Justin Burum

Date of Award

January 2016

Document Type


Degree Name

Doctor of Philosophy (PhD)



First Advisor

Steven Ralph


Insect herbivory is a major stress on plants that results in significant economic losses to forest plantations and natural forests. In response, Populus have evolved a suite of constitutive and inducible defenses to deter insect feeding. Despite advances in technologies and the sequencing of the Populus trichocarpa genome, defensive pathways against insect herbivores are poorly understood. With the creation of activation tagged (AT) Populus we can use a forward genetics approach to identify genes involved in insect resistance (IR). To explore the genetic architecture of tree defenses, we conducted a forward genetics screen of AT Populus tremula x Populus alba (Pt x Pa) trees using choice bioassays with the purpose of identifying mutants with altered constitutive and induced defenses to Orgyia leucostigma. 770 AT mutants from Michigan Technological University were screened using mutant-mutant pairs on unwounded and wounded leaves. Fourteen percent (108 mutants) were identified as candidates for additional choice bioassays against wildtype trees. From this screen we have identified a few candidates that can be investigated with additional bioassays. In addition, we conducted bioassays on nine AT mutants previously identified from a screen at Queen’s University to confirm that they had altered IR. One mutant from the QU population, E8-16, showed IR during two independent choice bioassays using unwounded leaves. Choice bioassays demonstrated that larvae consumed less area and less weight from E8-16 compared to wildtype leaf disks. No-choice bioassays, designed to determine if the AT mutant has an effect on insect weight gain and development, revealed larvae reared on E8-16 trees gained less wet and dry weight, consumed less leaf material, and delayed larval development.

The next step was to identify where the T-DNA inserted within the Populus genome for E8-16 and which gene(s) were “activated” in response to being in close proximity to cauliflower mosaic virus enhancers on the T-DNA. We used SiteFinding and thermal asymmetric interlaced (TAIL) PCR to map a T-DNA insertion on chromosome 10 for the E8-16 mutant. Real-time PCR of three genes within 20 kb of the T-DNA revealed 10s12800 had elevated expression in E8-16 versus Pt x Pa wildtype leaves (6.9-fold, nested ANOVA, p<0.00001).

To begin to characterize the gene a bioinformatic analysis of the 10s12800 amino acid sequence was conducted which identified a really interesting new gene (RING) domain. 10s12800 is a putative E3 RING-H2 ubiquitin ligase involved in the terminal step of the ubiquitin-proteasome pathway that marks target proteins with ubiquitin for degradation by the 26S proteasome. Real-time PCR was used to measure absolute abundance of 10s12800 mRNA in a tissue panel consisting of mature leaves, immature leaves, phloem, bark, xylem, petioles, and roots. This analysis revealed 10s12800 is ubiquitously expressed in all tissues with greatest expression in mature leaves. To identify potential defensive pathways influenced by increased 10s12800 expression, we analyzed leaf transcriptomes of E8-16 and Pt x Pa wildtype trees on an Agilent 4x44K Populus microarray. Using criteria of p<0.05 and a false discovery q value <0.15, only 22 of the 43, 803 gene probes showed expression differences between Pt x Pa wildtype and E8-16. Among those 22 probes were MADS-box transcription factors and chitinases. To confirm 10s12800 is responsible for the IR phenotype, 10s12800 full length cDNA was inserted into the pCAMBIA S1300 vector behind a partial superpromoter and introduced into two different Populus genetic backgrounds. qPCR screening of transgenic lines revealed one line, TL4, with a two-fold increase in expression compared to a vector control (Nested ANOVA, p<0.05, n=5). Choice assays revealed that Orgyia leucostigma larvae preferred to consume vector control trees compared to TL4 (Nested ANOVA, p<0.05, n=8 pairs). However, we did not see a statistical difference in insect growth and development in a no-choice assay where larvae were caged on either TL4 or vector control trees (t-test, p>0.05, n=9). Collectively, these results suggest that we have identified a putative E3 RING-H2 ubiquitin ligase that may be a regulator of plant defense against Orgyia leucostigma.