Abstract
Bacterial leaf blight incited by Pseudomonas cannabina pv. alisalensis (Pca) is a devastating disease with incidence reports worldwide and a wide host range capable of infecting all commercially valuable Brassica crops. With no chemical control options available, the most effective form of disease control is host plant resistance, but thus far resistant germplasm has only been identified in Brassica juncea L. (mustard greens). We report the first screening of Brassica oleracea L. var. viridis germplasm, including leafy green collard and collard-like accessions, for resistance to bacterial leaf blight by artificial inoculation of Pca in greenhouse trials. All commercial cultivars tested displayed an intermediate disease response resulting in leaf lesion development that renders the product unmarketable. Two sources of significant resistance were identified in the U.S. Department of Agriculture (USDA) viridis collection, which provides a valuable source of resistance alleles for collard cultivar development and introgression into other B. oleracea crops.
Bacterial blight emerged in the 1990s as a major disease of Brassicaceae crops and has since spread worldwide with disease incidence reports from Europe (Rubio et al., 2012; Sarris et al., 2010), Asia (Takahashi et al., 2013), Australia (Bull and Rubio, 2011), and both the East and West coasts of the United States (Cintas et al., 2002; Keinath et al., 2006; Wechter et al., 2010). Initial reports attributed the disease to Pseudomonas syringae pv. maculicola but were later determined to be incited by a new pathovar (pv. alisalensis; Pseudomonas syringae pv. alisalensis) responsible for bacterial blight of Brassicas (Bull et al., 2004; Cintas et al., 2002; Takahashi et al., 2013; Wechter et al., 2013). Pseudomonas syringae pv. alisalensis has since been reclassified as Pca (Bull et al., 2010a).
Pseudomonas cannabina pv. alisalensis is capable of infecting a wide range of Brassicas, including crop varieties of multiple species: Brassica rapa L. (Cintas et al., 2002; Keinath et al., 2006; Wechter et al., 2007), B. juncea (Keinath et al., 2006; Wechter et al., 2007, 2014), Brassica napus L. (Bull and Rubio, 2011; Koike et al., 2007), B. oleracea (Bull et al., 2010b; Cintas et al., 2002; Koike et al., 2000, 2006; Mauzey et al., 2011; Wechter et al., 2010), Raphanus sativus L. (Rubio et al., 2012), and Eruca sativa L. (Bull and DuToit, 2009; Bull et al., 2004; Sarris et al., 2010). A study comparing host plant resistance and the commonly used chemical bactericide acibenzolar-S-methyl found that host plant resistance was more effective at controlling bacterial blight in B. juncea (Keinath et al., 2016). However, levels of resistance to Pca, suitable for market production, have only been identified in B. juncea (Wechter et al., 2007, 2013), which led to the successful development of a resistant mustard green cultivar Carolina Broadleaf (Wechter et al., 2016). Herein, we report the first screening of B. oleracea germplasm for resistance to bacterial leaf blight incited by Pca.
Brassica oleracea consists of several economically important crop varieties, with a cash value of more than $2 billion from broccoli (var. italica), cabbage (var. capitata), cauliflower (var. botrytis), kale (var. acephala), Brussels sprout (var. gemmifera), and collard in the United States alone (U.S. Department of Agriculture, National Agricultural Statistics Service, 2014). Collard is a leafy green vegetable crop primarily produced in the southeastern United States with a yearly fresh market value of more than $125 million (U.S. Department of Agriculture, National Agricultural Statistics Service, 2014). Bacterial leaf blight was first found in collard in 2002 in commercial fields in South Carolina and has been detected every year since the initial report (Keinath et al., 2006). The disease is currently causing the highest yield losses of leafy green Brassicas in South Carolina (Keinath et al., 2006), and all tested commercial cultivars have been found to be susceptible to the pathogen (Keinath et al., 2011; Wechter et al., 2013). Thus, the objective of this study was to screen the USDA viridis PI collection for resistance to Pca, to identify resistant collard germplasm that can be used for collard cultivar development, and to provide resistance alleles for introgression into other economically important B. oleracea crop varieties.
Materials and Methods
Plant materials.
Germplasm for the disease screen included the USDA collection of viridis group PIs that are currently available from the National Plant Germplasm System in Geneva, NY, and 24 additional B. oleracea PIs of various varieties that produce collard-like leafy greens (germplasm information available from Dryad data repository, doi:10.5061/dryad.4851qq0). The Brassica leafy green commercial cultivars (obtained from Abbott and Cobb Seed Co., Feasterville, PA), ‘Topper’ (B. rapa turnip green) and ‘Top Bunch’ (B. oleracea var. viridis collard green), were used as susceptible controls (Wechter et al., 2013). ‘Carolina Broadleaf’ (B. juncea mustard green cultivar) and PI 195553 (Brassica spp. PI) were used as resistant controls based on previous reports of low disease occurrence in response to inoculation with Pca in both greenhouse and field studies (Wechter et al., 2007, 2013, 2016).
Initial disease screen and retests of selections.
The initial disease screen of the collard PI collection was completed as a series of five greenhouse tests of 19–24 PIs each between Oct. 2015 and Feb. 2016 in Charleston, SC. Each test included the same four controls (‘Carolina Broadleaf’, PI 195553, ‘Topper’, and ‘Top Bunch’) and followed a randomized complete block design with two replications of four plants per replication. Seeds were planted into Metro-Mix 200 soil mix (The Scott’s Co., Maryville, OH) in 5-cm square pots and thinned to one plant per pot after 10 d. At 7 d postinoculation, the two most affected leaves per plant were visually evaluated for percentage of leaf area with blight symptoms (necrosis, chlorosis, or both; Fig. 1). Each plant was given a score of 0% to 100% disease severity (DS), in 5% increments, based on the two-leaf average (Wechter et al., 2013).
Collard leaves with bacterial leaf blight symptoms depicting disease severity scores of 85% (A) and 5% (B). Images were taken 7 d postinoculation with Pseudomonas cannabina pv. alisalensis from PIs screened in greenhouse trials.
Citation: HortScience horts 53, 6; 10.21273/HORTSCI12347-17
Collard leaves with bacterial leaf blight symptoms depicting disease severity scores of 85% (A) and 5% (B). Images were taken 7 d postinoculation with Pseudomonas cannabina pv. alisalensis from PIs screened in greenhouse trials.
Citation: HortScience horts 53, 6; 10.21273/HORTSCI12347-17
Collard leaves with bacterial leaf blight symptoms depicting disease severity scores of 85% (A) and 5% (B). Images were taken 7 d postinoculation with Pseudomonas cannabina pv. alisalensis from PIs screened in greenhouse trials.
Citation: HortScience horts 53, 6; 10.21273/HORTSCI12347-17
Collard PIs with an average DS of less than 25% were chosen for retesting to confirm resistance to Pca. In addition, PIs with at least four individual plants displaying less than 25% DS were moved outside for vernalization. Vernalized selections were selfed through hand pollination, given a new ID (PI#-plant#), and the resulting progeny included in retesting along with the original PIs. Five additional lines were created by selfing and collecting seeds from individual selections from two PIs [PI 662820 (2) and PI 662833 (3)]. The most highly susceptible PI from the initial screening (PI 662819) was used as the susceptible control for the retests and ‘Carolina Broadleaf’ as the resistant control. Two commonly grown commercial collard cultivars, Top Bunch and Blue Max, were included in the retests to determine whether any of the PIs exhibited enhanced resistance as compared with commonly grown cultivars. After the initial screening of the collard PI collection, two retests of putatively resistant accessions were completed in Nov. 2016 and Apr. 2017 following the same protocols as described for the preliminary screening.
Bacterial isolate and inoculations.
Isolate T3C of Pca (Wechter et al., 2007, 2013) was used for all assays. Isolate T3C was streaked from a −80 °C glycerol stock onto Pseudomonas agar F (PAF) medium (Difco; Becton Dickenson and Co., Franklin Lakes, NJ) and grown for 24 h at 25 °C. A single colony was restreaked on a PAF plate and grown for 48 h at 25 °C. A 10-μL loop of cells was taken from the PAF plate and inoculated into 200 mL of King’s B broth (Difco; Becton Dickenson and Co.) and grown on a rotary shaker at 200 rpm at 25 °C for 24 h. Cell culture turbidity was determined using a biophotometer (Eppendorf, Hauppauge, NY) at a wavelength of OD600. Cell suspension was diluted to a final concentration of ≈2 × 106 colony-forming units (cfu) using sterile distilled water; based on previous studies, an OD600 of 0.1 is ≈1 × 106 cfu (data not shown). Cell suspension was applied to the adaxial and abaxial surfaces of the leaves using a Paasche airbrush (Paasche Airbrush Co., Chicago, IL) at a pressure of ≈160 kPa. Inoculum was applied from a distance of 15–20 cm from the leaf surface until incipient runoff. The plants were then placed in a dew chamber at 100% humidity and 25 to 27 °C for 24 h and then transferred to the greenhouse. The plants were left in the greenhouse for an additional 6 d and rated 7 d postinoculation.
Statistical analysis.
Each test of the initial screen was analyzed separately because of heterogeneity of variances (P value <2.2 × 10−16) as determined by Levene’s test (Levene, 1960) in the car package (Fox and Weisberg, 2011) of R (R Core Team, 2016). The two retests were analyzed both separately and in combination because they had homogeneity of variances (P value = 0.30). Analyses of variance (ANOVAs) of DS were conducted for all tests separately (N = 7) and the retests combined using a general linear model in the aov package (Chambers et al., 1992) in R. Models of the individual tests included genotype, replication, and plants nested within replication as fixed effects. The same model was used for analysis of the combined retests with the addition of test and the interaction of genotype and test as fixed effects. Fisher’s protected least significant differences (lsds) between genotypes were determined with the agricolae package (Mendiburu, 2016) in R with an alpha of 0.05. Accession means for each test (five total) of the initial screen and the two retests were calculated in R. Correlation between accession means of the retests was assessed by calculating Pearson’s correlation coefficient (r) using the stats package of R.
Results
Disease screen.
The USDA viridis PI collection (N = 110) was screened for resistance to bacterial leaf blight following inoculation with Pca. The results associated with the screening of the PI collection, including ANOVA tables, lsd, and DS of each accession tested, have been deposited in the Dryad data repository (doi:10.5061/dryad.4851qq0). Accession was the most significant factor (P value from 9.3 × 10−13 to 2 × 10−16) in all five tests of the screening. Significant differences were found among accessions evaluated in each test, with lsd values ranging from 16.8% to 24.4% DS. Disease severity of the accession means within each test varied widely, with a minimum difference between the most and the least resistant of 76%. Both cultivars had an intermediate disease response, with the mean DS of ‘Top Bunch’ ranging from 18.1% to 55.6% and ‘Topper’ from 11.3% to 35.6%. The resistant controls, ‘Carolina Broadleaf’ and PI 195553, had the lowest mean DS in every test, and eight accessions (PI 662586, PI 662668, G 30859, PI 662796, G 32556, PI 662805, G 32768, and G 33036) displayed similar results with no significant differences from these two resistant controls.
Retests of the most resistant PIs.
Fifteen of the most resistant PI accessions (mean DS <25%) and five selfed progeny were included in two additional tests (Table 1). Accession means across the two additional tests ranged from 2.6% to 39.7%. Accession means were significantly correlated between the tests with a P value of 0.02 and an r of 0.47. The commercial collard cultivars (Blue Max and Top Bunch) displayed an intermediate response to Pca as their DS means (17.5% and 16.9%, respectively) were significantly different from both the resistant and susceptible controls (Table 1). Three accessions (PI 662820-1, PI 662820-2, and G 33036) were identified as not significantly different from the resistant control with a mean DS of 9% to 11%. Both G 33036 and PI 662820 (the PI from which two selections were made) are classified as collards (var. viridis), but they have very different horticultural characteristics (Fig. 2). G 33036, named Purple Curly Collard, had a lower mean DS but produces a kale-like plant, undesirable for collard production. This accession also exhibited wide segregation for leaf size, shape, curliness, and color. Two resistant selections were derived from PI 662820, named Lyda Gibbs after the originating seed saver, which produces uniform, high-yielding horticulturally desirable collard plants with large, bright-green leaves.
Mean disease severity (DS) of accessions across two retests after inoculation with Pseudomonas cannabina pv. alisalensis.
Photographs of identified sources of Pseudomonas cannabina pv. alisalensis resistance, PI 662820 (A) and G 33036 (B), taken from a noninoculated field, lacking bacterial blight symptoms. Shown only to demonstrate the horticultural differences between the PIs. (Photo credit: Zachary Stansell.)
Citation: HortScience horts 53, 6; 10.21273/HORTSCI12347-17
Photographs of identified sources of Pseudomonas cannabina pv. alisalensis resistance, PI 662820 (A) and G 33036 (B), taken from a noninoculated field, lacking bacterial blight symptoms. Shown only to demonstrate the horticultural differences between the PIs. (Photo credit: Zachary Stansell.)
Citation: HortScience horts 53, 6; 10.21273/HORTSCI12347-17
Photographs of identified sources of Pseudomonas cannabina pv. alisalensis resistance, PI 662820 (A) and G 33036 (B), taken from a noninoculated field, lacking bacterial blight symptoms. Shown only to demonstrate the horticultural differences between the PIs. (Photo credit: Zachary Stansell.)
Citation: HortScience horts 53, 6; 10.21273/HORTSCI12347-17
Discussion
The rapid worldwide spread of Pca, combined with the devastating potential losses caused by bacterial leaf blight on Brassicas, highlights the importance of identifying sources of resistance in Brassica crops. In the studies described herein, resistance to Pca was found in two collard PI accessions, both collected in North Carolina, that vary drastically from one another in terms of horticultural characteristics. Selections from PI 662820 have resistance to Pca and produce horticulturally desirable leaves.
Even minor blemishes on collard leaves caused by bacterial leaf blight can result in an unmarketable product. Although the mean DS of the PI 662820 selections (≈10%) may be too high for fresh market production, previous experience with other Brassica species indicates that additional rounds of selection should provide gains in bacterial blight resistance. Indeed, a single round of selection has already decreased DS, with both PI 662820 selfed selections (-1 and -2) testing more resistant than the original PI. This significant gain from selection suggests polygenic inheritance for resistance in this B. oleracea germplasm that may be similar to what was found in B. juncea (Wechter et al., 2013). Additional generations of selection and selfing could be expected to increase the level of resistance in the PI 662820 lines. This strategy has already proven effective in the development of the cultivar Carolina Broadleaf (Wechter et al., 2016), which has marketable levels of resistance to Pca even though it originated from a PI that originally exhibited DS of more than 10% (Wechter et al., 2007). Inheritance of resistance in B. juncea is likely controlled by multiple recessive genes or by a single recessive gene with interaction of modifier genes (Wechter et al., 2013).
Genetic and cultural evidence suggest that collard may be most closely related to cabbage (Farnham et al., 2005, 2008; Pelc et al., 2015), which has a yearly cash value of almost half a billion dollars in the United States (U.S. Department of Agriculture, National Agricultural Statistics Service, 2014). In 2008, bacterial blight from Pca infection was found for the first time in commercial cabbage fields in Monterey County, CA, which caused reduced quality and marketability of the cabbage heads (Mauzey et al., 2011). The resistant germplasm identified here may provide alleles for the ready introgression of disease resistance into cabbage cultivars because of the close relationship between these crops (Farnham et al., 2005).
Here, we report the first screening of B. oleracea germplasm for resistance to bacterial leaf blight incited by Pca. We identified two significant sources of resistance within the USDA viridis PI collection, which is particularly important in light of the intermediate, but unmarketable, disease response of the commercial collard cultivars tested. Future breeding efforts will focus on continued rounds of selection from the initial selfed plants of PI 662820 to both attempt additional genetic gains and fix resistance to Pca for germplasm or cultivar release.
Literature Cited
Bull, C.T. & Du Toit, L.J. 2009 First report of bacterial blight on conventionally and organically grown arugula in Nevada caused by Pseudomonas syringae pv. alisalensis Plant Dis. 93 109
Bull, C.T., Goldman, P. & Koike, S.T. 2004 Bacterial blight on arugula, a new disease caused by Pseudomonas syringae pv. alisalensis in California Plant Dis. 88 1384
Bull, C.T., Manceau, C., Lydon, J., Kong, H., Vinatzer, B.A. & Fischer-Le Saux, M. 2010a Pseudomonas cannabina pv. cannabina pv. nov., and Pseudomonas cannabina pv. alisalensis (Cintas Koike and Bull, 2000) comb. nov., are members of the emended species Pseudomonas cannabina (ex Šutič & Dowson 1959) Gardan, Shafik, Belouin, Brosch, Grimont & Grimont 1999 Syst. Appl. Microbiol. 33 105 115
Bull, C.T., Mauzey, S.J. & Koike, S.T. 2010b First report of bacterial blight of brussels sprouts (Brassica oleracea var. gemmifera) caused by Pseudomonas cannabina pv. alisalensis in California Plant Dis. 94 1375
Bull, C.T. & Rubio, I. 2011 First report of bacterial blight of crucifers caused by Pseudomonas cannabina pv. alisalensis in Australia Plant Dis. 95 1027
Chambers, J.M., Freeny, A. & Heiberger, R.M. 1992 Analysis of variance; designed experiments. In: J.M. Chambers and T.J. Hastie (eds.). Chapter 5 of statistical models in S. Wadsworth & Brooks/Cole, New York, NY
Cintas, N.A., Koike, S.T. & Bull, C.T. 2002 A new pathovar, Pseudomonas syringae pv. alisalensis pv. nov., proposed for the causal agent of bacterial blight of broccoli and broccoli raab Plant Dis. 86 992 998
Farnham, M., Davis, E.H., Morgan, J.T. & Smith, J.P. 2008 Neglected landraces of collard (Brassica oleracea L. var. viridis) from the Carolinas (USA) Genet. Resources Crop Evol. 55 797 801
Farnham, M.W., Ruttencutter, G., Smith, J.P. & Keinath, A.P. 2005 Hybridizing collard and cabbage may provide a means to develop collard cultivars HortScience 40 1686 1689
Fox, J. & Weisberg, S. 2011 An {R} companion to applied regression. 2nd ed. Sage, Thousand Oaks, CA
Keinath, A.P., Farnham, M.W. & Wechter, W.P. 2011 Field evaluation of brassica lines for resistance to bacterial blight in Charleston, South Carolina, 2010 Plant Dis. Mgt. Rpt. 5 V082
Keinath, A.P., Wechter, W.P. & Smith, J.P. 2006 First report of bacterial leaf spot on leafy Brassica greens caused by Pseudomonas syringae pv. maculicola in South Carolina Plant Dis. 90 683
Keinath, A.P., Wechter, W.P. & Farnham, M.W. 2016 Suppression of bacterial blight on mustard greens with host plant resistance and acibenzolar-S-methyl Plant Dis. 100 1921 1926
Koike, S.T., Cintas, N.A. & Bull, C.T. 2000 Bacterial blight, a new disease of broccoli caused by Pseudomonas syringae in California Plant Dis. 84 370
Koike, S.T., Kammeijer, K., Bull, C.T. & O’Brien, D. 2006 First report of bacterial blight of romanesco cauliflower (Brassica oleracea var. botrytis) caused by Pseudomonas syringae pv. alisalensis in California Plant Dis. 90 1551
Koike, S.T., Kammeijer, K., Bull, C.T. & O’Brien, D. 2007 First report of bacterial blight of rutabaga (Brassica napus var. napobrassica) caused by Pseudomonas syringae pv. alisalensis in California Plant Dis. 91 112
Levene, H. 1960 Robust tests for equality of variances, p. 278–292. In: I. Olkin (ed.). Contributions to probability and statistics: Essays in honor of Harold Hotelling. Stanford Univ. Press, Stanford, CA
Mauzey, S.J., Koike, S.T. & Bull, C.T. 2011 First report of bacterial blight of cabbage (Brassica oleracea var. capitata) caused by Pseudomonas cannabina pv. alisalensis in California Plant Dis. 95 71
Mendiburu, F. 2016 Agricolae: Statistical procedures for agricultural research. R package version 1.2-4
Pelc, S.E., Couillard, D.M., Stansell, Z.J. & Farnham, M.W. 2015 Genetic diversity and population structure of collard landraces and their relationship to other crops. The Plant Genome 8:0
R Core Team 2016 R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria
Rubio, I., Hiddink, G., Asma, M. & Bull, C.T. 2012 First report of the crucifer pathogen Pseudomonas cannabina pv. alisalensis causing bacterial blight on Radish (Raphanus sativus) in Germany Plant Dis. 96 904
Sarris, P.F., Karri, I.V. & Goumas, D.E. 2010 First report of Pseudomonas syringae pv. alisalensis causing bacterial blight of arugula (Eruca vesicaria subsp. sativa) in Greece New Dis. Rpt. 22 22
Takahashi, F., Ogiso, H., Fujinaga, M., Ishiyama, Y., Inoue, Y., Shirakawa, T. & Takikawa, Y. 2013 First report of bacterial blight of crucifers caused by Pseudomonas cannabina pv. alisalensis in Japan J. Gen. Plant Pathol. 79 260 269
U.S. Department of Agriculture, National Agricultural Statistics Service 2014 Vegetable and pulses yearbook. USDA/NAAS, Washington, D.C
Wechter, W.P., Farnham, M.W., Smith, J.P. & Keinath, A.P. 2007 Identification of resistance to peppery leaf spot among Brassica juncea and Brassica rapa plant introductions HortScience 42 1140 1143
Wechter, W.P., Keinath, A.P., Farnham, M.W. & Smith, J.P. 2010 First report of bacterial leaf blight on broccoli and cabbage caused by Pseudomonas syringae pv. alisalensis in South Carolina Plant Dis. 94 132
Wechter, W.P., Keinath, A.P., McMillan, M., Smith, J.P. & Farnham, M.W. 2013 Expression of bacterial blight resistance in brassica leafy greens under field conditions and inheritance of resistance in a Brassica juncea source Plant Dis. 97 131 137
Wechter, W.P., Keinath, A.P., Smith, J.P., Farnham, M.W., Bull, C.T. & Schofield, D.A. 2014 First report of bacterial leaf blight on mustard greens (Brassica juncea) caused by Pseudomonas cannabina pv. alisalensis in Mississippi Plant Dis. 98 1151
Wechter, W.P., McMillan, M.M., Keinath, A.P., Smith, J.P. & Farnham, M.W. 2016 ‘Carolina Broadleaf’ mustard green (Brassica juncea) resistant to the bacterial leaf blight pathogen Pseudomonas cannabina pv. alisalensis HortScience 51 1062 1064
