Introduction

Araucaria heterophylla (Norfolk Island pine) is endemic to Norfolk Island, a small pacific Island 1500 km east of Brisbane, Australia (Green 1994). The tree is fully evergreen and is widely planted throughout Australia, especially in coastal areas. In the past few years, Norfolk Island pine trees in Perth, Western Australia, have been declining. Similar symptoms have been reported from eastern Australia (Varley 2008) and New Zealand (Anonymous 2007).

The Botryosphaeriaceae are common endophytes of a wide variety of woody plants worldwide (Smith et al. 1996; Slippers and Wingfield 2007; Pavlic et al. 2008; Taylor et al. 2009). Diseases associated with the Botryosphaeriaceae are often stress related requiring a predisposing incident to trigger disease expression. Environmental stresses include drought, extreme temperature fluctuations, nutrient deficiencies and mechanical injuries (Old and Davison 2000; Zhonghua et al. 2001; Desprez-Loustau et al. 2006; Slippers et al. 2009; Dakin et al. 2010). However, they can also be primary pathogens (Michailides 1991; Sánchez et al. 2002, 2003; Savocchia et al. 2007), especially when non-adapted hosts are exposed to an exotic pathogen (Denman et al. 1999).

Patterns of host association have been observed among the Botryosphaeriaceae (de Wet et al. 2008). In general, Diplodia species are predominantly associated with gymnosperms; Lasiodiplodia and Botryosphaeria species are associated with both angiosperms and gymnosperms; and Neofusicoccum and Dothiorella species are associated with angiosperms. There are, however, a few exceptions: a Neofusicoccum sp. has been reported from Araucaria cunninghamii and Wollomia nobilis in eastern Australia (Slippers et al. 2005).

Neofusicoccum parvum (Pennycook & Samuels) Crous, Slippers & A.J.L. Phillips (=Botryosphaeria parva) has been reported as one of the most common species on horticultural crops in Victoria and New South Wales (Cunnington et al. 2007). However, while common in eastern Australia, N. parvum is rarely encountered in Western Australia (Burgess et al. 2005). There has been no report of N. parvum associated with decline and canker of Norfolk Island pine in Australia and worldwide.

Identification of species in Botryosphaeriaceae has commonly been based on morphological characteristics of the anamorphs. N. parvum and N. ribis are closely related cryptic species, and for this reason identification of these species based on morphological characteristics is unreliable, and so differentiation relies on molecular phylogeny (Slippers et al. 2004; Pavlic et al. 2009b; Begoude et al. 2010).

In this study declined, cankered and asymptomatic materials were sampled from Norfolk Island pine around Perth, Western Australia. Fungi were isolated and identified using classic and molecular taxonomy techniques and pathogenicity trials were conducted.

Materials and methods

Fungal isolation and morphological observations

During the surveys, 200 Norfolk Island pine trees were inspected in Perth suburbs in 2009 and 2010. Samples were collected from 20 declining and cankered trees aged between 20 to over 80 years. Further samples were collected from five asymptomatic trees. Pieces of stem and branches were surface-sterilized by immersion in a 1.25% aqueous solution of sodium hypochlorite for 2 min, rinsed in sterile water and dried in a laminar flow cabinet. Segments were separately placed on potato dextrose agar (PDA), and malt extract agar (MEA) then incubated at 22 ± 3°C. Fungal colonies which developed from the segments were sub-cultured on PDA and MEA. In addition, in order to obtain further isolates and to supplement morphological observations, pieces of infected branches surrounding cankers were surface-sterilised as described above and placed on moist filter paper trays and incubated at 22 ± 3°C with a 12-h dark and light cycle. Conidia oozing from pycnidia, which formed on agar plates and branches on moist filter paper within 21 and 15 days, respectively, were spread on the surface of water agar (WA) in sterile drops of water. Single, germinating conidia were transferred to PDA and MEA plates. As fast growing sterile mycelia frequently developed, to induce sporulation, 85 isolates were inoculated onto sterilised pine needles (Slippers et al. 2005) and placed on the surface of 2% w/v water agar in petri dishes for 3 weeks at 25°C. The growth rate, colony morphology and micro-morphological characteristics of the fungi consistently isolated from diseased stem and branche tissue were determined.

Molecular identification and phylogenetic analysis

Three representative isolates, morphologically identified as N. parvum, were grown on PDA for 7 days at 25°C. Genomic DNA was extracted from 7-day-old cultures and a PCR conducted to amplify the internal transcribed spacer (ITS) of the rDNA, the translation elongation factor 1α (EF), β-tubulin (BT) and RNA polymerase subunit II (RPB2) as described previously (Pavlic et al. 2009a). For phylogenetic analysis, the dataset comprised of sequences obtained from this study and for closely related species in the N. ribis/parvum species complex available from GenBank. PAUP (Phylogenetic Analysis Using Parsimony) version 4.0b10 (Swofford 2003) was used to determine phylogenetic relationships from the combined data sets. Gaps were treated as a fifth character and all characters were unordered and of equal weight. A repetitive minisatellite region in the intron of the EF1-α in N. ribis was coded to represent a single evolutionary event.

Koch’s postulates and pathogenicity

Pathogenicity tests were performed by using non-diseased Norfolk Island pine branches (3–5 cm in diameter) excised into 25 cm lengths and the ends sealed with paraffin wax. A disk of bark (10 mm diam) in the mid length of each excised branches was removed to expose the cambium. Mycelial plugs of 7-day old cultures of the three representative isolates, morphologically identified as N. parvum, were placed surface down on the wound and covered again with the bark. The wounds were wrapped tightly with parafilm to prevent contamination and desiccation of the wounds and the inoculum. Negative controls were inoculated as described above but with sterile agar plugs (Burgess et al. 2005; Dakin et al. 2010). The excised branches were randomly placed in moist trays with three replicates at 25°C and a 12-h dark and light cycle. Two weeks after inoculation, bark was scraped from branches to reveal the cambium, and lesions were assessed by measuring the length of discolouration. Excised branches representing inoculated isolates or uninoculated controls were selected for fungal re-isolations from the lesion edges. A one-way analysis of variance (ANOVA) was performed using the Genstat 12 edition 2009 (VSN International Ltd).

Results

Fungal isolation and morphological observations

Cankers, bark cracking, die-back, gummosis and brown staining of the internal wood were observed on the Norfolk Island pine trees in the costal suburbs of Perth (Figs. 1, 2). Cankers were variable in size and often accompanied with clear to yellow sap exuding from stems and branches. Trees with advanced symptoms appeared to be in decline, with shoot death and dieback.

Fig. 1
figure 1

Cankers associated with Neofusicoccum parvum on Norfolk Island pine in Perth on a 50 year old tree (a) and a section showing internal wood decay (b). Bars = 20 and 10 cm respectively

Full size image
Fig. 2
figure 2

Decline of the branches associated with Neofusicoccum parvum on Norfolk Island pine trees in Perth (a) and Pycnidial structures formed on the infected foliage, two weeks after incubation in a moist tray (b). Bars = 25 and 0.2 cm respectively

Full size image

Of the 85 fungal isolates recovered, 62 (58 from diseased and 4 from asymptomatic tissues) produced morphological characteristics similar to that which has been reported for N. ribis and N. parvum. These isolates formed pycnidia after 2–3 weeks, both on sterilised pine needles on WA and on PDA and MEA. They also formed teleomoph structures on PDA, MEA and incubated branches. Colonies on PDA were initially white and turned to olivaceous gray with a thick aerial mycelium after 1 week. Based on isolations made from tissues surrounding cankers, infected and asymptomatic branches, the incidence of N. parvum was positively correlated (R 2 = 0.92) with the population of older trees; N. parvum was more consistently isolated from older trees than younger trees.

Individual or uni- to multi-locular pycnidial conidiomata formed on the infected plant tissues, initially embedded in the cortex, with a short neck opening through a non-periphysate ostiole (Fig. 2b). Conidia hyaline, thin-walled, ellipsoid, with subobtuse apex and truncate base; (15.5–)18–21.5(–22.5) × 4.5–7.5 μm (average 19 × 5.4) and mean of length/width ratio 3.5 (Fig. 3a–b). Older conidia became pale brown with one or two septa, the middle cell darker brown. Ascomata morphologically similar to conidiomata and occasionally formed on the same conidiomata. Asci bitunicate, clavate, stipitate, 80–145 × 17–20 μm. Ascospores broadly ellipsoid to fusoid, widest in the middle to upper third, (15–)18–25(–28.5) × (5–)7–10 μm, light brown, thin-walled (Fig. 3c). Morphological characteristics of the Norfolk Island pine isolates resembled those of species within the N. ribis/parvum species complex.

Fig. 3
figure 3

Neofusicoccum parvum (anamorph), conidia and conidiogenous cells (a, b); Botryosphaeria parva (teleomorph), asci and ascospores (c). Bars = 10 µm

Full size image

Molecular identification and phylogenetic analysis

Molecular analysis was required for species differentiation. The ITS dataset consisted of 1791 characters of which 50 were parsimony informative. The data included significant phylogenetic signal (g1 = −0.62) and after heuristic searches in PAUP, one most parsimonious tree of 51 steps was retained (CI = 0.98, RI = 0.99). The ex-type cultures (T) of all six species known from the N. ribis/parvum species complex were included in the analysis. Isolates from Norfolk Island pine in Perth were identified as N. parvum (Fig. 4).

Fig. 4
figure 4

Phylogram of the most parsimonious trees of 50 steps obtained from the combined ITS, β-tubulin, RPB2 and EF 1-α sequence data. Bootstrap support for the branch nodes is given above the branches (numbers in bold = isolates from this study)

Full size image

Koch’s postulates and pathogenicity

Analysis of the pathogenicity data indicated that isolates were significantly (P ≤ 0.05) pathogenic on the excised branches. The bark and cambium tissues of the control branches remained asymptomatic. No significant differences were obtained in lesion length between the three isolates of N. parvum used in this study (Fig. 5). Koch’s postulates were fulfilled by reisolation of N. parvum. Cultures of N. parvum were deposited in the Western Australia Plant Pathogen Collection (WAC13360, WAC13381, WAC13382, WAC13383). To our knowledge, this is the first report of decline and canker of Norfolk Island pine trees caused by N. parvum in Australia and worldwide. Fungal isolates other than N. parvum were also identified, belonging to the genera Alternaria, Epicoccum and Pestalotiopsis by in vitro observations, and no further study was carried out on them.

Fig. 5
figure 5

Mean lesion length (mm) of three N. parvum isolates, inoculated on excised branches of Norfolk Island pine tree, measured two weeks after inoculation (P ≤ 0.05)

Full size image

Discussion

The association of N. parvum with declining and cankered Norfolk Island pine is of interest for two reasons. Firstly, Neofusicoccum species are rarely associated with gymnosperms and secondly, despite extensive collections over several years, N. parvum has rarely been isolated in Western Australia. In contrast, N. parvum has a wide host range and has been reported from native and exotic hosts in other states in Australia and elsewhere in the Southern Hemisphere. (Slippers et al. 2005).

Woody trees which have been under environmental stress or with mechanical injuries are more vulnerable to attack by Botryosphaereiaceae (Old et al. 1990; Dakin et al. 2010). However, although they are considered to be opportunistic pathogens the diseases they cause are often aggressive and difficult to control, such as panicle and shoot blight of Pistachio (Michailides 2002), and Diplodia dieback in pines (Burgess and Wingfield 2002). Members of the Botryosphaeriaceae have been reported causing aggressive diseases on several southern hemisphere conifers such as Wollemia nobilis in Australia, Araucaria cunninghami in New Zealand and Widdringtonia nodiflora in South Africa (Slippers et al. 2005). These could potentially represent new encounters of these conifers with northern hemisphere fungi; however the natural distribution of most Botryosphaeriaceae is unknown. Northern hemisphere conifers such as Sequoia are also susceptible to blights caused by Botryosphaeriaceae (Worrall and Correll 1986).

The survey of Norfolk Island pines in Perth indicated that only older trees showed decline and canker symptoms, and that N. parvum was not consistently isolated from young trees. High temperature and drought stress seem to be important factors predisposing Norfolk Island pine trees to N. parvum in Western Australia. The increased susceptibility to fungi of drought stressed trees is well known, and it is assumed that drought-stressed trees are predisposed to disease because water stressed tissues act as a better substrate for pathogen growth, and/or because drought-stressed trees have decreased defense mechanisms (Desprez-Loustau et al. 2006).

Data from the Commonwealth Bureau of Meteorology (http://www.bom.gov.au/climate/averages) were obtained for the period 2001–2010. During the summer months in Perth, Norfolk Island trees experience dry conditions compared to their endemic environment on Norfolk Island. The summer average rainfall was 13.4 mm with temperatures 14–46°C during this period, compared to summer average on Norfolk Island of 219 mm rainfall and 19–24°C. Furthermore, a significant hail storm in March 2010 in Perth resulted in injury to Norfolk Island pines. These dry conditions in combination with hail damage injury are likely to have resulted in trees becoming susceptible to infection by N. parvum. However, further investigation is required to determine stress-related factors that associate with canker and decline of the Norfolk Island pines in Perth.