Hormesis: fitness effects, mixtures, and comparisons across species and toxicants

The field of ecotoxicology has traditionally used monotonic relationships to study dose-responses, such as the linear, threshold or quadratic models. However, hormesis describes the dose-response relationship of toxicants as a biphasic curve, with low-dose stimulation, followed by high-dose inhibition. This phenomenon is hypothesized to be the result of an overcompensation response to a disturbance in homeostasis. In this manuscript, several aspects of hormesis were investigated. The first portion of this work tested the phenomenon itself with two anthropogenic toxicants (sulfosulfuron and glyphosate, growth-inhibiting herbicides) applied individually on the grass Bromus madritensis ssp. rubens. Numerous growth parameters and fitness indicators were measured to test whether a stimulatory effect in one parameter causes a tradeoff in a different parameter. Finally, the two toxicants were combined at individual hormetic doses to test whether hormesis is cumulative. Results support the hormesis hypothesis in all cases, with toxicants applied individually and in combination. In most experiments, there was no evidence of a tradeoff associated with the stimulation, but in one case, the stimulation was accompanied by a tradeoff in seed quality. The combination of toxicants at hormetic doses does not appear to be cumulative in terms of stimulation; however, the doses needed to achieve stimulation may be cumulative. In addition, different low doses caused stimulatory effects in different parameters, underlining the importance of parameter selection when studying hormesis. Finally, the timing of toxicant application can drastically alter hormetic responses. The second portion of this work tested the phenomenon using naturally-occurring toxicants. Two similar toxicants were used (caffeic acid and p-coumaric acid, both phenolic allelopathic chemicals), on two closely related grasses (Bromus madritensis ssp. rubens and Briza maxima). Both toxicants were applied to B. madritensis ssp. rubens to test whether similar toxicants generate comparable responses at similar doses. p-coumaric acid was also applied to B. maxima to test whether closely related grasses demonstrate a similar hormetic response at similar doses of a toxicant. Hormesis was detected in all three experiments. However, while the hormetic dose was comparable for both toxicants applied to B. madritensis ssp. rubens, the magnitude of stimulation was different. Conversely, the maximum stimulatory dose for both plants treated with p-coumaric acid differed considerably, while the stimulation itself was fairly consistent. In short, the data indicate that the hormetic concentration (but not the response) is generalizable across similar toxicants for a given species, and that the hormetic response (but not the concentration) is generalizable across similar species for a given toxicant. The final phase of this work attempted to elaborate a model system to study hormesis in plants. This system should be highly reproducible, and could be used to correlate stimulatory morphological effects to underlying molecular changes, namely gene expression profiles, and eventually protein expression as well. This may eventually lead to a mechanistic understanding of the hormesis phenomenon. Arabidopsis thaliana Col-0 was treated with imazapyr, a growth-inhibiting herbicide which targets the CSR1 subunit of AHAS, the first enzyme in the pathway for the biosynthesis of leucine, isoleucine and valine. Hormesis was successfully identified at a low dose of imazapyr. However, the stimulatory effect could not be reproduced in two subsequent repeats. Nonetheless, total RNA extractions were successfully performed on individual seedlings at various stages of growth to test whether sufficient RNA could be extracted from individual seedlings at all time points. Since hormesis is widely hypothesized to be an overcompensation response to a stressor agent, a logical starting place to assess the hormetic response in this system is to determine if the target of imazapyr is being upregulated when hormesis is present. Although live specimens from a hormetic treatment group were not harvested here, the primers designed to amplify the genes expressing the target enzyme, as well as reference genes, were successfully tested on non-hormetic specimens. If hormesis can be induced reproducibly in this system, the current methodology can effectively assess changes in gene expression. A more tightly controlled protocol, and the inclusion of more doses in the future, may increase the reproducibility of the hormetic phenomenon. The hormesis phenomenon has important implications for a number of fields. It has been suggested to play a role in ecological risk assessment, agriculture, pharmacology, medicine, pest management and others. As the number of examples of hormesis increases, and with a greater understanding of the phenomenon, the acceptance and usefulness of hormesis is likely to increase as well. Hence, while hormesis has often been overlooked, it is a topic which needs to be much more heavily investigated.Ph.D.Includes bibliographical referencesIncludes vitaby John Francoi