Effective Management for Fire Blight for Sustainable Apple Production in Illinois

Field survey and sample collection. Incidence and severity of blossom blight and shoot blight caused by E. amylovora were assessed in 24, 35, and 39 apple orchards in Illinois in 2010, 2011, and 2012, respectively. Surveyed apple orchards were located in 24 counties, including Boone, DeKalb, Kane, McHenry, and Winnebago in northern; Calhoun, Champaign, Jersey, Macoupin, Marshall, Peoria, Piatt, Putnam, Sangamon, and Woodford in central; and Clinton, Jackson, Jefferson, Madison, Marion, Pope, Randolph, St. Clair, and Union in southern Illinois.

In each orchard, 50 asymptomatic flowers (two flower per tree) and 20 symptomatic shoots (one shoot per tree) were collected for isolation of the bacterium (E. amylovora). No symptomatic flowers were observed. Each infected shoot was cut 20 cm below the visible necrotic tissue and pruners were dipped in 95% ethanol between cuts. Collected flowers and shoots were placed in plastic bags, kept on ice during transport, and stored at 4ºC in the laboratory within 12 h of collection. In 2010, shoots were collected on 3 June and 22 July. In 2011, flowers were collected on 10 April and 16 May and shoots were collected on 7 June and 22 June. In 2012, flowers were collected on 23 March and 9 April and shoots were collected on 25 May and 8 June.
Isolation, purification, and maintenance of E. amylovora. Luria-Bertani (LB) medium was used for isolation and maintenance of E. amylovora (12). LB medium was amended with cycloheximide (50 mg/L) to make LBch. LBch was amended with 50 mg/L of streptomycin (Agri-Mycin 17WP; Nufarm Americas Inc., Burr Ridge, IL) to make LBcham.

All collected samples were processed within 36 h of collection. Five flowers were placed in a mesh bag (Agdia, Inc., Elkhart, IN) with 5 ml of sterilized 0.5x PBS buffer, on ice. Flower tissue was macerated for approximately 20 s or until tissue was well ground using a tissue homogenizer (Agdia, Inc., Elkhart, IN) attached to a drill press. For isolation of the bacterium from shoots, a 10-15 mm piece of shoot tissue was cut immediately below the visible necrotic lesion and bark was removed using sterile scalpel and tweezers. The shoot tissues were macerated as described for flowers. Then, underneath a sterile hood, extract from ground tissue was diluted (3:1) in sterilized 0.5x PBS buffer. Using a bacterial loop, the suspension was streaked to develop single colonies on LBch and LBcham. The cultures were incubated at 27ºC for 48 h. White colonies characteristic of E. amylovora on LBch and all white colonies growing on LBcham were sub-cultured on LB, and then each colony was transferred to a cryogenic vial containing 15% glycerol. The vials were stored at -20ºC and -80ºC for further studies.

PCR identification of E. amylovora. The identity of each E. amylovora isolate was confirmed by using a polymerase chain reaction (PCR). The PCR was conducted using primers AJ75 (5’CGC ATT CAC GGC TTC GCA GAT 3’) and AJ76 (5’AAC CGC CAG GAT AGT CGC ATA 3’) targeting ubiquitous plasmid pEA29 in E. amylovora (15). Using a sterile pipette tip, a colony was suspended in 100 ?l of sterilized distilled water (SDW). Then, 0.5 ?l of the bacterial suspension was added to 10 ?l DNA-free water, 1 ?l of each primer AJ75 and AJ76 (10 pmol/ 1 ?l) and 12.5 ?l Gotag Green Master Mix 2x (Promega Corporation, Madison, WI) for a final reaction mixture volume of 25 ?l. Cycling conditions for PCR (Model PCT-200, MJ Research Inc., Waltham, MA) included initial denaturation at 94ºC for 5 min; 37 cycles of denaturation for 1 min, annealing at 52ºC for 2 min, and extension at 72ºC for 2 min; followed by 15 min final extension at 72ºC and was kept at 4ºC until electrophoresis. The PCR product was run on 1% agarose gel containing ethidium bromide at 100 volts for 60 min, and photographed under UV light. The bands were compared to a known streptomycin-sensitive E. amylovora isolate, MK1, and a streptomycin-resistant E. amylovora isolate, Ea88, supplied by George Sundin (Michigan State University). The process was repeated at least once for each isolate.

Virulence of E. amylovora isolates. An immature pear fruit assay was used to determine virulence of 48 E. amylovora isolates from Illinois. Immature ‘Seckel’ pear fruit were surface disinfested in 75% ethanol, rinsed in SDW, and dried on a sterile lab bench. Using a sterile pipette, a 15-?l suspension (10,000,000 CFU/ml) of each isolate was inoculated into each pear fruit (approximately 2 mm deep). Erwinia amylovora isolate MK1 and SDW were used as positive and negative controls. Inoculated and control pears were incubated in a moist chamber at 27ºC in darkness. Diameters of necrotic lesions were measured perpendicularly 4 and 7 days post-inoculation. Each virulence test had five replications (fruits) and the experiment was repeated once. The virulence tests were complimented with a tobacco hypersensitivity assay to determine the presence of type III effectors.

Screening E. amylovora isolates for streptomycin sensitivity. Four-hundred and sixteen E. amylovora isolates were cultured on LB medium and LB medium amended with 50 mg/L of streptomycin (Agri-Mycin 17WP). The streptomycin-sensitive isolate MK1 and streptomycin-resistant isolate Ea88 were used as negative and positive controls. Colony development was recorded after 24 and 48 h. Each isolate was evaluated in two experiments with three replications per experiment.
One hundred thirty-three of 416 E. amylovora isolates were further tested for streptomycin sensitivity by zone of inhibition. Each isolate was grown on LB medium for 48 h. Colonies were washed in SDW and the bacterial suspensions were adjusted to 10,000,000 CFU/ml using a spectrophotometer (Smart Spec 3000; Bio-Rad, Philadelphia, PA). Using a sterilized bent glass rod, a 50-?l bacterial suspension was spread onto LB medium. A 12-mm filter disc was soaked in a 100 mg/L solution of streptomycin (Agri-Mycin 17WP), briefly dried, and placed onto the agar surface. The streptomycin-susceptible isolate MK1 and streptomycin-resistant isolate Ea88 were used as positive and negative controls. Clear zones of inhibition were measured at 24 and 48 h. The width of zone (? 1 mm) with no bacterial colony was considered a measure of sensitivity of E. amylovora to streptomycin. Each isolate was tested twice and each test had three replications and each replication had four filter discs.

In addition, 10 of the 416 E. amylovora isolates were compared for bacterial multiplication in LB broth (Miller, Arcos Organics, Geel, Belgium) amended with 0, 0.5, 1, 2, 3, 4 and 5 mg/L streptomycin (Agri-Mycin 17WP) and in nutrient broth (BD, Sparks, MD) amended with 0, 0.25, 0.5, and 1 mg/L streptomycin (Agri-Mycin 17WP). The tests were conducted using sterile 24-well plates (351147; BD Falcon, Franklin Lakes, NJ). Twenty microliters from each bacterial suspension (100,000,000 CFU/ml) was added to 2 ml of broth in each well and incubated on a shaker at 28ºC for 18 h. Bacterial cell density was assessed using a spectrophotometer at OD 600. The bacterial cell densities from the streptomycin amended cultures were compared to unamended cultures and expressed as a percentage. Each isolate was tested twice with four replications.

Non-Erwinia amylovora bacteria in Illinois orchards. While testing blossom and shoot samples for the presence of E. amylovora, 119 non-E. amylovora isolates were also collected that developed colonies at 50 mg/L streptomycin. All 119 non-E. amylovora isolates were screened for the presence of IS1133 region (associated with Tn5393) using primer set IS1133-F (‘GCG TGA TGC AGT TCG CAT AGC’) and IS1133-R (‘CAT ACG CGG CCT ACC ATA GCT’) (12). The IS1133 region of nine isolates was sequenced using primer IS1133-F, except the cycling parameters were modified as described below.

These nine non-E. amylovora isolates were also screened with primers strab01-F (‘TGG TGT CCC GCA ATG CCG TC’) and strab01-R (‘CCC GGA TCG GGA GAA GGG CA’) to amplify a portion of strB region on Tn5393. PCR cycling condition included initial denaturing for 4 min at 94°C; 40 cycles of denaturation at 94°C for 30 s, annealing at 55°C (IS1133 primer set) or 60°C (strab01 primer set) for 30 s, and extension at 72°C for 2.5 min; followed by a final extension at 72°C for 7 min. The PCR product was run in 1% agarose gel containing ethidium bromide at 100 volts for 60 min, and photographed under UV light. The bands and sequences were compared to results produced by E. amylovora isolates MI 5-1 and W4 that are known to contain Tn5393. The process was repeated twice for each isolate.

The PCR test was performed to amplify the 16S rRNA genes from the extracted DNA using the primers 27F (5’AGA GTT TGA TCM GGC TCA G 3’) and 1492R (5’ GGT TAC CTT GTT ACG ACT T 3’). PCR products were purified with the Wizard SV gel and PCR Clean-Up system (Promega, Madison, WI). Purified 16S rRNAs were single-end sequenced from 27F at the University of Illinois Core DNA Sequencing Facility. Edited sequences were compared using the BLASTn database (http://www.ncbi.nlm.nih.gov).

Oxytetracycline and kasugamycin sensitivity. Eighty-four of the 416 E. amylovora isolates were selected for in-vitro evaluation of antibiotics. LB medium (pH 7) was amended with oxytetracycline (Mycoshield 17WP; NuFarm Americas Inc., Burr Ridge, IL) at 0, 50, 100, and 200 mg/L. Also, two formulations of kasugamycin (Kasumin 2 L and ARY-4016-06; Arysta Life Science, Cary, NC) were evaluated at 0, 25, 50, and 100 mg/L in LB medium with bacterial suspensions streaked onto the agar surface. All experiments included unamended LB medium for controls. Colony development of E. amylovora was assessed after 24 and 48 h.

Ten E. amylovora isolates were further compared in nutrient broth (pH 7) amended with Kasumin 2L and ARY-4016-06 at 0, 3, 5, 10 mg/L of kasugamycin for bacterial multiplication. This test was conducted as described previously for the streptomycin test. The test was conducted using sterile 24-well plates, with 20 ?l from each bacterial suspension (100,000,000 CFU/ml) added to 2 ml of amended nutrient broth in each well and incubated on a shaker at 28ºC for 18 h. Bacterial cell density was assessed using a spectrophotometer at OD 600.

Copper sensitivity. The same 84 E. amylovora isolates were also evaluated for their sensitivity to copper sulfate using casitone-yeast extract (CYE) medium (1.7 g of casitone, 0.35g yeast extract, 2 g glucose, and 15 g agar in 1,000 ml deionized water), as reported by Loper et al. (11). CYE medium was amended with 0.08, 0.16, and 1.1 mM copper sulfate (Cuprofix Ultra 40DF; United Phosphorus, King of Prussia, PA). A 10-?l aliquot of bacterial suspension from each isolate was spotted onto the agar surface in each 110-mm dia Petri plate. All experiments included unamended agar plates for controls. Each isolate was tested twice with three replications (plates) per test. Colony development was assessed after 48 h and 72 h.

Eleven of the 84 E. amylovora isolates were further evaluated for their sensitivity to various copper formulations at 0 and 14 mg/L of metallic copper. CYE medium was amended with copper sulfate (Cuprofix Ultra 40DF); copper hydroxide (Kocide 3000 40.1DF; Dupont, Wilmington, DE); cuprous oxide (Nordox 75 WG; Nordox Industries AS, Oslo Norway); and a mixture of copper hydroxide and copper oxychloride (Badge X2 45DF; Isagro USA, Morrisville, NC). Colony development was assessed after 48 h and 72 h.

Field experiments. Field experiments were conducted in 2011 and 2012 to evaluate potential streptomycin-alternative compounds for management of fire blight. The experiments were conducted on a 16-year-old block of ‘Jonathan’ apple trees at the University of Illinois Fruit Research Farm, Urbana, IL. Inoculum of E. amylovora was prepared from 48-h-old colonies of the isolate (2A) that had been collected from the same orchard. Bacterial suspensions in 0.5x PBS buffer at 100,000,000 CFU/ml (in 2011) and 5,000,000 CFU/ml (in 2012) were used. The bacterial suspension was transported on ice and applied when MARYBLYT disease forecasting logger (Watchdog 400; Spectrum Technologies, Plainfield, IL) indicated high risk of fire blight infection. All treatments and inoculations were applied using a gas-powered backpack mist sprayer (SR400; Stihl, Virginia Beach, VA) at 1,400 L/ha. Data were collected on four sides of each tree from the upper, middle, and lower canopy. A total of 120 flower clusters and 120 shoots were examined on each tree when it was possible. In 2012, number of flower clusters per tree ranged from 6 to 120 (mean 70 clusters) per tree.

The 2011 experiment was performed in a randomized complete block design with eight treatments, and each treatment had two application times of 24 h pre- and 24 h post-inoculation. Each treatment had three replications for each application time, with one tree for each replication (48 trees in the experiment). The treatments included: (i) an uninoculated control; (ii) an inoculated control; (iii) streptomycin (Agri-Mycin 17WP) at 100 mg/L; (iv) oxytetracycline (Mycoshield) at 200 mg/L; (v) kasugamycin (Kasumin 2L) at 100 mg/L; (vi) kasugamycin (ARY-4016-06) at 100 mg/L; (vii) Pseudomonas florescences A506 (BlightBan A506) at 369 mg/L; and (viii) Bacillus subtillus QST713 (Serenade Max) at 525 mg/L.

Cuprofix Ultra 40D (7.8 kg/ha) and dormant oil were applied to all trees on 31 Mar 2011. Growth development stages of apple trees were full pink (18 April), 1% bloom (20 April), 40-60% bloom (25 April), and petal fall (11 May). The trees were inoculated with the pathogen twice on 26 April and 8 May, to ensure adequate inoculum was present. Pre-inoculation treatments (treatments iii-viii) were applied on 25 April and 7 May and post-inoculation treatments (treatments iii-viii) were applied on 27 April and 9 May. All plots received an additional spray of the respective test product on 29 April, a timing that was independent of the pre- and post-inoculation timings sprays. As a result, all test trees received a total of three fire blight control sprays during bloom. Blossom infection was evaluated on 12 May and 23 May and shoot infection was evaluated on 10 June, 30 June, and 27 July.

The 2012 experiment was a split-plot in a randomized complete block design with 11 main-plots (antibiotics and biocontrol agents), each with two sub-plots of with or without growth regulator prohexadione calcium (Apogee 27.5DF; BASF Corporation, Research Triangle Park, NC) at 1,200 mg/L. Each treatment had four replications (trees). The main plot treatments included: (i) inoculated control; (ii) copper hydroxide (Kocide 3000) at 240 mg/L + mancozeb (Dithane 75DF; Dow AgroSciences LLC, Indianapolis, IN) at 2,700 mg/L; (iii) streptomycin (Agri-Mycin 17WP) at 100 mg/L; (iv) oxytetracycline (Mycoshield) at 200 mg/L; (v) oxytetracycline (Mycoshield) at 200 mg/L+ copper hydroxide (Kocide 3000) at 240 mg/L; (vi) kasugamycin (Kasumin 2L) at 100 mg/L; (vii) kasugamycin (ARY-4016-06) at 100 mg/L; (viii) Pseudomonas florescences A506 (BlightBan A506 71) at 369 mg/L; (ix) Bacillus subtillus (Serenade Max QST713) at 525 mg/L; (x) kasugamycin (Kasumin 2L) at 100 mg/L and (xi) streptomycin (Agrimycin 17WP) at 100 mg/L. Treatments i-vii, were applied 24 h pre- and 24 h post-inoculation; treatments viii and ix were applied 48 h pre- and 24 h pre-inoculation; and treatments x and xi were applied 24 h post-inoculation only. All 11 treatment applications included a nonionic surfactant, Regulaid (Kalo, Inc., Overland Park, KS), at 1.25 ml/L.

Cuprofix Ultra 40DF (7.8 kg/ha) and dormant oil were applied to all trees on 13 March 2012. Growth development stages of apple trees in 2012 were full pink (23 March), 1% bloom (26 March), 40-60% bloom (28 March), and petal fall (12 April). A total of 3.363 kg/ha of Apogee (0.925 kg/ha of prohexadione calcium) in two sprays (1.68 kg/ha in each spray) plus Regulaid (1.25 ml/L) were applied on 29 March (60-80% bloom) and 12 April (petal fall), using 1,400 L of water per ha for each application. The trees were inoculated on 30 March and 3 April. Treatments were applied on 29 March (24 h pre-inoculation), 31 March (24 h post-inoculation), 2 April (24 h pre-inoculation), and 4 April (24 h post-inoculation). Additional applications of BlightBan A506 (treatment viii) and Serenade Max (treatment ix) were made on 28 March. Blossom infection was evaluated on 17 April, 24 April, and 1 May; and shoot infection was evaluated on 8 May, 22 May, 5 June, and 3 July.

Data analysis. All statistical analyses were performed using SAS 9.3 (SAS Institute Inc. Cary, NC). Homogeneity of variances was tested using the Brown-Forsyth test and normality was observed. Percent cell multiplication data from liquid amended media (streptomycin) were analyzed using analysis of variance (ANOVA) and means were separated by least significant difference (LSD) in PROC GLM at alpha = 0.05. All other data were analyzed using ANOVA in PROC MIXED. Percent cell multiplication data from liquid amended media (kasugamycin) were separated at alpha = 0.05 using macro pdmix800 (19) to indicate mean separation. Pear pathogenicity and field data were square root transformed [? (x + 1/6)] before analysis. In 2012, blossom data was analyzed as a covariate to account for uneven blossom development in the orchard. Dunnett’s test was used to compare blossom infection of field treatments to the inoculated control.