Xanthine (X) was obtained from Serva (Heidelberg, Germany). Buttermilk xanthine oxidase (XO), hydroethidine (HE), horseradish peroxidase (HRP), superoxide dismutase (SOD) from bovine erythrocytes, DNA (type III from salmon testes), horse heart cytochrome c, sodium dithionite, potassium superoxide, diethylenetriaminepentaacetic acid (DTPA), and Dowex 50X-8 (mesh 400) were obtained from Sigma (St. Louis, MO, USA). Anhydrous dimethyl sulfoxide (DMSO), chloroform, acetonitrile (ACN), absolute methanol (MetOH), hydrogen peroxide, Tris-HCl, and trifluoroacetic acid (TFA) were obtained from Merck (Darmstadt, Germany). Hydrophobic Oasis HLB 1 cm3 (30 mg) extraction cartridges were obtained from Waters (Milford, MA, USA). All reagents and solvents used were of the highest purity.
2.2. Superoxide Radical Sources
 The following superoxide radical sources are used in mixture with soil samples from Atacama: (I) DMSO-superoxide stock solutions produced (a) by alkalinization of DMSO and (b) by dissolving solid potassium superoxide in DMSO [Hyland and Auclair, 1981], and (II) solid potassium superoxide.
 I(a) Superoxide-DMSO stock solution produced by alkalinization of DMSO: Anhydrous DMSO (995 μl) is mixed with 5 μl 1 N NaOH (in a 1.5 ml-eppendorf tube) and is sonicated for 30–60 sec at 350 W cm−2, using sonicator (model UP-50 H, Dr. Hielscher, GmbH, Teltow, Germany) with 2 mm diameter microtip (type MS2) placed in the center of the NaOH//DMSO solution at 1.5 cm depth, followed by centrifugation at 15,000 g for 5 min. An identical superoxide stock solution, the water content of which is adjusted to 1.5% in order to induce superoxide destruction by dismutation, is used as a blank. Superoxide concentration in the supernatant (stock solution) is determined by using its extinction coefficient 2686 M−1 cm−1 at 256 nm [Hyland and Auclair, 1981]. The concentration of the superoxide radical stock solution was determined to be 4 × 10−5 M.
 I(b) Superoxide-DMSO stock solution produced by dissolving potassium superoxide in DMSO: Anhydrous DMSO (2 ml) is mixed with 1.27 mg potassium superoxide and sonicated as above for 30–60 sec, followed by centrifugation at 15,000 g for 5 min. A blank is prepared in an identical manner to that described above. The concentration of superoxide radical in supernatant (stock solution) was determined as in case (a) to be 1.4 × 10−4 M.
 Superoxide radical stock solutions (a) and (b) are stored in tightly stoppered bottles to prevent superoxide destruction due to humidity absorption by DMSO. Various volumes from these stock solutions are mixed with Atacama soil samples and treated as described in the ‘Soil Treatment’ subsection.
 II. Solid potassium superoxide: Various amounts of it are mixed with Atacama soil samples and treated as described in the ‘Soil Treatment’ subsection. For example, 1 g of soil sample is mixed with 1 mg potassium superoxide, which produces 1.1 × 10−4 moles superoxide.
2.4. HE-Superoxide Radical Assay
 Step A. Superoxide radical assay procedure: The superoxide/DMSO eluent (see ‘Soil Treatment’ subsection) is mixed with the following proportions HE and DTPA: 40μl 5 mM HE stock (1.58 mg HE dissolved in 1 ml 100% DMSO, N2-sparged and kept in a sealed 5-ml serum brown vial at −80°C for several weeks) and 3 μl 3 M DTPA aqueous stock per 1 ml DMSO-superoxide eluent. Under these conditions of low water content, superoxide radical rapidly reacts with HE to form 2-HO-ethidium. Specifically, in this mixture the final water content in DMSO is ∼0.4 % v/v (and DMSO 99.6%), and it is below the 0.9% water limit above which superoxide radical is very unstable since it reacts rapidly with water. Specifically, this water limit was determined by measuring superoxide in solutions with higher concentrations of water. The resulting 2-HO-ethidium/DMSO solution is treated as described in Step B below.
 To ensure that 2-HO-ethidium contained in the above 2-HO-ethidium/DMSO solution results solely from the superoxide radical of the soil sample, the following blank procedures are used to correct the superoxide radical value obtained from the soil sample in step E by subtraction.
 HE reagent-DMSO solvent-soil sample blank: This blank measures (a) 2-HO-ethidium present as contaminant in the solid HE commercial reagent, and (b) the extremely slight possibility that the soil might contain a contaminant that was isolated by this assay together with 2-HO-ethidium and has the same excitation/emission wavelengths as 2-HO-ethidium. This blank consists of the same soil amount used in its corresponding soil sample, impregnated with the corresponding volume of anhydrous DMSO. The soil/DMSO mixture is sonicated (as in the ‘Soil Treatment’ subsection), then water is added to a final DMSO/water ratio 98.5%/1.5% in order to destroy (by dismutation) any superoxide present, and finally it is mixed with HE and DTPA in proportions 40 μl HE stock and 3 μl DTPA stock per 1 ml DMSO. The resulting mixture is treated as in Step B.
 Sonication effect blank: This blank measures any artificially formed superoxide radical due to the sonication of the DMSO solvent (possibly containing dissolved molecular oxygen) as described in the ‘Soil Treatment’ subsection. This blank consists of the same volume of anhydrous DMSO used in its corresponding soil sample. DMSO is sonicated (as in the ‘Soil Treatment’ subsection), then water is added to a final DMSO/water ratio 99.1%/0.9% (in order for any superoxide radical formed during sonication to be preserved dissolved in the DMSO/water solvent), and finally it is mixed with HE and DTPA in proportions 40 μl HE stock and 3 μl DTPA stock per 1 ml DMSO. The resulting mixture is treated as in Step B.
 Step B. Isolation of 2-HO-ethidium by microcolumn cation exchange chromatography: The 2-HO-ethidium/DMSO solution from step A is brought to 70% DMSO with 0.2 M phosphate buffer, pH 7.0, and it is passed at a free flow rate through the Dowex 50X-8 cation exchange microcolumn prepared as described elsewhere [Georgiou et al., 2005]. Then, the microcolumn is washed in sequence with 1 ml 4 M NaCl (optional step), 1 ml distilled water, 2 ml 100% ACN, and 2 ml distilled water. The bound 2-HO-ethidium is eluted from the microcolumn with 1 ml 10 N HCl, and the eluent is diluted with distilled water to 3 N HCl.
 Step C. Purification of 2-HO-ethidium by microcolumn hydrophobic chromatography: The 2-HO-ethidium eluted from step B is subsequently passed through the HLB microcolumn at an approximate flow rate 2 ml min−1, after the column was activated as described elsewhere [Georgiou et al., 2005]. The microcolumn is first washed off with 1 ml 17% ACN-phosphate through the column to remove any fluorescent impurities. Subsequently, the bound 2-HO-ethidium is eluted with 1.5 ml 25% ACN-phosphate. The resulting eluent is mixed with 1.5 ml chloroform by vortexing, and the resulting chloroform/ACN layer (containing 2-HO-ethidium and possibly ethidium impurity from the HE stock solution) is collected. The 2-HO-ethidium in it is then concentrated by vacuum-evaporation of the chloroform/ACN solvent at room temperature.
 Step D. Fluorometric quantification of 2-HO-ethidium: The total fluorescence (F.U.total) of the vacuum-dried 2-HO-ethidium is due to both the 2-OH-ethidium collected from the soil sample plus any ethidium contaminant (from HE) being in mixture with 2-HO-ethidium. Therefore, the fluorescence of ethidium (F.U.Eth) in the mixture must be calculated after eliminating the fluorescence of 2-HO-ethidium by enzymically destroying it with HRP in the presence of hydrogen peroxide. The difference (F.U.total−F.U.Eth) between the two fluorescence values is due to the fluorescence of 2-HO-ethidium. For maximum assay sensitivity, the fluorescence of 2-HO-ethidium can be enhanced by complexation with DNA (it forms a fluorescent 2-HO-ethidium-DNA complex).
 Specifically, the 2-HO-ethidium/ethidium dry residue (from step C) is dissolved in 0.05 ml 50 mM phosphate buffer, pH 7.8, containing 1 mM DTPA and 6% DMSO. DTPA in this step (and in ‘Soil Treatment’ subsection) was used to chelate any metals from the soil sample that may inactivate the enzyme HRP used subsequently. In order to obtain maximum sensitivity for 2-HO-ethidium fluorescent quantification, its fluorescence (F.U.total) is measured in a final solution volume 0.3 ml using a fluorescence microcuvette with internal dimensions 4 × 4 × 45 mm in a Shimadzu RF-1501 spectrofluorometer, set at 10 nm excitation/emission slit width and high sensitivity. An additional increase in sensitivity is achieved by enhancing 25 fold the fluorescence of 2-HO-ethidium in the 0.3 ml solution by the addition of 0.02 ml 2 mg ml−1 DNA stock solution (to bring the final DNA conc. to approx. 0.15 mg/ml), and measuring the fluorescence (F.U.total) of the 2-HO-ethidium/ethidium-DNA complex at ex/em 515/567 nm. Subsequently, the same 0.3 ml solution is incubated for 1 min at RT after adding to it 0.025 ml 0.7 mM H2O2 stock (made in 50 mM phosphate buffer, pH 7.8, containing 1 mM DTPA) and 1 unit HRP (both required for 2-HO-ethidium fluorescence destruction), and the fluorescence (F.U.Eth) of the possibly present ethidium contaminant is measured. The fluorescence difference F.U.total−F.U.Eth corresponds to the actual fluorescence of 2-HO-ethidium, the concentration of which is determined by its fluorescence extinction coefficient (in the presence of DNA) described in the following step E.
 Step E. Conversion of the fluorescence of 2-HO-ethidium to superoxide radical concentration: The calculation of the fluorescence extinction coefficient of 2-HO-ethidium for the spectrofluorometer in use is done once in order to calibrate a known concentration of 2-HO-ethidium with its fluorescence in the presence of DNA. For this, a stock solution of 2-HO-ethidium is made in vitro by the X/XO superoxide generating system as follows: 0.1 ml 3.75 mM stock xanthine (made by dissolving 5.7 mg xanthine in 1 ml phosphate buffer containing 0.1 Í NaOH, and then diluting 10 times with phosphate buffer), 7 μl 5 mM HE stock (final HE conc. 35 μM) and 0.00825 units XO (i.e. 3 μl 33 Units/ml XO stock) are added to 0.9 ml 50 mM phosphate buffer, pH 7.8, and incubated for 30 min at RT. The concentration (35 μM) of the resulting 2-HO-ethidium in this stock solution is equal to the concentration of consumed HE, since it has been established that 2-HO-ethidium is the sole product of the reaction of superoxide with HE [Zhao et al., 2003]. Thus, the extinction coefficient of 2-HO-ethidium can be determined by measuring the fluorescence of various dilutions of the 2-HO-ethidium stock (in ±DNA), and the concentration of the 2-HO-ethidium in the 0.3 ml solution (step D) can be quantified from the fluorescence difference F.U.total−F.U.Eth measured in Step D.
 In order to ensure correspondence between the concentration of 2-HO-ethidium and the actual concentration of superoxide radical (in the 0.3 ml solution) it is necessary to construct a standard curve of 2-HO-ethidium versus superoxide radical concentration under the actual HE-superoxide radical assay reaction conditions (stated in Step A of this assay procedure, that is, in the presence of 99.1% DMSO and 0.9% water), using either of the superoxide radical sources I(a) or I(b) (described in the ‘Superoxide Radical Sources’ section above). This standard curve is shown in Figure 1. As long as the concentration of 2-HO-ethidium, determined from its fluorescence difference F.U.total−F.U.Eth converted to concentration by its fluorescence extinction coefficient, falls within the concentration range 0–120 nM (i.e., 0–40 pmoles in 0.3 ml solubilizate) where the molar stoichiometry ratio of formed 2-HO-ethidium/existing superoxide radicals is maintained constant at1/60 (Figure 1), the concentration of 2-HO-ethidium is multiplied by 60 and represents the actual concentration of superoxide radical in the 0.3 ml solution.
Figure 1. Standard curve of superoxide radical measured in soil by the HE-superoxide radical assay. It is an angled standard curve consisting of subcurves 1 and 2, which represent different ratios of measured/existing superoxide radicals (1/60 and 1/85, respectively).
Download figure to PowerPoint
2.5. Cytochrome c-Superoxide Radical Assay
 Superoxide radical in soil can be quantified spectrophotometrically by its equimolar reduction of cytochrome c absorbing at 550 nm, and it can be unambiguously identified by the inhibition of cytochrome c reduction utilizing superoxide dismutase (SOD) [Hyland and Auclair, 1981; McCord and Fridovich, 1969]. The assay was developed on Atacama Desert soil samples mixed with known quantities of DMSO-soluble and solid (potassium) superoxide radical from the sources stated in the ‘Superoxide Radical Sources’ subsection.
 Step A. Reduced cytochrome c extinction coefficient determination: Since in aqueous conditions 1 mole superoxide radical reduces 1 mole cytochrome c, for quantifying superoxide radical in the presence of DMSO (because, for this assay, DMSO is the solvent where superoxide radical is extracted from soil), it was necessary to determine the extinction coefficient of reduced cytochrome c within the range of the DMSO concentrations tested in this assay. For this, 1 mM oxidized cytochrome c stock (1.23 mg cytochrome c in 0.1 ml 50 mM phosphate buffer, pH 7.8) is prepared. A 250 × working dilution (4 μM) of the stock (made in final 1 ml 50 mM phosphate buffer, pH 7.8) was completely reduced with few grains solid sodium dithionite. The extinction coefficient of cytochrome c at 550 nm is calculated by the absorbance (ΔA550 nm) of its reduced 4μM solution in the presence of DMSO 0–40% v/v final concentration where it was found to be constant. As blank, a 4 μM solution of oxidized cytochrome c (in 0–40% DMSO) was used.
 Step B. Superoxide radical/cytochrome c reaction conditions and standard curve: Since superoxide radical is initially recovered from soil samples in 100% DMSO, it was necessary to determine the DMSO concentration range within which the 1:1 molar stoichiometric ratio for superoxide radical and reduced cytochrome c is constant. Therefore, a standard curve of reduced cytochrome c versus superoxide radical concentration was constructed in order to accurately quantify the concentration of superoxide radical in a 1 ml reaction volume. This volume is the minimum reaction volume that can be measured in 1 ml-sample cuvettes (with 1 cm light pathway) used by common laboratory bench-top spectrophotometers (in this study, the UV-VIS 1200 Shimadzu spectrophotometer was used). Moreover, the maximum limit of the DMSO concentration range (in conjunction with the 1 ml reaction volume) will determine the maximum soil sample size that can be used for superoxide radical detection by this assay. Specifically, in order to construct the standard curve, certain volumes (0.05–0.4 ml) of DMSO-soluble superoxide radical stock I(b) (containing 1.1 × 10−4 M in superoxide radical in 100% DMSO, see ‘Superoxide Radical Sources’ subsection) are brought to 1 ml final volume with 50 mM phosphate buffer, pH 7.8, containing 15 μM oxidized cytochrome c. The ΔA550 nm (absorbance of sample minus absorbance of blank) is then converted to reduced cytochrome c moles (Figure 2), using the extinction coefficient of cytochrome c determined in Step A. As a blank for the standard curve, the same volumes (0.05–4.0 ml) of DMSO-soluble superoxide radical stock are first mixed with water to a final DMSO/water ratio 98.5%/1.5%, in order to destroy (dismutate) the superoxide radical present, and mixed with oxidized cytochrome c as above. The standard curve shows that a constant 1:1 molar stoichiometry between reduced cytochrome c and superoxide radical is maintained within superoxide radical concentration range 0–4 μM (i.e., 0–4 nmoles in 1 ml reaction volume) for a DMSO concentration range of 0–40% v/v (Figure 2).
Figure 2. Standard curve of superoxide radical versus reduced cytochrome c concentration (solid symbol) as a function of DMSO concentration (open symbol).
Download figure to PowerPoint
 Step C. Assay procedure: Having established from Steps A and B that the constancy of the molar stoichiometric ratio 1:1 of superoxide radical to reduced cytochrome c is maintained up to a final DMSO concentration 40% (in 1 ml reaction volume, Figure 2), this means that the maximum DMSO volume that the superoxide radical can be recovered from any soil sample is 0.4 ml in order for its concentration to be determined accurately by this assay. Given the fact that the proportion of DMSO extraction volume/soil weight was established to be 0.3 ml/1 g (see ‘Soil Treatment’ subsection), the maximum amount of soil that this assay can test for extracting and measuring accurately the concentration of any superoxide radical on it is 1.3 g. Therefore, we used 1.3 g soil samples from Atacama in mixture with known quantities of superoxide radical from sources I and II (see ‘Superoxide Radical Sources’ subsection). The samples are extracted with a total amount 0.4 ml anhydrous DMSO as described in the ‘Soil Treatment’ subsection. The resulting DMSO-superoxide eluent is mixed with the following order: to 0.6 ml 50 mM phosphate buffer, pH 7.8, are added ±3.3 Units/ml SOD, and 15 μl 1 mM oxidized cytochrome c stock solution (15 μM oxidized cytochrome c final concentration), and its absorbance is measured at 550 nm against the following blanks. It should be noted that the use of SOD makes the assay specific for the detection and quantification of superoxide radical because this enzyme catalyzes its dismutation reaction to H2O2 and O2 [McCord and Fridovich, 1969]. Therefore, if superoxide radical is present in the 0.4 ml DMSO extract, SOD will dismutate all of it and it will inhibit the reduction of cytochrome c superoxide radical otherwise would have caused. The ΔA550 nm (absorbance of soil sample minus the sum absorbance of blanks) is then converted to reduced cytochrome c (and superoxide radical) moles from the standard curve of reduced cytochrome c versus superoxide radical concentration (Figure 2).
 DMSO solvent-soil sample blank: It measures whether any substances present in the soil absorb at 550 nm. It consists of the same soil amount used in its corresponding soil sample (1.3 g), extracted with 0.4 ml anhydrous 100% DMSO. The soil/DMSO mixture is sonicated (as in the ‘Soil Treatment’ subsection), the 0.4 ml DMSO extract is collected and mixed with water to a final DMSO/water ratio 98.5%/1.5% (in order to destroy any superoxide present by dismutation), it is mixed with 0.6 ml 50 mM phosphate buffer, pH 7.8, containing 15 μM oxidized cytochrome c, and finally its absorbance is measured at 550 nm.
 Sonication effect blank: It is similar to the corresponding blank of the HE-superoxide radical assay, and measures any artificially formed superoxide radical due to the sonication of the DMSO solvent as described in the ‘Soil Treatment’ subsection. This blank consists of the same volume of anhydrous DMSO used in its corresponding soil sample. DMSO is sonicated (as in the ‘Soil Treatment’ subsection), then water is added to a final DMSO/water ratio 99.1%/0.9% (in order for any superoxide radical formed during sonication to be preserved dissolved in the DMSO/water solvent), it is mixed with 0.6 ml 50 mM phosphate buffer, pH 7.8, containing 15 μM oxidized cytochrome c, and finally its absorbance is measured at 550 nm.