The strains used in this research were generously donated by the IAM Culture Collection, Center for Cellular and Molecular Research, the Institute of Molecular and Cellular Biosciences, the University of Tokyo (IAM), the Faculty of Engineering, Hiroshima University (HUT), and the Faculty of Agriculture, Hokkaido University (AHU). All chemicals (guaranteed reagents) used in this study were obtained from Nacalai Tesque (Kyoto, Japan).

The bacterial culture medium contained 3 g/L meat extract, 5 g/L peptone, and 5 g/L NaCl in deionized water [9] - [14]. The medium for growing actinomycetes, fungi, and yeasts contained 4 g/L yeast extract, 10 g/L malt extract, and 4 g/L glucose in deionized water with pH 7.1 (for actinomycetes) and pH 5.7 (for fungi and yeasts) [9] - [14]. The microorganisms were maintained on agar slants and grown in 300 mL of the medium in a 500-mL flask with continuous shaking (120 rpm) for 72 h at 30˚C. Cells were collected by centrifugation (for bacteria and yeasts) at 15,000 rpm for 3 times, or by filtration through a filter paper (No. 2, Advantec Co. Ltd., Tokyo, Japan) (for actinomycetes and fungi), which is washed thoroughly with deionized water, and then used in gold removal experiments.

Unless otherwise stated, the removal experiments were conducted as follows. Resting microbial cells [15 mg dry weight basis for tetrachloroaurate (III)] were suspended in 100 mL solution containing 50 mg/L (254 μM) gold (pH 3.0) containing hydrogen tetrachloroaurate (III). The suspension was shaken for 72 h at 30˚C. The resting microbial cells were then removed by filtration through a membrane filter (0.2 μm pore size). The gold removed by the cells was determined by measuring the gold content in the filtrate with an atomic absorption analysis quantometer (AA-6300, Shimadzu Corporation, Kyoto, Japan). Absorption spectrometry analysis of the filtrate of the gold removal using varying P. saccharophila IAM1504 amounts was measured by UV and visible spectrophotometer (V-650; JASCO Corporation, Tokyo, Japan) at wavelengths ranging from 200 - 870 nm.P. saccarophia and C. krusei cell surfaces were observed via scanning electron microscopy (SEM) and x-ray fluorescence (XRF) analysis (SEM-EDX S-4300, Hitachi High-Tech Corporation, Tokyo, Japan). Gold reduced by each microbial cell was frozen using a freeze dryer (FDU-830, EYELA Corporation, Tokyo, Japan) for 20 h in vacuo. The samples after deposition in vacuo for 24 h were observed SEM and XRF.

To determine the ability of different microbial cells to remove a larger amount of gold (III), 48 microorganism strains (5 actinomycetes, 19 bacteria, 13 fungi, and 11 yeasts) were screened. Gram-negative bacteria, such as P. maltophilia, have been reported to remove a large amount of gold (III) from an aqueous hydrogen tetrachloroaurate (III) solution (pH3) in 1 h at 30˚C by biosorption [9]. The highest adsorption of gold (III) occurred at pH 3.0 using P. maltophilia [9]. Therefore, the pH of the solution was adjusted to 3.0 in this screening process, additionally, we identified some microorganisms that removed a larger amount of gold (III) and reducted the gold in solution to gold (0) in the preliminary experiment.

The ability of the microbial cells to remove gold significantly varied (Tables 1-5). Of the tested microorganisms, high gold removal ability was observed in all microorganisms. Nocardiaerythropolis IAM1399 (gram-positive bacteria),Eschelichiacoli IAM1264, P. maltophilia IAM1554, and P. saccharophila IAM1504 (gram-negative bacteria), Aspergillusniger IAM2534,Chaetomium globosum IAM9272 and IAM9427 (fungi), and Candida. utilis IAM4220, Pichia farinosa IAM12223, and Saccharomyces cerevisiaeAHU3818 (yeasts) removed gold > 1200 μmol/g dry cell weight in 72 h at 30˚C. In the results of our previous study [9], the amount of gold (III) removed by gram-negative bacteria was higher than that by the gram-positive bacteria, actinomycetes, fungi, and yeasts. The maximum amount of gold (III) removal was 360 μmol/g dry cell weight in 1 h at 30˚C. Therefore, many microorganisms were able to remove large amounts of gold (III) after a long incubation time.

As these results were contact time-dependent, we hypothesized that the reaction mechanisms may be different. The solution was almost colorless for 1 h, however, it changed to dark colors like -violet or -dark green after longer removal time. Gold removal within 1 h likely occurred by biosorption, although longer incubation time (72 h) caused the reduction of gold (III) to zero-valent gold.

Many positively-charged metal ions can be removed at the neutral pH using gram-positive bacteria and actinomycetes [10] [11] [12] [13] [14]. Negatively-charged gold (III) ions can be removed at an acidic pH using gram-negative bacteria [9] through biosorption. However, all metal ions tested can be removed in small amounts using yeasts and fungi [9] - [14]. Therefore, we investigated the removal of gold (III) using P. saccharophila IAM1504 in detail because this microorganism removed the largest amount of gold (III) among the tested microorganisms by biomineralization.

Gold (III) removal as a function of time using P. saccharophila IAM1504 was examined (Figure 1); the amount of gold removed increased with incubation time. Importantly, gold removal reached two equilibria. The first equilibrium state was at approximately 6 h, and likely occurred by biosorption. Following this, the amount of gold removed increased again, and the solution color became darker, indicating biomineralization. The amount of gold removed using P. saccharophila IAM1504 by biosorption was relatively large [9], additionally the amount removed by biomineralization was much larger than biosorption.

Gold (III) removal byP. sacharophila cells was significantly affected by pH (Figure 2). The maximum amount of gold removal occurred at pH 3.0 (for 1 h) or pH 3.5 (for 72 h).

These results suggest that longer incubation time may change the reaction mechanism responsible for gold removal. The solution was nearly colorless after 1 h. However, the color changed to violet, green during the 72 h incubation period. Owing to the tetrachloroaurate ion having a negative charge, gold (III) can be effectively removed at pH 3 via biosorption [9]. It can also be reduced to atomic gold (0) by the activity of reductase in the presence of NADH [18] via biomineralization. Reduction occurred as shown in the following equation:

2 A u 3 + + 3 N A D H → 2 A u + 3 N A D + + 3 H +

The equilibrium in an acidic solution is driven to the left; thus, suitable pH changed from 3.0 to 3.5.

The amount of gold (III) removed (μmol/g dry wt. cells) by P. saccharophila cells decreased slightly with an increase in the cell amount (Figure 3). However, increasing the cell amount of P. saccharophila IAM1504 increased the total gold (III) removal. About 1300 μmol gold/g dry wt. cells were removed using 5.4 mg dry wt. of P. saccharophila cells after 72 h incubation. Although the solution color did not change after 1 h incubation, the color changed to violet after 72 h. Therefore, the amount of gold (III) removed by biomineralization was also much larger than that by biosorption.

To distinguish between the biosorbed and the biomineralized gold, we analyzed the ionic gold (III) and colloidal atomic gold (0) by absorption spectrometry. The 300 nm absorbance peak decreased with increasing cell amount, while no peak was observed from from 500 nm to 550 nm (Figure 4). Broad peak was observed usingC. krusei on similar experiment [19]. The 525 nm peak was identified as zero valent gold [20] and the 500 nm - 650 nm peak was small and broad because of the low solubility of gold (0). The amount of biosorbed gold (III) using P. saccharophilla cells was relatively large. Therefore, it can be inferred that gold reduction mainly occurred after being biosorbed on the cell surface ofP. saccharophilla cells. On the other hand, the amount of biosorbed gold (III) using C. krusei cells was relatively small, and hence, it can be inferred that gold reduction mainly occurred in solution using C. krusei cells.

To determine the maximum gold (III) removal ability at pH 4.0, we examined the mechanism by which the gold (III) concentration affected the gold removal by P. saccharophila cells. The amount of gold removed (μmol/g dry weight cells) by P. saccharophila cells increased with increased gold concentration, whereas the ratio of total amount of gold to the gold concentration decreased (Figure 5). For gold (III) concentration of 200 mg/L (1020 μM), 2500 μmol gold/g dry cell wt. was observed at pH 4.0.

To confirm the biomineralized gold condition, the cell surfaces of P. saccharophila and C. krusei were analyzed via SEM and XRF. As shown in Figure 6, many particles were observed on the P. saccharophila cell surface. It can be inferred that these small particles are reduced gold after adsorption by P. saccharophila (Figure 7). In contrast, no particles were observed on the cell surfaces ofC. krusei (Figure 8). However, as shown in Figure 9, peak was observed for reduced gold in the case of C. krusei. It can also be inferred that gold (III) is mainly reduced after being adsorbed on the P. saccharophila cells, whereas it is mainly reduced in the solution by C. krusei cells. This confirms the observations reported in Section 3.5.

Resting P. saccharophila cells (15 mg dry wt. basis) contacted with the hydrogen tetrachloroaurate (III) solution (Au 200 ppm, pH 4.0) for 72 h at 30˚C. Magnification times was 20,000x.

XRF analysis of dry cell surface of P. saccharophila were in contact with the the hydrogen tetrachloroaurate (III) solution (Au 200 ppm, pH 4.0) for 72 h at 30˚C.

Resting C. krusei cells (15 mg dry wt. basis) contacted with the hydrogen tetrachloroaurate (III) solution (Au 200 ppm, pH 4.0) for 72 h at 30˚C. Magnification times was 20,000x.

XRF analysis of dry cell surface of C. krusei were in contact with the hydrogen tetrachloroaurate(III) solution (Au 200 ppm, pH 4.0) for 72 h at 30˚C.