Bacterial Bioremediation of Selenium Oxyanions Using A Dynamic Flow Bioreactor And Headspace Analysis

S. L. McCarty , T. G. Chasteen , V. Stalder, and R. Bachofen *

Department of Chemistry, Sam Houston State University, Huntsville, Texas USA and
*Institute for Plant Biology and Microbiology, University of Z½rich, Zollikerstrasse 107, CH-8008 Z½rich, Switzerland

Proceedings of the Third International Symposium on In Situ and On-Site Bioreclamation, San Diego, CA, USA, April 1995, published in "Bioremediation of Inorganics", R. E. Hinchee, J. L. Means and D. R. Burris eds.; Battelle Press, Columbus, OH, 1995; 95-102.


The volatile products of the biological reduction and methylation of selenium's most common oxyanions, selenate and selenite, were determined using capillary gas chromatography and fluorine‚induced chemiluminescence detection. Dimethyl selenide and dimethyl diselenide were detected in the headspace above cultures of bacteria resistant to this metalloid using static and dynamic headspace sampling techniques. Fluorine-induced chemiluminescence detection was applied to determine the relative concentrations of the organo-sulfur and organo-selenium species released over many days of culture growth at a controlled temperature and purge rate (for dynamic headspace sampling). A selenium-resistant bacterium, Pseudomonas fluorescens K27, and a phototrophic bacterium Rhodobacter sphaeroides 2.4.1 were exposed to selenate, and the cultures' headspaces were examined over a period of several days for volatile selenium-containing products. The results show that the relative production of the volatile species over time depicts a pattern generally independent of the growth phase in the case of the phototrophic bacterium; the concentrations of metabolic dimethyl sulfide and dimethyl selenide determined in static headspace were highest after the microbe had been in stationary phase for 4 days. For P. fluorescens , bioremediation activity peaked soon after the end of the log phase (about 10 hours) and continuously decreased over the next 80 hours of this dynamic headspace experiment.

Experimental Details

Phototrophic Bacterial Growth, Headspace Sampling, and Analyses

All of the static headspace experiments used Rhodobacter sphaeroides 2.4.1 grown on minimal medium with 20 mM succinate as the sole carbon source. A 45-mL culture of the phototroph was grown in a 100-mL Schott flask sealed using our specially designed enclosure cap (see Stalder et al. Analytica Chimica Acta , 1995, 303, 91-97).

Growth of Selenium-Resistant Bacteria

K27 strain of Pseudomonas fluorescens was grown anaerobically in trypticase soy broth with 0.1 % KNO3 (TSN) amended with sodium selenate (10 mM Se).

The 170 mL liquid culture was sealed in the bioreactor, maintained at 30 degrees C, and purged with sterile-filtered nitrogen at 3.0 mL per minute for the entire experiment. When the headspace wasn't being sampled for cryogenic trapping it was purged to waste.

Dynamic Headspace Cryogenic Sampling

Sampling was accomplished using the apparatus diagrammed schematically in this Figure. [A smaller version of this figure is here.] Each gas-phase dynamic headspace sample was cryogenically trapped twice: once in a fused silica-lined loop and then again on the column in the GC oven (-20 degrees C).

Chromatographic Analyses

The fluorine-induced chemiluminescence detector used in this work responds selectively to organoselenium and organosulfur species with detection limits (at S/N = 3) of approximately 25 picograms organo-chalcogen on-column. This corresponds to about 1 part per billion by volume (ppbv) detectable in a 12 mL headspace sample (4 minutes trapping at 3 mL/min). For more data on this sensitive and selective chemiluminescence detector see Chasteen et al. Chromatographia, 1990, 30, 181-185.

Hypertext Figures

Selected chromatograms from this experiment are shown in the figures detailed below. All of the data in the first four figures was taken by Steve McCarty. The components labeled include: methanethiol (CH3SH), dimethyl selenide (DMSe, CH3SeCH3), dimethyl diselenide (DMDSe, CH3SeSeCH3), dimethyl sulfide (DMS, CH3SCH3), and dimethyl disulfide (DMDS, CH3SSCH3).

We are excited about the "real time" nature of this kind of experiment because it shows us that we can track the volatile products released by bacteria that are performing a detoxification procedure as it happens. DMDS and DMDSe headspace concentration in parts per billion by volume (ppbv) are shown here over a single 100+ hour experiment. The rise in headspace production roughly corresponds to, but actually lags behind, the bacterial growth curve; the highest culture production is just after stationary phase is achieved. While our prototype apparatus did not allow for the sampling of culture components (biomass, pH, redox potential, etc.), we still feel that these data clearly prove our concept is useful as a means of following the biodegradation of toxic metalloidal compounds using time course experiments.

Figure 1 is a chromatogram of the headspace above the culture after 13 hours of continuous purging. The culture is in the log phase of growth.

Figure 2 is a chromatogram of the headspace above the culture after 20 hours of continuous purging. The culture is in the stationary phase of growth.

Figure 3 is a chromatogram of the headspace above the culture after 106 hours of continuous purging. The culture is in the late stationary phase of growth.

Figure 4 is a plot of the analyses from a single experiment carried out over 100 hours. The relative headspace concentration of dimethyl disulfide (DMDS) and dimethyl diselenide (DMDSe) is shown. This time course experiment last over 100 hours.

The results of a single time course experiment for Rhodobacter sphaeroides are shown in Figure 5. The experiment lasted approximately 180 hours. The top graph in this figure shows the culture's optical density (OD) at 660 nm as a measure of biomass and the ratio of optical density at 875 and 660 nm as a measure of the amount of photosynthetic units per biomass. The middle and bottom graphs of this figure plot the relative variation in organo-selenium and organo-sulfur components in the culture's headspace over time: dimethyl selenide (DMSe), dimethyl diselenide (DMDSe), dimethyl sulfide (DMS), and dimethyl disulfide (DMDS).


The data in the triple plot of Figure 5 for the phototrophs show that the highest concentrations of DMS and DMSe were found in the stationary phase of growth where some growth requirement has become limited to Rb. sphaeroides . This reduction and methylation therefore are powered only by light and not by reducing power produced in metabolizing and anabolizing cells.

In the Pseudomonas fluorescens K27 culture with its headspace continuously purged with nitrogen, the concentrations of volatile organo-sulfur or organo-selenium species were not allowed to build up in the culture medium or headspace; therefore, the metabolic processes of these bacteria probably were not affected by the volatile selenium-containing species produced.

The production of the DMDS and DMDSe shows a maximum after the achievement of stationary phase and a gradual decrease in the relative headspace concentrations of these species over time as shown in Figure 4. Unlike the phototrophic bacteria examined in the static headspace experiments, with K27 the reduction and methylation of selenium appears to be at maximum at or near the end of the log phase of growth.

Results from this research are published in "Bioremediation of Inorganics", R. E. Hinchee, J. L. Means and D. R. Burris eds.; Batelle Press, Columbus, OH, 1995; 95-102.

This research was supported by an award from Research Corporation.
Thanks to Hakan Gurleyuk for calibration work.

Abstract and some images used with permission of Battelle Press

Return to Chasteen's Home Page.