Plant and Soil 260: 19–32, 2004. © 2004 Kluwer Academic Publishers. Printed in the Netherlands. 19 Zinc mobilisation from a contaminated soil by three genotypes of tobacco as affected by soil and rhizosphere pH Nancy Loosemore, Anne Straczek, Philippe Hinsinger & Benoît Jaillard 1 Institut National de la Recherche Agronomique, UMR INRA – ENSAM Rhizosph` ere & Symbiose, 2 Place Pierre Viala, 34060 Montpellier Cedex 2, France 1 Corresponding author ∗ Received 23 January 2003. Accepted in revised form 11 July 2003 Key words: exchangeable cations, pH, rhizosphere, tobacco, zinc Abstract The aim of this research was to evaluate the effect of soil and rhizosphere pH on the mobilisation of Zn by various tobacco genotypes. One-month-old tobacco plants were grown for 8 days on top of a thin layer of an arable soil that had been sampled near a Zn smelter. A range of rhizosphere pH values was obtained either by growing nitrate-fed tobacco on top of the soil amended with various amounts of acid or lime, or by growing tobacco on top of the unamended soil with nitrate or ammonium supply. In the latter case, we used three genotypes that were assumed to differ in their ability to accumulate Zn or acidify the rhizosphere and, hence, mobilise soil Zn. In spite of the moderate level of contamination of the soil, tobacco took up substantial amounts of soil Zn. No difference was found between the three genotypes. Exchangeable Zn steeply increased with decreasing soil pH, which could be adequately modelled with a simple model. Whatever the source of nitrogen supplied, a significant acidification occurred in the rhizosphere. This explains why the observed Zn mobilisation was larger than expected on the basis of bulk soil pH values. Taking account of the change of pH induced by tobacco roots is thus of prime importance for better predicting the actual amount of exchangeable Zn in the rhizosphere and, thereafter the bioavailability of soil Zn. Abbreviations: A1 – control genotype of tobacco (wild-type genotype neutrally transformed with a CAMV 35S promoter-GUS construct); C5 – genotype of tobacco genetically transformed to over-accumulate the Fe storage protein ferritin in the cytoplasm; CEC – cation exchange capacity; CAMV – cauliflower mosaic virus; DW – dry weight; Fe-EDTA – ethylene-diamine-tetra-acetic acid iron III sodium salt; GUS – β -D-glucuronidase (EC 3.2.1.31); H – genotype of tobacco genetically transformed to over-accumulate the H + -ATPase protein in roots; H + -ATPase – proton-translocating P-type ATPase (EC 3.6.3.6); pH cobalthexamine – soil pH in cobalthexamine extract Introduction Availability of Zn for plants depends in the first place on forms under which the element occurs in soil. As most metals, soil Zn can be (i) dissolved in soil solu- tion as various species of soluble Zn, (ii) bound as exchangeable Zn to negatively charged compounds, (iii) adsorbed onto metal oxides, (iv) complexed by organic matter, or (v) included as insoluble Zn in ∗ FAX No: +33-467-632-614. E-mail: benoit.jaillard@ensam.inra.fr the crystalline lattice of clay minerals or metal ox- ides (Kiekens, 1995). The distribution of Zn among these various soil compartments determines its solu- bility, and thereby its availability. The solubility of Zn varies considerably according to soil composition and properties. For instance, Lorenz et al. (1997) reported that, in 10 European, contaminated topsoils, the solid- solution partitioning K d ranged from 69 to 36 222 L kg −1 . In a critical review, Sauvé et al. (2000) compiled K d values from 300 soils from 70 studies: K d ranged from 1.4 to 320 000 L kg −1 , with a mean of 11 615