Article pubs.acs.org/est
Bacteriogenic Fe(III) (Oxyhydr)oxides Characterized by Synchrotron Microprobe Coupled with Spatially Resolved Phylogenetic Analysis Satoshi Mitsunobu,*,† Fumito Shiraishi,‡ Hiroko Makita,§ Beth N. Orcutt,∥ Sakiko Kikuchi,‡ Bo B. Jorgensen,∥ and Yoshio Takahashi‡ †
Institute for Environmental Sciences, University of Shizuoka, 52-1 Yada, Suruga-ku, Shizuoka 422-8526, Japan Department of Earth and Planetary Systems Science, Graduate School of Science, Hiroshima University, 1-3-1 Kagamiyama, Higashi-Hiroshima, Hiroshima 739-8526, Japan § Subsurface Geobiology Advanced Research (SUGAR) project, Japan Agency for Marine-Earth Science and Technology (JAMSTEC), 2-15 Natsushima-cho, Yokosuka 237-0061, Japan ∥ Center for Geomicrobiology, Department of Biological Sciences, Aarhus University, Ny Munkegade 114-116, Aarhus 8000, Denmark ‡
S Supporting Information *
ABSTRACT: Ubiquitous presence of microbes in aquatic systems and their inherent ability of biomineralization make them extremely important agents in the geochemical cycling of inorganic elements. However, the detailed mechanisms of environmental biomineralization (e.g., the actual reaction rates, the temporal and spatial dynamics of these processes) are largely unknown, because there are few adequate analytical techniques to observe the biogenic oxidation/reduction reactions in situ. Here, we report a novel technical approach to characterize specific biominerals associated with a target microbe on high spatial resolution. The technique was developed by combining directly in situ phylogenetic analysis, fluorescence in situ hybridization (FISH), with a synchrotron microprobe method, micro X-ray absorption fine structure spectroscopy (μ-XAFS), and was applied to iron mineral deposition by iron(II)-oxidizing bacteria (IOB) in environmental samples. In situ visualization of microbes revealed that in natural iron mats, Betaproteobacteria dominated by IOB were dominantly localized within 10 μm of the surface. Furthermore, in situ chemical speciation by the synchrotron microprobe suggested that the Fe local structure at the IOB accumulating parts was dominantly composed of short-ordered Fe−O6 linkage, which is not observed in bulk iron mat samples. The present study indicates that coupled XAFS−FISH could be a potential technique to provide direct information on specific biogenic reaction mediated by target microorganism.
■
INTRODUCTION Microorganisms in the environment critically impact global geochemical cycles and redox reactions of various elements. Many geochemically important redox reactions (e.g., methanogenesis, sulfate reduction, and Fe(II) oxidation) are largely associated with microbial activity and are energy sources for specific groups of microorganisms.1−3 For instance, recent studies suggest a significant relationship between Fe(II)oxidizing bacteria and ancient Banded Iron Formation, one of the large geochemical events in Earth’s history.4,5 In addition, microbes can mediate the formation of minerals by a process called biomineralization. Biominerals have unique morphology and characteristics such as nanoparticles, high surface area, and reactivity.3,6 Biominerals could be important sorbents for a range of metal(loid)s, and they often play a critical role as natural catalysts in adsorption and oxidation−reduction reactions for such elements.7−9 The general ecological importance of environmental microbial reaction and biomineralization has been well recognized; however, the specific mechanisms of the reactions in the © 2012 American Chemical Society
environment such as the reaction rate, spatial dynamics, and controlling factors are poorly understood, even in sediments and soils. For example, in sediments and soils, which are composed of heterogeneous mixture (e.g., water, organic materials, microorganisms, and primary and authigenic minerals), the local chemical profiles (e.g., pH and abundances of oxygen and nutrients) drastically change at the micrometer scale. Depending on such profiles, microbial reactions and habitability vary locally and form a complicated geochemical network in the environments. Another obstacle in understanding environmental biomineralization is that conventional cultivation-based methods are limited in understanding the whole suite of microbial processes and functions in the environments. The ecologically relevant microorganisms are often difficult to isolate in a pure culture. These facts make it Received: Revised: Accepted: Published: 3304
October 29, 2011 February 22, 2012 February 24, 2012 February 24, 2012 dx.doi.org/10.1021/es203860m | Environ. Sci. Technol. 2012, 46, 3304−3311
Environmental Science & Technology
Article
quite difficult to characterize the specific biogenic reactions, resultant biominerals, and the microorganisms involved in the processes in detail. Analytical techniques that allow for simultaneous determination of both microbial community composition and arrangement with characterization of the elements in biominerals associated with the microbes in high spatial resolution are needed to link microbial activity with the biogenic reaction and mineralization. In situ determination of chemical species in natural biogenic samples has advanced significantly in the past decade. Recently, organic components in biofilm were studied by Fourier transform infrared microspectroscopy at the cellular scale.10 However, the technique is available only for infrared-active functional groups such as hydroxyl and carbonyl groups. Scanning transmission X-ray microscopy (STXM) and X-ray photoemission electron microscopy (X-PEEM) were recently used to determine carbon and Fe species in bacteriogenic Fe(III) (oxyhydr)oxides at the submicrometer scale.11−13 Both of these techniques, however, also face some limitations. XPEEM must be used under ultrahigh vacuum (10−10 Torr), which may lead to unwanted alteration of biogenic hydroxide minerals.14 X-PEEM and STXM are mainly used for lighter elements with soft X-rays, and they require a relatively high concentration of target elements.15 In addition, both techniques are mainly based on an X-ray absorption near-edge fine structure (XANES) analysis only. Micro X-ray absorption fine structure spectrometry (μXAFS) using a synchrotron X-ray microbeam overcomes some limitations due to high element selectivity, nondestructive analysis, and analysis at atmospheric pressure. μ-XAFS can be used to identify in situ most of the elements on the periodic table at relatively low concentrations. Oxidation states and the coordination environment of elements are determined by using not only XANES and but also extended X-ray absorption fine structure (EXAFS). Based on these unique features, μ-XAFS recently has been applied to investigation of various biogenic reactions and mineralization in environmental samples.14,16−19 However, to the best of our knowledge, coupling of spatially resolved chemical speciation by μ-XAFS with spatially resolved phylogenetic characterization of microorganisms in the same environmental samples has not been described previously. Fluorescence in situ hybridization technique (FISH) offers a way to detect and identify microorganisms at the microscale without cultivation.20 In brief, target groups of microorganisms are resolved in natural samples by fluorescence microscopy based on using gene-targeted fluorescence probes that bind to cells of interest with high specificity.20 Recently, FISH has been applied to a number of types of natural samples, including soils, sediments, and microbial mats.21−25 Here, we directly coupled μ-XAFS and FISH to determine simultaneously the distributions of microbial groups and chemical species at micrometer scale (Figure 1). To achieve this approach, a new protocol for sectioning of natural samples was also developed. Coupling of XAFS and FISH provides more direct information on the identify and localization of microbially catalyzed redox processes and the associated minerals, which leads to a better understanding of the role of microorganisms on the geochemical cycling of elements. In this study, we applied the XAFS−FISH to one of the most ubiquitous and important environmental biomineralizations, Fe(III) mineral deposition by Fe(II)-oxidizing microbes. Our results provide new findings on microbially mediated Fe oxidation and biomineralization in the environment.
Figure 1. Schematic figure showing coupled XAFS−FISH technique.
■
EXPERIMENTAL SECTION Sampling and Analyses of Mat Samples. Natural Fe mats were collected near the Sambe hot spring (N 35.12, E 132.63) in Shimane Pref., Japan. The hot spring is located at the foot of an active volcano, Mt. Sambe. Iron-enriched hot water emanates from cracks within the lava bed. Yellow or orange microbial mats (Figure 2A,B) coat the floor and occur as meter-scale formations in the stream along the flow from the spring head with sedimentation rate of 5−8 μm/h. The mats are loosely aggregated with a soft structure (Figure 2B). Iron oxide mats used in this study were carefully collected around the source of Sambe hot spring with a sterile spatula. Then, the mat samples for 16S rRNA phylogenetic analysis and chemical analysis were immediately sealed in sterile test tubes, kept at −20 °C during transportation to our laboratory, and stored at −80 °C in a freezer prior to analyses. The mats for XAFS− FISH analysis were treated as described in following section. In situ measurements showed that the temperature and pH of the spring water were 34−38 °C and 5.7−6.3, respectively. Fluids above the mat were measured to have 1−6.5 mg/L dissolved oxygen and 30−60 μM total Fe. Details of chemical, mineralogical, and 16S rRNA phylogenetic microbial analyses of bulk Sambe Fe mats are described in the Supporting Information (SI). Preparation of Thin Sections Optimized for XAFS− FISH. As stated above, the mats are loosely aggregated. Hence, for the application of XAFS−FISH, a new sample preparation technique suited for soft and fragile samples was developed in the present study. The primary strategy of the method is to fix the mat samples after thin sectioning (“post-sectioning fixation”), while the previous FISH protocols for natural samples suggest an initial fixation step,21−25 which can cause significant disaggregation and deformation of soft solids. This modification has two advantages necessary for XAFS−FISH analysis. First, postsectioning fixation minimizes disaggregation and deformation prior to the analysis. Soft samples are too loose in the fixation solution, and once the original sedimentary structure is lost, there is no way to survey distributions of microbial communities by FISH and chemical species by μXAFS. Second, this new protocol enables us to apply μ-XAFS analysis to all possible intact samples. The fixation procedure was not conducted on the sections for μ-XAFS, while several sections for FISH were prepared and taken to the fixation step (Figure 1). The fixation solution typically includes significant amounts of formaldehyde, buffer reagents, and saline, which may cause unwanted alterations in chemical form (e.g., oxidation, reduction, and precipitation) in μ-XAFS analysis. 3305
dx.doi.org/10.1021/es203860m | Environ. Sci. Technol. 2012, 46, 3304−3311
Environmental Science & Technology
Article
Figure 2. (A, B) Photographs of sampling location and collected Fe mats at Sambe hot spring, respectively. (C) Scanning electron micrograph of the mats.
and Pernthaler et al.25 The detailed protocol for CARD-FISH is shown in SI. μ-XAFS Analysis. Fe K-edge μ-XAFS and μ-XRF analyses were conducted at both beamline BL37XU at SPring-8 (Hyogo, Japan) and beamline BL4A at KEK-PF (Tsukuba, Japan).17 At BL37XU, the incident beam was monochromated with a Si(111) double-crystal monochromator and focused to 0.9 (V) μm × 1.3 (H) μm beamsize at the sample position by a Kirkpatrick-Baez (K-B) mirror system. A thin-section sample was fixed on a sample holder oriented at 10° to the orthogonal direction of the beam. The fluorescence X-ray of Fe was measured by Si(Li)-SDD (SEIKO EG&G). At BL4A, the thin section was fixed on a sample holder oriented at 45° to the beam. The beam was focused by K-B optics, and the size was 5 (V) μm × 5 (H) μm, which is translated to 5 (V) μm × 7 (H) μm on the sample. The fluorescence X-rays were detected by a Ge-SSD. The energy calibration for Fe was performed using the pre-edge peak maximum of hematite fixed at 7.113 keV at both facilities. The Fe XAFS spectra of reference materials and the Sambe mat sample were measured in transmission detection mode. The energy calibration for Fe was performed using the pre-edge peak maximum of hematite fixed at 7.113 keV at both facilities. For Fe μ-XAFS measurements, the thin section sample was fixed to the XY stage controlled by X-Y pulse motors. The μXRF map of the mats was collected to determine points of interest in the sample by moving the sample with the pulse motor. The μ-XAFS spectrum was measured at each point at room temperature. Details for the EXAFS data analyses and preparation of reference materials for Fe XAFS are given in the SI.
This protocol can extend application of XAFS−FISH to other soft natural samples such as sediments and soils. For sample processing, a small volume (ca. 1 cm3) of mat trimmed by sterile blade was placed in a disposable plastic box, embedded in SCEM mounting medium (4% carboxymethyl cellulose; Leica Microsystems Japan, autoclaved prior to use), immediately frozen on sampling site by immersing in dry ice/ ethanol mixture, and then transported to our laboratory. In the laboratory, the frozen sample was sectioned by a cryotome (CM 1100; Leica) with disposable blades at −20 °C (our preliminary experiments show that the mounting medium is inert for chemical species of Fe based on Fe XAFS analyses). Thickness of the thin section used was around 5 μm for FISH and 20 μm for μ-XAFS in the present study. Sections obtained were mounted on thin films (Cryofilm Type 1; Leica Microsystems Japan) and used for further processing. It is important to note that adjacent sequential sections were used for coupled FISH and μ-XAFS (Figure 1), providing a way to conduct both analyses in the same area. Samples for μ-XAFS were thawed at room temperature, covered with Kapton adhesive tape in an N2 purged glovebox after the sectioning to minimize further reactions, and stored at −20 °C prior to the analysis. Section samples for FISH were fixed with Trisbuffered saline (TBS) buffered 3.7% formaldehyde, and kept cool at 5 °C in the dark overnight. Samples were then transferred to 50% ethanol in TBS, dehydrated by ethanol, and stored at −20 °C until further processing. Details for evaluation of the postsectioning fixation step are described in SI. FISH Analysis. To minimize the influence of mineral autofluorescence and/or unspecific binding of the oligonucleotide probe, FISH analysis was carried out by catalyzed reporter deposition FISH (CARD-FISH) using tyramide signal amplification, which enables 10- to 20-fold amplification of normal FISH. In addition, the strong signal of CARD-FISH is suitable for inspection of microbes in a wider region in the sectioned sample (millimeters to centimeters). The application of the CARD-FISH protocol to thin-section samples in this study is based on the procedure described by Shiraishi et al.22
■
RESULTS AND DISCUSSION Morphology, Mineralogy, and Microbial Ecology of Natural Fe Mats. XRD pattern of the bulk Fe mat samples showed broad reflections with no sharp peak, indicating that poorly crystalline Fe(III) (oxyhydr)oxides were dominant in the mats. SEM analysis of the bulk mat samples showed several morphologies of Fe oxides, including twisted stalk morpholo3306
dx.doi.org/10.1021/es203860m | Environ. Sci. Technol. 2012, 46, 3304−3311
Environmental Science & Technology
Article
suggest that the Betaproteobacteria which are abundant in the mats are closely associated to known Fe(II)-oxidizing bacteria, Gallionella spp. and may also perform this function. The dominance of Gallionella spp. is consistent with inspection by SEM showing the twisted stalk morphologies in the mats. Cultivation-Independent Analysis by FISH. The 16S rRNA gene phylogenetic analysis indicated that neutrophilic chemolithotrophic Gallionella related bacteria of the Betaproteobacteria may be involved in biogenically mediated oxidation of Fe(II) in the Sambe mats. We attempted to visualize the potential Fe(II) oxidizers in thin-sectioned mats by FISH technique using a specific probe for Betaproteobacteria, BET42a.29 DAPI staining to target all microorganisms in the mat was also performed, to observe where Betaproteobacteria and non-Betaproteobacteria resided in the mat. Bright-field and CARD-FISH with the Betaproteobacteria probe for the thin section are given in Figure 4A and 4B, respectively. A thin banded layer (10−15 μm thick) was formed on the surface of the Fe mat (Figure 4A), and cells bound by the BET42a were also abundant in this layer (Figure 4B). The magnified FISH image in Figure 4B shows that cells were vibrioid or curved rod shaped, and the cell diameters were approximately 1−2 μm (arrows in the image). These features agree well with those observed for Gallionella spp.,30 which also agrees with the 16S rRNA analysis. A depth profile of cell abundance from the surface of the mat into the interior (∼160 μm transect), estimated by direct counting (Figure 4D), indicates that cells bound to the BET42a probe show a maximum at the surface (∼10 μm depth, 500 cell/100 μm2), followed by a steep decrease with depth, and became constant in the deeper parts of >20 μm (less than 100 cell/100 μm2). In contrast, abundances of cells bound to the DAPI increased again in the depth of 80−120 μm. The significant increase may indicate an appearance of anaerobic Fe(II)-oxidizing bacteria (e.g., nitrate-reducing and phototrophic Fe(II) oxidizers) in the depth if it is assumed that anoxic condition was formed in the deep layer. Considering the close physical association of these Betaproteobacteria bacterial cells with dense layers of Fe oxides in the mat surface, it is possible that a microbial habitat suitable for Fe(II) oxidation was formed at the water/sediment interface. This microscale localization of Fe(II)-oxidizing microbes may be caused by the O2 abundance in the mats and water−mat interface, because the activities of neutrophilic Fe(II) oxidizers are limited by those chemical parameters such as O2 and Fe abundances.31,32 Previous studies using laboratory experiment reported that Gallionella spp. are microaerophile that normally grow at the O2 concentrations of 1−10% of the ambient saturation.33,34 However, our direct FISH image showed the localization of Gallionella spp. on the upper 10− 15 μm of the mat surface, where it is expected that O2 concentration is relatively higher than that at the inner part of the mat. This finding contrasts with the general perception of Gallionella being strictly microaerophilic.35 Thus, the localization of the bacteria on the mat surface observed in present study implies that (i) optimum low level O2 condition for the Gallionella spp. was formed on the surface at microscale and/or (ii) Gallionella spp. studied in present study may not be a true microaerophile and can grow in broader O2 concentration region. In fact, a recent publication has reported that no growth inhibition of Gallionella spp. was found in natural water under fully aerated condition.36 This report may support the hypothesis that Gallionella spp. grow under wider O 2 concentration conditions than generally assumed. The
gies (arrows in Figure 2C) that are characteristic of freshwater Gallionella relatives3,26 and marine Mariprof undus ferooxydans Fe(II)-oxidizing bacteria.27,28 The bacterial community of the Sambe Fe mats was obtained by PCR-based 16S rRNA gene sequence analysis. Seventy-four clones were obtained for the sample with the average length of 1360 bp. Figure 3A shows the distribution of bacteria in the
Figure 3. (A) Pie chart showing the percentage of clones in a phylum/ class in 16S rRNA clone libraries from Sambe Fe mats. (B) Phylogenetic tree showing the relationship between clones in βproteobacteria group based on Neighbor-Joining method. ζ-Proteobacteria group (Mariprof undus ferrooxydans) was chosen as the out group. In OTU1−2, clone and accession numbers of a representative clone are shown in each OTU in the tree.
mats at the phylum/class taxonomic level. Based on this analysis, the bacterial community of the mats is dominated by Proteobacteria (55% of clones) and Verrucomicrobia (34%). Within the Proteobacteria phylum, Betaproteobacteria (36% of clones, 27 clones) are more abundant than the other divisions (Figure 3A). The majority of the Betaproteobacteria clones (25 in 27 clones) in the mats were closely related with bacteria involved in Fe redox transformations. These included the Fe(II)-oxidizers Gallionella genus as shown in the phylogenetic tree of the Betaproteobacteria (Figure 3B). Two operational taxonomic units (OTU1−2) were observed in the Betaproteobacteria in the mats: OTU1 contains 20 clones with sequences that are similar to uncultured Gallionella sp. (EU266781) (individual sequence similarity = 95.31−96.62%; OTU mean = 96.40%) based on a BLAST search. OTU2 contains 5 clones that are similar to Gallionella ferruginea (L07897) (95.56− 97.10%; mean = 96.62%) and Gallionella capsiferriformans (DQ386262) (95.80−97.40%; mean = 96.85%). These findings 3307
dx.doi.org/10.1021/es203860m | Environ. Sci. Technol. 2012, 46, 3304−3311
Environmental Science & Technology
Article
Figure 4. (A) Bright-field image of thin-sectioned Sambe Fe mat. (B) CARD-FISH image of the same field stained by BET42a and the magnification of high population area. Arrows in the magnified image stand for curbed rod shaped cells typical of Gallionella relatives. (C) Fe chemical image of the same field collected by μ-XRF. (D) Depth profile (0−150 μm) of cell abundance in (B). White lined boxes in (B) and (C) indicate the area analyzed by μ-XAFS in Figure 4.
diminished abundance of BET42a bound cells in deeper parts of the mat suggests that biogenic Fe(II) oxidation dominantly proceeds within 10−15 μm of the surface of the mat. Spatially Resolved μ-XAFS Analysis. FISH analysis of the Fe mat thin section showed the localization of Betaproteobacteria (presumably related to Gallionella and involved in Fe oxidation) in the upper 10−15 μm of the mat. Assuming that the Betaproteobacteria were involved in Fe oxidation in this layer, we would expect that larger amounts of biogenic Fe oxides should be found in this layer as well due to microbial activity. Furthermore, we expected that the Fe oxides would carry a biogenic signature (i.e., short-range ordered Fe(III) (oxyhydr)oxides), as has been previously observed in the stalks produced by Fe(II)-oxidizing bacteria.18,19 To address these hypotheses, we characterized Fe chemical species and mineralogy in the mat by Fe μ-XRF-XAFS with high spatialresolution. Figure 4C shows the Fe μ-XRF map of the mirror thin section analyzed by FISH. The map shows that Fe was locally concentrated at the surface of the mat (where the abundance of the Betaproteobacteria was the highest) as well as in a deeper layer in the mat. We collected Fe μ-XAFS in two spots in the surface layer (spots 1 and 2, white circles in Figure 5A) and in one spot in the inner, Betaproteobacteria-poor section below the surface (spot 3 in Figure 5A). The Fe k3-weighted EXAFS spectra of reference materials, the Sambe spots 1−3, and bulk mats are given in Figure 5B. The spectral features of spots 1−3 and bulk sample were more similar to that of ferrihydrite than
to those of goethite or lepidocrocite, which indicates that the Fe mats are mainly composed of short-ordered Fe(III) (oxyhydr)oxides such as ferrihydrite. A small peak at k = 7.0−7.5 Å−1 was observed in ferrihydrite, Sambe bulk sample, and spot 3 (dotted line box in Figure 5B), whereas this peak was not found in spots 1 and 2 of the Fe mats. This implies that the coordination environment for Fe−Fe linkages in spots 1 and 2 may be different from the others, because this peak is dominantly derived from Fe−Fe coordination in the Fe(III) (oxyhydr)oxides (shown in Figure S2 in SI and ref 37). In the Fourier transform (FT) analysis of the EXAFS spectra of ferrihydrite, spots 1 and 3, two main shells are shown in R + ΔR = 1−3.5 Å (Figure 6, phase shift uncorrected). The first shell at 1.5 Å corresponds to the Fe−O coordination, and the intensity and position are approximately identical among the Sambe and ferrihydrite spectra. In contrast, second shell of spot 1 identified at R + ΔR = 2.3−3.5 Å corresponding to Fe−Fe coordination was significantly smaller than those of the ferrihydrite and spot 3. This indicates that Fe in the surface Fe(III) (oxyhydr)oxides makes a weaker Fe−Fe linkage than in ferrihydrite. Further analysis of the Fe−Fe linkage to determine structural properties was performed using quantitative simulation of EXAFS (Table S2 in SI). For the ferrihydrite, two Fe−Fe distances of ferrihydrite obtained by EXAFS analysis are 3.04 and 3.43 Å (Table S2 and Figure S3 in SI). The first value corresponds to the Fe−Fe distance along the edge-sharing linkage, and the second corresponds to the Fe−Fe corner3308
dx.doi.org/10.1021/es203860m | Environ. Sci. Technol. 2012, 46, 3304−3311
Environmental Science & Technology
Article
Figure 6. Fe K-edge FT spectra for spot 1 (surface area), spot 3 (inner area), and 2-line ferrihydrite.
100 μm deeper parts where the Fe oxidizer activity was low. These findings imply that biogenic Fe(III) (oxyhydr)oxides observed are a metastable phase, and subject to the mineral transformation with time. Chan et al.38 suggested that stalkforming Fe(II)-oxidizing bacteria such as Gallionella relatives initially form metastable and short-ordered Fe(III) (oxyhydr)oxides after metabolic Fe oxidation, and those gradually recrystallize to more crystalline Fe(III) (oxyhydr)oxides (e.g., lepidocrocite) as the stalks grow, which is consistent with our findings. Thus, this implies that in the present study, the metastable bacteriogenic Fe(III) (oxyhydr)oxides could be detected and characterized using the XAFS−FISH technique. Factors that cause lower crystallinity in Fe(III) (oxyhydr)oxides can include the rapid rate of hydrolysis reactions and presence of organic or inorganic ligands during Fe(III) polymerization.38−42 It is important to note that the abiotic Fe(II) oxidation (and Fe(III) (oxyhydr)oxides formation) rate is comparable to biotic oxidation in circumneutral pH waters such as in the present study.36 In addition, it is assumed that the Fe mats were successively deposited from the hot spring water with similar constituents of inorganic ligands. Previous studies suggested that microbial exopolysaccharides produced during metabolism control the crystallinity of Fe(III) (oxyhydr)oxides in microbe-associated biomineralization.18,38 We also examined effect of the microbial exopolymer in the mineralogical structure of Fe(III) (oxyhydr)oxides. The results showed that Fe(III) (oxyhydr)oxides synthesized in presence of microbial acidic polysaccharide also have short-ordered and limited structure observed in the spot 1 in the Fe mats (details are given in page 9 in SI), which infers that bacteriogenic organic ligands can be one of responsible factors for the limited structure found at the surface. XAFS−FISH Application in Biogeochemical Studies. Development of in situ, microscale techniques that can be used to investigate both the microbial community and the speciation of elements in natural samples is critical for increasing our knowledge of the role of microorganisms in the biogeochemical cycling of elements. In the present study, we developed a novel
Figure 5. (A) CARD-FISH image analyzed by Fe μ-XAFS. White circles in the image (spots 1−3) indicate points of interest in Fe μXAFS measurements. (B) Micro and bulk Fe K-edge EXAFS spectra of the points in Sambe thin section, bulk sample, and reference materials (lepidocrocite, goethite, and 2-line ferrihydrite).
sharing linkage (Figure S4 in SI). The coordination numbers (CNs) of both bonds are 3.1 and 1.3, respectively, which agree with the published structure of 2-line ferrihydrite.31 For spot 1, the presence of edge-sharing linkage was also observed similarly to ferrihydrite, but the CN (= 1.5) is smaller than that of ferrihydrite (CN = 3.1) as shown in Figure S3 and Table S2. In addition, the contribution of the corner-sharing link was also much smaller and was not detected in the simulation. These results suggest that Fe(III) (oxyhydr)oxides in spot 1 (surface part in Fe mats) are dominantly composed of edge-sharing linkages of Fe−O6 octahedra, showing the Fe(III) (oxyhydr)oxides with secondary structure, as has been previously suggested for biogenic Fe oxides.18 Implication of the Short-Ordered Fe(III) (Oxyhydr)oxides Observed. EXAFS analyses showed that Fe(III) (oxyhydr)oxides formed at Betaproteobacteria (presumably related to Gallionella)-accumulating parts have a short-ordered Fe(O,OH)6 structure compared with abiotically synthesized Fe(III) (oxyhydr)oxides. On the other hand, similar spectrum features were not found on the bulk Sambe mat samples and 3309
dx.doi.org/10.1021/es203860m | Environ. Sci. Technol. 2012, 46, 3304−3311
Environmental Science & Technology
Article
method using μ-XAFS and -XRF combined with FISH to determine directly microbial communities and chemical speciation of elements with high spatial resolution (1−5 μm), and applied it to bacteriogenic Fe deposition. We successfully conducted in situ visualization of potential Fe(II) oxidizers in a thin section of Fe mats by the phylogenetic FISH technique. The Betaproteobacteria presumably related to Fe(II)-oxidizing bacteria were highly localized within 10 μm of at the surface. There are many previous studies visualizing Fe-oxidizing bacteria in natural biofilms, soils, and sediments.19,23,27,30 However, these studies were performed in disassembled solids, which limit detailed understanding of biogenic Fe deposition and habitability of Fe oxidizers. This is the first study in situ visualizing vertical habitat of the Fe-oxidizing bacteria in natural Fe mats, to our knowledge. Furthermore, direct chemical speciation by μ-EXAFS revealed that Fe(III) (oxyhydr)oxides with the limited structure was deposited at the Betaproteobacteria-accumulating layer. As demonstrated in this study, our novel approach has many merits for investigating the relationship between microbes and chemical species: (i) simultaneous analysis by FISH and μ-XAFS enables us to characterize directly the biomineral associated with a targeted microbe, while most of the previous studies lack in situ phylogenetic information on a specific microbe; (ii) a nondestructive analytical technique, μ-XAFS with the high sensitivity, elemental specificity, and spatial resolution is useful to trace various biogenic reactions in natural environments and to characterize resultant biominerals; and (iii) both FISH and μ-XAFS are cultivation-independent methods, which expands the application of the coupled XAFS−FISH technique to various natural samples. Moreover, in present study, we also developed a new preparation method of thin section optimized for soft and fragile natural samples (see Experimental section and page 2 in SI). The technique can extend application of XAFS-FISH to other environmental samples. Based on these advantages, our approach can provide a means of further elucidating not only the characterization of the environmental biominerals (present study) but also geochemical function of microbes that control the cycling of elements in the environments using species-distinguishable method, μXANES. In addition, regarding a limitation of the XAFSFISH application to environmental samples such as soils and sediments, the μ-XAFS has various merits to the application (e.g., high element selectivity, nondestructive analysis, analysis at atmospheric pressure),15,16 which enable us to apply it easily. On the other hand, regarding FISH technique, we may have several limitations in its application to other environmental samples such as soils and sediments. It was reported the detection of FISH stained cells in the soils was mostly affected by autofluorescent soil particles and autotrophs.21 The effects are partly overcome by applying CARD-FISH using tyramide signal amplification, which allows 10−20 fold amplification of normal FISH signal.
■
Notes
The authors declare no competing financial interest.
■
ACKNOWLEDGMENTS We thank Dr. A. Kuwahara, Dr. S. Karaki, and Ms. T. Ishikawa for kind assistance and suggestions in preparation of thin section. This work is supported by funding from University of Shizuoka, the JSPS Research Fellowships for Young Scientists, the Researcher Exchange Program between JSPS and DU, the Danish National Research Foundation, and the Max Plank Society. This work has been performed with the approval of JASRI (Proposal 2010B1741, 2010B1363, and 2011A1223) and KEK (Proposal 2011G154).
■
(1) Habicht, K. S.; Canfield, D. E. Sulphur isotope fractionation in modern microbial mats and the evolution of the sulphur cycle. Nature 1996, 382, 342−343. (2) Orphan, V. J.; House, C. H.; Hinrichs, K. W.; McKeegan, K. D.; DeLong, E. F. Methane-consuming archaea revealed by directly coupled isotopic and phylogenetic analysis. Science 2001, 293, 484− 486. (3) Fortin, D; Langley, S. Formation and occurrence of biogenic ironrich minerals. Earth Sci. Rev. 2005, 72, 1−19. (4) Konhauser, K. O.; Hamada, T.; Raiswell, R.; Morris, R. C.; Ferris, F. G.; Southam, G.; Canfield, D. E. Could bacteria have formed the Precambrian banded iron formation? Geology 2002, 30, 1079−1082. (5) Kappler, A.; Pasquero, C.; Konhauser, K. O. Deposition of banded iron formations by anoxygenic phototrophic Fe(II)-oxidizing bacteria. Geology 2005, 33, 865−868. (6) Villalobos, M.; Lanson, B.; Manceau, A.; Toner, B.; Sposito, G. Structural model for the biogenic Mn oxides produced by Pseudomonas putida. Am. Mineral. 2006, 91, 489−502. (7) Stumm, W.; Morgan, J. J. Aquatic Chemistry, Chemical Equilibria and Rates in Natural Waters; John Wiley & Sons: New York, 1996. (8) Tessier, A.; Fortin, D.; Belzile, N.; DeVittre, R. R.; Leppard, G. G. Metal sorption to diagenetic iron and manganese oxyhydroxides and associated organic matter: Narrowing the gap between field and laboratory measurements. Geochim. Cosmochim. Acta 1996, 60, 387− 404. (9) Hohman, C.; Winkler, E.; Morin, G.; Kappler, A. Anaerobic Fe(II)-oxidizing bacteria show As resistance and co-precipitate As during Fe(III) mineral precipitation. Environ. Sci. Technol. 2010, 44, 94−101. (10) Ojeda, J. J.; Romeo-Ganzalez, M. E.; Banwart, S. A. Analysis of bacteria on steel surfaces using reflectance micro-Fourier transform infrared spectroscopy. Anal. Chem. 2009, 81, 6467−6473. (11) Chan, C. S.; Stasio, G. D.; Welch, S. A.; Girasole, M.; Frazer, B. H.; Nesterova, M. V.; Facra, S.; Banfield, J. F. Microbial polysaccharides template assembly of nanocrystal fibers. Science 2004, 303, 1656−1658. (12) Toner, B. M.; Fakra, S. C.; Manganini, S. J.; Santelli, C. M.; Marcus, M. A.; Moffett, J. W.; Rouxel, O.; German, C. R.; Edwards, K. Preservation of iron(II) by carbon-rich matrices in a hydrothermal plume. Nat. Geosci. 2009, 2, 197−201. (13) Miot, J.; Benzerara, K.; Morin, M.; Kappler, A.; Bernard, S.; Obst, M.; Ferard, F.; Skouri-Panet, F.; Guigner, J. M.; Posth, N.; Galvez, M.; Brown, G. E.; Guyot, F. Iron biomineralization by neutrophilic iron-oxidizing bacteria. Geochim. Cosmochim. Acta 2009, 73, 696−711. (14) Kikuchi, S.; Makita, H.; Mitsunobu, S.; Terada, Y.; Yamaguchi, N.; Takai, K.; Takahashi, Y. Application of synchrotoron μ-XRF-XAFS to the speciation of Fe on single stalk in bacteriogenic iron oxides (BIOS). Chem. Lett. 2011, 40, 680−681. (15) Tsuji, K.; Nakano, K.; Takahashi, Y.; Hayashi, K. X-ray spectrometry. Anal. Chem. 2011, 82, 4950−4987.
ASSOCIATED CONTENT
S Supporting Information *
Details of sampling and analysis and results as mentioned in the text. This material is available free of charge via the Internet at http://pubs.acs.org.
■
REFERENCES
AUTHOR INFORMATION
Corresponding Author
*E-mail:
[email protected]. 3310
dx.doi.org/10.1021/es203860m | Environ. Sci. Technol. 2012, 46, 3304−3311
Environmental Science & Technology
Article
(16) Manceau, A.; Nagy, K. L.; Marcus, M. A.; Lanson, M. G.; Nicolas, J. T.; Kirpichtchikova, T. Formation of metallic copper nanoparticles at the soil-root interface. Environ. Sci. Technol. 2008, 42, 1766−1772. (17) Mitsunobu, S.; Takahashi, Y.; Utsunomiya, S.; Marcus, M. A.; Terada, Y.; Iwamura, T.; Sakata, M. Identification and characterization of nanosized tripuhyite in soil near Sb mine tailings. Am. Mineral. 2011, 96, 1171−1181. (18) Toner, B. M.; Santelli, C. M.; Marcus, M. A.; Wirth, R.; Chan, C. S.; McCollom, T.; Bach, W.; Edwards, K. J. Biogenic iron oxyhydroxide formation at mid-ocean ridge hydrothermal vents: Juan de Fuca Ridge. Geochim. Cosmochim. Acta 2009, 73, 388−403. (19) Edwards, K. J.; Glazer, B. T.; Rouxel, O. J.; Bach, W.; Emerson, D.; Davis, R. E.; Toner, B. M.; Chan, C. S.; Tebo, B. M.; Staudigel, H.; Moyer, C. L. Ultra-diffuse hydrothermal venting supports Fe-oxidizing bacteria and massive umber deposition at 5000 m off Hawaii. ISME J. 2011, 5, 1748−1758. (20) Amann, R.; Ludwig, W.; Schleifer, L. H. Phylogenetic identification and in situ detection of individual microbial cells without cultivation. Microbiol. Rev. 1995, 59, 143−169. (21) Eickhorst, T.; Tippkötter, R. Improved detection of soil microorganisms using fluorescence in situ hybridization (FISH) and catalyzed reporter deposition (CARD-FISH). Soil Biol. Biochem. 2008, 40, 1883−1891. (22) Shiraishi, F.; Zippel, B.; Nue, T. R.; Arp, G. In situ detection of bacteria in calcified biofilms using FISH and CARD-FISH. J. Microbiol. Methods 2008, 75, 103−108. (23) Edwards, K. J.; McCollom, T. M.; Konishi, H.; Buseck, P. R. Seafloor bioalteration of sulfide minerals: Results from in situ incubation studies. Geochim. Cosmochim. Acta 2003, 67, 2843−2856. (24) Orcutt, B.; Boetius, A.; Elvert, M.; Samarkin, V.; Joye, S. B. Molecular biogeochemistry of sulfate reduction, methanogenesis and the anaerobic oxidation of methane at Gulf of Mexico cold seeps. Geochim. Cosmochim. Acta 2005, 69, 4267−4281. (25) Pernthaler, A.; Pernthaler, J.; Amann, R. Fluorescence in situ hybridization and catalyzed reporter deposition for the identification of marine bacteria. Appl. Environ. Microbiol. 2005, 68, 3094−3101. (26) Hallbeck, L.; Stahl, F.; Pederson, K. Phylogeny and phenotypic characterization of the stalk-forming and iron-oxidizing bacterium Gallionella ferruginea. J. Gen. Microbiol. 1993, 139, 1531−1535. (27) Emerson, D.; Rentz, J. A.; Liburn, T. G.; Davis, R. E.; Aldrich, H.; Chan, C.; Moyer, C. L. A novel lineage of proteobacteria involved in formation of marine Fe-oxidizing microbial mat communities. PLoS ONE 2007, 2, e667. (28) Chan, C. S.; Fakra, S. C.; Emerson, D.; Fleming, E. J.; Edwards, K. J. Lithotrophic iron-oxidizing bacteria produce organic stalks to control mineral growth: implication for biosignature formation. ISME J. 2010, 5, 717−727. (29) Manz, W.; Amann, R.; Ludwig, W.; Wagner, M.; Schleifer, K. H. Phylogenetic oligodeoxynucleotide probes for the major subclasses of proteobacteria-problems and solutions. Syst. Appl. Microbiol. 1992, 15, 593−600. (30) Hallberg, K. B.; Coupland, K.; Kimura, S.; Johnson, D. B. Macroscopic streamer growth in acidic, metal-rich mine waters in north Wales consist of novel and remarkably simple bacterial communities. Appl. Environ. Microbiol. 2006, 72, 2022−2030. (31) Franks, J.; Stolz, J. F. Flat laminated microbial mat communities. Earth Sci. Rev. 2009, 96, 163−172. (32) Hedrich, S.; Schlömann, M.; Johnson, D. B. The iron-oxidizing proteobacteria. Microbiology 2011, 157, 1551−1564. (33) Hanert, H. H. The genus Gallionella. The Prokaryotes, 2nd ed.; Springer-Verlag: New York, 1992; Vol. 4. (34) Emerson, D.; Revsbech, N. P. Investigation of an iron-oxidizing microbial mat community located near Aaruhus, Denmark: field studies. Appl. Environ. Microbiol. 1994, 60, 4022−4031. (35) Emerson, D.; Weiss, J. V. Bacterial iron oxidation in circumneutral freshwater habitats: Findings from the field and the laboratory. Appl. Environ. Microbiol. 2004, 21, 405−414.
(36) de Vet, W. W. J. M.; Dinkla, I. J. T.; Rietveld, L. C.; van Loosdrecht, M. C. M. Biological iron oxidation by Gallionella spp. in drinking water production under fully aerated conditions. Water Res. 2011, 45, 5389−5398. (37) Manceau, A.; Dritz, V. A. Local structure of ferrihydrite and feroxihyte by EXAFS spectroscopy. Clay Miner. 1993, 28, 165−184. (38) Chan, S. C.; Sirine, F. C.; Edwards, D. C.; Emerson, D.; Banfield, J. F. Iron oxyhydroxide mineralization on microbial extracellular polysaccharides. Geochim. Cosmochim. Acta 2009, 73, 3807−3818. (39) Vilge-Ritter, A.; Rose, J.; Masion, A.; Bottero, J. Y.; Laine, J. M. Chemistry and structure of aggregates formed with Fe salts and natural organic matter. Colloids Surf., A 1999, 147, 297−308. (40) Masion, A.; Doelsch, E.; Rose, J.; Moustier, S.; Bottero, J. Y.; Bertsch, P. M. Speciation and crystal chemistry of iron(III) chloride hydrolyzed in the presence of SiO4 ligands. 3. Semilocal scale structure of the aggregates. Langmuir 2001, 17, 4753−4757. (41) Voegelin, A.; Kaegi, R.; Frommer, J.; Vantelon, D.; Hug, S. J. Effect of phosphate, silicate, and Ca on Fe(III)-precipitates formed in aerated Fe(II)- and As(III)-containing water studied by X-ray absorption spectroscopy. Geochim. Cosmochim. Acta 2010, 74, 164− 186. (42) Mikutta, C. X-ray absorption spectroscopy study on the effect of hydroxybenzoic acids on the formation and structure of ferrihydrite. Geochim. Cosmochim. Acta 2011, 75, 5122−5139.
3311
dx.doi.org/10.1021/es203860m | Environ. Sci. Technol. 2012, 46, 3304−3311
