effect of season and different fungi on phenolics in response to xylem wounding and inoculation in...

16
Effect of season and different fungi on phenolics in response to xylem wounding and inoculation in Eucalyptus nitens BY K. M. BARRY 1,2 , N. W. DAVIES 3 and C. L. MOHAMMED 1,2,4 1 CRC for Sustainable Production Forestry, GPO Box 252–12, Hobart, Tasmania, Australia 7001. E-mail: [email protected]; 2 School of Agricultural Science, University of Tasmania, GPO Box 252–54, Hobart, Tasmania, Australia 7001; 3 Central Science Laboratory, University of Tasmania, GPO Box 252–74, Hobart, Tasmania, Australia 7001; 4 CSIRO Forestry and Forest Products, GPO Box 252–12, Hobart, Tasmania, Australia, 7001 Summary Pot-grown and plantation-grown Eucalyptus nitens trees (approximately 2 and 3 years old, respect- ively) were experimentally wounded and inoculated with different fungi and in different seasons. Decay lesion development and defence zones were assessed. Two zones were described, a narrow brown decay interface (interface reaction zone, IRZ) and a diffuse zone beyond this being either pale brown or purple (reaction zone, RZ). The total phenol levels in the reaction zone were determined. Selected phenolics (pedunculagin, tellimagrandin 1, tetragalloylglucose, pentagalloylglucose and catechin) were quantified by liquid chromatography–mass spectrometry (LC–MS). A range of fungi (mainly decay-causing) were used to inoculate wounds and the results indicated that more extensive decay lesions were generally associated with greater production of soluble phenols in response. Sterile inoculations and weakly aggressive fungi were associated with no or little xylem discoloration, whereas aggressive fungi elicited more discoloration and phenolic accumulation in advance of infection. This indicates that phenol accumulation is not a generalized response to wounding, but a variable response due to the interaction between microorganisms and sapwood. In plantation-grown trees examined 6 months after wounding, purple reaction zones were commonly associated with large decay lesions. Seasonal differences in decay column area caused by Ganoderma applanatum were not significant 1 month after wounding and inoculation. 1 Introduction Eucalyptus nitens (Maiden) is an important plantation-tree species being grown for solid wood products in Tasmania. As pruning is required, this is done early (beginning at age 3–4 years) to decrease the proportion of the knotty core. It is hoped that decay arising from the wounds will be restricted to the knotty core. Mechanisms of plant defence in woody xylem involve both a reaction zone, which forms in the tissue present at the time of wounding and a barrier zone which is formed by the cambium subsequently (SHIGO and MARX 1977; PEARCE 1996). Barrier zones are considered more resistant to fungal spread than reaction zones (SHIGO and MARX 1977; BAUCH et al. 1980) but are usually limited in extent above and below a wound. For decay columns spreading well beyond a wound, the reaction zone is likely to be more important long-term. The ability of different fungi to penetrate these defence zones (SCHWARZE and BAUM 2000) will determine the development of decay. Wounds offer a wide array of microorganisms the opportunity to colonize compromised plant tissue, in a process which has been termed ‘unspecialized opportunism’ (RAYNER and BODDY 1988). The ability of a fungus to invade beyond this compromised tissue may depend For. Path. 32 (2002) 163–178 ȑ 2002 Blackwell Verlag, Berlin ISSN 1437–4781 Received: 16.10.2001; accepted: 17.1.2002; editor: J. N. Gibbs U. S. Copyright Clearance Center Code Statement: 1437–4781/2002/3203–0163 $15.00/0 www.blackwell.de/synergy

Upload: k-m-barry

Post on 06-Jul-2016

212 views

Category:

Documents


0 download

TRANSCRIPT

Page 1: Effect of season and different fungi on phenolics in response to xylem wounding and inoculation in Eucalyptus nitens

Effect of season and different fungi on phenolics in response toxylem wounding and inoculation in Eucalyptus nitens

BY K. M. BARRY1,2, N. W. DAVIES

3 and C. L. MOHAMMED1,2,4

1CRC for Sustainable Production Forestry, GPO Box 252–12, Hobart, Tasmania, Australia 7001.

E-mail: [email protected]; 2School of Agricultural Science, University of Tasmania, GPO Box

252–54, Hobart, Tasmania, Australia 7001; 3Central Science Laboratory, University of Tasmania,

GPO Box 252–74, Hobart, Tasmania, Australia 7001; 4CSIRO Forestry and Forest Products, GPO

Box 252–12, Hobart, Tasmania, Australia, 7001

Summary

Pot-grown and plantation-grown Eucalyptus nitens trees (approximately 2 and 3 years old, respect-ively) were experimentally wounded and inoculated with different fungi and in different seasons.Decay lesion development and defence zones were assessed. Two zones were described, a narrowbrown decay interface (interface reaction zone, IRZ) and a diffuse zone beyond this being either palebrown or purple (reaction zone, RZ). The total phenol levels in the reaction zone were determined.Selected phenolics (pedunculagin, tellimagrandin 1, tetragalloylglucose, pentagalloylglucose andcatechin) were quantified by liquid chromatography–mass spectrometry (LC–MS). A range of fungi(mainly decay-causing) were used to inoculate wounds and the results indicated that more extensivedecay lesions were generally associated with greater production of soluble phenols in response. Sterileinoculations and weakly aggressive fungi were associated with no or little xylem discoloration, whereasaggressive fungi elicited more discoloration and phenolic accumulation in advance of infection. Thisindicates that phenol accumulation is not a generalized response to wounding, but a variable responsedue to the interaction between microorganisms and sapwood. In plantation-grown trees examined6 months after wounding, purple reaction zones were commonly associated with large decay lesions.Seasonal differences in decay column area caused by Ganoderma applanatum were not significant1 month after wounding and inoculation.

1 Introduction

Eucalyptus nitens (Maiden) is an important plantation-tree species being grown for solidwood products in Tasmania. As pruning is required, this is done early (beginning at age3–4 years) to decrease the proportion of the knotty core. It is hoped that decay arising fromthe wounds will be restricted to the knotty core. Mechanisms of plant defence in woodyxylem involve both a reaction zone, which forms in the tissue present at the time ofwounding and a barrier zone which is formed by the cambium subsequently (SHIGO andMARX 1977; PEARCE 1996). Barrier zones are considered more resistant to fungal spread thanreaction zones (SHIGO and MARX 1977; BAUCH et al. 1980) but are usually limited in extentabove and below a wound. For decay columns spreading well beyond a wound, the reactionzone is likely to be more important long-term. The ability of different fungi to penetratethese defence zones (SCHWARZE and BAUM 2000) will determine the development of decay.

Wounds offer a wide array of microorganisms the opportunity to colonize compromisedplant tissue, in a process which has been termed ‘unspecialized opportunism’ (RAYNER andBODDY 1988). The ability of a fungus to invade beyond this compromised tissue may depend

For. Path. 32 (2002) 163–178� 2002 Blackwell Verlag, BerlinISSN 1437–4781

Received: 16.10.2001; accepted: 17.1.2002; editor: J. N. Gibbs

U. S. Copyright Clearance Center Code Statement: 1437–4781/2002/3203–0163 $15.00/0 www.blackwell.de/synergy

Page 2: Effect of season and different fungi on phenolics in response to xylem wounding and inoculation in Eucalyptus nitens

upon its tolerance of the inherent nature of the wood environment (SHIGO 1974; GRAMSS

1992) and whether a defence response is elicited. There is evidence that different degrees ofreaction zone development occur when woody xylem is challenged with fungi of varyingaggression (PEARCE et al. 1994; PEARCE 2000). In Acer pseudoplatanus challenge of xylemwith fungi displaying weak aggression (such as Ustulina deusta or Ganoderma adspersum)resulted in rapid expression of defence responses including phytoalexin-like coumarins,metal ions and water accumulation (PEARCE 2000). In contrast, there was no evidence of thesedefence responses occurring in response to the invasive fungus Chondrostereum purpureum.As fungi will differ in their ability to invade wounds of different trees, this system may not bereplicated in other trees. For example, Ganoderma adspersum is a strongly aggressive invaderof beech and other broad-leaved trees (SCHWARZE and BAUM 2000).

The season in which wounds are created is an important variable influencing host–pathogen interactions. In E. nitens grown in Tasmania, MOHAMMED et al. (2000) found thatafter 12 months there was a slightly higher number of decay lesions in the sapwoodassociated with wounds pruned in spring and summer than in wounds pruned in autumnand winter. In addition, decay columns were longest following pruning in summer andautumn. In wounded Eucalyptus regnans also in Tasmania, spring and summer wounds hadsignificantly greater defect volume than autumn wounds during the first 6 months (WHITE

and KILE 1993). This seasonal effect was not apparent after 12 and 24 months. Therefore itmay be expected that differences in defence responses are also likely to be mostdistinguishable at the early stages following wounding. MIREKU and WILKES (1989) foundthat extent of infected tissue resulting from spring and summer stem wounds in Eucalyptusmaculata was less than for autumn and winter wounds. In summer, higher phenolproduction was detected and wounds were sealed with kino.

Previous studies of antimicrobial defence in sapwood of E. nitens have involved thedescription of a number of morphologically distinguishable zones (BARRY et al. 2000).Decayed tissue was immediately interfaced by a narrow band of dark discoloured tissue (2–3 mm wide), commonly followed by a purple-coloured zone (approximately 5 mm wide).Decay fungi were commonly isolated from the narrow band of dark discoloured tissue, butless commonly from the purple zone (BARRY 2001). As reaction zones have been describedas being brightly coloured, ranging from 1 to 10 mm in thickness and being free fromfungal hyphae (SHAIN 1967, PEARCE and WOODWARD 1986), the purple zone in E. nitenshas been described as a reaction zone.

In wounding experiments in which E. nitens was inoculated with Ganoderma adspersum,a narrow band of enriched phenol deposition was observed in stained sections (BARRY et al.2001b). This relates in position to the band of dark discoloration described above and willbe referred to as the interface reaction zone (IRZ) in this paper. Phenolics were sampled inthe region up to 10 mm beyond the IRZ, which would correspond with the position of thetissue termed the reaction zone (RZ) described above. In the initial stages of defenceresponse, this reaction zone was not purple but pale brown. This will be further discussedin the course of the paper.

Ganoderma adspersum proved to be weakly aggressive and elicited increased levels of totalphenols up to six-fold over those found in unwounded tissue (BARRY et al. 2001b). In thepresent study another species of this genus, G. applanatum, was used for further inoculationstudies to determine the effect of season on the defence response in young plantation trees.The response of E. nitens to the challenge of different fungi was also examined using bothpot-grown and plantation trees. The fungi used included G. applanatum, anothercosmopolitan decay species (Phellinus robustus) and a non-decay species (Botrytis cinerea).In addition, fungi associated with decay of pruned plantation-grown E. nitens were used,including an Aleurodiscus sp. and two species that remain unidentified.

At sampling, any lesions present were measured and a sample of wood was removed forphenol analysis in a manner comparable to previous studies, in which sapwood from the

164 K. M. Barry, N. W. Davies and C. L. Mohammed

Page 3: Effect of season and different fungi on phenolics in response to xylem wounding and inoculation in Eucalyptus nitens

region beyond the immediate discoloured interface of decay (IRZ) was sampled, termed theRZ. The current paper quantifies total phenol levels and some key phenols (hydrolysabletannins) by liquid chromatography–mass spectrometry (LC-MS).

2 Materials and methods

2.1 Plant material

2.1.1 Experiment 1

Pot-grown 22-month-old E. nitens trees were used. Seedlings were raised in a glasshouse inseed-raising mix and then transferred to tubes (4 cm diameter) containing commercialpotting mix (80% composted bark, 20% coarse sand) supplemented with slow-releasefertilizer (Osmocote, Scotts Australia Pty Ltd., Baulkham Hills, NSW, Australia). Seedlingswere then transferred to 17 cm pots and placed outdoors under shade cloth. Atapproximately 16 months of age, they were transferred to 25 cm diameter pots, and placedin full sun. Pots were watered by a drip-irrigation system (up to four times a day in summer).Additional liquid fertilizer (Aquasol, Hortico Ltd, Melbourne, Vic, Australia) was also addedin the growing seasons. The 6-week experiment was conducted in spring (October–mid-November) 1999. During this period, average daily minimum temperatures were 9.2�C andthe maximum values were 17.9�C (based on Hobart data, Bureau of Meteorology, Hobart).

2.1.2 Experiments 2 and 3

Young E. nitens trees grown in a plantation in southern Tasmania were used. This site was aForestry Tasmania coupe (Arve 26 E) at 43�08¢ S latitude, 146�49¢ E longitude and 230 maltitude, approximately 10 km from Geeveston. The trees were of Toorongo provenance andwere planted in December 1996. The experiments were begun in April 1999 (autumn). At thistime, trees of approximately 3 m height, without showing phase change to adult foliage, wereselected over an area of approximately 100 m · 100 m. Treatments within the experimentswere then distributed randomly. During the 6-month experimental period average dailyminimum temperatures were 4.4�C, maximum values were 14.6�C and average daily rainfallwas 1.6 mm (based on Geeveston data, Bureau of Meteorology, Hobart – see Fig. 1).

2.2 Fungal material

The origin of the fungal isolates is given in Table 1. Fungi were identified to species levelwhere fruit bodies were present. The Aleurodiscus sp. was tentatively identified asA. botryosus, based on the presence of botryose acanthophyses of the hymenial layer(CUNNINGHAM 1963). This fungus has not been previously recorded in Tasmania (ANON

2000). Two unidentified decay isolates were also used, and both were classified asbasidiomycetes. Isolate D had abundant clamp connections and while no clampconnections were observed for isolate R, both isolates grew on a selective media forbasidiomycetes (described below) and produced peroxidase and laccase (enzymesfrequently associated with white-rot fungi). Enzyme assays were performed by drop-tests on the margin of the fungal cultures (STALPERS 1978). Isolate D has been found at arange of sites associated with large decay columns resulting from pruning wounds (K.HARRISON, M. HALL and C. MOHAMMED unpublished).

All fungi were maintained on 3% malt agar in the dark at 20�C. They were characterizedby growth rate, cultural morphology, hyphal structures (presence and type of clamps) andenzyme production. This information was required for recognition of the fungi when re-isolated.

165Effect of season and different fungi on wounds in Eucalyptus nitens

Page 4: Effect of season and different fungi on phenolics in response to xylem wounding and inoculation in Eucalyptus nitens

2.3 Wounding and inoculation

2.3.1 Experiment 1

For the pot-grown trees, single wounds were made on each stem at approximately 30 cmabove ground level. Wounds (18 mm long, 6 mm wide and 3–4 mm deep) were made witha sterile chisel. Rectangular pieces of fungal agar culture (or sterile agar) were adpressed tothe wound and sealed with parafilm, followed by PVC tape. The five treatments includedfour different fungi cultured on malt agar (Ganoderma applanatum, Aleurodiscus sp.,isolate D and isolate R) and sterile malt agar. Five trees per treatment were wounded andinoculated.

Fig. 1. Meteorological data (temperature and rainfall) for Geeveston over the seasonal periods(1 month in the middle of each season) for experiment 3

Table 1. Details of fungal isolates used to inoculate E. nitens trees

Fungal isolate Obtained from Site Date isolated

Ganodermaapplanatum

Fruit body on Nothofaguscunninghamii

Maydena, SW Tasmania February1999

Phellinusrobustus

Fruit body on Eucalyptusamygdalina

Molesworth, SW Tasmania February1999

Botrytis cinerea Culture collection University of Birmingham, UK NAAleurodiscus sp. Decayed Eucalyptus nitens

stemCalder, NW Tasmania February

1998Isolate D Eucalyptus nitens white-rot

decayEvercreech, NE Tasmania June 1999

Isolate R Eucalyptus nitens reactionzone

Evercreech, NE Tasmania June 1999

166 K. M. Barry, N. W. Davies and C. L. Mohammed

Page 5: Effect of season and different fungi on phenolics in response to xylem wounding and inoculation in Eucalyptus nitens

2.3.2 Experiment 2

Each tree to be inoculated received two chisel wounds. These wounds were 24 mm long,12 mm wide and 10 mm deep. Both wounds were between 50 and 100 cm above groundlevel, at least 10 cm apart and on opposite sides of the stem. There were three inoculationtreatments (G. applanatum, P. robustus and B. cinerea) and five trees per treatment. Theinoculation procedure was as for experiment 1. Trees were harvested after 6 months. All 30wounds were assessed for lesion morphology and re-isolation of the challenge fungus, butonly one wound per tree was utilized for phenol extraction. The second wound wasreserved for anatomical and histochemical studies.

2.3.3 Experiment 3

Each tree to be inoculated received one wound and all wounds were inoculated with G.applanatum. Wound size and inoculation procedure was as for experiment 2. Five treeswere inoculated on each of the dates: 12 April 1999 (autumn), 12 July 1999 (winter), 11October 1999 (spring) and 10 January 2000 (summer). Trees were harvested after 1 month.

2.4 Analysis of inoculated wounds

At the completion of each experimental period, the stem segment containing the woundwas cut from the tree and analysed. Using a sterile chisel, stem segments were split axiallythrough the centre of the wound. The extent of lesion development was recorded by atracing and included any bleached decay region. Zero lesion development was recorded if ableached decay region had not developed, even if discoloration was observed. The area ofthe lesion was later determined by making an enlarged copy of the tracing. The area of thelesion was carefully cut out and weighed to ± 1.0 mg. This weight was compared to theweight of a standard area of known size.

2.5 Re-isolation of fungi

For experiments 1 and 2, half of each stem segment was immediately transferred to alaminar flow cabinet and wood-chips were removed from areas of apparently infectedtissue both above and below the wound. If bleached decayed material was present,isolations were made from the edges of the decay. In the absence of bleached decaymaterial, any discoloured tissue present was sampled. A small number of chips (3–4) wereplaced on both 1% malt agar selective for basidiomycetes and 3% malt agar. The selectiveagar contained 50 mg/l penicillin, 50 mg/l streptomycin, 25 mg/l polymixin and 230 mg/lthiabendazole. The plates were checked at intervals and subcultures made as appropriate.The number of wood chips from which the challenge fungus was successfully isolated wasdetermined and expressed as a percentage.

2.6 Anatomy and histochemistry

For experiments 1 and 2, wood blocks were prepared from selected stems of eachtreatment and sectioned while fresh with a sliding microtome. Both transverse and radiallongitudinal sections were prepared (10–50 lm thick). Sections were stained with 1%toluidine blue in 0.2 M phosphate buffer (pH 6.5) for general wood anatomy and todiscriminate fungal hyphae. An aqueous solution of 0.5% fast blue RR salt (FBRRS;Sigma Chemical Co., Sydney, Australia) was used to detect phenolics which stained a redcolour.

167Effect of season and different fungi on wounds in Eucalyptus nitens

Page 6: Effect of season and different fungi on phenolics in response to xylem wounding and inoculation in Eucalyptus nitens

2.7 Extraction of phenolics

For all experiments, wood shavings (approximately 0.5 mm thick) were prepared with asmall chisel from an area of approximately 1 cm above and below the discoloured IRZboundary of the lesion. This tissue was often pale brown (but sometimes purple) and hasbeen referred to in our studies as the reaction zone.

Healthy sapwood was sampled away from the wound (on the opposite side of the stemand approximately 5 cm below the wound). The fresh weight of the shavings(approximately 30–90 mg) was determined and then extracted in 750 ll of 70% acetonefor 24 h at 4�C in the dark, followed by a second repeat extraction. Extracts were pooled(making 1.5 ml) and stored at )20�C in the dark. Extracts were centrifuged before analysis.The extracted shavings were dried at 85�C for 3 days and dry weight determined.

2.8 Determination of total phenols

Total phenols were estimated by the Folin-Ciocalteu method, adapted from BONELLO andPEARCE (1993). Either 50 or 20 ll of extract was diluted to 3000 ll water. To this, 1500 llof 1 : 1 diluted Folin and Ciocalteu’s reagent (Sigma Chemical Co.) was added and left for3 min, followed by adding 1500 ll of 1 M aqueous Na2CO3. The solution was shaken andleft to react for 1 h. The absorbance was measured at 725 nm with a spectrophotometer.Concentrations of total phenols were calculated with reference to a gallic acid standardcurve (10–200 lg/ml dissolved in acetone) by applying regression analysis and results wereexpressed as gallic acid equivalent (lg) per mg extracted dry weight (DW) of wood.

2.9 Analysis of phenols by LC-MS

2.9.1 Internal standard

As an internal standard, rutin (Sigma Chemical Co.) was prepared in pyridine (5 mg/ml)and added to each extract prior to analysis.

2.9.2 LC-MS system

Samples were analysed by LC-MS (sample chamber maintained at 5�C) by the followingsystem. High performance liquid chromatography separations were carried out on aWaters Alliance 2690 (Waters Corporation, Milford, MA, USA) using a Waters Nova PakC18 column (150 mm · 3.9 mm). The mobile phase consisted of water/acetic acid 98 : 2(solvent A) and methanol/acetic acid 98 : 2 (solvent B). Initial conditions were 5% solventB and 95% solvent A. The programme was a linear gradient to 54% solvent B and 46%solvent A over 40 min with a flow rate of 0.8 ml per min, before returning to initialconditions with 12 min re-equilibration between samples.

Mass spectrometry was carried out on a Finnigan LCQ (San Jose, CA, USA) with anelectrospray ion source, using LCQ Navigator Version 1.2 software (Finnegan, San Jose,CA, USA). The instrument was operated in negative ion mode, scanning from m/z 125–1500 with an AGC target value of 2 · 107 and maximum ion injection time of 100 ms.Operating conditions were as follows: sheath gas 90 psi, aux gas 50 psi, ESI needle voltage4.5 kV, capillary temperature 270�C and capillary voltage )30 V. Data-dependent MS-MSspectra were routinely acquired from the most intense ion in the spectrum, with a defaultcollision energy of 30% and a peak isolation width of 3 amu.

2.9.3 Monitoring and quantification

In all experiments, two ellagitannins, two galloylglucoses and one flavonoid were selectedfor monitoring by LC-MS. Previous identification of E. nitens reaction zone extractives

168 K. M. Barry, N. W. Davies and C. L. Mohammed

Page 7: Effect of season and different fungi on phenolics in response to xylem wounding and inoculation in Eucalyptus nitens

(BARRY et al. 2001a) revealed that of a number of ellagitannins, pedunculagin andtellimagrandin I were dominant and considerably increased compared to the sapwood.Both compounds consisted of two anomers and therefore required summation of twopeaks. Of the gallotannins, tetragalloylglucose and pentagalloylglucose showed someincrease and were also selected. There is a range of isomers of tetragalloylglucose but onlythe main peak showed an increase (BARRY et al. 2001a). This peak was composed of twoisomers including 1,2,3,6-tetragalloylglucose (BARRY et al. 2001a). For pentagalloylglucose,a peak with equal retention time to the standard (one of several at [M-H]) 939) was chosen.Catechin levels (while generally low) showed an inverse trend, which warranted furtherexamination. The complexity of compounds in the reaction zone made some othercompounds of interest difficult to quantify with good repeatability from sample to sample(due to slight changes in retention time confusing identification of a range of isomers).

To quantify peak areas of individual compounds, characteristic ions were selected. Thisrequired inclusion of not just the singly charged de-protonated molecular ions, but doublycharged molecular ions, daughter ions and adducts formed during ionization. This then simulatedthe total ion current for each compound while allowing the exclusion of co-eluting peaks.

2.9.4 Correction factors

Peak areas were converted from ‘rutin equivalents’ to appropriate units by using responsefactors to hydrolysable tannin standards (non-commercial) and a commercially suppliedstandard of catechin (Sigma Chemical Co.). The rutin to 1,2,3,6-tetragalloylglucose standardresponse factor was 1.45. The response factors to equate weights of 1,2,3,6-tetragalloylglucoseto pedunculagin, tellimagrandin I, pentagalloylglucose and catechin were 3.83, 3.36, 1.92, and4.15, respectively. These response factors were calculated from an average of eight runs ofstandards (prepared in a mix). The purity of non-commercial standards was also calculatedand incorporated into the results. This was complicated by the fact that some impuritiescontributed to the total of other compounds (for example, tellimagrandin 1 had aapproximately 13% impurity of pentagalloylglucose which was incorporated into the totalamount calculated). Concentration of compounds was expressed as lg/mg extracted DW.

2.10 Statistical analysis

For all data sets, a Fishers t-test was applied using SAS software version 6.11 (SAS Institute;Cary, NC, USA). All statistics were assessed at the 5% probability level. Where two-factortests were used, significant p-values for factor interactions were analysed.

3 Results

3.1 Re-isolation of fungi

Re-isolation of challenge fungi from the pot-grown trees of experiment 1 was relativelysuccessful. Where extensive decay lesions were present (isolate D and Aleurodiscus sp.),the inoculated fungi were successfully re-isolated from each wound. Overall isolationsuccess for these fungi was 71 and 44% of wood-chips, respectively. Failure to recoverAleurodiscus was frequently associated with the presence of ‘contaminant’microorganisms, whereas with isolate D, no other fungi were obtained. Isolate R wasisolated from four of the five wounds and G. applanatum from two of them. Theproportion of ‘contaminant’ microorganisms was high from these wounds. Contaminantmicroorganisms, principally Penicillium spp. were isolated from all the woundsinoculated with sterile agar.

169Effect of season and different fungi on wounds in Eucalyptus nitens

Page 8: Effect of season and different fungi on phenolics in response to xylem wounding and inoculation in Eucalyptus nitens

With the exception of two isolates of B. cinerea, the challenge fungi were not re-isolatedfrom the plantation trees of experiment 2. Twelve of 25 ‘contaminant’ fungi had whitecord-forming mycelium that were morphologically similar to xylariaceous fungi(ascomycetes). None of these isolates grew on the agar selective for basidiomycetes orhad clamp connections.

3.2 Morphology and microscopic examination of lesions and response

3.2.1 Experiment 1

The average lesion area after 6 weeks is shown in Fig. 2. Lesions were much larger forisolate D than with any of the other treatments and this effect was significant at p ¼ 0.0026.

At the interface of lesions, discoloration varied substantially in shade and intensity.Discoloration was most prominent at the boundary of the lesions formed by isolate D(Fig. 3) and Aleurodiscus sp. Discoloration was brown/orange for wounds inoculated withG. applanatum and typically had a water-soaked appearance (Fig. 4). There was littlediscoloration associated with wounds inoculated with sterile agar and isolate R. Callus hadcommonly formed around wounds inoculated with sterile agar, isolate R and G.applanatum, but not around wounds inoculated with isolate D and Aleurodiscus sp.

Sections were examined from a number of representative wounds. For sterile wounds,hyphae were rarely visible. For lesions caused by Aleurodiscus sp., isolate D and G.applanatum, hyphae with clamp connections were regularly observed within the vessels,fibres and parenchyma.

3.2.2 Experiment 2

Quite extensive decay columns were formed after 6 months. Because of the disappointingresults with re-isolation, it was considered inappropriate to analyse the data in terms of theresponse to particular challenge fungi. However, scrutiny of the samples suggested thatuseful information could be obtained by classifying the wounds into those with and

Fig. 5. Decay lesions from a wound of experiment 2. Above the wound the lesion is associated withdistinct purple zones (pRZ) beyond the dark IRZ. Below the wound discoloration is not as dark. The

barrier zone (BZ) is also associated with purple discoloration. Scale bar ¼ 10 mm

Fig. 6. Transverse section showing the decayed (D), discoloured interface (IRZ) and purple reactionzone (pRZ) tissues from a wound of experiment 2, stained with toluidine blue. Scale bar ¼ 100 lm

Fig. 2. Lesion area (cm2 ± SE) 6 weeks after wounding and inoculation with a variety of treatments(S, sterile; G, G. applanatum; R, isolate R; A, Aleurodiscus sp.; D, isolate D)

170 K. M. Barry, N. W. Davies and C. L. Mohammed

Page 9: Effect of season and different fungi on phenolics in response to xylem wounding and inoculation in Eucalyptus nitens

3 4 5

6

Fig. 3. Decay lesion caused by isolate D, with prominent IRZ and the RZ region beyond, both aboveand below the infection. The dashed-line boxes indicate the region of phenol sampling. Scale

bar ¼ 10 mm

Fig. 4. Small amount of orange/brown discoloration below a wound inoculated with G. applanatum.Dashed-line boxes represent ‘RZ’ tissue sampled. Scale bar ¼ 8 mm

171Effect of season and different fungi on wounds in Eucalyptus nitens

Page 10: Effect of season and different fungi on phenolics in response to xylem wounding and inoculation in Eucalyptus nitens

without purple reaction zones (Fig. 5). Thirteen of the 30 wounds fell into this categoryand it was soon apparent that there was an association between purple zones and largerlesions (Table 2). In addition to its occurrence in wood present at the time of wounding,the purple coloration was often also found in the position of the barrier zone and beyondthis to wood formed after wounding (Fig. 5).

Examination of material from experiment 2 showed that the IRZ that bounded the largerlesions was characterized by the presence of brown-golden deposits (Fig. 6). Thesedeposits stained red with FBRRS (indicating phenolics). In purple reaction zones, rayparenchyma and vessel tyloses stained light red with FBRRS.

3.2.3 Experiment 3

Lesion areas formed after 1 month were generally small and are shown in Fig. 7. Althoughlesions formed after wounding in summer tended to be the largest, the effect was notstatistically significant.

3.3 Phenolics

3.3.1 Experiment 1

There was a marked effect of different challenge fungi on the accumulation of phenolics inthe RZ compared to levels in HS (Fig. 8). This was apparent for both total phenols and forthe individual compounds. In general, inoculation with isolate D and Aleurodiscus sp.resulted in the induction of greater levels of phenols than did the other treatments.Exceptions can be seen for pedunculagin, where large but very variable amounts wereinduced by G. applanatum, and for catechin. Catechin induction was lowest withAleurodiscus sp.

Fig. 7. Lesion area (cm2 ± SE) one month after wounding and inoculation with G. applanatum in eachseason for experiment 3

Table 2. Average lesion area (cm2 ± SE) for experiment 2 classed by whether a purple RZ wasformed

Decay column Number of wounds Average lesion area (cm2 ± SE)

Total 30 1.24 ± 0.29Purple RZ present 13 2.34 ± 0.62No purple present 17 0.39 ± 0.07

172 K. M. Barry, N. W. Davies and C. L. Mohammed

Page 11: Effect of season and different fungi on phenolics in response to xylem wounding and inoculation in Eucalyptus nitens

3.3.2 Experiment 2

Here the data were analysed in relation to the presence or absence of a purple RZ. Fifteenof the totals of 30 wounds were sampled for phenols and in eight cases some purplecoloration was present. Analysis of individual compounds revealed that the concentrationwas extremely similar beyond the IRZ interface regardless of whether it was purple (pRZ)or pale brown (bRZ) (Fig. 9). Increases in pedunculagin in these tissues were highest, being

Fig. 8. Total phenol levels (lg gallic acid equivalent/mg extracted DW) and concentration ofindividual tannins and catechin (lg/mg extracted DW) between healthy sapwood (HS) and reactionzone (RZ) for the pot-grown trees challenged with different fungal treatments (S, sterile; G, G.

applanatum; R, isolate R; A, Aleurodiscus sp.; D, isolate D)

173Effect of season and different fungi on wounds in Eucalyptus nitens

Page 12: Effect of season and different fungi on phenolics in response to xylem wounding and inoculation in Eucalyptus nitens

23- and 22-fold greater than healthy sapwood in the pale brown RZ tissue and purple RZtissue, respectively.

3.3.3 Experiment 3

Total phenol concentrations did not vary according to season in healthy sapwood, butvariations of the individual compounds were apparent. This was particularly evident fortetragalloylglucose, pentagalloylglucose and tellimagrandin 1 which showed markedseasonal alterations in healthy sapwood levels, which were significantly different(p < 0.0001, p ¼ 0.005 and p ¼ 0.002, respectively). In contrast, concentrations ofpedunculagin and catechin did not vary greatly in the healthy sapwood with season.

Induction of total phenols in the RZ by G. applanatum was most pronounced in the autumn(Fig. 10). This was matched by increased concentrations of tellimagrandin 1 and catechin.Catechin was particularly increased in response to autumn (four-fold) and summer wounding(three-fold) and there was a significant interaction effect (p ¼ 0.004) as well as highly significantdifferences between tissue types (p < 0.0001). There was a significant difference in pedunculaginlevels between tissue types (p ¼ 0.02) but no significant seasonal effect. Tetragalloylglucose andpentagalloylglucose were poorly induced in the reaction zone in all seasons.

4 Discussion

4.1 Definitions of the reaction zone

A number of observations have been made in this study which require a modification ofprevious descriptions of the reaction zone in E. nitens. A typical feature of reaction zonesdescribed in other trees (e.g. PEARCE 1996; PEARCE et al. 1994; SCHWARZE et al. 2000) is thepresence of insoluble polyphenolic deposits, which are present in the E. nitens discoloured

Fig. 9. Concentration of phenols (lg/mg extracted DW) for three different tissues of 15 wounded andinoculated plantation-grown trees of experiment 2. The tissues are healthy sapwood (HS), pale brownreaction zone (bRZ) and purple reaction zone (pRZ). NB. As purple zones never extended around thewhole decay column, pale brown zones were also present and so could be sampled from all 15 wounds

174 K. M. Barry, N. W. Davies and C. L. Mohammed

Page 13: Effect of season and different fungi on phenolics in response to xylem wounding and inoculation in Eucalyptus nitens

interface but lacking in the purple zone. On this basis the discoloured interface has beentermed a reaction zone (IRZ), even though decay fungi can be isolated from it (BARRY 2001).This zone requires more investigation. It is likely that both the IRZ and tissue just beyond(either purple or pale brown reaction zones) are involved in confining fungal spread. The‘reaction zone’ as described in this paper may be similar in role to the transition zonesoriginally described by SHAIN (1979), serving as the location for the synthesis of monomericphenols which are then polymerized at the IRZ. However, the purple zone has been shown tobe antifungal in character (BARRY 2001) and therefore warrants the term reaction zone. Thesituation in E. nitens appears similar to Cryptomeria japonica in which fungal hyphae wereabundant in a discoloured reaction zone but not the transition zone (YAMADA et al. 1988;YAMADA 1992) and both zones were included within the concept of a ‘reaction zone barrier’.

4.2 Effect of different fungi

Studies of differential defence responses in E. nitens according to the challenge fungi testedshowed that the least infection (Fig. 2) was associated with the least production ofphenolics in the reaction zone and vice versa (Fig. 8). Less phenolic production anddiscoloration were observed in wounds inoculated with sterile agar than in othertreatments. This indicates fungal challenge is necessary to elicit a response.

Increases in total phenols in the RZ tissue for E. nitens wounds inoculated with G.applanatum (approximately two-fold) were not as great as those found previously with G.adspersum, which were up to six-fold greater than healthy sapwood (BARRY et al. 2001b).

Fig. 10. Concentration of total phenols (lg gallic acid equivalent/mg extracted DW) individualcompounds (lg/mg extracted DW) in response to wounding and inoculation with G. applanatum in

different seasons

175Effect of season and different fungi on wounds in Eucalyptus nitens

Page 14: Effect of season and different fungi on phenolics in response to xylem wounding and inoculation in Eucalyptus nitens

Increases in total phenols for decay columns caused by isolate D and Aleurodiscus sp. wereeight-fold and six-fold, respectively (Fig. 8). These two fungi appeared to be comparativelypathogenic as the cambium was killed around the wound and no callus developed. It isinteresting that catechin levels were greatly increased (approximately seven-fold) in responsesto isolate D, whereas they were only marginally increased for Aleurodiscus sp. (Fig. 8).

Studies of Acer pseudoplatanus reveal that green reaction zone deposits were formed whenwounds were inoculated with sterile agar and a range of non-aggressive fungi (PEARCE et al.1994; PEARCE 2000). Lesions caused by the aggressive, pathogenic fungus Chondrostereumpurpureum were not associated with green deposits or other features of the reaction zone(PEARCE et al. 1994). The situation in E. nitens seems to be converse. A noticeable reaction zone(IRZ and the purple RZ) was only produced with fungi that showed some pathogenic ability,such as Aleurodiscus sp. and isolate D. It should be recognized that if a highly pathogenicfungus were to be used on E. nitens it might also be able to progress through the tissues withoutprompting the production of a reaction zone. However, there remains the real possibility thatthe defence system operating in E. nitens is quite different from that in A. pseudoplatanus.

Where greater phenol production is associated with the presence of more aggressive fungi itis reasonable to conclude that this is because xylem cells are being continually challenged andconsequently defence mechanisms are constantly elicited. This fits the model of reaction zoneformation as a dynamic process (SHAIN 1979). Further studies of the major decay-causingfungi in Tasmanian plantation forests are required and this is currently being attempted.

4.3 Effect of season

Although experiment 3 revealed no significant evidence that decay is more serious followingwounding in summer than at other times of year, it may be noted that such an effect wouldbe consistent with that found in previous studies of E. nitens in Tasmania (MOHAMMED et al.2000). Results of these studies with E. nitens contrast with seasonal studies of E. maculata byMIREKU and WILKES (1989), who found that decreased decay in summer was correlated withincreased phenol production. As their study was completed in mainland Australia, seasonalchanges may be different to Tasmania and it would be expected that different decay fungiwere present. Defence response and capacity probably differ within the Eucalyptus genus.

A number of factors may contribute to seasonal differences in decay and defenceresponse. Differences in rainfall and temperature may influence the hydraulic status ofxylem (rendering it more or less compromised by wounding). Distribution of resourcesmay change with varied environmental factors and climate to effect phenol production(WAINHOUSE et al. 1998). Factors affecting fungal infection could include seasonaldifferences in sporulation and dissemination for various fungi (GADGIL and BAWDEN

1981) as well as factors affecting subsequent growth rate.Although decay extent was not significantly different with seasonal changes in this study

of E. nitens, analysis of extractives from healthy sapwood and reaction zone tissue fromtrees wounded at different times of year showed some striking differences. For example,the greatest increase in pedunculagin, tellimagrandin I and catechin was in autumn(Fig. 10). With only data available for 1 year it is not possible to determine whether thisdifference is linked to the growth changes taking place at this time of year or other effects.

The biosynthetic pathway between ellagitannins was first indicated by seasonal studiesof deciduous leaves in Liquidambar formosana (HATANO et al. 1986). Although woodyxylem is a perennial tissue, seasonal balances between compounds may also reflectbiosynthetic relationships. For example, levels of pentagalloylglucose (the immediateprecursor to ellagitannins) is highest in winter whereas the ellagitannins are reasonably low.Therefore conversion may not be occurring in winter due to low temperatures, allowingthe precursor levels to increase. Conversely, tetragalloylglucose is highest in summer whentellimagrandin 1 levels are lowest.

176 K. M. Barry, N. W. Davies and C. L. Mohammed

Page 15: Effect of season and different fungi on phenolics in response to xylem wounding and inoculation in Eucalyptus nitens

Acknowledgements

K.M.B. would like to thank Glen KILE and Carolyn RAYMOND for reviewing the manuscript and toJohn GIBBS and reviewers for substantial input. The study was part of a Ph.D. programme and fundingfrom an Australian Postgraduate Award with industry funding from the Tasmanian Forest ResearchCouncil was greatly appreciated.

Resume

Effet saisonnier et de differents champignons sur les phenols en reponse a la blessure du xyleme et al’inoculation, chez Eucalyptus nitens

Des Eucalyptus nitens en pots et en plantation, ages respectivement de 2 et 3 ans, ont ete blesses etinocules avec differents champignons, a differentes saisons. Le developpement des alterations et deszones de defense a ete evalue. Deux zones ont ete distinguees: une etroite bande brune (interface avec lazone de reaction, IRZ), et, au dela de celle-ci, une zone diffuse de couleur brun pale ou pourpre (zonede reaction, RZ). Les niveaux de phenols totaux ont ete determines. Les phenols suivants ont etequantifies par chromatographie liquide et spectrometrie de masse (LC-MS): pedunculagine,tellimagrandine 1, tetragalloylglucose, pentagalloylglucose et catechine. Un certain nombre dechampignons, principalement des agents d’alteration, ont ete inocules sur blessure. Les resultats ontmontre que les alterations les plus etendues etaient generalement associees a une plus grandeproduction de phenols solubles dans les reponses de l’arbre. Les inoculations steriles et leschampignons peu agressifs etaient associes a une coloration faible ou absente. Les champignonsagressifs induisaient plus de coloration et une plus grande accumulation de phenols en avant del’infection. Ceci indique que l’accumulation de phenols n’est pas une reponse generale a la blessuremais qu’elle depend de l’interaction entre les microorganismes et l’aubier. Chez les arbres enplantation, 6 mois apres la blessure, des zones de reaction pourpres etaient couramment associees a degrandes alterations. La surface de l’alteration induite par Ganoderma applanatum variait selon la saisond’inoculation, mais les differences n’etaient pas significatives un mois apres la blessure et l’inoculation.

Zusammenfassung

Auswirkung von Jahreszeit und Pilzart auf die Phenolbildung nach Xylemverletzung und Inokulationbei Eucalyptus nitens

Ungefahr zweijahrige Eucalyptus nitens im Topf und ca. dreijahrige Pflanzen in einer Plantage wurdenexperimentell verletzt und mit verschiedenen Pilzen in verschiedenen Jahreszeiten inokuliert. DieEntwicklung der Faulelasionen und Abwehrzonen wurde untersucht. Es wurden zwei Zonenbeschrieben, eine schmale braune Ubergangszone an der Peripherie der Faule (Interface Reaction ZoneIRZ) und eine diffuse Zone mit blassbrauner oder purpurner Verfarbung distal davon (Reaction Zone,RZ). In der RZ wurden die Gesamtphenolgehalte bestimmt. Ausgewahlte Phenole (Pedunculagin,Tellimagrandin 1, Tetragalloylglucose, Pentagalloylglucose und Catechin) wurden quantitativ mitFlussigchromatographie-Massenspektrometrie (LC-MS) erfasst. Verschiedene Pilze (hauptsachlichFauleerreger) wurden fur die Wundinokulation verwendet. Die Ergebnisse zeigen, dass grossereFaulebereiche im Allgemeinen mit einer erhohten Bildung loslicher Phenole assoziiert waren. SterileInokulationen und schwach aggressive Pilzarten hatten keine oder nur schwache Verfarbungen desXylems zur Folge, wahrend aggressive Pilze eine starkere Verfarbung und Phenolanhaufung vor demInfektionsbereich auslosten. Das zeigt, dass Phenolanhaufung keine unspezifische Reaktion aufVerletzung ist, sodern eine variable Antwort aufgrund der Interaktion zwischen Mikroorganismus undSplintholz. Sechs Monate nach der Verletzung waren bei den Plantagenpflanzen purpurneReaktionszonen gewohnlich mit grossen Faulebereichen assoziiert. Inokulationen mit Ganodermaapplanatum zu verschiedenen Jahreszeiten bewirkten einen Monat nach der Verletzung keinesignifikanten Unterschiede im Fauleausmass.

References

ANON, 2000: Records of the Authenticated List of Plant Diseases in Tasmania. Hobart, Tasmania:Department of Primary Industry, Water and Environment.

BARRY, K. M., 2001: Antimicrobial defence in Eucalyptus nitens sapwood. Ph.D. Thesis. University ofTasmania, Australia.

177Effect of season and different fungi on wounds in Eucalyptus nitens

Page 16: Effect of season and different fungi on phenolics in response to xylem wounding and inoculation in Eucalyptus nitens

BARRY, K. M.; DAVIES, N. W.; MOHAMMED, C. L., 2001a: Identification of hydrolyzable tannins in thereaction zone of Eucalyptus nitens wood, by high performance liquid chromatography/electrosprayionization mass spectrometry. Phytochem. Anal. 12, 120–127.

BARRY, K. M.; PEARCE, R. B.; EVANS, S. D.; HALL, L. D.; MOHAMMED, C. L., 2001b: Initial defenceresponses in sapwood of Eucalyptus nitens (Maiden) following wounding and fungal inoculation.Physiol. Molec. Plant Path. 58, 63–72.

BARRY, K. M.; PEARCE, R. B.; MOHAMMED, C. L., 2000: Properties of reaction zones associated withdecay from pruning wounds in plantation-grown Eucalyptus nitens. For Pathol. 30, 233–245.

BAUCH, J.; SHIGO, A. L.; STARCK, M., 1980: Wound effects in the xylem of Acer and Betula species.Holzforschung. 34, 153–160.

BONELLO, P.; PEARCE, R. B., 1993: Biochemical defence responses in primary roots of Scots pinein vitro with Cylindrocarpon destructans. Plant Pathol. 42, 203–211.

CUNNINGHAM, G. H., 1963: The Thelephoraceae of Australia and New Zealand. NZ Department Sci.Industr. Res. Bull. 145. Wellington, New Zealand: Government Printer.

GADGIL, P. D.; BAWDEN, A. D., 1981: Infection of wounds in Eucalyptus delegatensis. NZ J. For. Sci.11, 262–270.

GRAMSS, G., 1992: Invasion of wood by basidiomycetous fungi. VI. Quantitative but not qualitative dif-ferences in the pathovirulence of pathogenic and saprophytic decay fungi. J. Basic Microbiol. 32, 75–90.

HATANO, T.; KIRA, R.; YOSHIZAKI, M.; OKUDA, T., 1986: Seasonal changes in the tannins ofLiquidambar formosana reflecting their biogenesis. Phytochemistry 25, 2787–2789.

MIREKU, E.; WILKES, J., 1989: Seasonal variation in the ability of the sapwood of Eucalyptus maculatato compartmentalize discolouration and decay. For Ecol. Manage. 28, 131–140.

MOHAMMED, C.; BARRY, K.; BATTAGLIA, M.; BEADLE, C.; EYLES, A.; MOLLON, A.; PINKARD, E., 2000:Pruning-associated stem defects in plantation E. nitens and E. globulus grown for sawlog andveneer in Tasmania, Australia. In: Proceedings of the Future of Eucalypts for Wood Products,19–24 March, Launceston, Australia. Launceston, Australia: International Union of ForestryResearch Organisations, pp. 357–364.

PEARCE, R. B., 1996: Antimicrobial defences in the wood of living trees. New Phytol. 132, 203–233.PEARCE, R. B., 2000: Decay development and its restriction in trees. J. Arboriculture 26, 1–11.PEARCE, R. B.; SUMER, S.; DORAN, S. J.; CARPENTER, T. A.; HALL, L. D., 1994: Non-invasive imaging of

fungal colonization and host response in the living sapwood of sycamore (Acer pseudoplatanus L.)using nuclear magnetic resonance. Physiol. Molec. Plant Pathol. 45, 359–384.

PEARCE, R. B.; WOODWARD S., 1986: Compartmentalization and reaction zone barriers at the margin ofdecayed sapwood in Acer saccharinum L. Physiol. Mol. Plant Path. 29, 197–216.

RAYNER, A. D. M.; BODDY, L., 1988: Fungal Decomposition of Wood. Its Biology and Ecology.Chichester, UK: John Wiley & Sons, 587 pp.

SCHWARZE, F. W. M. R.; BAUM, S., 2000: Mechanisms of reaction zone penetration by decay fungi inwood of beech (Fagus sylvatica). New Phytol. 146, 129–140.

SCHWARZE, F. W. M. R.; ENGELS, J.; MATTECK, C., 2000: Fungal Strategies of Wood Decay in Trees.Berlin: Springer-Verlag, 185 pp.

SHAIN, L., 1976: Resistance of sapwood in stems of loblolly pine to infection by Fomes annosus.Phytopathology 57, 1034–1045.

SHAIN, L., 1979: Dynamic responses of differentiated sapwood to injury and infection. Phytopathology69, 1143–1147.

SHIGO, A. L., 1974: Relative abilities of Phialophora melinii, Fomes connatus and F. igniarius to invadefreshly wounded tissues of Acer rubrum. Phytopathology 64, 708–710.

SHIGO, A. L.; MARX, H. G., 1977: Compartmentalization of Decay in Trees. For. Serv. Agric. Info.Bull. US Department of Agriculture, p. 405.

STALPERS, J. A., 1978: Identification of wood-inhabiting Aphyllophorales in pure culture. StudiesMycol. 16, 1–248.

WAINHOUSE, D.; ASHBURNER, R.; WARD, E.; ROSE, J., 1998: The effect of variation in light and nitrogenon growth and defence in young sitka spruce. Funct Ecol. 12, 561–572.

WHITE, D.; KILE, G. A., 1993: Discolouration and decay from artificial wounds in 20-year-oldEucalyptus regnans. Eur. J. For. Path. 23, 431–440.

YAMADA, T., 1992: Biochemistry of gymnosperm xylem responses to fungal invasion. In: DefenseMechanisms of Woody Plants Against Fungi. Ed. by BLANCHETTE, R. A.; BIGGS, A. R. Berlin:Springer Verlag, pp. 147–164.

YAMADA, T.; TAMURA, H.; MINEO, K., 1988: The responses of sugi (Cryptomeria japonica D. Don)sapwood to fungal invasion following attack by the sugi bark borer. Physiol. Molec. Plant Pathol.33, 429–442.

178 K. M. Barry, N. W. Davies and C. L. Mohammed