Uptake of zwitterionic antibiotics by rice (Oryza sativa L.) in contaminated soil

Highlights

• Response of rice plants to varying soil antibiotic concentrations investigated.

• Maximum concentrations in plants a linear function of initial soil concentration.

• Time to maximum concentration independent of initial soil concentration.

• Maximum concentrations and times required not related to compound hydrophobicity.

• Activity-based model successfully used with zwitterions.

Abstract

Antibiotics, including members of the tetracycline and fluoroquinolone families, are emerging organic environmental contaminants. Uptake from soil by plants is a means for antibiotics to enter terrestrial food chains. Chemical exchange between plant and the soil/water matrix occurs simultaneously with degradation in the soil/water matrix. In this study, the comparative temporal behaviour of rice (Oryza sativa L.) towards the zwitterionic antibiotics oxytetracycline, chlortetracycline and norfloxacin at initial soil/water concentrations of 10, 20 and 30 μg g⁻¹ (dry weight) is investigated. This is accomplished within the framework of an activity-based mass-conserving dynamic model. Plant antibiotic concentrations are observed to increase to a maximum then decline. Maximum concentrations in rice are compound-dependent linear functions of initial soil/water concentrations, but the relationships are not related to the compound octan-1-ol/water distribution ratio (D_OW). The times required to attain maximal concentrations are independent of initial soil/water levels for a given antibiotic, but again vary between antibiotics and are not related to D_OW values. Translocation from root to other tissues is not observed. The magnitudes of Root Concentration Factors (RCFs), the ratio of root and soil/water concentrations, are consistent with significant sorption to soil and consequent relatively low concentrations in interstitial water.

Introduction

Pharmaceuticals such as antibiotics have been categorized as organic contaminants of emerging concern [1]. Antibiotics are used both in human medicine and for prophylactic and therapeutic veterinary purposes [2]. It is well documented that absorption following administration is often poor, and significant proportions, in the order of 70–90%, may be excreted unmetabolized [3]. Incomplete removal of excreted antibiotics in wastewater treatment plants and contributions from animal wastes result in contamination of water, sediment and soils [4]. The fate and behaviour of antibiotics in soils has received increasing attention [5], [6]. However one aspect that has not received the same scrutiny is uptake from soil by plants [7]. This is important because there is the potential for antibiotics to enter the animal and human food chains in this way, representing an exposure pathway to humans and other biota. Further research on the behaviour of antibiotics at the soil/root interface, and subsequent plant uptake, has been advocated [8].

Common-use veterinary antibiotics include members of the tetracycline and fluoroquinolone families. These compounds are generally non-volatile solids, possessing acidic and/or basic functional groups and so can exist as anions, zwitterions or cations in aqueous solution, depending on the pK_a values of these functional groups and solution pH [9]. The zwitterionic form is typically predominant in neutral aqueous solutions. Physicochemical properties such as aqueous solubility and the octan-1-ol/water distribution ratio (D_OW) are pH dependent. The latter is often employed in preference to K_OW for such compounds in order to take into account the different hydrophobicities of the various species present as pH changes, and their relative concentrations [10], [11].

Trapp et al. [12] point out that almost half of the compounds preregistered for REACH, the European Community Regulation on chemicals and their safe use, are acids, bases or zwitterions. However knowledge of the environmental behaviour of ionized compounds and zwitterionic species in particular is less developed than that for unionized organic compounds. For example, the plant uptake studies that have been carried out with organic contaminants have largely focused on compounds such as polychlorinated biphenyls, polyaromatic hydrocarbons and pesticides [13].

Since the physicochemical properties of antibiotics differ from these more traditional organic contaminants, current experimental and modelling approaches for measuring or predicting fate and subsequent exposure do not necessarily work well for antibiotics [11]. Antibiotics are labile compounds, being subject to various abiotic (e.g. photolysis, hydrolysis) and biotic (e.g. microbial transformation) processes [4], but are continuously being released into the environment. Therefore concentrations in soil and water are unlikely to be constant with time. In response to locally decreasing contaminant levels in the external environment, previous work has shown that plant concentrations initially increase with time, to a maximum, then decrease [14]. Knowledge of potential maximum concentrations of antibiotics in plants, and how long it takes to reach these concentrations are of particular interest.

It is useful to investigate the uptake and accumulation of antibiotics by plants within the framework of a mathematical model. Reviews note that models can range in complexity and sophistication from simple assumptions of equilibrium attainment to dynamic models that consider plants to comprise a number of compartments (often physiologically based). The latter models are often extremely intensive in terms of data requirements [15]. It is generally recognized that accumulation of contaminants from soil by plants takes place through an intermediate water phase [16]. An obvious approach then would be a three-compartment (soil/interstitial water/plant) dynamic model. However, it is generally not possible to derive an analytic or closed form expression for the maximum concentration in the plant or time to maximum concentration from these models [14]. Numerical solutions are available but do not provide an understanding of underlying principles governing accumulation behaviour. A simpler two-compartment model that represents the soil/water matrix as a single peripheral compartment (Fig. 1) does allow the derivation of analytic solutions to these questions. Preliminary investigation has shown that such an approach can successfully describe the behaviour of a fluoroquinolone antibiotic in rice (Oryza sativa L.). In many parts of South East Asia, farmers employ an alternating rice-shrimp farming system producing shrimp in the dry season and rice in the wet season in the same plot [17]. Rice, the world's most important cereal crop [14], can thus be exposed to antibiotics from antecedent aquaculture activities under conditions such as those investigated in this work.

The general aim of this work was to investigate the temporal behaviour of antibiotic uptake by rice for a range of antibiotics at differing initial soil concentrations. The analytic solution of the two-compartment model was fit to the experimental data to enable investigation of the effects of initial soil/water concentrations and chemical compound on the maximum concentration in rice, the time to maximum concentration and the plant concentration factor. The results may also act as a general template for accumulation of organic contaminants from soil by plants where those contaminants are undergoing concomitant first order loss processes in the environment external to the plant.

The antibiotics selected were oxytetracycline, chlortetracycline and the fluoroquinolone norfloxacin. These are hereafter abbreviated as OTC, CTC and NORFLOX respectively and their structures in zwitterionic form are shown in Fig. 2.

These antibiotics were selected on the basis of their widespread usage as veterinary antibiotics [18] together with their possession of a representative range of physicochemical properties. These or similar compounds have also recently been found to be the highest risk for contamination of soils in the European Union and subsequent introduction of adverse ecotoxicological effects [6].