Abstract
Drippy blight is an emergent disease of red oaks, caused by the interaction between a kermes scale insect (Allokermes galliformis) and a bacterium (Lonsdalea quercina subsp. quercina). Multi-locus sequence analysis was used to confirm the bacterial pathogen’s identity and its relationship to other phylogenetically related Enterobacteriaceae species. Further, Koch’s postulates were performed on sapling red oaks. Prior to the discovery of drippy blight disease in Colorado, in the United States, the bacterium was reported on oak trees in California but was limited to acorn infections. The scale insect, A. galliformis, was previously known to occur on pin oak in the eastern United States but was not previously associated with either this bacterium or the production of significant branch dieback associated with drippy blight. In addition to a description of this new disease, this research documents a host range expansion of L. quercina subsp. quercina to northern red oak (Quercus rubra), Shumard oak (Q. shumardii), and pin oak (Q. palustris) and extends the reported host range of A. galliformis to include northern red and Shumard oaks.
Northern red oak (Quercus rubra), pin oak (Q. palustris), and Shumard oak (Q. shumardii) comprise a small but important component of the urban tree landscape along the eastern urban corridor adjacent to the Rocky Mountains in Colorado, U.S. Since the early 2000s, some municipalities within this region have noted sites where all ages of these oak species and their hybrids have experienced significant dieback of unknown origin. The disease incidence has accelerated in some locations. For example, in Boulder, which has the largest concentration of northern red oaks on public property in the state of Colorado, trees showing extensive twig dieback throughout the crown more than doubled over three years, from 11% (50/450) in 2012 to 24% (110/450) in 2015. As the trees have shown progressive decline, there has been a resultant increase in tree removals (K. Alexander, personal communication, 2016).
Affected trees initially exhibit leaf scorching and leaf drop, followed by dieback of small-diameter twigs throughout the canopy (Figure 1a; Figure 1b). As branch cankers form or as twigs die, tree parts become brittle and snap off the tree, especially during windstorms. At the point of breakage, new shoot growth often results in small witch’s brooms, or twig dieback from successive years may result in major limb dieback. Another feature associated with the disease, particularly in northern red oak, is copious gummosis that drips from cankered, damaged twigs, and onto sidewalks and parked cars, creating a nuisance (Figure 1c; Figure 1d). This combination of symptoms has led to this condition being described as “drippy blight” of red oak.
Signs and symptoms of drippy blight disease: a) a symptomatic pin oak showing witch’s brooms and dieback as a result of drippy blight disease, b) northern red oak branches exhibiting flagging and dieback from drippy blight disease, c) an adult kermes scale insect (Allokermes galliformis, arrow) next to bacterial exudates (Lonsdalea quercina subsp. quercina, star), and d) dead (shriveled) and live (round) kermes scales surrounded by bacterial cankers.
Initially, it was considered that these symptoms resulted from a species of kermes scale insect, Allokermes galliformis, consistently found on damaged branches (Figure 1c). Kermes scales are insect gall mimics that develop almost exclusively on oak (Gill 1993). Allokermes galliformis has been reported from several Quercus species but is most often found in association with pin oak, leading to it sometimes being called “pin oak kermes” (Bullington and Kosztarab 1985).
Kermes scales are generally considered minor pests of oak, occasionally causing some twig dieback, and can be associated with excretion of waxy substances (Bullington and Kosztarab 1985). However, the extensive production of gummosis and the presence of small cankers noted regularly at scale feeding sites (Figure 1c; Figure 1d) led to suspicions that there may be pathogen involvement. The objectives of this report were to identify the microbe(s) associated with the twig cankers, establish their potential pathogenicity, and to better understand interactions of pathogens and A. galliformis in producing the symptoms of drippy blight disease on red oaks.
MATERIALS AND METHODS
Pathogen Identification
Cream-colored bacterial colonies were isolated from canker margins in northern red and pin oaks at several locations in Boulder and Denver, Colorado. For molecular identification of the bacteria, single-spore cultures were grown in nutrient broth for 24 hours on a rotary shaker, and genomic DNA was extracted using an Invitrogen™ Easy-DNA™ kit (Invitrogen, Carlsbad, California, U.S.). The 16S ribosomal DNA was amplified using the methods of Jiang et al. (2006). Amplified products were Sanger-sequenced at Colorado State University’s Proteomics and Metabolomics Facility (Fort Collins, Colorado, U.S.), and compared to the NCBI Nucleotide Database using BlastN search (Altschul et al. 1990) in order to identify the pathogen.
To ensure each sequence belonged to Lonsdalea quercina subsp. quercina, multiple genomic regions of other closely related species were also used for comparison (Table 1). Sequences were retrieved from the NCBI Nucleotide Database, either from individual gene submission (Genbank accession numbers in Table 1) or from complete genome sequences (Caballero et al. 2014; Zhao et al. 2014). These were trimmed and aligned using ClustalW (Larkin et al. 2007) and Mega5 (Tamura et al. 2011), respectively. Bayesian phylogenetic analyses were performed in MrBayes using the general time-reversible model, with inverse-gamma rates of evolution for 500,000 generations, and a burn-in equal to 0.25 (Huelsenbeck and Ronquist 2001). Erwinia piriflorinigrans was used as the outgroup.
GenBank Accession Numbers. The species, isolate number, and GenBank accession numbers are provided for the species used to compare the pathogen Lonsdalea quercina subsp. quercina to closely related Enterobacteriaceae species. The species comparison was based off of several loci, including 16S ribosomal RNA, DNA gyrase subunit B genes (gyrB), beta subunit of ATP synthase (atpD), and translational initiation factor genes (infB) sequences.
Pathogenicity Studies
One-year-old potted northern red, pin, and Shumard oaks were grown in a shade house at Colorado State University, and used to confirm microbe pathogenicity. Trees were inoculated with a northern red oak isolate of L. quercina subsp. quercina (NCCB100490) collected in Boulder, Colorado. Two experiments were conducted with different inoculation sites: emerging leaf whorls and one-year-old stems. A single colony was streaked onto full-strength nutrient agar, incubated for two days at 26°C, then diluted in 1 mL of sterile water for each inoculation experiment. In the first experiment, five trees of each species were used, and each tree was inoculated three times (n = 45). Emerging leaf whorls were mechanically wounded with a 1 cc syringe needle as trees broke dormancy, then 50 µl of the 5x108 CFU bacterial suspension was injected between the layered whorls. In a second experiment, three trees of each species were inoculated three weeks post budbreak with 5 µl of a bacterial suspension injected into one-year-old stems. Each tree was inoculated twice (n = 18). In each experiment, control inoculations using sterile water were also made twice, slightly (3–4 cm) above the inoculation sight, or on a different twig. Each site was wrapped (Parafilm®, Bemis Company, Inc., Oshkosh, Wisconsin, U.S.). The canker lengths were recorded and the pathogen was re-isolated from cankers on nutrient agar five to fifteen days after inoculation.
Data were analyzed using JMP software (11.1.0v; SAS Institute). For each inoculation experiment, a mixed model was fit using canker lesion length as the response variable with tree species as a fixed effect. Tukey’s HSD was used to obtain pairwise comparisons between lesion lengths on the different tree species.
Disease Description
Three drippy-blight-diseased trees measuring 50–75 cm diameter at breast height located in Boulder, Colorado, parks were felled in December 2010. Approximately 100 twigs <1 cm diameter were collected from each tree, and evaluated for the presence of kermes scales and cankers. In 2015 and 2016, diseased trees throughout Boulder were monitored weekly from May to October to document the disease progression.
RESULTS AND DISCUSSION
The red oak (NCCB100490) and pin oak (NCCB100489) isolates of L. quercina subsp. quercina, shared 100% of nucleotide identity when the four gene sequences were compared (Caballero et al. 2014), and formed a monophyletic clade with L. quercina subsp. quercina ATCC 29281, based on the multi-locus sequence analysis (Figure 2). The Colorado strains are closely related to, but genetically distinct from, L. quercina subsp. quercina ATCC 29281 found in California. Strains are less closely related to L. quercina subsp. iberica (LMG 26264T, LMG 26265, LMG 26266, R-43277) or L. quercina subsp. britannica (LMG 26267T, LMG 26268, LMG 26269, R-43661, LMG 6054). Lonsdalea quercina subsp. iberica causes bark cankers and drippy bud on holm oak and Pyrenean oak in Spain (Q. ilex and Q. pyrenaica, respectively) (Biosca et al. 2003; Poza-Carrión 2008), whereas in the United Kingdom, L. quercina subsp. britannica is thought to contribute to acute oak decline of Q. robur and Q. petraea (Brady et al. 2012; Denman et al. 2012). Oozing bark cankers on hybrid poplar (Populus × euramericana) have been attributed to L. quercina subsp. populi (Tóth et al. 2013; Li et al. 2014).
Phylogenetic tree of Enterobacteriaceae species based on 16S ribosomal RNA, DNA gyrase subunit B genes (gyrB), beta subunit of ATP synthase (atpD), and translational initiation factor genes (infB) sequences. A Bayesian analysis was performed for 500,000 generations using a GTR / gamma distributed with invariant sites model of evolution. Bayesian probabilities are shown next to each branch. Species: Cronobacter sakazakii; Erwinia piriflorinigrans (CFBP 5882); Lonsdalea quercina subsp. iberica (LMG 26264T, LMG 26265, LMG 26266, R-43277); Lonsdalea quercina subsp. quercina (red oak isolate NCCB100490, pin oak isolate NCCB100489, ATCC 29281); Lonsdalea quercina subsp. britannica (LMG 26267T, LMG 26268, LMG 26269, R-43661, LMG 6054); Pantoea calida; and Erwinia toletana.
Following inoculation with L. quercina subsp. quercina, canker development associated with copious production of bacterial exudates was noted in tested trees involving all three red oak species (northern red, pin, and Shumard), except for one of eight Shumard oak saplings. Lonsdalea quercina subsp. quercina was consistently re-isolated from canker margins on each oak species. Trees inoculated on leaf whorls exhibited bacterial ooze within five to seven days of inoculation (Figure 3a), and small lesions formed as the shoots elongated. Shoot and leaf dieback were also observed. Cankers on red oak were significantly longer than cankers on pin oak (P = 0.04), with lengths of 1.02 cm and 0.55 cm, respectively. Shumard oak cankers measured 0.77 cm and did not differ from cankers on northern red or pin oaks (P = 0.53 and P = 0.35, respectively). Oaks inoculated on stems three weeks post budbreak developed bacterial exudates within ten to fifteen days of inoculation, and small cankers formed on stems within 21 days of inoculation (Figure 3b). Although the cankers on northern red oak trees averaged 1.25 cm and were over twice as long as cankers on Shumard and pin oak trees, there were no statistical differences in the canker lengths (P = 0.12 and P = 0.47, respectively). Cankers found on Shumard and pin oaks were also similar in size (0.53 cm and 0.57 cm, respectively, P = 0.76).
Inoculation experiment: a) Shumard oak, 14 days post leaf whorl inoculation with Lonsdalea quercina subsp. quercina (note the bacterial ooze at inoculation sites, arrows); b) canker formation in northern red oak on one-year-old stems.
In the course of field collections of three northern red oak trees in 2010, A. galliformis was found on 70%–81% of the twigs. Cankers typical of those produced by L. quercina subsp. quercina were present at one or more feeding sites of scales on 51%–57% of the kermes-scale-infested twigs. Further, cankers were only present at scale insect feeding sites, which strongly supports an interaction between A. galliformis and L. quercina subsp. quercina. This combination of kermes scale feeding and subsequent bacterial canker formation resulted in 30%–42% twig dieback.
The occurrence of twig cankers associated with the production of large amounts of bacterial ooze from infection with L. quercina has not previously been reported. In California, L. quercina subsp. quercina infections are limited to acorns of coast live oak (Q. agrifolia) and interior live oak (Q. wislizenii), producing a condition sometimes described as “drippy nut of acorns” because of the copious bacterial ooze manifesting at acorn wound sites (Hildebrand and Schroth 1967).
Although L. quercina causes infections of acorns in California (Hildebrand and Schroth 1967), a canker disease of Populus in China and Hungary (Tóth et al. 2013; Li et al. 2014), and contributes to oak decline in Britain and Spain (Biosca et al. 2003; Brady et al. 2012; Denman et al. 2012), drippy blight of red oak differs from these previous documentations. Branch dieback, canker formation, twig abscission at the junction to the current season’s growth, leaf drop, epicormic branching, and witch’s brooms are symptoms associated with this newly described condition. Furthermore, during the spring and summer, bacterial oozing occurs throughout the canopy. Exudates may be so copious that dripping ooze results in large sticky areas on sidewalks and other surfaces under the canopy throughout the middle of the summer. Twig cankers, indicated by maroon discoloration and clear margins (Figure 1d), appear in the late summer near kermes-scale-feeding sites and wounds. If the bacterium is present in the fall, it dries and hardens.
Historically, tree damage accompanying kermes scale infestations was attributed exclusively to the insect feeding (Turner and Buss 2005; Turner et al. 2005; Kosztarab 1996; Pellizzari et al. 2012). In contrast, these results document a situation in which the combined activity of a kermes scale insect and a bacterium produce the tree decline symptoms described as drippy blight of red oak (Snelling et al. 2011; Caballero et al. 2014). The exact manner of how these two organisms interact to produce drippy blight disease remains unclear. In drippy nut of acorns, the bacterium establishes at wound sites caused by seed-feeding weevils, filbertworms, and cynipid gall wasps (Swiecki and Bernhardt 2006). Similarly, in drippy blight of red oak, the wounding associated with kermes-scale-feeding injuries may provide entry or exit courts for the pathogen. Alternatively, the interaction may be more indirect, where kermes scales act as a stressor to facilitate growth and spread of the pathogen within the host.
Additional outstanding questions on drippy blight disease remain. For example, efforts to manage this disease complex have been disappointing and are complicated by the presence of two causal agents. Although effective scale management programs, such as removing scales by hand, pruning and destroying infested materials, treating with horticultural oils, and applying contact or systemic insecticides are likely the best way to control drippy blight disease (Turner and Buss 2005; Turner et al. 2005), studies to quantify the efficacy of control treatments are pertinent to maintaining red oaks in drippy-blight-diseased regions. If horticultural oils or insecticides are used to control drippy blight disease, monitoring for the most susceptible life stage of A. galliformis is necessary to determine the proper timing of applications. The life history of A. galliformis is similar to A. kingii (Hamon et al. 1976; Kosztarab 1996), but checking for A. galliformis egg hatch is pivotal to ensuring that the timing of application targets susceptible life stages (Turner et al. 2005).
Drippy blight disease of red oaks is an emergent disease caused by the interaction between A. galliformis and Lonsdalea quercina subsp. quercina. In addition to a description of this new disease, this report documents a host range expansion of L. quercina subsp. quercina to northern red oak, Shumard oak, and pin oak, and extends the reported host range of A. galliformis to include northern red and Shumard oaks.
Acknowledgments
Funding for this project was provided by the Colorado State University Agricultural Experiment Station Project 618A, and the TREE Fund grant 15-JD-01. We thank Boulder City foresters Pat Bohin and Tom Read, as well as University of Colorado arborist Vince Aquino for their help in locating and obtaining samples from diseased trees. Alison Hall and Erika Peirce provided laboratory help, and Emily Luna provided technical assistance. We thank the four expert scale insect taxonomists, Ian Stocks from Florida Department of Agriculture and Consumer Services, as well as Raymond Gill, Gillian Watson, and Natalia von Ellenrieder from the California Department of Agriculture, who were approached to provide identification of the Allokermes species associated with drippy blight disease. In addition, we are grateful for the comments from the editor and anonymous reviewers to improve this manuscript.
- © 2018, International Society of Arboriculture. All rights reserved.
Jump to section
Related Articles
Cited By...
- No citing articles found.