A Survey of Phytoalexin Induction in of the by Copper Ions Tetsuo Kokubun and Jeffrey B. Harborne Department of Botany, University of Reading, Whiteknights, Reading, RG6 2 AS, U.K. Z. Naturforsch. 49c, 628-634 (1994); received April 28/May 26, 1994 Phytoalexins. Constitutive Fungitoxins, Hydroquinone, Aucuparin The leaves of 130 of Rosaceae were surveyed for phytoalexin induction. Both biotic and abiotic induction was examined and antifungal compounds were detected in 47 species. However, these compounds appeared to be constitutive metabolites, released from bound phenolic materials already present in the . In Pyrus, hydroquinone was produced from the hydrolysis of arbutin present in the vacuole before inoculation. In most other species, the fungitoxic agents were mainly catechin-like derivatives, apparently released from the tannins present within the leaf. By contrast, the synthesis in the leaf of the characteristic biphenyl or benzofuran phytoalexins which are produced in sapwood, was confined to a very few species. The biphenyl aucuparin was identified as a phytoalexin from the leaves of Sorbus aucuparia.

Introduction clear how universal within the angiosperms this Phytoalexin production is now well established mechanism is. We therefore decided to survey as one of several defence systems which provide members of the Rosaceae for phytoalexins in the with resistance to microbial infections. Sev­ leaves. Since the family is related both m orpho­ eral hundred phytoalexins have now been charac­ logically and chemically to the Leguminosae terized from over thirty families (Grayer and (Bate-Smith, 1961), which regularly produces iso- Harborne, 1994). The response has been mainly flavonoids as phytoalexins (Ingham, 1981), we studied using the drop diffusate technique on leaf expected to find some similarities in response tissue, but , shoot, stem and wood tissues have between the two families. also been used as well as cell culture. The type of Relatively little is known about the disease re­ compound formed de novo in a plant is usually sistance mechanisms in the Rosaceae, in spite of characteristic at the family level, in that isofla- its economic importance as a source of many culti­ vanoid phytoalexins are commonly produced in vated . The first phytoalexin to be reported legumes, sesquiterpenoids in the Solanaceae, in the family was benzoic acid, which is formed in furanocoumarins in the Umbelliferae and so on apple following infection by Nectria galligena (Harborne and Turner, 1984). As many as thirty (Browne and Swinburne, 1971). Since then, bi­ compounds may be formed in a given plant-fungal phenyl or benzofuran phytoalexins have been interaction, as happens in infected carnation Dian- characterized from sapwood of Cotoneaster lactea thus cciryophyllus, tissue (Niemann, 1993). (Burden et al, 1984), Malus pumila (Kemp et al., The protective value of phytoalexin synthesis in 1985) and Pyrus communis (Kemp and Burden, warding off fungal pathogens has been established 1984) and from the leaves of Eriobotrya japonica by the experiments of vanEtten and co-workers (Mikayado et al, 1985), Photinia glabra (Widya- (1989) on the Nectria haematococca-Pisum sati­ stuti et al, 1992) and Rhaphiolepsis umbellata vum interaction and by the successful genetic (Watanabe et al, 1990). Additionally some ses­ transfer of phytoalexin production (e.g. stilbene quiterpenoids have been characterized as anti­ synthase) from one plant, Vitis vinifera, to another, microbial agents in damaged Rosa rugosa leaves Nicotiana tabacum (Hain et al., 1993). In spite of (Hashidoko et al, 1989). Also, flavan-3-ols have these and many other experiments, it is not yet been reported to accumulate in fungus-infected leaves of several Rosaceae (Treutter and Feucht, 1990). The results of a representative survey of the Reprint requests to Dr. J. B. Harborne. family for phytoalexin synthesis in leaf tissue is Telefax: 0734-753671. presented here.

0939-5075/94/0900-0628 $ 06.00 © 1994 Verlag der Zeitschrift für Naturforschung. All rights reserved.

Dieses Werk wurde im Jahr 2013 vom Verlag Zeitschrift für Naturforschung This work has been digitalized and published in 2013 by Verlag Zeitschrift in Zusammenarbeit mit der Max-Planck-Gesellschaft zur Förderung der für Naturforschung in cooperation with the Max Planck Society for the Wissenschaften e.V. digitalisiert und unter folgender Lizenz veröffentlicht: Advancement of Science under a Creative Commons Attribution Creative Commons Namensnennung 4.0 Lizenz. 4.0 International License. T. Kokubun and J. B. Harborne ■ A Survey of Phytoalexin Induction in Leaves of the Rosaceae 629

Materials and Methods solution became maximal. Controls were run with Plant material 0.05% Tween-20 solution; they were incubated in the same manner as described above. Most plant material was botanically verified b) Hammer mill ground flakes of chitin and chi­ from the Harris Garden, the botanical garden of tosan, both of crab shell origin, were used as elici­ the University of Reading. Several species were tors. Suspension of 4000|igml~1 (Keen et al., grown from seed in pots in the glasshouse (see 1983) of fine powder were applied on the leaf sur­ Table I). In all cases, material treated with chemi­ face floating on water in Petri dish. cal insecticides and fungicides was avoided. Care was taken to use only healthy, non-damaged young Fungitoxin detection leaf tissue throughout the experiment. Plant material were collected between early May and Only cupric sulphate-treated leaves and the cor­ mid-August, 1992. responding controls were further analyzed as fol­ lows. The supporting C uS0 4 solution and water Fungal inoculum were decanted and extracted (x 2 ) separately with ca. 2 ml EtOAc. The extracts were brought to dry­ Fungal species used as phytoalexin inducers ness in an air flow. The residue were dissolved in (Alternaria alternata, Aspergillus niger, Botrytis a small amount of MeOH and analyzed on silica cinerea, Cladosporium herbarum, Fusarium cul- gel TLC with M eO H -C H C l 3 (49:1) and n-hex- morum, Geotricum candidum and Trichoderma ane-EtOAc-MeOH (60:40:1) (HEM) as solvent viride) were obtained from stock cultures main­ systems. At the same time, the leaves showing tained at the School of Plant Sciences, the Univer­ extensive necrosis were extracted with MeOH for sity of Reading. They were grown on a dish con­ overnight. This extract was concentrated and the taining potato-dextrose agar medium, pH 5.6. EtOAc soluble fractions were chromatographed Cultures vigorously producing spores or conidia on TLC plates similarly. The developed TLC were flooded with deionized water containing plates were sprayed with a dense spore suspension 0.05. Tween-20 and gently scraped. The spore of Cladosporium herbarum in a nutrient solution suspensions so obtained were passed through (KH 2 P 0 4, 4.7 g; K N 03, 2.7 g; Na2 H P 0 4 -2H 2 0 , three layers of lens tissue to remove mycelial frag­ 2 g; M gS0 4 -7H 2 0 , 0.7 g, glucose, 100 g, in 11 of ments and were then adjusted to a density of ap­ tap water, pH 5.7). After three days incubation in prox. l.OxlO 6 spores/ml with the aid of haemo- the moist condition at 22 °C, the presence/absence cytometer. of fungitoxins and their R f values recorded.

Stress treatment Further characterization of induced fungitoxin 1. Biotic induction Once induced fungitoxins, i.e. those not detected Freshly excised leaves were immediately floated in the control or in healthy tissue, were detected, on ca. 1 0 ml of deionized water, adaxial side up­ a larger scale induction was performed. Typically permost, in a plastic Petri dish (9 cm in diam.). a few tens of grams of fresh leaves were treated Fungal spore suspension droplets were placed on with 100 ml of 0.1% C uS 0 4 -5H20 solution. After the leaf surface and incubated under diffused light TLC purification with suitable solvent system, the at 20 ± 1 °C for 3 days (72-84 h). Controls re­ UV spectra with/without shift reagents (NaOAc, ceived only water droplets containing 0.05% NaOH) were measured. Also, some specific “phe­ Tween-20. nolic” reagents such as Folin-Ciocalteu’s and Gibbs were employed to determine their identity. 2. Abiotic induction Results a) A similar amount of leaf material was floated on ca. 10 ml of 0.1% of CuS04. 5H20 solution At the beginning of the survey, spore suspen­ containing 0.05% Tween-20 in a plastic Petri dish sions of several pathogenic and non-pathogenic (9 cm), so that the leaf area contacting with CuS0 4 fungi were used to cause phytoalexin induction. 630 T. Kokubun and J. B. Harborne • A Survey of Phytoalexin Induction in Leaves of the Rosaceae

Table I. Production of necrosis and of antifungal agents Table I. (Continued). in leaves of the Rosaceae. Species Necrotic Antifungal Species Necrotic Antifungal reaction agents+ reaction agents+ R. centifolia L. + - Subfamily Spiraeoideae R. felicite + - Tribe Spiraeae R. fdipes + - Aruncus sylvester Kneffii ++ R. hugonis Hemsley + - Neillia thibetica ++ R. hugonis x xanthina + - Physocarpus malvaceus R. iberica ++ - (Greene) O. Kuntze B, E R. laevigata Michx. + - Sibiraea altaiensis R. macrophylla + - (Laxm.) Schneid. + R. maximowicziana Sorbaria arborea + Regel “Jackii” + - Spiraea bella Sims + C, E R. mollis Sm. + - S. betulifolia Pallas ++ C, D. E R. multiflora Thunb. ++ - S. japonica L.** + R. nanothamnus x willmottiana ++ - S. nipponica “Tosaensis” ++ C, D. F R. nutkana Presl. S. pubescens ++ (R. muricata Greene) + - S. thunbergii ++ A. D, F R. primula Boulengep + - R. primula x hugonis + - Tribe Exochordeae R. primula x (primula x Exochorda racemosa omeiensis) + - (Lindl.) Rehder R. roxburghii Tratt + - Subfamily Rosoideae R. rubiginosa L. + - Tribe Kerrieae R. rugosa Thunb. + - R. salicitorum Lydberg + - Kerria japonica (L.) DC. “Picta” - B R. sericea Lindley ++ - K. japonica (L.) DC. “Pleniflora” - - R. virginiana - - scandens R. woodsii Lindley + - (Thunb.) Makino Tribe Potentilleae Tribe Sanguisorbeae Fragaria vesca L. Acaena buchananii Hooker, f.* var. americana Porter + A. hieronymi Kuntze* F. virginiana Duchesne + A. sanguisorba Vahl.* F. x ananassa Duchesne + Alchemilla conjuncta ++ Geum macrophyllum Willd ++ A. glabra + F, G G. montanum L. ++ Sanguisorba minor L.* + - G. pyrenaicum Miller ++ H S. officinalis L.* G. rivale L.* + B Sarcopoterium spinosum (L.) G. roylei Bolle + Spach G. x intermedium Ehrh. + Tribe Ulmarieae Potentilla anserina L. P. argyrophylla Filipendula hexapetala Gilib. P. fruticosa L. ++ “Flore Pleno” + P. nepalensis Hook.* + B F. ulmaria (L.) Maxim.* + P. reptans L. + C, G Subfamily Prunoideae P. x russelliana Lindley ex Sweet ++ Osmaronia cerasiformis (Torrey & Gray) Greene ++ - Rubus fruticosus + + R. idaeus L. + Prinsepia uniflora Batai. I, K, L Prunus armeniaca R. loganobaccus + ++ J R. odoratus L. ++ “Farmingdale” ++ K R. tricolor + P. autumnalis P. avium “Plena” ++ K. L Tribe Roseae P. cerasoides D. Don ++ O Rosa acluibrensis Chrsham. + - P. divaricata Ledeb. ++ L R. agrestis Savi + - P. domestica L. “Plum” ++ - R. alba L. - - P. domestica L. “Stella" ++ L R. beggeriana Schrenk P. domestica L. “Victoria” ++ - ex Fisch. & May + F P. laurocerasus L. + - R. blanda Aiton + - P. lusitanica L. + O R. brunonii Lindley ++ - P. padus L. ++ I. J, K. L R. celeste + - P. persica "Duke of York” ++ I. K, L T. Kokubun and J. B. Harborne • A Survey of Phytoalexin Induction in Leaves of the Rosaceae 631

Table I. (Continued). However, these spores germinated rather poorly on leaves, even after three days incubation, while Species Necrotic Antifungal reaction agents+ those in water germinated well. Under the micro­ scope, it was apparent that there was hardly any P. persica var. nectarine necrotic response in the cells underneath the drop­ “Pineapple” ++ I. K, L P. rufa Hooker Fil. ++ I, L lets. On the other hand, cupric sulphate-treated P. serrula Frachet ++ K, L, M, N leaves produced a necrotic response and released P. spinosa L. ++ - some antifungal compounds at the same time. A P. tenella Batsch. ++ I, L, M P. yedoensis + - necrotic response was observed with most species P. “Ichiyo” ++ J, K, L (Table I). The identity of these active compounds P. “Mikurama-Gaeshi” ++ K is not entirely clear; however, the UV spectra P. “Taihaku” ++ K strongly suggest that they are probably catechins Subfamily Maloideae and gallocatechins. Amelanchier alnifolia Nuttal + G, S, T Plants of the subfamily Prunoideae often pro­ A. ova/is Medickus + G Aronia arbutifolia (L.) Elliott duce strong fungitoxins, whereas plants of the Ma- var. atropurpurea Robins. + _ loideae, do this less often. The Spiraeoideae also Chaenomeles cathayensis produce fungitoxins, but the UV spectra of the (Hemsley) Schneid. ++ T predominant fungitoxins (Table II) suggest inter­ C. japonica (Mast.) Lavallee + - Cotoneaster acutifolius Turcz. + - esting differences between those of the Pru­ C. divaricatus Rehder & Wilson + - noideae and of the Maloideae. The Rosoideae C. horizontalis Decne. + L seldom produce fungitoxins as stress metabolites. C. splendens Flinck & Hylmö + - Crataegus monogyna Jacq. ++ - Most of the antifungal compounds detected re­ C. pontica C. Koch ++ L sponded positively to phenolic spray reagents on C. prunifolia (Poir.) Pers. ++ - TLC plates. The data available including chroma­ Cydonia oblonga Miller** ++ L, Q Malus baccata (L.) Borkh. var. tographic behaviour for the fungitoxic compounds, mandshurica Schneid. ++ _ designated A to W, are listed in Table II. M. coronaria ++ - Apart from the catechin-like compounds, two M. domestica Borkh. ++ - distinctive antifungal compounds were identified. M. fusca (Raf.) Schneid. + L M. hupehensis (Pamp.) Rehder ++ - One is hydroquinone (compound R) from all M. orientalis ++ - Pyrus species examined. Hydroquinone is well M. sieboldii (Regel) Rehder ++ B, P, S known as the aglucone of arbutin, which is ubiqui­ M. sieversii (Ledeb) Roem ++ - M. silvestris Miller ++ P tous in Pyrus as a constitutive compound (Bate- M. toringoides (Rehder) Smith, 1961). The other is aucuparin (compound Hughes ++ U) from Sorbus aucuparia. Although aucuparin Mespilus germanica L. ++ and its derivatives were discovered first in the Photinia davidiana - Pyracantha coccinea Roemer heartwood of this plant (Erdtman et al., 1963), the “Lalandii” • W leaves are apparently a new source. Aucuparin was P. coccinea “Golden Sun” ++ V not detected in the whole extract of healthy, non­ P. coccinea “Mohave” ++ V Pyrus communis L. treated leaves and appears therefore to be a true “Doyenne du comice” ++ s, w phytoalexin in the leaves of this plant. P. communis L. Finally, in attempting a chemical elicitation by “Glou Morceau” ++ P. R, V P. elaeagrifolia Pallas ++ R, V biological means, treatment with chitin and chito­ P. pyraster ++ Q, R, V san caused virtually no response in the host plants. P. ussuriensis Maxim. ++ - Among those examined, only Malus toringoides Sorbus aucuparia L. “Edulis” ++ u showed an extensive necrotic “lesion”. In this case, S. commixta Hedl. + - S. x intermedia (Ehrh.) Pers. + - the flakes of chitin and chitosan became yellow, perhaps due to deposition of colouring material * Grown in glasshouse. leaked from the leaves. A similar necrotic re­ ** Plant material provided from local gardens. + For properties, see Table II. sponse and the leakage of yellow substance to the supporting water was also observed on treatment 632 T. Kokubun and J. B. Harborne • A Survey of Phytoalexin Induction in Leaves of the Rosaceae

Table II. Rf values. UV/VIS spectra and chemical nature of fungitoxins.

UV/VIS spectral maxima [nm] Spray reagent* Com­ Rf (xlOO) MeOH +NaOH +NaOAc F-C Gb pN pound in HEM

A 0 _ B 26 282 - 280 maroon C 34 - D 48 268 268 E 57 - F 66 268 264 264 G 13 - yellow H 84 - I 17 274. 319 275 274, 312 - purple J 30 275 275 275 - K 43 270 - 268 L 56 226, 272, 272 sh. 226, 272, - 280 sh 279 sh 280 sh ± purple M 66 288, 292 -- ± N 77 274 -- O 91 226, 282 grey mauve P 6 - grey grey yellow Q 14 224 sh - - + blue R+ 36 227, 295 304 sh, 312, 219 + mauve yellow S 43 -- T 52 272 275 sh 275 sh U++ 58 273 + V 63 225, 285 - w 74 226, 267 sh, 275, 282 sh

* Key: F-C, Folin-Ciocalteu’s reagent with ammonia vapour; Gb, Gibbs reagent; pN, diazotized p-nitroaniline with 20% Na2C 0 3 overspray. + Identified as hydroquinone by direct comparison (UV spectra, co-TLC) with an authentic marker. ++ Identified as aucuparin by direct comparison (UV spectra, MS, co-TLC, co-PLC) with an authentic marker. with cupric sulphate solution, but no fungitoxic Feucht, 1990 a, b). Indeed, Rosaceous plants are in compound was detected. Two Prunus species general rich in tannins based on procyanidin and (P. padus and P. yedoensis) showed only a slight ellagic acid, and are recognized as a “tanniferous necrotic spot underneath the droplets, but other­ family”, beside possessing high amounts of phe­ wise no response. nolics such as p-coumaric acid and caffeic acid (Bate-Smith, 1961). The toxicities of catechins and Discussion tannins against microorganisms are well docu­ mented (Scalbert, 1991). The results of a survey of leaves of 130 plants We employed a variety of both biotic and from the Rosaceae (12, Spiraeoideae; 62, Roso- abiotic techniques to induce phytoalexin synthesis ideae; 22, Prunoideae; 34, Maloideae) revealed so that our results cannot be simply due to exper­ only one species, Sorbus aucuparici, that gave a imental error (see also Hashidoko et al, 1989). genuine phytoalexin response. Instead of phyto­ Our techniques were successful in inducing the ex­ alexin production, over a third of the species pected phytoalexins in several non-rosaceous boosted the production of constitutive phenolic plants, including the pea, groundnut and rice. A compounds such as catechins. Our results are in similar study on Cucurbitaceae earlier revealed line with other reports of flavan-3-ols as major the lack of a phytoalexin response in that family antifungal compounds in this family (Treutter and (Deverall. 1976). It is apparent therefore that T. Kokubun and J. B. Harborne • A Survey of Phytoalexin Induction in Leaves of the Rosaceae 633 plants in certain families do in fact lack an active Disease resistance in plant relies on several phytoalexin induction system in the leaves. Macro- complex mechanisms, and phytoalexin production molecular barriers, such as chitinases and other is only a part of an integrated natural defence sys­ glucanohydrolyases of plant origin (Ghaouth et al, tem. The results of our survey show that in the 1991) which degrade fungal cell walls, may be leaves of rosaceous plants, “phytoalexins” are not present instead. often produced as major protective compounds. Overall, the resistance mechanisms of the rosa­ This is further exemplified by the fact that when ceous plants to fungal pathogens are still unclear. they are occasionally produced in the leaves, as Possessing antifungal compounds does not simply they are in Photinia glabra (Widyastuti et al, mean that the plants are resistant (Hunter, 1971; 1992), the amount formed is quite insufficient to Oydvin and Richardson, 1987; Sierotzki and halt the advancement of pathogens. Our survey Gessler, 1993). If a fungitoxin, or any other ward­ has more recently been extended to cover fruit ing off agent, is involved in a plant’s defence and sapwood tissues. Preliminary results reveal mechanism, it has to obey important criteria (Har­ that the fruit tissue produce many phenolic anti­ borne, 1987). In fact, some pathogens on the rosa­ fungal compounds especially when they are imma­ ceous plants grow without contacting the intracel­ ture, but these are constitutive components of lular components. They establish hyphal growth healthy tissue. On the other hand, sapwood tissues between the cuticle and epidermal cells or inter­ respond dynamically, and a range of phytoalexin­ cellular spaces, without penetrating the cells like compounds have been detected, as will be re­ (Dickinson, 1982; Valsangiacomo and Gessler, ported later. We have also detected phytoalexin 1992). This is true for Venturia inaequalis, Diplo- induction in the roots of a herbaceous member of carpon rosae, a black spot fungus on Rosa and the family, Sanguisorba minor (Kokubun et al, Taphrina deformans, a hypertrophic fungi causing 1994). the leaf curl on Prunus species.

Bate-Smith E. C. (1961), Chromatography and Grayer R. and Harborne J. B. (1994), A survey of anti­ in the Rosaceae, with special reference fungal compounds in higher plants 1982-1993. to Potentilla and Prunus. J. Linn. Soc. Bot. 58, Phytochemistry, in press. 39-54. Hain R., Reif H. J., Krause E., Langebartels R., Brown A. E. and Swinburne, T. R. (1971), Benzoic acid: Kindl H., Vornam B., Wiese W., Schmelzer E., an antifungal compound formed in Bramley’s seed­ Schreier P. H., Stocker R. H. and Stenzel K. (1993), ling apple fruits following infection by Nectria galli­ Disease resistance results from foreign phytoalexin gena Bres. Physiol. Plant Pathol. 1, 469-475. expression in a novel plant. Nature 361, 153-156. Burden R. S., Kemp M. S. and Wiltshire C. W. (1984), Harborne J. B. (1987), Natural fungitoxins, in: Biologi­ Isolation and structure determination of cotone- cally Active Natural Products (K. Hostettman and furan, an induced dibenzofuran from Cotoneaster P. J. Lea, eds.). Ann. Proc. Phytochem. Soc. Eur. lactea. J. Chem. Soc. Perkin Trans. I, 1445-1447. Clarendon Press, Oxford, pp. 195-211. Deverall R. J. (1976), Current perspectives in research Harborne J. B. and Turner B. L. (1984), Plant Chemo- on phytoalexins, in: Biochemical Aspects of Plant- systematics. Academic Press, London. Parasite Relationships (J. Friend and D. R. Threlfall, Hashidoko Y., Tahara S. and Mizutani J. (1989), Anti­ eds.). Academic Press, London, pp. 207-223. microbial sesquiterpene from damaged Rosa rugosa Dickinson C. H. (1982), Plant pathology and plant leaves. Phytochemistry 28, 425-430. pathogens, in: Basic Microbiology, Vol. 6 (J. F. Hunter L. D. (1971), Is sieboldin a factor in apple scab Wilkinson, ed.). Blackwell Scientific Publications, resistance? Rep. E. Mailing Res. Stn. for 1970, Oxford. pp. 134-135. Erdtman, H., Eriksson G. and Norin T. (1963), Aucupa- Ingham J. (1981), Phytoalexin induction and its taxo­ rin and methoxyaucuparin, two phenolic biphenyl nomic significance in the Leguminosae, in: Ad­ derivatives from the heartwood of Sorbus aucuparia vances in Legume Systematics (R. M. Polhill and (L.). Acta Chem. Scand. 17, 1151-1156. P. H. Raven, eds.). HMSO, London, pp. 599-626. Ghaouth A. E., Arul J., Grenier J. and Asselin A. (1991), Keen N. T., Yoshikawa M. and Wang M. C. (1983), Glucanohydrolases and inhibitory activity to Botry­ Phytoalexin elicitor activity of carbohydrates from tis cinerea in extracts from strawberry fruit. Can. J. Phytophthora megaspora f. sp. glycinea and other Plant Pathol. 13, 315-320. sources. Plant Physiol. 71, 466-471. 634 T. Kokubun and J. B. Harborne • A Survey of Phytoalexin Induction in Leaves of the Rosaceae

Kemp M. S. and Burden R. S. (1984), Isolation and Sierotzki H. and Gessler C. (1993), Flavan-3-ol content structure determination of y-pyrufuran, a third in­ and the resistance of Malus x domestica to Venturia duced antifungal dibenzofuran from the wood of inaequalis (Cke) Wint. Physiol. Mol. Plant Pathol. Pyrus communis L. infected with Chondrostereum 42, 291-297. purpureum (Pers. ex Fr.) Pouzar. J. Chem. Soc. Treutter D. and Feucht W. (1990a), Accumulation of Perkin Trans. I, 1441-1443. flavan-3-ols in fungus-infected leaves of Rosaceae. Kemp M. S., Holloway P. J. and Burden R. S. (1985), Z. Pflanzenkr. Pflanzenschutz 97, 634-641. A pentacyclic triterpene induced in wood of Malus Treutter D. and Feucht W. (1990b), The pattern of pumila infected with Chondrostereum purpureum. J. flavan-3-ols in relation to scab resistance of apple Chem. Res., 154-155. . J. Hortic. Sei. 65, 511-517. Kokubun T., Harborne J. B. and Eagles J. (1994), 2',6'- Valsangiacomo C. and Gessler C. (1992), Purification Dihydroxy-4'-methoxyacetophenone, a phytoalexin and characterization of an exopolygalacturonase from the roots of Sanguisorba minor. Phytochemis­ produced by Venturia inaequalis, the causal agent of try 35, 331-333. apple scab. Physiol. Mol. Plant Pathol. 40, 63-77. Miyakado M., Watanabe K., Ohno N., Nonaka F. and VanEtten H. D., Matthews D. E. and Mathews P. S. Morita A. (1985), Isolation and structural determi­ (1989), Phytoalexin detoxification: importance for nation of eriobofuran, a new dibenzofuran phyto­ pathogenicity and practical applications. Annu. Rev. alexin from leaves of loquat, Eriobotrya japonica L. Phytopath. 27, 143-164. J. Pestic. Sei. 10, 101-106. Watanabe K., Widyastuti S. M. and Nonaka F. (1990). Niemann G. J. (1993), The anthranilamide phytoalexins Two biphenyl compounds from Rhaphiolepsis um- of the Caryophyllaceae and related compounds. bellata as its phytoalexins. Agric. Biol. Chem. 54, Phytochemistry 34, 319-328. 1861-1862. Oydvin J. and Richardson D. G. (1987), A paper chro­ Widyastuti S. M., Nonaka F., Watanabe K., Sako N. and matographic survey of the phenol content of apple Tanaka K. (1992), Isolation and characterization of peel from scab-resistant and scab-susceptible geno­ two aucuparin-related phytoalexins from Photinia types. Nowr. J. Agric. Sei. 1, 7-14. glabra. Ann. Phytopath. Soc. Japan 58, 228-233. Scalbert A. (1991), Antimicrobial properties of tannins. Phytochemistry 30, 3875-3883.