www.uhasselt.be
DSpace

Document Server@UHasselt >
Education >
Archive >
Applied Statistics: Master theses >

Please use this identifier to cite or link to this item: http://hdl.handle.net/1942/3746

Title: Determination of effective concentration 50% (EC50): case of uranium toxicity on carrot root grown in vitro cropping device
Authors: RENOLAYAN, Nanette
Advisors: THIJS, H.
Issue Date: 2007
Abstract: Uranium is a naturally existing heavy metal found in low levels in rocks, soil and water. In soil, the normal concentration of Uranium is 300 μg Kg-1 to 11.7 mg Kg-1 (Wikipedia 2006). In exceptional situations, Uranium concentrations in soils can reach tens to hundreds of milligrams per kg of soil, mostly because of mining and milling ores activities (Plant et al. 1999). High Uranium concentrations in soil can be toxic and therefore poses danger to the living organisms. Because of the undesirable effects of chemicals in soil, the evaluation of their toxicity becomes paramount. Toxicological tests are conducted, for instance by measuring the decrease in the rate of soil respiration upon increasing the concentration of heavy metals (Haanstra and Doelman 1985). Another way is by measuring the growth of terrestrial plants at increasing chemical concentrations. The part of plant that is first exposed to the chemical is the root so that in toxicological studies, root length is measured for different chemical exposures at certain points in time after planting. The toxicity of chemicals is commonly expressed in terms of dosage which gives 50% effect to the response (such as soil respiration or growth of a plant eg. root length) compared to the control. The effect can be either an increase or a decrease in response. This is called EC50 or Effective Concentration 50. The latter is also termed Effective Dose 50(ED50) or RD50 for dosage causing 50% reduction. In animal systems, it is referred to as LD50, the dosage lethal to 50% of the subjects (Schabenberger et al. 1999). The EC50 is usually estimated by fitting a log-logistic curve to the data. The model is a sigmoidal relation on a logarithmic scale rather than linear relation. The logistic model can be applied to dichotomous data such as survival or death and to continuous data for example weight or biomass, and in terms of length for growth. Several studies of doseresponse in herbicide application experiments have used the log-logistic function to model dose-response relationships (e.g. Streibig 1980; Laerke and Streibig 1995; Seedfeldt et al. 1995; Hsiao et al. 1996; Sandral et al. 1997 as cited in Schabenberger et al. 1999) Some studies with growth as response (continuous response) have shown that at some low concentrations of the toxic substance, growth is stimulated instead of being suppressed. This stimulus is called hormesis (from the Greek for ‘setting into motion’). A definition of hormesis derived from Stebbing (1982) is low-dose stimulation followed by higher-dose inhibition. The most common form of hormesis follows the widely recognized ß-curve shown in Figure 1. The use of the ß-curve follows principally from the widespread use of growth as a principal end point in hormesis research. Hormetic dose-response relationships are also seen in the form U-shaped curves. U-shaped dose-response curves would most appropriately be applied when the end point relates to a traditional toxicologically based health end point such as cancer incidence (Davis and Svendsgaard 1990) or a response for instance, the proportion of affected fetuses (Hunt and Bowman 2004). Reference to hormesis can be traced back to Schulz in 1888 who first expressed what is known today as the Arndt-Schultz law that every toxicant is a stimulant at low levels (Schabenberger et al. 1999). Several studies have shown that for low dosages of herbicide, the hormetic effect can occur that raises the average response at low dosages above the control value (Miller et al. 1962,; Freney, 1965; Wiedman and Appleby, 1972 as cited in Schabenberger et al. 1999). An investigation done by Calabrese and Baldwin in 1998 revealed that chemical hormesis is a reproducible and a relatively common biological phenomenon. Evidence of chemical hormesis was judged to have occurred in approximately 350 of the 4000 studies evaluated. Chemical hormesis was observed in a wide range of taxonomic groups and involved agents representing highly diverse chemical classes, many of which are of potential environmental relevance. Studies with chemical hormesis use different biological endpoints. Growth responses were found to be the most prevalent followed by metabolic effects, longevity, reproductive responses, and survival. If hormesis occurs, the standard log-logistic model does not fit the data. The usual practice was to still use the log-logistic model or drop part of the data. A solution was proposed in 1989 by Brain and Cousens by extending the log-logistic model. This modification naturally implements hormesis in the log-logistic model (Van Ewijk and Hoekstra 1993).
Notes: Master in Applied Statistics
URI: http://hdl.handle.net/1942/3746
Category: T2
Type: Theses and Dissertations
Appears in Collections: Applied Statistics: Master theses

Files in This Item:

Description SizeFormat
N/A267.01 kBAdobe PDF

Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.