Year of Publication


Degree Name

Doctor of Philosophy (PhD)

Document Type

Doctoral Dissertation




Plant Pathology

First Advisor

Dr. Lisa J. Vaillancourt


Anthracnose, caused by the fungus Colletotrichum sublineola, is one of the most important diseases on sorghum (Sorghum bicolor) in the United States (U.S.) and worldwide. The production of sweet sorghum for feedstock has been increasing in the Southeastern U.S. (SE), and anthracnose has emerged as a significant production constraint. Anthracnose is also common on the ubiquitous wild sorghum relative, johnsongrass (Sorghum halepense). The degree to which the pathogen population on johnsongrass contributes to disease epidemics on cultivated sorghum (S. bicolor) in the SE is unknown. Genetic and pathological diversity was characterized among a collection of Colletotrichum strains recovered from S. bicolor and S. halepense in Kentucky, Alabama, Georgia, and Florida. The pathogenicity of five C. sublineola isolates from sweet and grain sorghum and from johnsongrass was measured on the susceptible sweet sorghum inbred Sugar Drip in the field. Isolates from cultivated sorghum were more aggressive than isolates from S. halepense, which generally caused little or no disease. The disease levels observed in the field had no effect on the yields of sorghum biomass, grain, or juice, or on Brix levels. Removal of sorghum seed heads increased sugar levels in the plants, but this had no effect on susceptibility to anthracnose. Greenhouse and laboratory assays were developed that gave rankings of relative strain aggressiveness that were consistent with the field results. Marker analyses with repetitive fingerprinting probes were used to evaluate several hundred Colletotrichum strains isolated from S. bicolor and S. halepense across the SE. Results revealed that, with a few exceptions, isolates from cultivated sorghum were genetically distinct from isolates from S. halepense. A restriction fragment length polymorphism (RFLP) analysis based on probes against individual sequences presumed to encode effectors and secondary metabolism enzymes confirmed that most isolates from johnsongrass grouped separately from most isolates from cultivated sorghum. The RFLP analysis revealed the presence of three distinct groups within the population that were distinguished by fixed allelic variations, or by presence-absence polymorphisms, of some of these putative pathogenicity genes. Phylogenetic trees were inferred based on a sampling of isolates from both host species and representing each of the three groups by using the internal transcribed spacer (ITS) sequence of the ribosomal DNA; portions of the DNA lyase gene (Apn2) and the manganese superoxide dismutase gene (Sod2); and a region between the Mat1 and the Apn2 genes (Mat1/Apn2). The trees were found to be congruent, and to identify three distinct species, including C. sublineola and two previously undescribed species. One of these novel species was found associated only with S. halepense throughout the SE. This new species was named Colletotrichum halepense (Xavier & Vaillancourt). There was evidence for cross-infection of S. bicolor and S. halepense by C. sublineola and the second new species, which was named C. caselae (Xavier & Vaillancourt). These findings have significant implications for the development and deployment of resistant sweet sorghum varieties in areas where johnsongrass is common. This information will help to evaluate the potential for Colletotrichum spp. to cause epidemics in sweet sorghum if acreages continue to expand in the SE in the future.

Digital Object Identifier (DOI)