Biosignatures and Bacterial Diversity in Hydrothermal ... - CiteSeerX

Geomicrobiology Journal, 21:529–541, 2004 Copyright C Taylor & Francis Inc. ISSN: 0149-0451 print / 1362-3087 online DOI: 10.1080/01490450490888235

Biosignatures and Bacterial Diversity in Hydrothermal Deposits of Solfatara Crater, Italy Mihaela Glamoclija,1 Laurence Garrel,1 Jonathan Berthon,2 and Purificación López-Garc´ıa2 1

International Research School of Planetary Sciences, Università d’Annunzio, Viale Pindaro 42, 65127 Pescara, Italy 2 Unité d’Ecologie, Systématique and Evolution, CNRS UMR 8079, Université Paris-Sud, bâtimet 360, France

We have combined mineralogy, organic geochemistry and molecular microbiology to study hydrothermal deposits from Solfatara Crater, a geologically young volcanic formation (∼4,000 years old) displaying hot (45–95◦ C) and acidic (pH 1.7) mud pools and fumaroles. The search for inorganic (mineral) biosignatures revealed the presence of delicate structures, most likely mineralized extracellular polymers (EPSs), and the presence of potential biologically induced minerals: sulfides, sulfates (barite and alunite), elemental sulfur, and iron oxides. Geochemical analyses revealed a low total organic carbon content, 0.13 to 0.53%, displaying δ 13 C values from −17.09 to −27.39‰, and total nitrogen contents from 0.03 to 0.12%, which are characteristic of hydrothermal systems and suggest the presence of autotrophic carbon fixation. Lipid biomarker analysis showed the presence of hopanoids and linear alkanes, and the absence of detectable steroids, implying the occurrence of bacteria in our samples. We constructed 16S rRNA gene libraries from the environmental samples. Most environmental sequences obtained were affiliated to the Alphaand Betaproteobacteria (Hydrogenophilus-like), the Acidobacteria, and to a lesser extent, the Gammaproteobacteria and Actinobacteria. When known, the closest cultivated relatives were often thermophilic or thermotolerant bacteria oxidizing iron, hydrogen, or

Received 1 July 2004; accepted 9 September 2004. We thank A. Baliva for XRD analysis, A. Traini for the possibility of using SEM, and L. Tonucci for NMR analysis. The FT-IR analysis was done by G. de Matia, “Parco Scientifico e Tecnologico d’Abruzzo,” Chieti. Special thanks are given to Prof. R. Barbieri, University of Bologna, for his hospitality and useful advisory to M. Glamoclija during two months. We want to thank A. P. Rossi, IRSPS, for help during fieldwork, as well as D. Moreira for assistance during purification of samples for biological purposes. A special thanks to Vulcano Solfatara s.r.l. for the hospitality and interest in this research. This work was funded by the Agenzia Spaziale Italiana (ASI), the exobiology program, and the French CNRS-INSUE program Géomicrobiologie des environnements extrêmes (GEOMEX). Address correspondence to Mihaela Glamoclija, International Research School of Planetary Sciences, Università d’Annunzio, Viale Pindaro 42, 65127 Pescara, Italy. E-mail: [email protected]

methane/methanol, suggesting an important microbial contribution to the formation of biominerals. Keywords

16S rRNA, bacterial diversity, biosignatures, hydrothermal, Solfatara Crater, thermoacidophile

INTRODUCTION Since life appeared on Earth and for most of its history, microorganisms have been the lone inhabitants of our planet. Microbes can live in a wide variety on environments, including those exhibiting the most extreme conditions (Rothschild and Mancinelli 2001). Microbial genetic diversity is huge, as has been increasingly revealed by molecular ecology surveys over the past fifteen years (Pace 1997; Hugenholtz et al. 1998), yet prokaryotic microorganisms display a restricted number of morphotypes. This, together with the fact that microorganisms are rarely preserved in fossil form, has hampered the reconstruction and timing of early evolutionary diversifications. The search for diagnostic biosignatures from past microorganisms is not only crucial to understand early evolution on our planet, but might also help to reveal traces of ancient biological activity on planets such as Mars, where physical–chemical conditions were similar to those of the Archaean Earth. In addition to fossils discernible by their morphology, microorganisms and microbial communities influence and modify their environment during their lifetime, both at the micro and macroscale, and may thus leave traces of their existence. The most easily recognizable are stromatolites and permineralized biofilms. At a smaller scale, remnants of mineral-microbe interactions may remain but, unfortunately, surface chemistry can often yield a wide variety of mineral alterations that can be easily misinterpreted as derived from biological activities. The difficulty of finding unmistakable microbial biosignatures is evidenced by two recent controversies. The first concerns the nature 529

530

M. GLAMOCLIJA ET AL.

of carbon in Isua’s rocks, claimed initially to be of biological origin because of its light isotopic composition (Schidlowski 1988). This isotopic fractionation may also be the consequence of hot fluids reacting with older crustal rocks (metasomatism), however (van Zuilen et al. 2002). The second concerns the nature of the “earliest microfossils” (3.4–3.5 Ga) described by Schopf and coworkers in the Australian Warrawoona and Apex Formations as cyanobacteria (Schopf and Packer 1987; Schopf 1993), which have been later reinterpreted as possible artifacts (Brasier et al. 2002). The origin of usually microbially derived minerals, such as magnetite (Schuler and Frankel 1999), has also been put into question (Buseck et al. 2001; Treiman 2003; Weiss et al. 2004). Organic fossilized biomarkers, such as fossil lipids (e.g., hopanes, steranes) are certainly biogenic, but these are not exempt of contamination problems. To overcome all these difficulties, the identification of nonambiguous traces of ancient microbial activity will most likely demand the combination of several concurrent biosignatures. In this sense, the study of contemporary model system analogous to past environments is essential to understand the fossilization process and correlate present-day biosignatures with old putative biogenic traces. Various recent models on the origin of life propose that it originated some time between 4 and 3.7 Ga ago in moderately hot to hot environments, possibly linked to hydrothermal fluid activity (Kasting and Ackerman 1986; Wächtershäusser 1988; Nisbet and Sleep 2001; Martin and Russell 2003). Modern hydrothermal biotopes colonized by thermophilic and hyperthermophilic microorganisms constitute therefore potential model systems to identify biosignatures and link them to a particular microbial diversity and activity. Among the earliest studied contemporary

geothermal areas, together with Yellowstone in the U.S., is the Solfatara Crater close to Naples, Italy (Figure 1), where the thermoacidophilic archaea Sulfolobus solfataricus and S. brierleyi were first isolated (DeRosa et al. 1974, 1975; Zillig et al. 1980). Later, other hyperthermophilic archaea belonging to the genera Acidianus, Pyrobaculum and Metallosphaera have been isolated from Solfatara as well (Huber et al. 2000a, 2000b). The Solfatara Crater, characterized by its subareal activity, is located in the Mid–Eastern part of the Campi Phlegrei Caldera, which is a nested, 12-km wide structure, formed by two main collapses corresponding to Campanian Ignimbrite and Neapolitan Yellow Tuff eruptions (37,000 and 12,000 years ago, respectively) (Rosi and Sbrana 1987). Early studies revealed a major upheaval of the mantle beneath the area (Ferrucci et al. 1989). The top of the Campi Phlegrei Caldera magma chamber lies at 5 km depth, probably within carbonate sequences (Rosi and Sbrana 1987). Volcanological, petrological and geophysical data suggest that the activity at Campi Phlegrei Caldera was once fed by a large magmatic reservoir (Panichi and Volpi 1999). Solfatara’s volcanism belongs to the last volcanic epoch of the Campi Phlegrei Caldera. In past times, both explosive and effusive eruptions occurred within short-time intervals, but extensive explosive volcanism finished ∼4,000 years before present. Today, the Solfatara activity is marked by continuous hydrothermal emissions within mud pools in the central sector of the crater and fumarolic activity in the Northern side, possibly fed by a 1.5-km deep, low-permeability, geothermal aquifer of mixed magmatic-meteoric origin (Chioni et al. 1984). The geothermal system is vapor dominated, and temperatures range from very high (the hottest fumaroles ∼160◦ C; Tedesco et al. 1988) to moderate (40–50◦ C for areas around emission points). The pH of

Figure 1. Location of Solfatara Crater and sampling sites.

BACTERIAL BIOSIGNATURES AND DIVERSITY IN SOLFATARA CRATER

Solfatara fluids is ∼1.7, making this environment not only extreme by its temperature but also by its acidity. Despite the isolation of a few thermoacidophilic archaea from Solfatara hot mud, systematic studies on the prokaryotic diversity have not been carried out yet in this biotope. In this study, we have combined mineralogy, organic geochemistry and molecular ecology techniques to identify possible biosignatures and to characterize the bacterial community associated to Solfatara fumarole-adjacent crusts and mud samples. We have identified a number of minerals, including barite and alunite crystals as probable inorganic biosignatures, hopanoids as organic biomarkers, and a relative wide diversity of putative meso- to thermoacidophilic bacteria. MATERIALS Samples from the Solfatara Crater were collected on May 2001 from both types of emission points (fumaroles and mud pools). A total of 25 samples were selected to study their petrology, mineralogy and organic chemistry. Two of these were additionally used to characterize the associated microbial diversity. Samples for microbiological and biogeochemical analysis were collected following protocols to avoid contamination. Plastic gloves were worn all throughout the sampling procedure, and elementary precautions, such as avoiding any physical contact with samples and sealing the collected material immediately after removal, were taken. Surface crust samples were selected from intact well preserved areas. The rest of samples were collected after the removal of ∼10 cm of surface deposits aseptically. Samples were collected and stored in sterile 50-ml Falcon tubes and/or, in the case of larger volumes, in sterile Ziplock plastic bags. Collected samples were placed in a thermal-isolated container for their transport to the laboratory. Temperature was measured at each sampling point. pH values of water samples were first estimated in situ using pH strips (Carlo Erba), and subsequently determined in the laboratory using a pH-meter (GLP 22 Crison) equipped with a pH-electrode (n◦ 52-09 Crison). Calibration of the electrode has been done by Crison standard buffer solutions 7.0 and 9.0 pH respectively. Both buffer solutions and water samples that we measured were at the room temperature (25◦ C). METHODS Petrology and Mineralogy The identification of minerals in thin sections was done using a petrological microscope (Nikon E400). X-Ray diffraction (XRD; Siemens D 5005) analysis was performed on powdered samples of consolidated crusts collected around mud pools and fumaroles. The search for possible mineral biosignatures was first done by optical microscopy (Orthoplan Leitz Wetzlar) and, for potential interesting structures, by scanning electron microscopy (SEM; LEO 435VP). Energy-dispersive X-ray spectrometry (EDS) coupled to SEM (Philips XL-30; X-

531

EDS EDAX ECON IV, Microanalytical system 9900) was used to determine the elementary composition of identified minerals. Electron microscopy observations were done on thin sections prepared from intact samples and from samples that had been etched by 1% and 10% HCl, and 1% and 3% HF for different time periods. Thin sections for SEM and SEM/EDS were gold-coated previous to their observation. Organic Chemistry and Lipid Biomarker Analysis Analyses of total organic carbon (TOC), isotopic δ 13 C and total nitrogen content of Solfatara samples were done at the CNR (Bologna, Italy). Organic carbon and nitrogen were determined on duplicate samples using a FISONS NA2000 Elemental Analyzer (EA) after removal of the carbonate fraction by dissolution in 1.5N HCl. Stable isotope analyses of organic C were carried out by using a FINNIGAN Delta Plus mass spectrometer, which was directly coupled to the FISONS NA2000 EA by means of a CONFLO interface. The IAEA standards NBS19 were used as calibration materials for carbon. Stable isotopic data are expressed in the conventional delta (δ) notation in which the 13 C/12 C isotopic ratios are reported relative to the international PDB standard. The content of organic elements N, C, H, S and O in samples SOL7 and SOL18 used for microbial diversity analysis was done by the Service d’Analyse des Roches et Minéraux du CNRS (Nancy, France). Lipid biomarker analyses were done on samples corresponding to non-consolidated crust from a mud pool rim and consolidated crust. All samples were initially crushed to small pieces and carefully cleaned by repeated washing with 3M HCl and acetone, dilute HCl, methanol and dilute NaClO. After each step, samples were again carefully cleaned with organic-free UV-oxidized water dispensed by a Milli-Q Gradient instrument (Millipore). All glassware used was also treated by soaking in a 10% HCl bath for 24 h, rinsed thoroughly with water and dried at 200◦ C for 24 h. Fragmented samples were dried and then crushed into fine powder. Nuclear Magnetic Resonance (NMR) and Fourier-Transform Infrared (FT-IR) analyses were performed after demineralization of the samples using the method described by Gélinas et al. (2001). Solid state 13 C NMR spectroscopy was performed on demineralized fractions with a Brucker Avance 300 MHz spectrometer using power decoupling, cross polarization and magic angle spinning (CPMAS). Spectra were recorded at different spin rates in order to discriminate spinning side bands. FTIR spectra were recorded on a Perkin Elmer Spectrum 2000 equipped with an MIR source and a MIR-TDGS detector as 5 mm KBr pellets. For Gas Chromatography-Mass spectrometry (GC/MS) analyses, cleaned samples were saponified in 6% KOH in methanol, the supernatant was decanted and the residue subjected to Soxhlet and/or ultrasonically extracted with different dichloromethane/ methanol solutions. The combined supernatants were extracted with dichloromethane vs. water (pH 2). After concentration of the organic phase by rotatory evaporator under reduced pressure, sulfur was removed by freshly prepared activated copper.

532


The organic components of the CH2 Cl2 phase were fractionated by column chromatography (i.d. 15 mm, length 35 mm, Merck Silica gel 60, 70–230 mesh) and eluted with 2 column volumes of n-hexane (“total hydrocarbon”) and 3 column volume of CH2 Cl2 (alcohol/ketone fraction). The latter was derivatized with acetic acid anhydride in an equal volume of pyridine (14 h at room temperature) and fractions were analyzed by a Fisons Instruments MD 800 GC 8000 series GC/MS spectrometer equipped with a 50-m fused silica capillary column (DB5HT, 0.32 mm i.d. 0.25-µm film thickness) using He as carrier gas. Temperature program: 5 min 80◦ C to 310◦ C at 4◦ C/min: 20 min at 310◦ C. Nucleic Acid Extraction and 16S Ribosomal RNA Gene Libraries Samples (water from mud pool, hot mud, and nonconsolidated crusts) devoted to the biological study of microbial diversity by molecular methods were kept in 80% ethanol at 4◦ C until DNA extraction. The DNA used in this study was extracted from nonconsolidated black crust at the edge of a mud pool in the central part of the crater (SOL7, 40◦ C) and from white, melted deposits adjacent to a chimney in the Northeast fumarole area of the crater (85◦ C, SOL18) by two methods, the SoilMaster DNA extraction kit (Epicentre) and a classical phenol-chloroform extraction. For the latter, prior to DNA extraction, samples were rehydrated with phosphate saline buffer (130 mM NaCl, 10 mM phosphate buffer, pH 7.7, PBS). PBS was also added to the sediment to a same final volume of 0.5 mL. Samples were then subjected to 6 freezing/thawing cycles in liquid nitrogen to facilitate cell lysis. Subsequently, 80 µg mL−1 proteinase K, 1% SDS, 1.4 M NaCl, 0.2 β-mercaptoethanol and 2% hexadecyltrimethylammonium bromide (CTAB) (final concentrations) were added sequentially. Lysis suspensions were incubated overnight at 55◦ C. Lysates were extracted once with hot phenol (65◦ C), once with phenolchloroform-isoamylalcohol, and once with chloroform-isoamylalcohol. Nucleic acids were concentrated by ethanol precipitation. 16S rRNA genes were amplified by PCR using the bacterial-specific primer 63F (CAGGCCTAACACATGCAAGTC) and the prokaryote-specific reverse primer 1397R (GGGCGGWGTGTACAAGGC). Polymerase chain reactions (PCR) were performed under the following conditions: 30 cycles (denaturation at 94◦ C for 15 s, annealing at 50◦ C for 30 s, extension at 72◦ C for 2 min) preceded by 2 min denaturation at 94◦ C, and followed by 7 min extension at 72◦ C. Dimethyl sulfoxide was added to a final concentration of 3–5% to the PCR reaction mix. rDNA clone libraries were constructed using the Topo TA Cloning system (Invitrogen) following the instructions provided by the manufacturers. After plating, positive transformants were screened by PCR amplification of inserts using flanking vector primers. Sequence and Phylogenetic Analyses A total of 63 (34 from SOL18 and 29 from SOL7) expectedsize amplicons from these libraries was partially sequenced

(lengths from 849 to 1050 nucleotides) with the primer 1387R (Genome Express). Closest relatives to our sequences were identified in databases by BLAST (Altschul et al. 1997) and retrieved from GenBank http://ncbi.nlm.nih.gov/. Sequences were automatically aligned using the program BABA (H. Philippe, personal communication) to a 16S rRNA gene alignment containing ∼17,000 sequences. The multiple alignment was then manually edited using the program ED from the MUST package (Philippe 1993). A preliminary phylogenetic analysis of all partial sequences was done by distance methods (neighbor-joining, NJ) using the program MUST, allowing the identification of identical or nearly identical sequences and the selection of representative clones for subsequent analysis. We then selected 15 representative sequences to be included in a phylogenetic tree, together with their closest relatives in GenBank and some cultivated species. A total of 710 positions were used in our analysis after removal of gaps and ambiguously aligned positions. The maximum likelihood (ML) tree was done using TREEFINDER (Jobb 2002) applying a general time reversible model of sequence evolution (GTR), taking among-site rate variation into account by using an eight-category discrete approximation of a gistribution (invariable sites are included in one of the categories). The α parameter of the distribution estimated from the sequence set was 0.27. ML bootstrap proportions were inferred using 500 replicates. The sequences reported in this study were submitted to GenBank with accession numbers AY629323 to AY629339 (see also Figure 6).

RESULTS AND DISCUSSION Petrological Characteristics of Solfatara Samples The studied material from the Solfatara Crater is geologically young (∼4,000 years) which, in principle, should facilitate the identification of unaltered biosignatures left by recent and present-day microbial communities. Hydrothermal springs in the center of the crater have their foundation on local faults that run radially through the crater, whereas the fumarolic activity is mainly dependent on the regional fault along the Northeast crater wall (Figure 2A, 2B). Products of Solfatara activity consisted mainly of breccia and stratified deposits with layers of pisolitic and coarse ashes, the basal surge structure being composed of beds of well-sorted pumice lapilli that were a few decimeters in size. The surges covered the East-Northeast wall of the crater and overflowed the Eastern and Western crater edges. Most volcanic products in the area were hydrothermally altered, giving raise to disordered trachytic deposits. Samples collected for this study included highly porphyritic scoria with trachytic appearance composed of sanidine, plagioclase, clinopyroxene, biotite, and opaque phenocrystals. Samples of emitted material around mud pools were mainly fine-grained clays, consolidated to some extent. All samples were very rich in sulfur, sulfates, and a variety of iron minerals. In the Northeastern zone of the crater, where fumarolic activity is intense, gypsum was also present. From the petrological and mineralogical characteristics, the observed

533


Figure 2. Sampling sites inside Solfatara Crater: (A) mud pool with boiling water in the central part of the crater; (B) one of the associated fumaroles chimney, close to the North-Eastern wall of the crater. and sampled material represent an environment with reducing chemistry.

Inorganic (Mineral) Biosignatures In many extreme environments, such as hot and acid biotopes, microbial communities derive their energy from oxidationreduction reactions made possible by the coexistence of adequate electron donors and acceptors along a geochemical gradient. Different types of surfaces and cracks in mineral deposits where electron donors and acceptors co-exist are suitable environments for microbial colonization. Biological redox reactions can lead to the direct or indirect formation of “bio-minerals,” which are potential indicators of microbial metabolic activity (Banfield et al. 2001). Among known minerals that can have a biologically induced origin, elemental sulfur, sulfides, sulfates, and iron oxides/hydroxides are highly abundant in Solfatara. Aiming at establishing the biogenic nature of some minerals found in small cracks within Solfatara samples, we looked for the co-occurrence of bio-alteration signs in different samples. Two types of crack samples in Solfatara material were found to be biologically-altered and associated to likely biologically induced mineral precipitation. A first type of fractures (Type 1) was observed within partially consolidated stratified deposits of pisolitic ashes from the Northeastern sector of the crater close to fumarola emissions (Figure 2B). Most ground minerals in this area displayed different levels of alteration. In particular, these deposits contained ∼1 mm-wide fractures, which were traversed by fluids rich in H2 S, CO2 , and, to a lesser extent, HCl, CH4 , and H2 , while they were still unconsolidated (Valentino et al. 1999). A narrow, reddish altered zone was observed under the microscope on one of the fracture walls in the sample shown in Figure 3B. Since the appearance of this alteration could be suggestive of biological activity, rather than thermal alteration induced by hot fluids, we used electron microscopy to detect any further indication of biogenic origin. A closer look revealed that the altered zone exhibited irregular masses of elemental sulfur (S0 ) accompa-

nied by a layer of pyrite crystals and, in lower quantity, small barite crystals (Figure 3B, C, D). S0 is unstable in nature and usually gets readily oxidized. It can precipitate around rims of fumaroles in orthorhombic and monoclinic form. However, the characteristic widespread sulfur interlayering generally associated to abiotic crystallization, which is otherwise observed on fumarola edges, was not seen in this case (Figure 3A and B). One possible explanation for the presence of S0 along this fracture is, therefore, the occurrence of sulfur-metabolizing bacteria (sulfide-oxidizers or sulfur disproportionating bacteria) whose activity may at some point lead to the accumulation of locally important amounts of S0 . 0 2− SO2− 4 ⇐ S ⇒ S sulfur-disproportionating bacteria [1] 0 2− SO2− 4 ⇐ S ⇐ S sulfur-oxidizing bacteria and phototrophic sulfur bacteria [2]

Pyrite (FeS2 ) was also present along the altered area, but just on one of the fracture walls (Figure 3B). When sulfide oxidation occurs in the presence of adsorbed oxidized metal ions, there is a concomitant reduction and precipitation of these metal species. Several chemolithotrophic and mixotrophic acidophiles use elemental sulfur and reduced sulfur components as electron donors, and ferric iron (Fe3+ ), as electron acceptor, to support growth (Johnson 1998). This may lead to pyrite formation as follows: 2+ S0 + 6Fe3+ + 4H2 O ⇒ HSO− + 7H+ 4 + 6Fe

Fe2+ + S2−

⇒ FeS (soluble)

FeS (soluble) + S2 O2− 3

⇒ FeS2 + SO2− 3

Since pyrite was only observed on one of the fracture walls, processes involving sulfide oxidation by microorganisms differentially attached to one side could have influenced this asymmetric distribution of pyrite. Small (a few microns in size) barite (BaSO4 ) grains were also observed associated to this same crack (Figure 3C and D). Barite

534


Figure 3. Layered sulfur- and iron-rich crust sample collected near the associated fumarol chimney (Figure 2B). (A) Overall view of the layered sulfur-iron crust. (B) Micrograph of a thin section showing a ∼1-mm wide crack with dark, iron-rich bio-alteration (arrows). “S” indicates the position of elemental sulfur precipitates and “Py” pyrite crystals. (C) Scanning electron micrograph from bio-altered zone showing barite crystals (ba) and mineralized extracellular polymers (EPS). (D) Scanning electron micrograph of irregular elemental sulfur (S) and barite (ba) precipitates in the same area. The inset between panels C and D corresponds to EDS spectra of sulfur and barite obtained from this samples. Scale bars: 100 µm (B), 1 µm (C), and 10 µm (D).

frequently results from hydrothermal alteration by substitution of calcium by barium in gypsum (CaSO4 ), and precipitates at low pH (∼1.2) in the presence of large amounts of sulfates (Africano and Bernard 2000). Though the precipitation of barium and sulfates is clearly an abiotic reaction, sulfide- and sulfur-oxidizers may have contributed to increase the local sulfate concentration. Microbial production of sulfate is known to occur in Solfatara, as different archaeal (Sulfolobales) species able to oxidize elemental sulfur (Segerer and Stetter 1999) have been isolated from Solfatara mud pools, and it is also suggested by isotopic S fractionation (Valentino et al. 1999). Since barite minerals appeared to co-occur with extracellular polymer-like structures (see later), a direct induction of BaSO4 crystallization might have occurred in association with microbial cells. The precipitation of barite indirectly produced by some bacterial species has been documented (González-Muñoz et al. 2003). In addition to these minerals, delicate reddish net-like structures were also observed in this type of fractures. In most cases, they were associated with irregular sulfur masses and barite minerals (Figure 3C, D). The reddish rust color is due to the presence of iron oxy-hydroxide compounds which may replace microbial

soft parts during fossilization (Butterfield et al. 1996; Provencio and Polyak 2001). The location of these delicate iron-rich structures in the fracture, together with their morphology (irregular net appearance) and morphometry (size in the nanometer range), suggests that they correspond to permineralized extracellular polymers (EPS). Many microorganisms synthesize EPSs to attach to the substrate and to other biofilm-forming microbes. Because of their negative charge, EPSs are among the first macromolecules to permineralize (Butterfield et al. 1996; Barker et al. 1998; Gehrke et al. 1998; Karthikeyan and Beveridge 2002; Hockin and Gadd 2003). The presence of net-like structure reminiscent of EPS is one of the (morphological) criteria employed to identify ancient microfossils (Cady et al. 2003). A second type of fractures (Type 2) displayed alunite minerals that were always covering one of the walls (Figure 4). Opaque pyrite minerals usually followed the alunite distribution. This type of mineralization was observed in samples of consolidated crusts collected both near the hydrothermal springs and fumaroles. Alunite is a hydrous potassium aluminosulfate, from the jarosite mineral group, forming irregular masses and euhedral crystals. In Type 2 cracks (Figure 4), pyrite constituted


535

Figure 4. Thin section taken under optical microscope showing a fracture inside a crust sample collected close to Solfatara mud pool. The fracture contains alunite minerals (indicated by arrows) that have precipitated on the upper edge along with secondary pyrite “Py.” X-ray spectrometry (EDS) spectrum of alunite is at the bottom. a secondary mineral phase, i.e., it was formed after precipitation of alunite. In contrast to Type 1 cracks, morphologically identifiable microbial structures were not observed. Alunite forms by sulfuric acid solutions acting on deposits (rocks) containing potassium feldspar at acid pH (∼1.3) in deep subsurface hydrothermal anoxic environments. The formation of alunite decreases the concentration of potassium (K+ ) and aluminum (Al3+ ) cations in solution. The concomitant rise in pH is compensated by the production of H+ during the reaction. + K+ + 3Al3+ + 2HSO− 4 + 6H2 O ⇔ KAl3 (SO4 )2 (OH)6 + 8H .

This has been also documented in other settings displaying hydrothermal alteration (Bove and Hon 1990). Although alunite is abiotically formed, sulfide oxidizers might have contributed to increase local concentrations of sulfuric acid (H2 SO4 ), thus triggering the formation of this mineral. Indeed, the latter is supported by the fact that alunite and pyrite were specifically formed on the same wall in this crack.

Finally, in addition to the above potential biominerals, high quantities of magnetite were locally observed on several places within samples. Even though the presence of magnetite is often correlated with biological activity, the observed minerals were too large (∼4 µm in size) to constitute intracellular products, and magnetite appeared to be the product of biotite oxidation. Indeed, the magnetite observed was associated to oxybiotite minerals. Oxybiotite is actually a melted biotite that has released its iron content (necessary for magnetite formation) while passing through processes of oxidation. Titanium, often found as an impurity in hydrothermal fluids, was detected in oxybiotite and magnetite by EDS analysis. The precipitation of magnetite requires circumneutral pH settings (>5.5) and sufficiently oxidizing conditions (Bell et al. 1987; Tor et al. 2001). The association of magnetite with oxybiotite suggests that, locally, the redox potential may have been relatively high. Although pH values are generally very low at Solfatara (∼1.7 on average), circumneutral pH could have been maintained at microniches by the activity of particular bacteria (Dong et al. 2000), which would

536


eventually contribute indirectly to magnetite formation by creating the appropriate environmental conditions for the abiotic process to occur.

Organic Geochemistry and Biosignatures Total organic carbon (TOC) was present at very low concentrations, 0.13 to 0.53%, in the different Solfatara samples. δ 13 C values in those samples were also low, −17.09 to −27.39 ‰, which would be in agreement with chemoautotrophic organic synthesis. Total N varied from 0.04 to 0.12%, implying a C/N ratio between 1.63 and 7, which would also be compatible with microbial organic production (Des Marais et al. 1992; Des Marais 1996; Strauss et al. 1992; Meyers 1997; Schidlowski 2000; Twichell et al. 2002; House et al. 2003). Kerogen, the naturally occurring insoluble organic matter in rocks, may retain primary biogeochemical information in the form of lipid biomarker compounds (Ourisson et al. 1987; Summons et al. 1996). We looked for the presence of lipid biomarkers in two Solfatara samples: black crust from the rim

of the central mud pool (corresponding to SOL7, see later), and gray endured crust near the fumarola chimney where type 1 cracks were characterized. Nondestructive spectroscopic methods like NMR and IR spectroscopy constitute suitable techniques for the examination of organic matter in heterogeneous macromolecular mixtures and yield good results concerning the gross chemical composition. The CPMAS 13 C NMR (Figure 5A) spectra of Solfatara samples (after demineralization) exhibited peaks at the same chemical shift. They were dominated by an intense peak due to aliphatic carbon that reached its maximum at 30 ppm (carbon from polymethylenic chains CH2 ). This peak showed shoulders at 15 and 35 ppm due to methyl groups and substituted carbons, respectively. The second characteristic signal (110–160) with a maximum at 130 ppm was due to unsaturated carbon, which can be, a priori, in olefinic or aromatic units. However, the maximum at 130 ppm suggests that the major part of unsaturated carbon corresponds to alkenes. In any case, solid state (CPMAS) 13 C-NMR can highly overestimate the aliphacity of heterogeneous compounds materials, due to a more efficient transfer of

Figure 5. Analysis of organic matter from Solfatara samples: (A) Solid state CPMAS 13 C-NMR spectrum (after demineralization). (B) FTIR spectrum (after demineralization). (C) Distribution of products identified by GC/MS (after saponification).


polarization to protonated aliphatic carbon than to aromatic carbons, or carbon involved in highly cross-linked structures. The FT-IR spectra (Figure 5B) of the samples studied (after demineralization) showed similar functions. According to NMR, the presence of CH2 and CH3 is confirmed by their stretching bands at respectively 2,920 and 2,850 cm−1 as well as their asymmetric C H bending bands at 1,455 and 1,375 cm−1 . Absorption of olefinic and aromatic carbon C C stretching vibration fell in the range 1,570 to 1,680 cm−1 . However, the bands centered at 1,628 and 1,637 cm−1 (C C nonconjugated), and the weak broad shoulder detected in the 3,000–3,100 cm−1 range (aromatic C H stretching centered vibration), suggest that alkenes are more abundant than aromatics, in accordance with the NMR data. Oxygen containing functions were detected as a broad band of great intensity from 3,400 to 3,600 cm−1 (O H). C O stretching vibration are almost absent (band at 1,707 cm−1 is very weak). The presence of C O is confirmed by intense bands from 1,050 to 1,150 cm−1 . Thus, the organic matter in Solfatara samples is mainly composed of: •

Saturated carbon, mostly formed by secondary (R-CH2R ) and, in less quantity, tertiary carbon (R-CH-RR ) with a few methyl groups (15 ppm) and few quaternary carbons. This suggests the occurrence of straight-chain skeletons with few branched cyclic alkanes skeletons. • Olefins, constituting the majority of unsaturated carbon. • Oxygen-containing functions, mostly alcohols and ethers. In accordance with these results, GC-MS analysis of the alkane-alkene soluble fraction showed a major contribution of linear alkanes (m/z = 57), hopenes C27 H43 R (m/z = 191) along with few branched alkanes (m/z = 43, 57), linear alkenes and linear alkanols (Figure 5C). The wide range of alkane carbon numbers and their distribution (even/odd number predominance) excludes a terrestrial plant wax origin. Hopanoid lipids are particularly important biomarkers for bacteria (De Las Heras et al. 1997; Tritz et al. 1999; Farrimond et al. 2000). They derive from the diagenesis of bacterial cell membranes and are extremely stable over geological time. Their presence attests for the presence of contemporary and/or past bacterial cells. The occurrence of essentially two types of lipids, linear alkanes and especially hopanoids, as well as the absence of steroids, suggests that the microbial community is essentially composed of bacteria. Bacterial Diversity in Hydrothermal Deposits A first inspection under the optical microscope of different Solfatara samples showed the presence of various prokaryotic morphotypes (cocci and short rods), already suggesting a certain microbial diversity. Therefore, in addition to the search for biosignatures, we carried out molecular diversity surveys in two Solfatara samples. One corresponded to the same nonconsolidated black crust collected near the central mud pool used

537

for lipid biomarker analyses (SOL7, 40◦ C), and the second to melted deposits at a chimney in the Northeast Solfatara fumarole area, adjacent to the samples in which type 1 cracks were found (SOL18, 85◦ C). The concentrations of organic elements in this two samples showed the following values for SOL7 and SOL18, respectively: 0.1 and 0.02 (N%), 0.34 and 0.33 (C%), 0.93 and 0.32 (H%), 17.38 and 44.50 (S%), and 18.34 and 4.37 (O%). There was a similar weak C content suggesting that biomass was low. Despite various attempts using different PCR conditions and primer pairs specific for archaea, we always failed to amplify 16S rRNA genes for this prokaryotic domain. However, thermoacidophilic archaea have been isolated from central Solfatara boiling mud pools (DeRosa et al. 1975; Huber et al. 2000a, 2000b; Zillig et al. 1980). Biases due to DNA extraction or primer utilization could explain this apparent discrepancy, although DNA was extracted by various methods and different PCR conditions were used to minimize possible biases. Therefore, this result may actually reflect the absence or a very low density of archaea in these particular samples. This explanation would be in agreement with lipid biomarker analysis, which shows a clear dominance of bacterial hopanoids. Bacterial 16S rRNA genes were successfully amplified from the two Solfatara samples, and 63 clone inserts were sequenced from the generated 16S rDNA libraries. The diversity in terms of major bacterial taxa was relatively low (Figure 6 and Table 1). In the case of SOL7, phylotypes identified belonged only to the alpha and beta subdivisions of the Proteobacteria, and to the Acidobacteria, whereas in SOL18 phylotypes affiliated to the Gammaproteobacteria and the Actinobacteria were found as well (Figure 6). Some phylotypes displayed sequences nearly 100% identical to those of known thermophilic or thermotolerant cultivated species, such as Hydrogenophilus thermoluteolus, a facultative hydrogen-oxidizer growing optimally at 50–52◦ C

Figure 6. Taxonomic distribution of bacterial phylotypes in 16S rDNA libraries corresponding to soft deposits adjacent to a chimney (fumarolic Northeast area), and to iron-sulfur-rich crust adjacent to a mud pool in the center of the Solfatara crater.

538


Table 1 Phylogenetic affiliation of representative bacterial 16S rDNA clones obtained from Solfatara crust and mud samples as deduced from BLAST searches

Phylotype

Sequence length (bp)

Number of similar sequences∗

Phylogenetic ascription

SOL18-11

997

11

Acidobacteria

SOL18-20

901

2

Acidobacteria

SOL7-39

849

3

Acidobacteria

SOL18-50

944

5

Acidobacteria

SOL18-51 SOL7-22 SOL7-1

1,001 1,025 1,050

1 5 13

Actinobacteria Alphaproteobacteria Alphaproteobacteria

SOL18-52

1,001

2

Alphaproteobacteria

SOL18-16

1,000

5

Alphaproteobacteria

SOL7-40 SOL7-4 SOL18-39 SOL18-21 SOL18-37 SOL18-38

993 1,030 1,025 985 984 1,018

1 12∗∗

∗

1 1 1

Betaproteobacteria Betaproteobacteria Betaproteobacteria Gammaproteobacteria Gammaproteobacteria Gammaproteobacteria

Closest relative in database, environmental origin (GenBank accession number) Uncultured bacterium WD257, polychlorinated biphenyl-polluted soil (AJ292583) Uncultured bacterium WD228, polychlorinated biphenyl-polluted soil (AJ292578) Uncultured bacterium WD277, polychlorinated biphenyl-polluted soil (AJ292587) Uncultured bacterium WD277, polychlorinated biphenyl-polluted soil (AJ292587) Agrococcus jenensis (X92492) Paracoccus marcusii (Y12703) Iron-oxidizing acidophilic methylotrophic isolate Y005, Yellowstone (AY140237) Iron-oxidizing acidophilic methylotrophic isolate Y005, Yellowstone (AY140237) Methylosinus sp. NCIMB 13214, methane-utilizing bacterium from a bacterial consortium that rapidly degrades trichloroethylene (AB007840) Delftia acidovorans (AY367028) Hydrogenophilus thermoluteolus (AB009828) Hydrogenophilus thermoluteolus (AB009828) Acinetobacter sp. (Z93446) Pseudomonas thermotolerans (AJ311980) Pseudomonas thermotolerans (AJ311980)

% identity (BLAST) 94

98

98

97

99 99 92 92 95

99 99 98 96 98 99

Number of sequences >98% identity. ∗∗ 12 corresponds to the total number of Hydrogenophilus-like sequences.

(Hayashi et al. 1999) or Pseudomonas thermotolerans, growing optimally at 47◦ C (max 55◦ C) (Manaia and Moore 2002). Indeed, most betaproteobacterial clones from both SOL samples affiliated to the genus Hydrogenophilus (Table 1 and Figure 7). Together betaproteobacteria, alphaproteobacteria and acidobacteria were the most represented in our libraries. Interestingly, within the alphaproteobacteria, a majority of clones were most closely related to the iron-oxidizing strain Y005, which was isolated from acidic geothermal areas in Yellowstone National Park. Y005 is related to methylotrophic species and, indeed, it can also be grown in ferrous iron/methanol medium (Johnson et al. 2003). Finally, members of the Acidobacteria were profuse in both SOL7 and 18 libraries (Figure 6). The Acidobacteria/Holophaga represents a broad bacterial division occupying various environments from activated sludge to marine sediments where most

members remain uncultivated so far (Ludwig et al. 1997). Members of this division have been identified in high proportion in shallow submarine vents near Milos (Greece) (Sievert et al. 2000) and around deep-sea vents (López-Garc´ıa et al. 2003). Remarkably, members of the Acidobacteria were also found to be very abundant in soils following a geothermal heating event at Yellowstone. Their amount increased in soils incubated at 50◦ C indicating that various members of this group are thermophilic (Norris et al. 2002). The Solfatara acidobacterial sequences are relatively varied, but they are all more closely related to the genus Acidobacterium and quite distant from Holophaga. The few available cultivated members of this group are heterotrophs, and this has also been suggested for lineages found in shallow thermal areas. However, not only the Solfatara samples are very poor in organic content, but it is very difficult to predict the physiology of the Solfatara lineages on the basis of sequence


539

Figure 7. Maximum likelihood (ML) phylogenetic tree showing the position of representative bacterial 16S rDNA sequences from iron-sulfur crust close to a mud pool (SOL7) and melted deposits in the Northern fumarolic area (SOL18) samples. Only ML bootstrap values above 50% are shown at nodes. The scale bar corresponds to 10 substitutions for a unit branch length. CFB, Cytophaga-Flexibacter-Bacteroides group.

comparison, since they are quite distant from the few currently isolated species (Figure 7). Furthermore, the lack of more isolated strains from this broad division may imply that current culture media are not suitable for their growth and that these organisms may display novel metabolic strategies. Since many of these lineages are identified in metal-rich, acidic thermal areas, they may possibly rely on some type of strict or chemolithoautotrophic metabolism.

CONCLUDING REMARKS We have applied a multidisciplinary analysis to the study of hydrothermal deposits from the Solfatara crater, involving geology, micropaleontology, organic chemistry and molecular microbiology methods. The hydrothermal deposits were young (∼4,000 years old), mildly hot to hot (40–95◦ C) and acidic (pH ∼1.7). Chemical conditions were predominantly anoxic and reducing, although oxic/anoxic transition zones existed in

540


surface areas, and rock fractures. These features could favor the development of chemolithotrophic microbes by allowing access to different electron and acceptor donors along physicalchemical gradients. Potential biomineral signatures were identified in altered fractures within sulfur-rich Solfatara deposits, along with lipid biomarkers. These included particles of elemental sulfur, pyrite and barite crystals, which were sometimes associated to permineralized EPS net-like structures. Though barite minerals are usually affiliated with marine or with deep underground hydrothermal environments, here they seem to have formed in the surface settings. The organic C content of Solfatara deposits was low, yet lipid biomarkers including bacterial hopanoids were identified. Finally, a molecular diversity survey carried out on melted deposits in a fumarolic area and on iron-rich crust, for which lipid and mineral signatures were detected, allowed the detection of different bacterial phylotypes. The diversity was relatively low, and the phylotypes detected affiliated to known thermophilic or thermotolerant bacteria, some of which are iron- and hydrogenoxidizers, or methane/methanol consumers. This is in agreement with the substrate’s nature and with the content of fumarolic emissions. Nevertheless, many phylotypes, such as the acidobacterial lineages detected, are too distant from cultivated species to safely conclude anything about their metabolic capabilities. Given the nature of the habitat and the low content in organics, some of these could correspond to chemolithotrophic bacteria obtaining energy from redox reactions involving species of iron or sulfur, similarly to the sulfur-oxidizing archaea that have been isolated from the hottest Solfatara mud pools. The mineral and biochemical signatures observed could therefore attest for the activity of past (recent) microbial communities, which were likely very similar to those observed today, as well as contemporary microorganisms. A better understanding of the genesis of these potential biosignatures from the autochthonous microbial communities will be further needed to interpret most of ancient signals from similar environments. REFERENCES Africano F, Bernard A. 2000. Acid alteration in the fumarolic environment of Usu volcano, Hokkaido, Japan. J Volcan Geoth Res 97:475–495. Altschul SF, Madden TL, Schaffer AA, Zhang J, Zhang Z, Miller W, Lipman DJ. 1997. Gapped BLAST and PSI-BLAST: a new generation of protein database search programs. Nucleic Acids Res 25:3389–3402. Banfield JF, Moreau JW, Chan CS, Welch SA, Little B. 2001. Mineralogical biosignatures and the search for life on Mars. Astrobiology 1:447–465. Barker WW, Welch SA, Chu S, Banfield JF. 1998. Experimental observations of the effects of bacteria on aluminosilicate weathering. Amer Mineral 83:1551– 1563. Bell PE, Mills AL, Herman JS. 1987. Biogeochemical conditions favoring magnetite formation during anaerobic iron reduction. Appl Environ Microbiol 53:2610–2616. Bove DJ, Hon K. 1990. Compositional changes induced by hydrothermal alteration at the red mountain alunite deposit, Lake City, CO. US Geol Surv Bull 1936:1–21. Brasier MD, Green OR, Jephcoat AP, Kleppe AK, Van Kranendonk MJ, Lindsay JF, Steele A, Grassineau NV. 2002. Questioning the evidence for Earth’s oldest fossils. Nature 416:76–81.

Buseck PR, Dunin-Borkowski RE, Devouard B, Frankel RB, McCartney MR, Midgley PA, Posfai M, Weyland M. 2001. Magnetite morphology and life on Mars. Proc Natl Acad Sci USA 98:13490–13495. Butterfield NJ, Knoll AH, Swett K. 1996. Paleobiology of the Neoproterozoic Svanbergfjellet formation, Spitsbergen. Fossils Strata 34:3–83. Cady SL, Farmer JD, Grotzinger JP, Schopf JW, Steele A. 2003. Morphological biosignatures and the search for life on Mars. Astrobiology 3:351–368. Chioni R, Corazza E, Marini L. 1984. The gas/steam ratio as indicator of heat transfer at the Solfatara fumarole, Phlegrean Fields (Italy). Bull Volcanol 47:295–302. De Las Heras FXC, Grimalt JO, Lopez JF, Albaiges J, Damste JSS, Schouten S, De Leeuw JW. 1997. Free and sulphurized hopanoids and highly branched isoprenoids in immature lacustrine oil shales. Org Geochem 27:41–63. DeRosa M, Gambacorta A, Bu’Lock JD. 1975. Extremely thermophilic acidophilic bacteria convergent with Sulfolobus acidocaldarius. J Gen Microbiol 86:156–164. DeRosa M, Gambacorta A, Millonig G, Bu’Lock JD. 1974. Convergent characters of extremely thermophilic acidophilic bacteria. Experientia 30:866– 868. Des Marais DJ. 1996. Stable light isotope biogeochemistry of hydrothermal systems. In: Bock GR, Goode JA, editors. Evolution of Hydrothermal Ecosystems on Earth (and Mars?), New York: John Wiley and Sons. p. 83–98. Des Marais DJ, Bauld J, Palmisano A, Summons RE, Ward DM. 1992. The biogenicity of carbon in modern microbial mats. In Schopf JW, Klein C, editors. The Proterozoic Biosphere, A Multidisciplinary Study. NY: Cambridge University Press. P. 299–308. Dong H, Fredrickson JK, Kennedy DW, Zachara JM, Kukkadapu RK, Onstott TC. 2000. Mineral transformation associated with the microbial reduction of magnetite. Chem Geol 169:299–318. Farrimond P, Head IM, Innes HE. 2000. Environmental influence on the biohopanoid composition of recent sediments. Geochim Cosmochim Acta 64:2985–2992. Ferrucci F, Gaudiosi G, Pino NA, Luongo G, Hirn A, Mirabile L. 1989. Seismic detection of a major upheaval beneath the Campanian volcanic area, Naples, Southern Italy. Geophys Res Lett 16:1317–1320. Gehrke T, Telegdi J, Thierry D, Sand W. 1998. Importance of extracellular polymeric substances from Thiobacillus ferrooxidans for bioleaching. Appl Environ Microbiol 64:2743–2747. Gélinas Y, Baldock JA, Hedges J. 2001. Demineralization of marine and freshwater sediments for CP/MAS 13 C NMR analysis. Org Chem 32:677–693. González-Muñoz MT, Fernández-Luque B, Mart´ınez-Ruiz F, Chekroun KB, Arias JM, Rodr´ıguez-Gallego M, Mart´ınez-Cañamero M, De Linares C, Paytan A. 2003. Precipitation of barite by Myxococcus xanthus: possible implications for the biogeochemical cycle of barium. Appl Environ Microbiol 69:5722–5725. Hayashi NR, Ishida T, Yokota A, Kodama T, Igarashi Y. 1999. Hydrogenophilus thermoluteolus gen. nov., sp. nov., a thermophilic, facultatively chemolithoautotrophic, hydrogen-oxidizing bacterium. Int J Syst Bacteriol 49:783–786. Hockin SL, Gadd GM. 2003. Linked redox precipitation of sulfur and selenium under anaerobic conditions by sulfate-reducing bacterial biofilms. Appl Environ Microbiol 69:7063–7072. House CH, Schopf WJ, Stetter KO. 2003. Carbon isotopic fraction by Archaeans and other thermophilic prokaryotes. Org Geochem 34:345–356. Huber R, Huber H, Stetter KO. 2000a. Towards the ecology of hyperthermophiles: biotopes, new isolation strategies and novel metabolic properties. FEMS Microbiol Rev 24:615–623. Huber R, Sacher M, Vollmann A, Huber H, Rose D. 2000b. Respiration of arsenate and selenate by hyperthermophilic archaea. Syst Appl Microbiol 23:305–314. Hugenholtz P, Pitulle C, Hershberger KL, Pace NR. 1998. Novel division level bacterial diversity in a Yellowstone hot spring. J Bacteriol 180:366–376. Jobb G. (2002). TREEFINDER, Distributed by the author at www.treefinder. de. Johnson DB. 1998. Biodiversity and ecology of acidophilic microorganisms. FEMS Microbiol Ecol 27:307–317.

BACTERIAL BIOSIGNATURES AND DIVERSITY IN SOLFATARA CRATER Johnson DB, Okibe N, Roberto FF. 2003. Novel thermo-acidophilic bacteria isolated from geothermal sites in Yellowstone National Park: physiological and phylogenetic characteristics. Arch Microbiol 180:60–68. Karthikeyan S, Beveridge TJ. 2002. Pseudomonas aeruginosa biofilms react with and precipitate toxic soluble gold. Environ Microbiol 4:667–675. Kasting JF, Ackerman TP. 1986. Climatic consequences of very high carbon dioxide levels in the Earth’s early atmosphere. Science 234:1383– 1385. López-Garc´ıa P, Duperron S, Philippot P, Foriel J, Susini J, Moreira D. 2003. Bacterial diversity in hydrothermal sediment and epsilonproteobacterial dominance in experimental microcolonizers at the Mid-Atlantic Ridge. Environ Microbiol 5:961–976. Ludwig W, Bauer SH, Bauer M, Held I, Kirchhof G, Schulze R, Huber I, Spring S, Hartmann A, Schleifer KH. 1997. Detection and in situ identification of representatives of a widely distributed new bacterial phylum. FEMS Microbiol Lett 153:181–190. Manaia CM, Moore ER. 2002. Pseudomonas thermotolerans sp. nov., a thermotolerant species of the genus Pseudomonas sensu stricto. Int J Syst Evol Microbiol 52:2203–2209. Martin W, Russell MJ. 2003. On the origins of cells: a hypothesis for the evolutionary transitions from abiotic geochemistry to chemoautotrophic prokaryotes, and from prokaryotes to nucleated cells. Phil Trans R Soc Lond B Biol Sci 358:59–83. Meyers PA. 1997. Organic geochemical proxies of paleoceanographic, paleolimnologic, and paleoclimatic processes. Org Geochem 27:213–250. Nisbet EG, Sleep NH. 2001. The habitat and nature of early life. Nature 409:1083–1091. Norris TB, Wraith JM, Castenholz RW, McDermott TR. 2002. Soil microbial community structure across a thermal gradient following a geothermal heating event. Appl Environ Microbiol 68:6300–6309. Ourisson G, Rohmer M, Poralla K. 1987. Microbial lipids betrayed by their fossils. Microbiol Sci 4:52–57. Pace NR. 1997. A molecular view of microbial diversity and the biosphere. Science 276:734–740. Panichi C, Volpi G. 1999. Hydrogen, oxygen and carbon isotope ratios of Solfatara fumarole (Phlegrean Fields, Italy): further insight into source processes. J Volcanol Geotherm Res 91:321–328. Philippe H. 1993. MUST, a computer package of management utilities for sequences and trees. Nucleic Acids Res 21:5264–5272. Provencio PP, Polyak VJ. 2001. Iron oxide-rich filaments: possible fossil bacteria in Lechuguilla Cave, New Mexico. Geomicrobiol J 18:297–309. Rosi M, Sbrana A. 1987. Phlegrean fields. Consiglio Nazionale Delle Ricerche Quaderni De “La Ricercha Scientifica” 114(9):175. Rothschild LJ, Mancinelli RL. 2001. Life in extreme environments. Nature 409:1092–1101. Schidlowski M. 1988. A 3,800-million-year isotopic record of life from carbon in sedimentary rocks. Nature 333:313–318. Schidlowski M. 2000. Carbon isotopes in microbial sediments. In: Riding RE, Awramik SM, editors. Microbial sediments, Berlin: Springer Verlag. P. 84–96.

541

Schopf JW. 1993. Microfossils of the Early Archean Apex chert: new evidence of the antiquity of life. Science 260:640–646. Schopf JW, Packer BM. 1987. Early Archean (3.3-billion to 3.5-billion-year-old) microfossils from Warrawoona Group, Australia. Science 237:70–73. Schuler D, Frankel RB. 1999. Bacterial magnetosomes: microbiology, biomineralization and biotechnological applications. Appl Microbiol Biotechnol 52:464–473. Segerer AH, Stetter KO. 1999. The Order Sulfolobales. In: Dworkin M, editors, The prokaryotes: an evolving electronic resource for the microbiological community. Berlin: Springer-Verlag, http://link.springer-ny.com/link/service/ books/10125/. Sievert SM, Kuever J, Muyzer G. 2000. Identification of 16S ribosomal DNAdefined bacterial populations at a shallow submarine hydrothermal vent near Milos Island (Greece). Appl Environ Microbiol 66:3102–3109. Strauss H, Des Marais DJ, Hayers JM, Summons RE. 1992. Concentrations of organic carbon and maturities and elemental compositions of kerogens. In: Schopf, JW, Klein C, editors. The Proterozoic Biosphere, A Multidisciplinary Study. NY: Cambridge University Press. P. 95–101. Summons RE, Jahnke LL, Simoneit BR. 1996. Lipid biomarkers for bacterial ecosystems: studies of cultured organisms, hydrothermal environments and ancient sediments. Ciba Found Symp 202:174–193. Tedesco D, Pece R, Sabroux JC. 1988. No evidence of the new magmatic gas contribution to the Solfatara volcanic gases during bradiseismic crisis at Campi Phlegrei, Italy. Geophys Res Lett 15:1441–1444. Tor JM, Kashefi K, Lovley DR. 2001. Acetate oxidation coupled to Fe(III) reduction in Hyperthermophilic microorganisms. Appl Environ Microbiol 67:1363–1365. Treiman AH. 2003. Submicron magnetite grains and carbon compounds in Martian meteorite ALH84001: inorganic, abiotic formation by shock and thermal metamorphism. Astrobiology 3:369–392. Tritz JP, Herrmann D, Bisseret O, Connan J, Rohmer M. 1999. Abiotic and biological hopanoid transformation: towards the formation of molecular fossils of the hopane series. Org Geoch 30:499–514. Twichell SC, Meyers PA, Diester-Haass L. 2002. Significance of high C/N ratios in organic-carbon-rich Neogene sediments under Benguela current upwelling system. Org Geochem 33:715–722. Valentino GM, Cortecci G, Franco E, Stanzione D. 1999. Chemical and isotopic compositions of minerals and waters from Campi Flegrei volcanic system, Naples, Italy. J Volc Geoth Res 91:329–344. van Zuilen MA, Lepland A, Arrhenius G. 2002. Reassessing the evidence for the earliest traces of life. Nature 418:627–630. Wächtershäusser G. 1988. Before enzymes and templates: theory of surface metabolism. Microbiol Rev 52:452–484. Weiss BP, Kim SS, Kirschvink JL, Kopp RE, Sankaran M, Kobayashi A, Komeili A. 2004. Magnetic tests for magnetosome chains in Martian meteorite ALH84001. Proc Natl Acad Sci USA 101:8281–8284. Zillig W, Stetter KO, Schulz W, Priess H, Scholz I. 1980. The Sulfolobus“Caldariella” group: taxonomy on the basis of the structure of DNAdependent RNA polymerases. Arch Microbiol 125:259–260.