destructans (MJ Wingf & Crous) Crous, Ferreira & Sutton, Phaeophleospora eucalypti (Cooke & Massee) Crous, Ferreira & Sutton, Phaeophleospora lilianie (Walker, Sutton & Pascoe) Crous, Ferreira & Sutton, Phaeophleospora delegatensis Park & Keane (Crous, 1998) and the recently described Phaeophleospora toledana Crous & G. Bills (Crous et al., 2004). Of these species, P. epicoccoides, P. destructans and P. eucalypti are considered important pathogens (Park et al., 2000). Phaeophleospora lilianie has been found only on yellow bloodwood (Eucalyptus eximia) in New South Wales and little is known regarding its importance (Chippendale, 1988). Phaeophleospora delegatensis is the anamorph of Mycosphaerella delegantesis (Park & Keane, 1984) isolated from the leaves of Eucalyptus delegantensis and Eucalyptus obliqua in Australia. It occasionally causes premature defoliation if the infection levels are severe. Both P. liliane and P. delegatensis have poor survival in culture and they have thus have never been successfully stored. Phaeophleospora toledana is the anamorph of Mycosphaerella toledana (Crous et al., 2004) named for its location of origin and it is not considered as a serious leaf pathogen.

Phaeophleospora destructans is an aggressive and often devastating pathogen that causes distortion of infected leaves and blight of young leaves, buds and shoots (Wingfield et al., 1996). This pathogen was first discovered in Indonesia in 1996 and has subsequently spread to Thailand, China, Vietnam and Timor (Old et al., 2003a, b; Barber 2004; Burgess et al., 2006). While most Phaeophleospora species infecting Eucalyptus leaves are known from Australia, P. destructans, the most pathogenic of these fungi has not been found in this country. Thus, the potential impact of P. destructans on native eucalypt forests is unknown, but of concern.

Phaeophleospora epicoccoides is the anamorph of Mycosphaerella suttoniae (Crous et al., 1997) and it occurs worldwide infecting almost all eucalypt species (Sankaran et al., 1995). This species is well known on native Eucalyptus species in Australia and it has most likely been spread to other countries with germ-plasm used to establish plantations. Phaeophleospora epicoccoides is a relatively weak pathogen typically infecting older leaves and stressed trees (Knipscheer et al., 1990). Phaeophleospora eucalypti, a native pathogen in Australia, has in the past resulted in complete defoliation of juvenile leaves of Eucalyptus nitens in New Zealand, the only country where it is known to have been introduced (Dick, 1982; Hood et al., 2002a, b). The worst affected E. nitens stands in New Zealand are currently being converted back to farmland (Hood et al., 2002b).

The appearance and severity of lesions on Eucalytpus leaves are generally used to recognize the species of Phaeophleospora responsible for disease. However, depending on host and climate, the symptoms associated with infection by P. epicoccoides, P. eucalypti and P. destructans can be almost identical (Fig. 1) and incorrect diagnosis is a common problem. In addition, identification of P. eucalypti and P. destructans based on conidial morphology can be difficult because spore size varies depending on host species. A simple and accurate molecular diagnostic technique to distinguish between these important species would compliment traditional morphological diagnosis.

Figure 1

Comparison of symptoms produced on juvenile Eucalyptus grandis leaves infected with (a) Phaeophleospora destructans, (b) Phaeophleospora eucalypti and (c) Phaeophleospora epicoccoides showing the similarity of symptoms associated with these fungi.

The aim of this study was to construct multiple gene genealogies for P. epicoccoides, P. destructans and P. eucalypti, the most common and destructive species occurring on Eucalyptus. Thus, partial sequences for six protein coding genes were generated to elucidate the phylogenetic relationships between these Phaeophleospora species. Following the construction of the phylogenies, species specific primers were then designed for diagnostic purposes.

Materials and methods

Fungal isolates

Phaeophleospora species were isolated under a dissecting microscope by collecting conidia exuding from single pycnidia, on the tip of a sterile needle. The spores were placed on malt extract (20 g L−1) agar (MEA), in a single spot and allowed to hydrate for 5 min. Conidia were then drawn across the agar surface with a sterile needle and single spores were picked off the agar and transferred to new MEA plates. Spores were left to germinate, which usually occurred within 24 h. Cultures were maintained at 20°C on MEA. Isolates made for this study were compared with those of other closely related species (Table 1). All isolates are maintained in the collection (CMW) of the Forestry and Agricultural Biotechnology Institute (FABI), University of Pretoria, South Africa or the Murdoch University culture collection (MUCC), Perth, Western Australia.

View this table:
Table 1

Species and isolates considered in the phylogenetic study

GenBank accession nos
Culture no.TeleomorphAnamorphHostLocationCollectorITSβ-tubulinEF-1αCHS
STE-U 1454 CMW 5351Phaeophleospora eugeniaeEugenia unifloraBrazilMJ WingfieldAF309613 DQ632710
STE-U1366 CMW 5219P. destructansEucalyptus grandisSumatra, IndonesiaMJ WingfieldAF309614 DQ632699
CMW 7127P. destructansEucalyptus spSumatra, IndonesiaMJ WingfieldDQ632698
CMW 19906P. destructansE. grandisSumatra, IndonesiaPA BarberDQ632700
CMW 22553P. destructansE. grandisSumatra, IndonesiaPA BarberDQ632667DQ632625DQ632732DQ632646
CMW 17918P. destructansE. grandisSumatra, IndonesiaPA BarberDQ632666DQ632624DQ632731DQ632645
CMW 19832P. destructansE. grandisSumatra, IndonesiaPA BarberDQ632665DQ632623DQ632730DQ632644
CMW 17919P. destructansE. urophyllaGuangzhou, ChinaTI BurgessDQ632701DQ632622DQ632729DQ632643
MUCC 433P. eucalyptiE. nitensVictoria, AustraliaPA BarberDQ632661DQ632631DQ632726DQ632650
CMW 17915P. eucalyptiE. nitensVictoria, AustraliaPA BarberDQ632664DQ632626DQ632727DQ632653
MUCC 432P. eucalyptiE. grandis x E. tereticornisNew South WalesAJ CarnegieDQ632660DQ632627DQ632724DQ632648
MUCC 434P. eucalyptiE. grandis x E. tereticornisNew South WalesAJ CarnegieDQ632662DQ632632DQ632728DQ632651
CMW 17917P. eucalyptiE. grandis x E. tereticornisNew South WalesAJ CarnegieDQ632711DQ632630DQ632725DQ632649
MUCC 435P. eucalyptiE. grandis x E. camaldulensisQueenslandAJ CarnegieDQ632663DQ632629DQ632723DQ632652
CMW 17916P. eucalyptiE. grandis x E. camaldulensisQueenslandAJ CarnegieDQ632659DQ632628DQ632722DQ632647
CMW 11687P. eucalyptiE. nitensNew ZealandM DickDQ240001DS890168DQ235115DQ890167
NZFS85C/23P. eucalyptiE. nitensNew ZealandM DickAY626988
NZFS85C/1P. eucalyptiE. nitensNew ZealandM DickAY626987
MUCC 422M. suttoniaeP. epicoccoidesE. grandis x E. camaldulensisQueenslandG HardyDQ632656
MUCC 424M. suttoniaeP. epicoccoidesE. grandis x E. camaldulensisQueenslandG HardyDQ632703DQ632617DQ632712DQ632633
MUCC 428M. suttoniaeP. epicoccoidesE. grandis x E. camaldulensisQueenslandTI BurgessDQ632707DQ632618DQ632717DQ632638
MUCC 430M. suttoniaeP. epicoccoidesE. grandisQueenslandG WhyteDQ632708
MURU 327M. suttoniaeP. epicoccoidesE. globulusWestern AustraliaS JacksonDQ632702DQ632619DQ632716DQ632639
MUCC 426M. suttoniaeP. epicoccoidesE. globulusWestern AustraliaS JacksonDQ632704DQ632620DQ632715DQ632637
CMW 22482M. suttoniaeP. epicoccoidesE. grandisSumatra, IndonesiaPA BarberDQ632658DQ632621DQ632719DQ632636
MUCC 425M. suttoniaeP. epicoccoidesE. grandisNew South WalesTI BurgessDQ632655DQ632613DQ632713DQ632634
MUCC 429M. suttoniaeP. epicoccoidesE. grandisNew South WalesTI BurgessDQ530226
MUCC 431M. suttoniaeP. epicoccoidesE. grandisNew South WalesTI BurgessDQ530227
CMW 22484M. suttoniaeP. epicoccoidesE. urophyllaChinaTI BurgessDQ632705DQ632616DQ632714DQ632635
CMW 22486M. suttoniaeP. epicoccoidesE. urophyllaChinaTI BurgessDQ632706DQ632615DQ632720DQ632642
CMW 17920M. suttoniaeP. epicoccoidesE. urophyllaChinaTI BurgessDQ632654DQ632612DQ632721DQ632641
CMW 22483M. suttoniaeP. epicoccoidesE. grandisIndonesiaPA BarberDQ632709
CMW 5348 STE-U 1346M. suttoniaeP. epicoccoidesEucalyptus sp.IndonesiaMJ WingfieldAF309621DQ240117DQ240170DQ890166
SA12M. suttoniaeP. epicoccoidesE. fragrataSouth AfricaMN CortinasDQ632657DQ632614DQ632718DQ632640
STE-U 10840CPC 10840M. toledanaP. toledanaE. globulusSpainPW CrousAY725580
CBS 113313 CMW 14457M. toledanaP. toledanaE. globulusSpainPW CrousAY725581DQ658235DQ235120DQ658226
AMR 051M. nubilosaE. globulusWestern AustraliaA MaxwellAY509777
AMR 057M. nubilosaE. globulusWestern AustraliaA MaxwellAY509778
CMW 11560M. nubilosaE. globulusTasmaniaA MilgateDQ658232DQ658236DQ240176DQ658230
CMW 6211M. nubilosaE. nitensSouth AfricaG HunterAF449094
CMW 9003M. nubilosaE. nitensSouth AfricaG HunterAF449099
AMR 118M. crypticaColletogloeopsis nubilosumE. globulusWestern AustraliaA MaxwellAY509753
AMR 115M. crypticaC. nubilosumE. globulusWestern AustraliaA MaxwellAY509754
CMW 3279M. crypticaC. nubilosumE. globulusAustraliaAJ CarnegieAY309623DQ658234DQ235119DQ658225
CMW 4915C. zuluensisE. grandisSouth AfricaMJ WingfieldAY244421
CBS 117262 CMW 7449C. zuluensisE. grandisSouth AfricaL Van ZylDQ240021DQ240102DQ240155DQ658224
CBS 113399 CMW 13328C. zuluensisE. grandisSouth AfricaL Van ZylDQ240018DQ658233DQ240172DQ658223
CBS 110499 CMW 13704M. ambiphyllaPhaeophleospora sp.E. globulusWestern AustraliaA MaxwellAY150675DQ240116DQ240169DQ658229
STE-U 784M. mollerianaC. mollerianaEucalyptus sp.USAAF309619
CMW 4940 CPC1214M. mollerianaC. mollerianaEucalyptus sp.PortugalMJ WingfieldDQ239969DQ240115DQ240168DQ658228
A/1/8M. vespaConiothyrium ovatumEucalyptus sp.TasmaniaA MilgateAY045499
CMW 11588M. vespaCo. ovatumE. globulusTasmaniaA MilgateDQ239968DQ240114DQ240167DQ658227
CMW 6210M. vespaCo. ovatumE. globulusNew South WalesMJ WingfieldAF449095
CBS 110906Coniothyrium sp.E. cladocalyxSouth AfricaPW CrousAY725513
CBS 111149Coniothyrium sp.E. cladocalyxSouth AfricaPW CrousAY725514
CBS 113621Coniothyrium sp.E. cladocalyxSouth AfricaPW CrousAY725515
CBS 116427Coniothyrium sp.Eucalyptus sp.South AfricaPW CrousAY725516
CPC 18Coniothyrium sp.E. cladocalyxSouth AfricaPW CrousAY725517
CBS 116428Coniothyrium sp.Eucalyptus sp.South AfricaPW CrousAY725518
CBS 113265CMW 13333M. punctiformisRamularia endophyllaQuercus roborNetherlandsAY490763
CMW 9091M. marksiiPseudocercospora epispermogoniaEucalyptus sp.South AfricaG HunterAF468871
STE-U 796 CBS 680.95M. africanaEucalyptus sp.South AfricaPW CrousAF173314
STE-U 1084M. keniensisEucalyptus sp.KenyaMJ WingfieldAF173300
CBS 110500AMR 221M. aurantiaE. globulusWestern AustraliaA MaxwellAF509743
CBS 110969 STE-U1106M. colombiensisPs. colombiensisEucalyptus sp.ColombiaMJ WingfieldAF309612
CBS 110503 AMR 251M. parvaE. globulusWestern AustraliaA MaxwellAF509782
NZsM. suberosaA MilgateAY045503
CBS 110949M. ohnowaE. grandisSouth AfricaMJ WingfieldAY725575
STE-U 1225M. ellipsoideaUwebraunia ellipsoideaEucalyptus sp.South AfricaMJ WingfieldAF173303
CMW 9098M. ellipsoideaU. ellipsoideaEucalyptus sp.South AfricaMJ WingfieldAF468874
CMW 7774Botryosphaeria obtusaRibes sp.New York, USAB SlippersAY236953
CMW 7773B. ribisRibes sp.New York, USAB SlippersAY236936AY808170AY236878DQ658231
  • Designation of isolates and culture collections: CBS, Centraalbureau voor Schimmelcultures, Utrecht, Netherlands; CMW, Tree Pathology Co-operative Program, Forestry and Agricultural Biotechnology Institute, University of Pretoria, South Africa; STE-U, Stellenboch University, South Africa; MUCC, Murdoch University, Perth, Western Australia.

  • Sequences in bold were obtained during this study.

DNA extraction

Isolates were grown on 2% MEA at 20°C for 4 weeks and the mycelium was harvested, frozen in liquid nitrogen, ground to a fine powder and genomic DNA extracted using a hexadecyl trimethyl ammonium bromide (CTAB) protocol from Graham (1994) modified by the addition of 100 µg mL−1 Proteinase K and 100 µg mL−1 RNAse A to the extraction buffer.

PCR amplification

This study included partial amplification of the18S gene, the complete internal transcribed spacer (ITS) region 1, the 5.8S rRNA gene and the complete ITS region 2 and the 5′ end of the 26S (large subunit) rRNA gene, part of the β-tubulin gene region, part of elongation factor 1α gene (EF-1α), part of Chitin synthase 1 gene (CHS), part of the RNA polymerase II subunit (RPB2) and part of ATPase gene (ATP-6). Primers used for amplification of these regions are listed in (Table 2). The PCR reaction mixture (25 µL), PCR conditions and visualization of products were as described previously (Cortinas et al., 2006) except that 1 U of Taq polymerase (Biotech International, Needville, TX) was used in each reaction. For failed amplifications, the Mg concentration was increased to 4 mM, and primer concentration to 0.9 pmol and the following PCR conditions were used; 7 min at 94°C, followed by 35 cycles of 1 min at 94°C, 1 min at 45°C, 2 min at 72°C and final elongation step of 10 min at 72°C. RPB2 degenerate primers were tested at a range of temperatures, but failed to amplify the DNA of some representative isolates. Therefore, two successful amplicons were sequenced and primers redesigned and named RPB2-myco-6F and RPB2-myco-7R (Table 2). The PCR products were purified with Ultrabind®DNA purification kit (MO BIO Laboratories, Solana Beach, CA) following the manufacturer instructions. Amplicons were sequenced as described previously (Burgess et al., 2005)

View this table:
Table 2

Primer sets and annealing temperature used to amplify Phaeophleospora spp

RegionOligosOligo Sequence (5′–3′)Amplicon size (bp)AT (°C)Reference
β-tubulin (P. eucalypti)Pey-Bt-F Pey-Bt-RGTAACCAAATCGGTGCTGCT GAGTACAAGTGGCTGCTTAG20362This study
β-tubulin (P.epicoccoides)Pep-Bt-F Pep-Bt-RCGACGGCTCAGGCGTGTATG GCGTTAGTGGTGTTGCTTGA21862This study
  • Base codes: R (AG), Y (CT), K (GT), W (AT).

Phylogenetic analyses

In order to compare Phaeophleospora isolates used in this study with other closely related species, additional sequences were obtained from GenBank (Table 1). Sequence data were assembled using sequence navigator version 1.01 (Perkin Elmer) and aligned in clustalx (Thompson et al., 1997) Manual adjustments were made visually by inserting gaps where necessary. All sequences obtained in this study have been deposited in GenBank and accession numbers are shown in Table 1.

The initial analysis was performed on an ITS dataset alone and subsequent analyses were performed on a combined dataset of ITS, β-tubulin, CHS and EF-1α sequence, after a partition homogeneity test (PHT) had been performed in phylogenetic analysis using parsimony (paup) version 4.0b10 (Swofford, 2003) to determine whether sequence data from the four separate gene regions were statistically congruent (Farris et al., 1995; Huelsenbeck et al., 1996). The most parsimonious trees were obtained using heuristic searches with random stepwise addition in 100 replicates, with the tree bisection-reconnection branch-swapping option on and the steepest-descent option, off. Maxtrees were unlimited, branches of zero length were collapsed and all multiple, equally parsimonious trees were saved. Estimated levels of homoplasy and phylogenetic signal (retention and consistency indices) were determined (Hillis & Huelsenbeck, 1992). Characters were unweighted and unordered, branch and branch node supports were determined using 1000 bootstrap replicates (Felsenstein, 1985), characters were sampled with equal probability. Trees were rooted to Botryosphaeria ribis and Botryosphaeria obtusa, which were treated as the outgroup taxa.

Baysian analysis was conducted on the same aligned combined dataset. First mrmodeltest v2.2 (Nylander, 2004) was used to determine the best nucleotide substitution model. Phylogenetic analyses were performed with mrbayes v3.1 (Ronquist & Heuelsenbeck, 2003) applying a general time reversible (GTR) substitution model with gamma (G) and proportion of invariable site (I) parameters to accommodate variable rates across sites. The Markov Chain Monte Carlo (MCMC) analysis of four chains started from random tree topology and lasted 10 000 000 generations. Trees were saved each 10 000 generations, resulting in 10 000 saved trees. Burn-in was set at 500 000 generations after which the likelihood values were stationary, leaving 9950 trees from which the consensus trees and posterior probabilities were calculated. paup 4.0b10 was used to reconstruct the consensus tree and maximum posterior probability assigned to branches after a 50% majority rule consensus tree was constructed from the 9950 sampled trees.

Specific primer design and validation

To design species-specific primers, the gene regions with the greatest sequence difference between P. epicoccoides, P. eucalypti and P. destructans were targeted. Only two gene regions, β-tubulin and EF-1α, were sufficiently variable between P. eucalypti and P. destructans to allow for primer design.

Repeatability of the specific primers was tested using at least 10 isolates of each Phaeophleospora species (P. destructans, CMW17918, 17919, 19832, 19844, 19886, 19906, 19909, 19910, 19936, 22553; P. eucalypti, CMW17912, 17915, 19916, MUCC432, 433, 434, 435, 436, 437, 438; P. epicoccoides, CMW5348, 22482, 22984, 22485, 22486, MUCC327, 424, 425, 426, 427). The isolates were amplified using specific β-tubulin and EF-1α primers (Table 2) and the same PCR conditions as (Cortinas et al., 2006). Thereafter, primers were tested for their specificity, primarily to closely related species, but also to four less related Mycosphaerella spp. (Table 3).

View this table:
Table 3

Specific primers test results

P. destructansP. eucalyptiP. epicoccoides
Test speciesCodeβ-tubulin 198 bpEF1-α 204 bpβ-tubulin 203 bpEF1-α 229 bpβ-tubulin 218 bpEF1-α 173 bp
P. destructansCMW17919+++
P. eucalyptiCMW17916++
P. epicoccoidesCMW5348++
M. crypticaCMW3279
M. vespaCMW11588
M. toledanaCMW14457+
C. zuluensisCMW7449+(500 bp)
M. nubilosaCMW11560
M. mollerianaCMW4940+
M. ambiphyllaCMW13704+
P. eugeniaeCMW5351+(400 bp)+
M. aurantiaMUCC258
M. marksiiMUCC214Multiple bands
M. grandisMUCC216+
M. lateralisMUCC436+
  • Shaded cells indicate where the primers amplified nonspecific DNA.

The ability of the primers to amplify DNA directly from fruiting bodies from infected leaves was determined. The samples were frozen in liquid nitrogen, ground and DNA extracted with CTAB as described previously (Wittzell, 1999). DNA was then subjected to nested PCR, first using general β-tubulin and EF-1α primers and then the initial PCR product was diluted 1 : 5 and nested PCR conducted using the specific primers.


DNA sequence comparisons

Initially, 57 isolates representing 24 Mycosphaerella species and their anamorphs, including five species of Phaeophleospora found on Eucalyptus species and Phaeophleospora eugeniae the type species of the genus, were compared based on ITS sequence data (Table 1). The aligned data set consisted of 709 characters of which 127 bp were due to a large indel in two isolates of P. epicoccoides (MUCC327 and MUCC424) and this indel was excluded from the analyses. Of the remaining characters, 261 were parsimony informative. These data contained significant phylogenetic signal (P<0.01; gl=−0.41) to allow for meaningful analysis. Initial heuristic searches of unweighted characters in paup resulted in three most parsimonious trees of 910 steps (CI=0.56, RI=0.85). The Phaeophleospora species from Eucalyptus; P. destructans, P. eucalypti, P. epicoccoides, P. toledana and Mycosphaerella ambiphylla (which has a Phaeophleospora anamorph) grouped together in a strongly supported clade. This clade also included Mycosphaerella nubilosa, Mycosphaerella cryptica, Mycosphaerella vespa, Mycosphaerella molleriana, Colleteogloeopsis zuluensis and various undescribed ‘Coniothyrium’ spp. (Fig. 2). The ex-type culture of P. destructans (STEU1336=CMW5219) was resequenced in this study and was distant from the isolate of P. destructans on GenBank (AF309614) (Crous et al., 2001). It was also distant from P. eugeniae, which is the type species of the genus, but close to P. eucalypti (Fig. 2, TreeBASE SN2884). The ex-type culture of P. eugeniae (STEU1454=CMW5351) was also resequenced and, while the new sequence was similar to that on GenBank (AF309613), it differed in the first 50 bp of the ITS1 region. Based on results obtained for analysis of ITS sequence data, only species from the ‘nubilosa clade’ were retained for further study.

Figure 2

One of three most parsimonious phylogenetic trees of 977 steps obtained from analysis of ITS sequence data. Branch support (bootstrap values) is given above the branches. The sequences of the ex-type cultures of Phaeophleospora eugineae and Phaeophleospora destructans from Crous et al. (2001) are in a shaded box and those from the present study are in bold type. The tree is rooted to Botryosphaeria ribis and Botryosphaeria obtusa.

The multiple gene genealogies compared 31 isolates, including five Phaeophleospora species from Eucalyptus. The data set for the ATP6 region could not be completed because of difficulties encountered in amplifying DNA for all isolates. The RPB2 region proved not to be informative and these two regions were excluded from the combined analysis. The aligned data set for the combined ITS, β-tubulin, CHS and EF-1α sequences consisted of 1259 characters of which 352 were parsimony informative and were included in analysis. The PHT showed significant difference (P=0.001) between the data from the different gene regions (sum of lengths of original partition was 902, range for 1000 randomizations was 902–921). When the data sets were compared in pairs, the incongruence in the complete combined data set was actually due to incongruence between CHS and both the ITS and EF-1α datasets. On closer examination of the individual tree topography, the incongruence was due to the relationship of M. cryptica and C. zuluensis and not to the positions of the Phaeophleospora species (data not shown, sequence alignments are available from TreeBASE SN2884). Despite the fact that the PHT showed significant difference between data sets, they were nonetheless combined as suggested previously (Hognabba & Wedin, 2003).

The combined data set contained significant phylogenetic signal (P<0.01, gl=−0.29). Heuristic search of unweighted characters in paup resulted in 18 most parsimonious trees of 937 steps (CI=0.68, RI=0.90). In the resultant tree (Fig. 3, TreeBASE SN2884), M. vespa, M. molleriana and M. ambiphylla grouped together, while P. destructans and P. eucalypti were separated with 100% bootstrap support. The four isolates of P. destructans were identical and no polymorphisms were observed in any of the gene regions. There were eight fixed polymorphic sites in the ITS region, nine in the β-tubulin region and 24 in the EF-1α region separating P. destructans and P. eucalypti. The variable sites in the β-tubulin and EF-1α regions were used to design specific primers (Table 2). A table of polymorphic sites is available at http://path.murdoch.edu.au/downloads/Andjicetal_Additionals.pdf

Figure 3

Consensus phylogram of 9950 trees resulting from Baysian analysis of the combined ITS-2, β-tubulin, EF-1α and CHS sequence data of Phaeophleospora isolates. Posterior probabilities of the node are indicated above the branches and bootstrap values from the parsimony analysis are indicated below branches in italics. Not all nodes with high posterior probabilities also have bootstrap support. The tree is rooted to Botryosphaeria ribis.

Phaeophleospora eucalypti isolates were further separated in three subgroups, corresponding to isolates from (a) Queensland, (b) New South Wales and (c) Southern New South Wales, Victoria and New Zealand (Fig. 3). There were 18 polymorphic positions across the four gene regions among isolates of P. eucalypti with 2–3 distinct profiles corresponding to geographic regions. Phaeophleospora epicoccoides was the basal species of the group and has three strongly supported subgroups (Fig. 3). Although there were 26 polymorphic sites across the four gene regions, there was no geographic association linked to these polymorphisms. A table showing polymorphic sites between isolates of P. eucalypti and P. epicoccoides is available at http://path.murdoch.edu.au/downloads/Andjicetal_Additionals.pdf

Validation of species-specific primers

Gel photos showing reproducibility of the specific primers for P. destructans, P. eucalypti and P. epicoccoides are given at http://path.murdoch.edu.au/downloads/Andjicetal_Additionals.pdf

Phaeophleospora destructans

DNA for 10 isolates of P. destructans was amplified using the primers specific for β-tubulin and EF1-α. These primers were then tested on 10 closely related Mycosphaerella spp. and five less related species and none gave amplification products for the β-tubulin primers specific to P. destructans. The EF1-α primer specific to P. destructans also amplified DNA of C. zuluensis, P. eugeniae, Mycosphaerella marksii, but the amplicons either contained multiple bands or were larger than the amplicon for P. destructans (Table 3). Both specific primer sets detected P. destructans directly from spores scraped from the surface of leaves. The β-tubulin primers specific for P. destructans also detected the presence of P. eucalypti, but the amplicon was larger than that obtained for P. destructans and it contained a double band.

Phaeophleospora eucalypti

DNA for all 10 isolates of P. eucalypti was amplified using specific primers for β-tubulin and EF1-α. None of Mycosphaerella spp. tested in this study gave amplification products for the EF1-α primers designed to be specific to P. eucalypti (Table 3). The β-tubulin primers designed for P. eucalypti were not specific and amplified seven other species, amplifying bands of the same size as those for P. eucalypti (Table 3). Only the EF1-α primers detected P. eucalypti from spores scraped from leaves.

Phaeophleospora epicoccoides

All ten isolates of P. epicoccoides gave amplification products using the β-tubulin and EF1-α primers developed for this species. None of the Mycosphaerella spp. tested gave amplification products using these primers (Table 3). In planta, the EF1-α primer set detected the presence of P. eucalypti as well as P. epicoccoides and the β-tubulin primer set detected presence of P. epicoccoides and P. destructans on leaf material.


The current phylogenetic study has unequivocally shown P. destructans to be closely related to P. eucalypti and specific primers have been developed to easily distinguish between these two species. Phaeophleospora destructans is unknown in Australia and is considered a major biosecurity threat. However, based on symptoms it is hard to distinguish between P. eucalypti, which is well-known in Australia, and P. destructans. Thus the specific primers will be very useful for detection and surveillance activities.

In a former study, Phaeophleospora species emerged in two separate clades (Crous et al., 2001). One of these clades included P. eucalypti and P. epicoccoides and the other accommodated P. eugeniae and P. destructans (Crous et al., 2001). All the isolates of P. destructans that have been examined, including the ex-type culture (STE-U1366=CMW5219), had identical ITS sequence data, which was different to the single sequence previously lodged in GenBank (isolate STE-U 1366, AF309613). Consequently, all Phaeophleospora species from Eucalyptus species cluster together and they are closely related to the important Eucalyptus pathogens, C. zuluensis, M. cryptica and M. nubilosa. In contrast, these fungi are distantly related to P. eugeniae. A taxonomic re-evaluation of species of Phaeophleospora and Colletogloeopsis associated with Eucalyptus species is currently underway (unpublished data).

The sequence data obtained in this study for four isolates of P. destructans, three from Indonesia and one from China, were identical for all six gene regions examined. This finding is unusual as some variability is usually observed in sequence data between isolates of the same species, especially when more than one region of origin is considered. The limited variability among isolates of P. destructans supports the hypothesis of selection pressure resulting in the adaptation of a limited number of genotypes to a new host (Eucalyptus in Sumatra, Indonesia) followed by dispersal of these genotypes throughout Asia. In the present study, no informative characters in the RPB2 and CHS regions were found that could separate P. destructans from P. eucalypti. There were, however, a few stable differences between the two species in the sequences for the ITS2 and β-tubulin regions. The most variable gene region was EF1-α where a 22 bp indel separated these species. For ITS2, β-tubulin and CHS gene regions there were more polymorphic sites among isolates of P. eucalypti than between P. destructans and P. eucalypti. This suggests that while P. destructans emerged as a major Eucalyptus pathogen in Asia, it may have very recently evolved from P. eucalypti, to which it is very closely related. Where this adaptation could have occurred, however, remains a mystery as P. eucalypti has not been detected in Asia.

The sequence data for different P. eucalypti isolates was variable and analysis resulted in the isolates residing in different subgroups based on their origin. As isolates from New Zealand grouped with isolates from Victoria and southern New South Wales, P. eucalypti might have been moved to New Zealand from this region. Phylogeographic studies are required to test this hypothesis appropriately (Carbone & Kohn, 2001; Kasuga et al., 2003).

Many polymorphic sites were observed amongst the sequence data sets for isolates of P. epicoccoides, but the groupings did not reflect any obvious pattern relating to origin or other characteristics of the isolates. Unlike P. eucalypti, this species is widely distributed throughout most Eucalyptus growing regions of the world. The lack of phylogenetic grouping amongst isolates with variable sequence data, probably reflects anthropogenic movement of germplasm and multiple introductions of the fungus into new areas. Phaeophleospora epicoccoides is known to be a morphologically variable species and it may represent a species complex rather than a single taxon (Crous & Wingfield, 1997). Population genetic studies and large numbers of isolates from different locations, especially in Australia are required to resolve this question.

Efforts to develop species specific primers for P. destructans, P. eucalypti and P. epicoccoides reflected the close relatedness between these species and the variably within the species. Nonetheless a suite of species specific primers have been developed that allow for simple distinction between these species. Primers based on the EF1-α region distinguished between all three species and primers for the β-tubulin regions provided reliable detection of P. destructans and P. epicoccoides. Specific primers based on EF1-α sequences were able to detect P. eucalypti and P. destructans directly from plant samples. The β-tubulin primers developed to detect P. epicoccoides also showed a faint positive band for P. destructans, while EF1-α primers developed to detect P. epicoccoides showed a faint band for P. eucalypti from leaf material. While this result may be considered confusing, it is believed that this reflects duel infection as P. epicoccoides is very often present on the same lesion together with P. eucalypti and P. destructans (Burgess et al., 2006).

Phaeophleospora destructans is a devastating pathogen of Eucalyptus as yet undetected in Australia. Since the fungus has been detected in East Timor, which is very close to the Australian border, it is a potential threat to the biosecurity and biodiversity of Australia's vast native Eucalyptus forests. Its early detection in Australia is important and the Australian Quarantine and Inspection Service (AQIS) regularly inspects Eucalyptus species in Australia and neighbouring countries for pathogens including P. destructans. Because the symptoms caused by P. destructans can be almost identical to those associated with P. eucalypti and P. eppicocoides, unequivocal identification procedures are important. The DNA sequence data for many gene regions and the specific markers produced in this study should assist in this process.


This work was funded in part by the Australian Research Council DP0343600, ‘Population genetics of fungal pathogens that threaten the biosecurity of Australia's eucalypts’. Vera Andjic is a recipient of a Murdoch University Doctoral Research Scholarship. This work also acknowledges funding from various grants to the University of Pretoria linked to tree protection research and a collaborative research agreement linking the University of Pretoria and Murdoch University. Dr Angus Carnegie is thanked for providing samples of various Phaeophloespora species used in this study.


  • Editor: Michael Bidochka


View Abstract