Study of Xanthomonas arboricola pv. juglandis Population Dynamics in French Walnut Orchards over Three Years M. Giraud 1,a, J.P. Prunet 2, A. Chevallier 2, S. Ramain 3, V. Thiriaud 4, I. Santrac 5 and O. Bray 6 1 Ctifl, Centre de Lanxade, 24130 La Force, France 2 Station Expérimentale de la Noix, Creysse, 46600 Martel, France 3 ENITA Clermont-Fd, d Amiens, France 4 ENSA Montpellier, France 5 Université, France 6 Université Reims, France Keywords: Juglans regia, Xanthomonas arboricola pv. juglandis, walnut blight, epidemiology Abstract Walnut blight, caused by Xanthomonas arboricola pv. juglandis is a major disease of Walnut in France, mainly responsible for necrosis on fruits and early nut drop before harvest. The population dynamics were studied over three years in orchards in the South-West of France, with comparison among orchards with different infestation levels. Samples of buds, leaves, flowers, pollen and fruits were collected from the beginning of April till the end of August, and the epiphytic population size of X. a. pv. juglandis was measured by plating dilutions of water extract on Tween Modified Medium. Population size broadly varies during the year from 10 3 to 10 7 cfu/g, increasing from budbreak, and remains stable after flowering. The level is a little lower in less infected orchards, suggesting that the expression of disease is more affected by pedo-climatic conditions than bacterial population level. INTRODUCTION Walnut blight, caused by Xanthomonas arboricola pv. juglandis is a major disease of Walnut in France, mainly responsible for necrosis on fruits and early fruit drop before harvest (Gardan et al., 1986). The use of copper, the only registered chemical for blight control, should be reduced for environmental reasons. A walnut blight forecasting model is currently under development by Ctifl (French Technical Institute for Fruits and Vegetables) with the collaboration of the French Experimental Station of Creysse, IRTA and the University of Girona in Spain. In order to validate the model, a research program has been carried out annually since 2004, within our working group. The main objectives are: i) to study the population dynamics of X. a. pv. juglandis in walnut orchards; ii) to establish the possible relationship between the bacterial level and the expression of the disease; iii) to supply missing data in the epidemiology of walnut blight, including the incidence under varying climatic conditions. The present paper concerns mainly the first point of this program. MATERIALS AND METHODS Orchards From 2006 to 2008, different walnut orchards growing cv. Franquette, were selected in the south-west growing region of walnut in France located close to the Experimental Station of Creysse: Creysse D1 is a trial plot on the Experimental Station of Creysse, planted in 1993 with a density of 8 8 m, untreated against walnut blight, but highly infected by this disease. The population dynamics in this plot was only studied in 2006. Bouscarel is a grower orchard, planted in 1995 (density 11 11 m), located nearby Creysse Station in a situation (bottom of a small valley) and with soil conditions a giraud@ctifl.fr Proc. VI th Intl. Walnut Symposium Ed.: D.L. McNeil Acta Hort. 861, ISHS 2010 439
favourable to walnut blight. The orchard is usually untreated against walnut blight, except for a part where a trial of chemical treatments (combination copper-mancozeb) was carried out by Creysse Station. In 2007, the untreated part was divided into 3 blocks to study the heterogeneity of the plot. Soursac is a grower orchard, planted in 1993 (density 11 11 m), and located at Queyssac-les-vignes. The location is on a chalky plateau and the soil is not favourable to walnut blight. In 2007, the orchard was also divided into 3 blocks to study the heterogeneity of the plot, in comparison with Bouscarel. Meyronne is a symptomless orchard, planted with a density of 8 10 m. The population dynamics in this plot was only studied in 2006. Sampling In 2006, only dormant, open and new summer buds were sampled. Five to eight buds per tree were randomly collected, using a sterile scalpel or disinfected (70% ethanol solution) pruning shears, to obtain a single sample of 50 buds per orchard. In 2007 and 2008, samples consisted of buds (dormant, open and summer buds), leaflets, flowers, pollen and fruits. In 2007, 7 trees were randomly chosen in each block and repeatedly sampled during the season. On each tree at breast height, 7 to 8 organs were collected to obtain the final number of 50 organs per block. After the results obtained in 2007 provided information on the heterogeneity of both studied orchards, 10 trees were randomly chosen in the most homogenous part of each entire orchard in 2008, and 2 organs were sampled per tree. Only healthy organs without any symptom of blight were sampled. According to the phenological stage, entire leaves or leaflets were used, and the number of leaflets per leaf was determined to get 5 grams of fresh matter per sample. Samples were stored at 4 C until the next day before analysis. Population Size Analysis Buds and flowers were cut longitudinally in two parts, leaves or leaflets in small pieces, fruits were peeled in 2007, but not in 2008. All organs (or parts of organs) were weighed and washed in 10 ml of sterile demineralised water per gram of fresh matter, on a rotary shaker for 45 minutes at 150 rpm. In 2008, flowers were washed in 1 ml of sterile demineralised water per organ, and 5 fruits within the sample of 20 fruits were randomly chosen to be washed in 4 to 30 ml of sterile demineralised water per fruit depending on the size of the fruit. Pollen was analyzed by our partner C. Moregraga (University of Girona, Spain), by washing 100-150 mg in 1 ml of sterile distilled water. Water extract was diluted from 10-1 to 10-5, and each dilution plated with 2 replicates on Tween Modified Medium (by S. Lindow, in Schaad et al., 2001). Colonies were counted after 5 days of incubation at 28 C, and population size expressed as log (cfu+1) per gram or per organ. RESULTS AND DISCUSSION In 2006, X. arboricola pv. juglandis population size broadly varied from 10 5 to 10 6 cfu/g in the buds of the infected orchards Bouscarel and Creysse D1, increasing from bud break to June, according to rain and temperature, and remaining stable after flowering. In the Meyronne orchard, where walnut blight symptoms had never been seen, an epiphytic population of X. a. pv. juglandis is naturally present inside the scales of the buds, with log values ranging from 5.1 to 5.4, but this level remained lower than in infected orchards (Fig. 1). We also observed that there was no difference between X. a. pv. juglandis population sizes in buds, in treated (3 sprayings with combination coppermancozeb before bloom) and untreated plots of Bouscarel after bloom, despite the incidence of the disease on nut drop being lower in the treated plot (Table 1). This observation confirms that vegetative buds are the main storage sites for the bacteria that remain protected inside the bud (Garcin et al., 2001). The only difference between population size in buds of treated and control plots was observed before bloom (Ff2), in flower buds from Af (dormant) to Df stage. It represents the potential of bacterial 440
inoculum before the growing season, and we suppose that it could be the consequence of the previous year situation. In 2007 and 2008, the objective was to compare the epiphytic population size of X. a. pv. juglandis in buds, on leaves and fruits, in two similar orchards, one highly infested: Bouscarel (untreated), and another less infested: Soursac. In 2007, X. a. pv. juglandis population size broadly ranged from 10 3 to 10 7 cfu/g on various organs of both orchards. At the dormant stage, the potential of bacterial inoculum in the buds was identical for both orchards, but became lower in Soursac at budbreak (Fig. 2A). On leaves, curiously, the population size was higher in Soursac than in Bouscarel before bloom, and the opposite occurred after fruit setting; the same observations was made on flowers and fruits. Conversely, pollen was significantly more infected on Bouscarel trees than on Soursac ones (Table 2), meaning that pollen is probably an important dissemination means for the pathogen, as previously reported (Ark, 1943). The heterogeneity of the results was generally very low, showing that the bacteria dispersion is homogenous in the orchard, if pedo-climatic conditions are homogenous. In 2008, meteorological conditions were less favourable to walnut blight because of colder temperatures during the spring, although it was an exceptionally rainy year. X. a. pv. juglandis population size ranged from 10 3 to 10 6 cfu/g in buds and on leaves, and from 4.10 2 to 10 5 cfu/organ on flowers and fruits of both orchards. There was no difference between bacterial levels in buds and on leaves of Bouscarel and Soursac plots (Fig. 2B), but on flowers and fruits, the level was lower in the Soursac orchard. Very few differences were observed in pollen infestation level (Table 2). CONCLUSION In buds of symptomless trees, we found that epiphytic X. a. pv. juglandis population size was around 10 5 cfu/g, but could reach 10 7 cfu/g on infected trees. Bacteria are protected inside the buds, and could not be affected by chemical treatments. The level of bacterial population on flowers and fruits was higher in Bouscarel orchard, during bloom in 2008 and in summer in 2007. Since first infection by X. a. pv. juglandis seems to occur around bloom, the relation between the level of bacterial population at this period and the incidence of disease was only observed in 2008. The X. a. pv. juglandis population size is a little lower in less infected orchards, suggesting that the expression of disease is probably more affected by pedo-climatic conditions than bacterial population level. Literature Cited Ark, P.A. 1943. Pollen as a source of walnut bacterial blight infection. Phytopathology 34:330-334. Garcin, A., El Maataoui, M., Prunet, J.P., Ginibre, T. and Penet, C. 2001. La bactériose du noyer: nouvelles connaissances pour une vieille maladie, synthèse de travaux réalisés (1995, 2000). Infos Ctifl 171:27-30. Gardan, L., Luisetti, J. and Gaignard, J.L. 1986. La bactériose du noyer. Phytoma-La Défense des végétaux 11:35-41. Schaad, N.W., Jones, J.B. and Chun, W. 2001. Laboratory Guide for identification of Plant Pathogenic Bacteria, 3 rd Ed. APS Press, St Paul. 441
Tables Table 1. Fruit drop due to walnut blight in the orchards Bouscarel and Soursac, respectively highly infected and less infected by X. a. pv. juglandis, during 3 years (% of total harvest yield). Orchards 2006 2007 2008 Bouscarel (untreated) 48% 50% 35% Bouscarel treated part 26% 36% - 1 Soursac <10% <10% 10% 1 no data available Table 2. X. a. pv. juglandis population size in pollen, expressed in log (cfu+1) per gram (Source: C. Moregraga, 2007, pers. comm.). Orchards 2007 2008 Bouscarel 6.30±0.28 6.33 Soursac 3.54±0.70 1 6.07 1 significantly different (P>0.05) Figurese 7.0 6.5 Af Df Ff2 6.0 log(cfu+1)/g 5.5 5.0 4.5 4.0 3.5 Bouscarel (untreated) Creysse D1 Bouscarel treated Meyronne 10/4 20/4 30/4 10/5 20/5 30/5 9/6 19/6 29/6 9/7 19/7 29/7 8/8 18/8 28/8 Fig. 1. Dynamics of X. a. pv. juglandis population size in buds of two infected orchards, Bouscarel (treated and untreated part) and Creysse D1, and the symptomless plot Meyronne, in 2006. Phenological stages: Af: dormant bud, Df: open bud, Ff: bloom. From bloom period, the values are corresponding to new summer buds. 442
8 7 Af Bf Ff2 Gf A 6 5 4 3 2 23/3 2/4 12/4 22/4 2/5 12/5 22/5 1/6 11/6 21/6 1/7 11/7 21/7 31/7 10/8 log (ufc+1)/g 20/8 30/8 8 7 Af Bf Ff2 Gf B log (ufc+1)/g 6 5 4 3 2 23/3 2/4 12/4 22/4 2/5 12/5 22/5 1/6 11/6 21/6 1/7 11/7 21/7 31/7 10/8 20/8 30/8 Bouscarel / buds Bouscarel / leaves Bouscarel / flowers, fruits Soursac / buds Soursac / leaves Soursac / flowers, fruits Fig. 2. Dynamics of X. a. pv. juglandis population size in buds, on leaves, flowers and fruits of Bouscarel and Soursac, respectively infected and less infected, in 2007 (A) and 2008 (B). Mean values for 3 blocks with standard error in graph A. Af: dormant, Bf: budbreak, Ff: bloom, Gf: fruit setting. In 2008 (graph B), the values for flowers and fruits are expressed in log (cfu+1) per organ. 443
444