Effects of Antibiotic Applications on Epiphytic in the Apple Kiersten Tancos and Kerik Cox Department of Plant Pathology and Plant-Microbe Biology New York State Agricultural Experiment Station, Cornell University, Geneva

This research was supported in part by the NY Apple Research and Development Program

rwinia amylovora, the causal agent of fire blight, is the spread of SmR a destructive bacterial pathogen capable of causing E. amylovora to disease in numerous rosaceous plant species throughout other apple growing E t h e w o r l d regions. Subsequent “[This work was conducted] to monitor ( B o n n 2 0 0 0 ) . fire blight surveys This pathogen in 2004 and 2006 the presence of streptomycin resistance i s e s p e c i a l l y did not result in in epiphytic bacterial populations devastating in the detection of following applications of streptomycin apple and pear SmR E. amylovora, and kasugamycin after bloom.” a n d l e a d s t o leading to the c o n s i d e r a b l e assumption that p r o d u c t i o n eradication efforts losses in the United States annually. E. amylovora causes were successful blossom blight during bloom which, in turn, often leads to a in containing more devastating infection of young shoots and rootstocks the outbreak (Vanneste 2000). Blighted tissues typically become blackened (Russo 2008). and seemingly burnt, resulting in the appearance for which However, in recent the disease is named (Figure 1A). The loss of blossoms and investigations, SmR Figure 1. Typical fire blight symptoms on fruit-bearing shoots leads to a decrease in fruit production E. amylovora was an Idared apple tree. A) Blossom and potentially loss of bearing wood and entire trees. This is recovered from cluster 15 days after infection. The especially the case for moderate to highly susceptible apple apple production cluster shows blackened aborted fruitlets that have become stiff and cultivars, which are widely planted and in high demand (Bonn operations in several will fail to abscise during thinning. 2000). While there are several horticultural management western New York B) A current season’s shoot tip 7 days practices and chemical management options, the most effective counties spanning after infection. The underside of management practices for controlling infection at bloom are three of the state’s the leaves have an orange cast, and orange droplets of ooze are visible the aminoglycoside antibiotics streptomycin and kasugamycin. six apple growing on the wilting young shoot tissue. While use of kasugamycin is relatively new to the management regions (Bekoscke of fire blight, streptomycin has been used to control fire blight 2014). in the eastern United States for over 50 years. Aside from The development of streptomycin resistance has been at- these aminoglycoside antibiotics, there are no viable chemical tributed to streptomycin overuse after bloom to control the management alternatives that provide a comparable and shoot blight phase of the disease. Several survey studies have acceptable level of blossom blight control in the temperate correlated the history of streptomycin use in production or- production conditions of the eastern United States. chards with the recovery of streptomycin-resistant isolates of The reliance of the apple industry on streptomycin in E. amylovora (Loper 1991; Yashiro 2012). Aside from the afore- particular has become an issue of concern in recent years, due mentioned studies, the majority of the evidence for the develop- the emergence of streptomycin resistance in populations of E. ment of streptomycin resistance remains anecdotal. To bridge amylovora the United States (Moller 1981). The first reports the knowledge gap between antibiotic applications and resis- of streptomycin resistance occurred in California in 1972 and tance development, we wished to monitor the presence of strep- shortly afterwards in Washington (Miller 1972;Coyier 1975). tomycin resistance in epiphytic bacterial populations following Currently, streptomycin resistance is found in several western applications of streptomycin and kasugamycin after bloom. and midwestern states such as Missouri and Michigan (McManus We hypothesized that increasing applications of streptomycin 1994). The most recent reports of streptomycin-resistant (SmR) would lead to an increased recovery of streptomycin-resistant E. amylovora occurred in New York in 2002 (Russo 2008). epiphytic bacteria, while increased applications of kasugamycin These SmR isolates originated from a single orchard site, which would not preferentially select for streptomycin-resistant epi- prompted immediate eradication efforts in order to prevent phytic bacteria.

NEW YORK FRUIT QUARTERLY . VOLUME 23 . NUMBER 4 . WINTER 2015 23 Materials and Methods To determine the effects of streptomycin and kasugamycin application schedules on the selection of populations of E. amylovora and other epiphytic non- with streptomycin resistance, field trials were conducted in two plantings of ‘Idared’ apples at the New York State Agricultural Experiment Station. Both plantings consisted of tall spindle ‘Idared’ apples on B.9 rootstocks approximately 15 (Orchard 1) and 7 (Orchard 2) years in age. At 80% bloom in both orchard plantings, treatment programs consisting of varying applications of either streptomycin or kasugamycin were implemented as follows: 1. No applications of streptomycin or kasugmycin during full bloom or terminal elongation 2. A total of three applications of streptomycin or kasugmycin with one application during bloom, and two applications during early terminal elongation (1–3 inches) 3. A total of five applications of streptomycin or kasugmycin with one application during bloom, and four more applications through terminal elongation (15 inches) 4. A total of ten applications of streptomycin or kasugmycin with one application during bloom, and nine more applications through terminal elongation and beyond Applications of streptomycin were made using Firewall 17WP (AgroSource, Inc., Cranford, NJ) at a rate of 24 oz/A and applications of kasugamycin were made using Kasumin Figure 2. The recovery of bacterial epiphytes from the apple phyllosphere 2L (Arysta LifeScience, Cary, NC) at rate of 64 fl oz/A. All following streptomycin treatment programs in a 15-year old ‘Idared’ planting (Orchard 1) (A) and a 7-year old ‘Idared’ applications were made at weekly intervals using a gas-powered planting (Orchard 2) (B). Bars represent the mean percent and backpack sprayer (200 PSI) to four replicate single tree plots. the standard error of the total CFU/L for Pantoea agglomerans, Streptomycin programs were implemented in both the 7- Pseudomonas sp. and miscellaneous epiphytic bacteria (Misc.). and 15-year old ‘Idared’ planting (Orchards 1 and 2), but the kasugamycin program was only implemented in the 15-year old planting (Orchard 1). Calendar-based applications of protectant colonies were also tested for the presence of the gene pair strA/ fungicides, insecticides, and herbicides were applied to all trees strB, which is commonly responsible for streptomycin resistance for plot maintenance throughout the duration of this trial. in Pseudomonas sp., Pantoea agglomerans, and E. amylovora One week after the completion of the final antibiotic in NY (Russo 2008). A subset of the remaining miscellaneous application in each treatment program, 50 leaves were collected epiphytic bacteria with distinctive colony morphology on throughout the canopy from each of the four replicate trees. the two selective media was identified through PCR and Leaves were subsequently sonicated in 10X phosphate- sequencing of the internal transcribed spacer region (ITS) in buffered saline solution to remove epiphytic bacteria. From bacterial ribosomal DNA using the primers and procedures each sonicated leaf rinsate sample, ten replicate aliquots of 100 described previously (Jensen et al. 1993). The effect of antibiotic ul were dilution plated to an enumerable amount of bacterial treatment program on bacterial count data was determined colony forming units when necessary on Cross-Goodman (CG) using Generalized Linear Mixed Models with the GLIMMIX media at 28°C for 2 days, to select for Erwinia spp. and closely- procedure of SAS v9.3 (SAS Institute). Differences in the mean related epiphytic bacterial species. To detect the presence of CFUs and mean percent of CFUs between antibiotic treatment antibiotic-resistant epiphytic bacteria, resulting colonies from programs were determined using the ‘lsmeans’ statement of the dilution plating were also spot-plated on CG media that was GLIMMIX at the 5% level of significance. amended with 100 ppm analytical grade streptomycin sulfate, or 100 ppm analytical grade kasugamycin hydrochloride and Results and Discussion grown at 28°C for 48 hrs. To detect the presence of epiphytic In both orchard locations and for all antibiotic treatment Pseudomonas sp., colonies from the dilution plating were also programs, Pantoea agglomerans and fluorescent Pseudomonas spot-plated on King’s B medium. The resultant colony forming species were the most frequently recovered epiphytic bacteria. units were identified on the basis of characteristic morphology This observation was not surprising given that these bacteria on CG media and fluorescence on King’s B media. To confirm are commonly recovered from apple blossoms and leaves in scoring by morphology, a subset of colonies from each plate, similar studies on apple epiphytes (McGhee 2010; Yashiro selected on the basis of colony morphology, were subjected to 2012). Despite the fact that there was active shoot blight confirmation by PCR using primers specific to the 16S rRNA of throughout the orchard planting, Erwinia amylovora was rarely Pseudomonas sp. (Widmer, 1998), the pag2R gene of Pantoea recovered from apple leaves. Across all treatment programs, E. agglomerans (syn. Erwinia herbicola) (Braun-Kiewnick 2012), amylovora was present in less than 0.1% of leaf rinsate samples. and the pEA29 plasmid of E. amylovora (Russo 2008). Selected Given that E. amylovora is a poor epiphyte (Bonn 1981), the

24 NEW YORK STATE HORTICULTURAL SOCIETY were recovered from leaf rinsate plated on kasugamycin- amended medium. Hence, excessive application of kasugamycin (i.e., 10 applications) after bloom had no impact on the selection or recovery of kasugamycin-resistant epiphytes. Certainly, additional years of field trials would be needed to validate such an observation. Despite the fact that little to no Erwinia amylovora was recovered from shoot tissues, increasing applications of streptomycin (Fire Wall 17 WP) did have an impact on the overall populations of P. agglomerans and Pseudomonas in the phyllosphere. In Orchard 1, the percentage of the total epiphytic population represented by P. agglomerans decreased with increasing applications of streptomycin (Figure 2A). Similar results were obtained for Orchard 2 (Figure 2B). By comparison, the percentage of the total epiphytic population identified as Pseudomonas sp. increased as streptomycin applications increased in both Orchard 1 and Orchard 2 (Figure 2). No significant differences were observed within the percentage of the total epiphytic population represented by the miscellaneous epiphytic bacteria (Figure 2). In terms of total epiphytic bacteria recovered, the number of streptomycin applications had no bearing on the CFU/L for each of three types of epiphytic bacteria (data not shown). Interestingly, increasing applications of kasugamycin (Kasumin 2L) had a slightly different impact on the overall Figure 3. The recovery of bacterial epiphytes from the apple phyllosphere populations of P. agglomerans and Pseudomonas in the following kasugamycin treatment programs in a 15-year old phyllosphere. In Orchard 1, the percentage of the total ‘Idared’ planting (Orchard 1) (A) and a 7-year old ‘Idared’ planting (Orchard 2) (B). Bars in (A) represent the mean percent and the epiphytic population represented by P. agglomerans increased standard error of the total CFU/L for Pantoea agglomerans, with increasing applications of kasugamycin (Figure 3A). By Pseudomonas sp. and miscellaneous epiphytic bacteria comparison, the percentage of the total epiphytic population (Misc.). Bars in (B) represent the mean total CFU/L for Pantoea represented by Pseudomonas species and the remaining agglomerans, Pseudomonas sp. and miscellaneous epiphytic bacteria (Misc.). miscellaneous epiphytic bacteria decreased with increasing applications of kasugamycin (Figure 3A). In terms of total epiphytic bacteria recovered, trees receiving ten applications lack of E. amylovora in leaf rinsate samples was somewhat of kasugamycin had considerably lower CFUs/L for each of expected. In addition to the aforementioned bacteria, a variety three types of epiphytic bacteria (Figure 3B). Moreover, all the of miscellaneous epiphytic bacteria were recovered from leaf bacteria colonies recovered for trees receiving ten applications rinsate. Across all treatment programs, they represented less of kasugamycin were identified asP. agglomerans. than 20% of the total CFUs recovered from dilution plating (Figs. Despite the fact the total number of epiphytic bacteria 2, 3). These included several Pantoea spp. (e.g., P. ananatis), declined with increasing applications of kasugamycin, the and Erwinia spp. (e.g., E. rhapontici), and Sphingomonas spp. increase in the proportion of P. agglomerans in the phyllosphere Irrespective of the species, nearly all of the epiphytes recovered following applications of kasugamycin may be disconcerting. had streptomycin resistance. Across all treatment programs and In the case of streptomycin resistance development in E. orchards, less than 5–20% of the epiphytic bacteria recovered amylovora, it is believed that streptomycin resistance genes strA were streptomycin sensitive. In both orchards, the recovery of and strB were transferred to E. amylovora from P. agglomerans streptomycin-resistant epiphytes was significantly lower (P < (Chiou 1993; McGhee 2010). Hence, this epiphytic bacterial 0.05) for the trees that received no applications of streptomycin species could be abundant following kasugamycin applications than those receiving three, five, or ten applications. Although to serve as a source of horizontally transferred kasugamycin we believed that applications of kasugamycin would not exert resistance genes should they arise in the future. Regardless, equivalent selection for streptomycin resistance and sensitivity, it’s important to note that no kasugamycin-resistant bacteria increased application of kasugamycin resulted in a decrease in were recovered in these trials, and that the population of the percentage of streptomycin-resistant epiphytes recovered P. agglomerans was reduced by nearly 1000-fold after ten from the apple phyllosphere. While one might be tempted to applications of kasugamycin (Figure 3A). suggest that there is a fitness cost that may be associated with In summary, our results further confirm that E. amylovora streptomycin resistance, this observation may be an artifact is not present in the phyllosphere in high abundance, even if of the decrease in total epiphytic bacteria following increased there is active fire blight in the planting. Along these lines, application of kasugamycin (Figure 3B). Further field and three, five, or ten applications of streptomycin after bloom did laboratory studies would be necessary to elucidate the effects of not result in an increased recovery of streptomycin-resistant E. kasugamycin on total streptomycin-resistant bacteria. amylovora. However, other common bacterial epiphytes of apple In contrast, in streptomycin-amended medium, no bacteria (P. agglomerans & Pseudomonas) with streptomycin resistance

NEW YORK FRUIT QUARTERLY . VOLUME 23 . NUMBER 4 . WINTER 2015 25 did increase with increasing application of streptomycin, Moller, W., Schroth, M., and Thompson, S. V. 1981. The scenario further suggesting that use of streptomycin after bloom should of fire blight and streptomycin resistance. Plant Disease. 65: be minimal and reserved for trauma events. Finally, increasing 563–568. application of kasugamycin appeared to reduce the overall Russo, N. L., Burr, T. J., Breth, D. I., and Aldwinckle, H. S. 2008. number of bacterial epiphytes and the percentage of bacterial Isolation of streptomycin-resistant isolates of Erwinia epiphytes with streptomycin resistance. amylovora in New York. Plant Disease. 92: 714–718. Widmer, F., Seidler, R. J., Gillevet, P. M., Watrud, L. S., Di Acknowledgements Giovanni, G. D. 1998. A highly selective PCR protocol We would like to acknowledge the funding support for for detecting 16s rRNA genes of the genus Pseudomonas this project provided by the Apple Research and Development (sensu stricto) in environmental samples. Applied and Program. We would like to thank Catherine Miller and Jason Environmental . 64: 2545–2553. Hagel for assistance in sample collection and processing, and the Yashiro, E., McManus, P. S. 2012. Effect of streptomycin New York Agricultural Field Research Unit for their efforts in treatment on bacterial community structure in the apple maintaining the experimental test orchards. phyllosphere. PLoS ONE 7(5): e37131. doi:10.1371/journal. pone.0037

Literature Cited Kerik Cox is a research and extension professor who Bekoscke, K., Breth, D., Kuehne, S., Borejsza-Wysocka, E., leads Cornell’s program in disease control of tree fruits Aldwinckle, H., Villani, S., and Cox, K. 2014. Status of and berries. Kiersten Tancos is a graduate student who streptomycin resistant fire blight in New York orchards. works with Kerik Cox. New York Fruit Quarterly 22: 5–8. Bonn, W.G. 1981. Monitoring of epiphytic Erwinia amylovora and the incidence of fire blight of apple and pear in southwestern Ontario. Acta Horticulturae 117: 31–36. Bonn, W. G., and van der Zwet, T. 2000. Pages 37–53 in: Fire Blight: the Disease and Its Causative Agent, Erwinia amylovora. J. L. Vanneste, ed. CAB International. Wallingford, UK. Braun-Kiewnick, A., Lehmann, A., Rezzonico, F., Wend, C., Smits, T.H.M., and Duffy, B. 2012. Development of species-, strain- and antibiotic biosynthesis-specific quantitative PCR assays for Pantoea agglomerans as tools for biocontrol monitoring. J. Microbiol. Methods. 90: 315–320. Burr, T. J., Norelli, J. L., Katz, B., Wilcox, W. F., and Hoying, S. A. 1988. Streptomycin resistance of Pseudomonas syringae pv. papulans in apple orchards and its association with a conjugative plasmid. Phytopathology 78: 410– 413. Chiou, C. S., and Jones, A. L. 1993. Nucleotide sequence analysis of a transposon (Tn5393) carrying streptomycin resistance genes in Erwinia amylovora and other gram-negative bacteria. J. Bacteriol. 175: 732–740. Coyier, D. L., and Covey, R. P. 1975. Tolerance of Erwinia amylovora to streptomycin sulfate in Oregon and Washington. Plant Disease. 59: 849–852. Jensen, M. A., Webster, J. A., and Straus, N. 1993. Rapid identification of bacteria on the basis of polymerase chain reaction-amplified ribosomal DNA spacer polymorphisms. Applied and Environmental Microbiology. 59: 945–952. Loper, J. E., Henkels, M. D., Roberts, R. G., Grove, G. G., Willet, M. J., and Smith, T. J. 1991. Evaluation of streptomycin, oxytetracycline, and copper resistance of Erwinia amylovora isolated from pear orchards in Washington State. Plant Disease. 75: 287–290. McGhee, G. C., and Sundin, G. W. 2010. Evaluation of kasugamycin for fire blight management, effect on nontarget bacteria, and assessment of kasugamycin resistance potential in Erwinia amylovora. Phytopathology. 101: 192–204. Miller, T. D., and Schroth, M. N. 1972. Monitoring epiphytic population of Erwinia amylovora on pear with a selective medium. Phytopathology 62: 1175–1182.

26 NEW YORK STATE HORTICULTURAL SOCIETY