Mercury stable isotope fractionation during bacterial reduction of Hg(II) to Hg 0 Introduction The extreme toxicity of mercury (Hg) compounds warrants the search for new methods that can be used to track the sources of Hg and the dominant pathways leading to its bioaccumulation. Hg has seven stable isotopes (Fig. 1; 0.15 – 30% abundance; mass spread of 4%), is redox sensitive, and its compounds have a high degree of covalent character. Moreover, in recent years, a number of groups have reported significant and measurable Hg isotope ratio variation in natural samples 1,2 from hydrothermal ores, sediment cores and fish tissues--but the causes of fractionation are not clear. Thus Hg seems to be undergoing stable isotopic fractionation and the isotopic signatures of Hg may attest to its origin and/or redox history. This study investigates the naturally occurring processes that cause Hg to undergo reproducible and systematic mass dependent stable isotopic fractionation. Research Questions I.Is there any fractionation associated with the reduction of Hg(II) to Hg 0 by the mercuric reductase, an enzyme found in a broad range of Hg-resistant bacteria from diverse environments? II. If yes, is it mass dependent? Is it kinetic fractionation? What is the value of alpha? What is the effect of changing temperature? III. Is this fractionation phenomenon limited to pure cultures grown under laboratory condition or does it occur when naturally occurring bacterial consortium reduce Hg(II)? Methods Hg(II) reduction by a pure culture NIST 3133 was used as a source of 3 µM (600 ppb) Hg(II). Hg 0 volatilized during the growth of E.coli/pPB117 cells at 37 0 C (or 22 0 C) in M9-based minimal media and was purged into a trapping solution by air stripping (Fig. 2). In order to determine the change in isotopic composition as a function of the extent of the reaction, traps were replaced every min for a period of 320 min (and every 90 minutes for a period of 900 minutes for the experiment at 22 0 C) to collect products corresponding to different stages of the reaction. Hg reduction by naturally occurring bacteria : NIST 3133 was added to water samples from an uncontaminated source after a 4 day long pre-exposure 4 and Hg o produced was purged into a trapping solution (See Fig. 2 & 4). 250 ppb NIST was added to the control given no exposure. MC-ICPMS analysis 3 Sample introduction: Cold vapor generation was employed using Sn(II) reduction. The cold vapor sample introduction has a >99% efficiency and generates a signal of ~600 mV/ppb at a sample consumption rate of 0.75 mL/min. Mass bias correction: Addition of thallium (NIST 997) to the Hg vapor using a desolvating nebulizer. Precision: Fractionation was measured relative to the NIST 3133 Hg standard run before and after each sample and data are presented as δ 202 Hg/ 198 Hg (hereafter δ 202 Hg). Typical in-run precision of better than ±0.05‰ (2σ) and external reproducibility of δ 202 between NIST 3133 and a secondary standard was ±0.08‰ (2σ). The kinetic fractionation factor () was determined from the results of our experiments using the Rayleigh Distillation Equation: R Vi /R Lo = (1/) f (1/ -1) 1/ = 1 + [Slope of ln (R Vi /R Lo ) vs. ln(f)] (Equation 1) K. Kritee 1, B. Klaue 2, J. D. Blum 2, T. Barkay 1 1 Rutgers University, 76 Lipman Drive, New Jersey 08901, 2 University of Michigan, 1100 N. University Avenue, Michigan References 1. Smith C. et al. (2004), Eos Trans. AGU, 85(47), Fall Meet. Suppl., V51A Xie Q. et al. (2004), Eos Trans. AGU, 85(47), Fall Meet. Suppl., V51A Lauretta et al. (2001) Geochim.Cosmochim. Acta 65, Barkay T. (1987) Appl. & Env. Microbiology 53(12), Anbar(2004) Earth & Planet. Sci. Lett. 217, Johnson C. M. et al. (Ed.) (2004) Geochemistry of non-traditional isotopes. Reviews in Mineralogy & Geochemsitry 55. Acknowledgements Authors wish to acknowledge funding by NSF and NJWRRI. We thank John Reinfelder, Paul Falkowski, Robert Sherrell, Constantino Vetriani & Ariel Anbar for their helpful inputs at different stages of this project. Results At 37 0 C, Hg(II) undergoes mass dependent (Fig. 3b) Rayleigh fractionation (Fig. 3a) with fractionation factor () = / per amu during its reduction to Hg 0 by E. coli. At 22 0 C, modeled isotope ratios (& corresponding ) based on  (~ ) estimated by using Equation 1 (see methods & Fig. 3d) does not match with measured isotope ratios (Figure 3c). Plausible explanation: A constant offset (~0.0009) between measured ratios and ratios modeled assuming rayleigh fractionation could mean that net Hg fractionation at lower temperature is a result of combination of kinetic fractionation by Hg(II) reductase and equilibrium fractionation by Hg transport proteins (Fig. 1). Hg(II) transport across bacterial cell could be the rate limiting step in the reduction of Hg(II) at 22 0 C and not at 37 0 C due to increased rigidity of cell membrane at lower temperature. For Manipulated naturally occurring bacteria: When Hg 0 was produced after being pre-exposure to Hg(II) conc. of 250 & 175 ppb: 100% of surviving bacterial cells were Hg resistant (Fig. 5c) &  ~ (similar to pure culture) was observed (Fig 5b). But at low or no pre-exposure: Much lower % of total cells (10%) were Hg resistant & lower extent of fractionation (Fig. 5a and 5b) was observed. Plausible explanation: At high exposures, reduction by bacteria which have a unique & efficient (Fig. 5d) Hg reducing mechanism leads to fractionation similar to its extent in pure cultures. At lower exposures, reduction by variable non- specific mechanisms like reduction by light or weak organic acids results in mixed/weaker signal. Conclusions  Systematic Hg stable isotope fractionation does happen, both in pure cultures of bacteria and naturally occurring bacterial consortia!  Hg is the heaviest metal for which biological fractionation has been detected to date. In spite of the reduced % mass spread of its isotopes and increased molecular weight, the extent of fractionation found lies in the same range as for much lighter elements (Table 1).  Use of Hg isotope ratios for identifying sources and sinks, in situ pathways leading to its toxicity, and/or the nature and evolution of redox reactions in both modern and paleo environments is plausible.  Future work will determine how the change in physico-chemical parameters (T, pH, e - donor etc.) can change the extent of fractionation during Hg(II) reduction and other Hg transformations. Figure 5. 5a. Extent of isotopic fractionation (in per mil) measured in the trapping solution vs. exposure 5b.  factor/amu calculated assuming kinetic isotope fractionation vs. exposure 5c. Percentage of colony forming units (CFU) after 4 days of exposure which are Hg resistant vs. Hg exposure 5d. Hg 0 produced (in ppt) per Hg(II) resistant colony forming unit (Hg R CFU) vs. Hg exposure.