Towards the ecology of hyperthermophiles: biotopes, new isolation strategies and novel metabolic properties


*Corresponding author. Tel.: +49 (941) 943-3182; Fax: +49 (941) 943-2403, E-mail address:


Ecological studies have shown that water-containing terrestrial, subterranean and submarine high-temperature environments harbor a great diversity of hyperthermophilic prokaryotes, growing fastest at temperatures of 80°C or above. The investigations included cultivation, isolation and detailed analysis of these hyperthermophiles as well as in situ 16S rRNA gene sequence analysis and in situ hybridization studies. For a safe and fast isolation of novel hyperthermophiles from mixed cultures, a new, plating-independent isolation technique was developed, based on the use of a laser microscope (‘optical tweezers’). This method, combined with 16S rRNA gene sequence analysis and whole-cell hybridization using fluorescently labelled oligonucleotide probes, even allows the recovery of pure cultures of phylogenetically predicted organisms harboring novel 16S rRNA gene sequences. In their natural habitats, hyperthermophiles form complex food webs, consisting of primary producers and consumers of organic material. Their metabolic potential includes various types of aerobic and anaerobic respiration and different modes of fermentation. In hydrothermal and geothermal environments, hyperthermophiles have important ecological functions in biogeochemical processes. Members of the Sulfolobales are able to mobilize heavy metals from sulfidic ores like pyrite or chalcopyrite. Biomineralization processes of hyperthermophiles include the formation of magnetite from iron or the precipitation of arsenate as realgar, a reaction performed by a novel hyperthermophile that was isolated from Pisciarelli Solfatara, Naples, Italy.


During the last two decades, a great diversity of hyperthermophilic prokaryotes have been isolated from hydrothermal, geothermal and anthropogenic high-temperature ecosystems [1,2]. So far, about 75 species of hyperthermophilic archaea and bacteria, representing 32 genera and 10 orders, have been described [3]. By definition, hyperthermophiles grow fastest at temperatures of 80°C or above. As a rule, they are unable to propagate below about 60°C in contrast to moderate thermophiles [2]. Pyrolobus fumarii, the most extreme hyperthermophile, is even unable to grow below 90°C and exhibits the highest growth temperature observed so far at 113°C (Fig. 1a) [4]. Cultures with (vegetative) cells of Pyrolobus and Pyrodictium are even able to survive autoclaving [5]. Both in terms of their phylogeny, biochemical and physiological properties, hyperthermophiles are very divergent [3,5]. Physiologically, they cover a broad spectrum, ranging e.g. from obligate chemolithoautotrophs to strict organotrophs, from aerobes to strict anaerobes or from highly acidophiles to alcaliphiles [3]. Phylogenetically, hyperthermophiles appear in both kingdoms of the domain archaea and in the deep branching phyla of the bacterial domain [3,6]. By comparison of 16S rRNA gene sequences, hyperthermophiles appear as the most primitive life forms on Earth [7,8]. Due to their metabolic flexibility, hyperthermophiles function as primary producers and/or consumers of organic matter in their natural environment and play an important role in biogeochemical processes. In addition, their outstanding heat resistance makes them interesting both for basic research and for biotechnological applications [3,9,10].

Figure 1.

a–d: Electron micrographs of hyperthermophilic archaea (a: freeze-etched; b–d: platinum-shadowed; bar: 1.0 μm). a: Lobed cell of P. fumarii, b: grape-like cell aggregate of T. aggregans, c: single, irregular cell of F. placidus, d: rod-shaped cell of Pyrobaculum sp. PZ6*; formation of black precipitates after cultivation with arsenate.

In this paper, we report on hyperthermophiles and their ecology including their biotopes, physiology, diversity and their possible role in biogeochemical processes.

2Biotopes of hyperthermophiles

2.1General considerations

So far, hyperthermophiles have been isolated mainly from water-containing terrestrial and marine high-temperature areas, where they form complex microbial communities [3]. Marine biotopes of hyperthermophiles are various hydrothermal systems, located at shallow and abyssal depth, and at active seamounts, e.g. Teahicya and Macdonald in the Tahiti area, Polynesia [11]. These ecosystems are characterized by high concentrations of salt (about 3%) and by pH values which range from slightly acidic to slightly alkaline (pH 5.0–8.5). In active volcanic environments, large amounts of steam are formed, containing carbon dioxide, hydrogen sulfide, and variable quantities of hydrogen, methane, nitrogen, carbon monoxide and traces of ammonia or nitrate. Furthermore, sulfur, thiosulfate and sulfate are the major sulfur compounds present in solfatara fields and in marine hot areas.

Shallow marine hydrothermal systems exist in many coastal areas around the world, e.g. Vulcano island (Italy), Ribeira Quente (Azores), Sangeang Island (Indonesia), Kolbeinsey Ridge (north of Iceland) or Lihir island (Papua New Guinea). Abyssal hot vents (‘black smokers’) are located in the deep sea along submarine tectonic fracture zones. They are found within the Guaymas basin, gulf of California (depth about 2000 m), at the East Pacific Rise at 21°N (depth about 2500 m) or at the Mid Atlantic Ridge in a depth between 3000 and 4000 m. Temperatures of up to 350°C can be measured within the ‘black smoker’ chimneys. As a characteristic feature, smoker walls exhibit steep temperature gradients due to the strong cooling by the surrounding cold sea water (about 3°C). Hyperthermophiles can be isolated both from the porous smoker walls and from the surrounding hot sediments.

On land, natural biotopes of hyperthermophiles are hot springs, mud holes and solfataric fields. They are characterized by low salinity (0.1–0.5%) and pH values ranging from about 0.5 to 9.0. Depending on the altitude, the maximum temperature is 80–100°C. Terrestrial high-temperature environments are widespread on Earth and are often situated around active volcanoes. Examples of such solfataric fields are Krisuvik, Hveragerthi, Kerlingarfjöll (all Iceland), Yellowstone National Park (USA), White Island (New Zealand), Kamchatka peninsula (Russia), Hokkaido (Japan) or Pisciarelli Solfatara, Naples (Italy).

Deep subterranean, geothermally heated oil reservoirs about 3500 m below the bottom of the North Sea and the permafrost soil at the North Slope, northern Alaska are non-volcanic biotopes for hyperthermophiles [12,13]. These may have entered the reservoirs by sea water injection or through natural routes such as faults and oil seeps. A major part of the oil-field hyperthermophiles are sulfidogens and may, therefore, participate in ‘reservoir souring’ at temperatures previously considered too high for biochemical reactions [14].

Artificial high-temperature biotopes are smoldering coal refuse piles (e.g. Ronneburg, Thüringen, Germany) [15] and hot outflows from geothermal power plants.

3Ecological studies in Pisciarelli Solfatara

The continental solfatara field of Pisciarelli Solfatara is located close to the Solfatara crater near Naples, Italy. It is only about 800 m2 in size, but contains more than 20 physically and chemically different springs and mud holes. A typical feature of continental solfataric fields is the existence of two dominating zones in the soil. The upper oxidized zone is often rich in ferric iron and appears therefore orange-colored. Generally, the thickness depends on the volcanic activity and exhalation of reducing volcanic gases (e.g. H2S, H2), and is between a few centimeters and about 50 cm in Pisciarelli Solfatara. Below the oxidized layer, an anoxic zone exists, often black-colored due to ferrous sulfide. The interphase between these zones is characterized by the presence of elemental sulfur and is therefore slightly yellow-colored. In Pisciarelli Solfatara, this sulfur layer has a width of a few millimeters. The sulfur is formed by chemical oxidation of H2S coming from below by molecular oxygen penetrating from the surface into the soil.

Further characteristic features of such solfataric areas are steep temperature gradients and the presence of highly acidic soils and springs in close vicinity to soils, exhibiting neutral (or even slightly alkaline) pH values. As an example, Table 1 summarizes pH and temperature measurements at different sites in Pisciarelli Solfatara from the soil surface to a depth of about 30 cm. The chemical composition of soil samples of Pisciarelli Solfatara (by ICP measurements) revealed that the main element was iron, an element which is involved in several energy-yielding reactions of hyperthermophiles (Table 2). As in other geothermal settings, arsenate is an important component in Pisciarelli Solfatara (Table 2). Therefore, novel hyperthermophiles may thrive in such biotopes, using this heavy metal for energy-yielding reactions (see Section 7).

Table 1.  Temperature and pH profiles of different soils in Pisciarelli Solfatara, Naples, Italy
 Acidic soil; hot surface; pH constant (three sites)Acidic soil; warm surface; pH constant (three sites)Acidic soil; warm–hot surface; pH increasing (two sites)Neutral soil; warm surface (five sites)
 Temp. (°C)pHTemp. (°C)pHTemp. (°C)pHTemp. (°C)pH
Depth: 0 cm60–701.5–2.530–371.5–2.035–702.534–574.0–6.0
1 cm80–851.5–2.035–551.5–2.055–854.5–5.551–655.5–6.0
5 cm86–871.551–651.5–2.070–854.5–6.058–925.0–6.0
10 cm88–901.578–955.0–6.061–934.0–6.0
30 cm921.592–985.5–6.075–985.0–6.0
Table 2.  Heavy metal content of soil samples from Pisciarelli Solfatara, Naples, Italy (ICP analysis)
 Acidic soil (g kg−1)Neutral soil (g kg−1)
Cd, Co, Cr, Cu, Mo, Ni, Se, Sn

The physical and chemical parameters of the sampling sites (Tables 1 and 2) had a strong influence on the occurrence of metabolically diverse hyperthermophiles, as determined by in situ 16S rRNA gene sequence analysis, in situ hybridization experiments and cultivation attempts. Phylogenetic analyses of a highly acidic sample (original temperature: 95–97°C) showed new archaeal 16S rRNA sequences, belonging to the Sulfolobales and Thermoplasmatales. Members of these two orders are known to grow optimally between pH 1.0 and 3.0. In contrast, from neutral samples (original temperature: 82–96°C), members of the Archaeoglobales and Thermococcales were identified by 16S rRNA gene sequence analyses, which is in line with their physiological properties. Furthermore, from the neutral samples, 16S rRNA gene sequences were identified which exhibited high similarity to a group of deep branching crenarchaeal sequences from the ‘Obsidian Pool’ (pSL1, pSL44, pSL69, pSL123, pJP89) [16].

The phylogenetic analyses were verified by semi-quantitative in situ hybridization experiments of different acidic and neutral sites in Pisciarelli Solfatara (Table 3) [17,18]. Cell densities of more than 107 irregularly lobed cells per g sediment and per ml hot spring water were determined in a total of 13 acidic samples from the upper and central area of Pisciarelli Solfatara. As expected, they were representatives of the Sulfolobales order (Table 3). The neutral samples harbored archaeal coccoid and rod-shaped cells in densities up to about 107 cells g−1 sediment or per ml hot spring water (Table 3). Interestingly, short, curved rods in two samples from the upper area showed a positive hybridization signal with a probe designed for deep branching crenarchaeal 16S rRNA gene sequences (see above).

Table 3.  Results of the in situ hybridization experiments in Pisciarelli Solfatara using different oligonucleotide probes
Genetic probe for:Six acidic samples (central area), pH 1.5, Temp. 70–97°CSeven acidic samples (upper area), pH 1.5–2.0, Temp. 75–92°CThree neutral samples (central area), pH 5.5–6.5, Temp. 80–97°CFour neutral samples (upper area), pH 5.5–6.0, Temp. 82–100°C
Bacteriashort rods ++short curved rods ++short rods +++
Archaeairregular cocci +++irregular cocci +++small cocci ++curved long rods ++
Crenarchaeotairregular cocci +++irregular cocci +++
Sulfolobalesirregular cocci +++irregular cocci +++
Deep branching Crenarchaeotashort curved rods ++
+++=more than 107 cells ml−1.
++=between 106 and 107 cells ml−1.
+=between 105 and 106 cells ml−1.
−=below detection limit (about 105 cells ml−1).

In line with the molecular results, members of the Sulfolobales were isolated only from different acidic soils (Tables 1 and 3). In contrast, neutral samples contained strictly anaerobic hyperthermophiles growing optimally around pH 6.0, such as Thermoproteus, Pyrobaculum or Thermofilum. However, direct microscopic inspection indicated that there are many more morphotypes yet to be cultivated and isolated. Therefore, it is important to develop and apply new isolation procedures and isolation strategies to get a handle on such organisms.

4Isolation procedures


The study of pure cultures is extremely important for a deeper understanding of the properties and functions of organisms in their natural environments. Traditionally, strains are purified from enrichment cultures by plating on solid surfaces. However, due to the high growth temperatures of hyperthermophiles, most solidifying agents (e.g. agar) are not suitable. Therefore, more thermostable gelling agents like polysilicate or gellan gum, Gelrite (Kelco, USA), are used for plating [19]. The gel stability of Gelrite plates is highly dependent on the Gelrite concentration, on the soluble salt concentration and the type of salt added. Gelrite solidifies rapidly in the presence of divalent cations such as magnesium or calcium and, to a lesser extent, in the presence of monovalent ions [19].

However, most likely due to the inability to grow on solid surfaces, plating of different hyperthermophiles was not successful [2]. Therefore, serial dilution in liquid medium repeated at least three times was used as an alternative but less safe isolation method.

4.2Selected single cell isolation using the ‘optical tweezers trap’

For a safe, efficient and fast isolation of hyperthermophiles from mixed cultures, a novel, plating-independent isolation procedure was developed recently in our lab [20]. The method is based on separation of a single cell from enrichment cultures by the use of a laser microscope and subsequent growth of this cell. The ‘optical tweezers trap’ consists of a computer-controlled inverted microscope, equipped with a continuously operating neodymium-doped yttrium aluminum garnet laser (Nd:YAG laser). The emission wavelength of the laser is in the near infrared at 1064 nm, the maximum output power is 1 W. The laser can be focused to a spot size of less than 1 μm in diameter by the use of a high numerical aperture oil immersion objective (100×). As a consequence of the strong intensity of the laser light, optical trapping and manipulation of single cells in μm-size in three dimensions is possible (the ‘optical tweezers trap’ or ‘laser trap’) [21,22]. The micromanipulation of the cells is performed by keeping the laser beam at a fixed position and moving the motor-driven mechanical stage.

For isolation of single cells, a cell separation unit was designed. It consists of a rectangular microslide as observation and separation chamber (inside dimension: 0.1×1 mm2; length: 10 cm), which is connected by a tube to the needle of a 1-ml syringe (Fig. 2a) [20,23]. A cutting line separates the microslide into two compartments (Fig. 2a). After sterilization of the cell separation unit, about 90% of the microslide is filled with sterile medium. Afterwards, the mixed culture is soaked into the remaining volume of the microslide. A single cell is selected under 1000-fold magnification and is optically trapped in the laser beam by activation of the laser (Fig. 2a). The cell can be separated by at least 6 cm from the mixed culture within 3–10 min, being transferred into the sterile compartment by moving the microscopic stage (Fig. 2b) [23]. At the cutting line, the microslide is gently broken and the single cell is flushed into sterile medium (the ‘selected cell cultivation technique’).

Figure 2.

a,b: Isolation of a single cell from an enrichment culture by use of a laser microscope; schematic drawing, reprinted [23] with permission from Spektrum Akademischer Verlag. a: A single cell is optically trapped within the focus of the laser beam, b: the single cell is separated from the mixed culture into the sterile compartment. 1: Microslide, 2: tube, 3: needle, 4: syringe, 5: cutting line, 6: mixed culture, 7: objective.

This ‘selected cell cultivation technique’ was applied to a variety of hyperthermophilic archaea and bacteria. Meanwhile, representatives of all orders and families harboring hyperthermophiles have been successfully isolated by this technique [23]. Depending on the strain, 20–100% of the separated cells from exponentially growing enrichment cultures generated pure cultures. Pure cultures with cell titers of at least 107 cells ml−1 are available within only 1–5 days, depending on the generation time of the isolated cells.

The application of the ‘selected cell cultivation technique’ is only successful if the trapped cell is alive and divides after isolation. Therefore, the cultivation efficiency can be enhanced if the physiological state of each single cell is determined before trapping and isolation. Using bis-(1,3-dibutylbarbituric acid) trimethine oxonol (=DiBAC4(3)), a membrane potential-sensitive dye, a safe and rapid discrimination of viable and dead cells of hyperthermophilic archaea and bacteria was possible [24]. Therefore, the combination of the ‘selected cell cultivation technique’ with the use of DiBAC4(3) can significantly enhance the isolation of novel hyperthermophiles either from enrichment cultures or directly from natural samples.

5A new isolation strategy

In recent years, an increasing number of different 16S rRNA gene sequences from environmental samples have been published [25]. In order to obtain the organism predicted from such a phylogenetic sampling, a new isolation strategy was developed recently in our lab [20]. This procedure combines the analysis of 16S rRNA gene sequences from the environment, specific whole-cell hybridization within enrichment cultures, and isolation of the morphologically identified single cell by the ‘optical tweezers trap’[20]. This approach allowed us for the first time to isolate a novel hyperthermophilic archaeum, tracked by 16S rRNA analysis from a hot pond in Yellowstone National Park (‘Obsidian Pool’). In the future, this new strategy could also be applied to isolate uncultivated organisms from different biotopes, phylogenetically predicted by in situ 16S rRNA gene sequence analysis.

The new coccoid isolate from the ‘Obsidian Pool’, Thermosphaera aggregans, grows in grape-like aggregates (Fig. 1b) and belongs to the Desulfurococcaceae within the Crenarchaeota branch of the archaea domain [26]. A characteristic feature of T. aggregans is its inhibition of growth in the presence of sulfur. This inhibition was unexpected because members of the Desulfurococcaceae are typically sulfur-respirers [27]. Also within other main phylogenetic branches of the Crenarchaeota, the mode of metabolism is not uniform. Within the otherwise strictly anaerobic sulfur-respiring Thermoproteales and Pyrodictiaceae[27], Pyrobaculum aerophilum[28] and P. fumarii[4] utilize aerobic respiration or anaerobic respiration with nitrate. Furthermore, the strictly anaerobic Stygiolobus azoricus[29] is the closest relative of the strict aerobe Sulfolobus acidocaldarius[30] within the Sulfolobales order [31]. In view of such fundamentally different physiological properties of phylogenetically closely related organisms, it is not possible to predict the metabolic type from the phylogenetic position.

6Metabolic potential of hyperthermophiles

6.1General considerations

The unique physiological property of hyperthermophiles is their ability to grow fastest at very high temperatures (≥80°C). In their ecosystems, they form complex food webs. Within these communities, hyperthermophiles function as primary producers and/or as consumers of organic material. The primary producers are chemolithoautotrophs, using a variety of inorganic compounds as electron donors and acceptors (Table 4). Under aerobic growth conditions, molecular hydrogen is converted to water (‘Knallgas reaction’) and elemental sulfur, sulfides or thiosulfate are oxidized to sulfuric acid. In the absence of oxygen, elemental sulfur, sulfate, thiosulfate, sulfite, nitrate, nitrite and CO2 are suitable electron acceptors for growth, while molecular hydrogen is the electron donor in these energy-yielding reactions (Table 4). Instead of hydrogen, Ferroglobus placidus is able to use ferrous iron as electron donor for nitrate reduction (Fig. 1c; Table 4) [32].

Table 4.  Energy conservation in chemolithoautotrophic hyperthermophiles
  1. aFacultatively heterotrophic.

Energy-yielding reactionGenera
H2+1/2O2→H2OAquifex, Thermocrinis Sulfolobusa, Acidianusa, Metallosphaeraa, Pyrobaculuma
2S°+3O2+2H2O→2H2SO4Aquifex, Sulfolobusa, Acidianusa, Metallosphaeraa
2FeS2+7O2+2H2O→2FeSO4+2H2SO4=‘metal leaching’Sulfolobusa, Acidianusa, Metallosphaera
H2+HNO3→HNO2+H2OAquifex, Pyrobaculuma
H2+6FeO(OH)→2Fe3O4+4H2OPyrobaculum sp. nov.
H2+S°→H2SIgnicoccus, Acidianus, Stygiolobus, Pyrodictiuma, Pyrobaculuma, Thermoproteusa
4H2+CO2→CH4+2H2OMethanopyrus, Methanothermus, Methanococcus

Recently, it was shown that heavy metal compounds are further suitable electron acceptors for anaerobic growth of hyperthermophiles. Several members of Pyrobaculum can reduce ferric iron to form magnetite [5]. By coexistence with anaerobic iron oxidizers like F. placidus, a hot iron cycle can be postulated, which may have existed already within early life forms on Earth [33].

Autotrophic hyperthermophiles are able to obtain cellular carbon through the reduction of carbon dioxide via the reductive tricarboxylic acid cycle (performed by e.g. Aquifex pyrophilus, Thermoproteus neutrophilus) or the reductive acetyl CoA pathway (e.g. Archaeoglobus lithotrophicus, F. placidus) [34,35]. Recently, a new CO2 fixation pathway, the 3-hydroxypropionate cycle, was described for members of the Sulfolobales (Metallosphaera sedula, Sulfolobus metallicus and Acidianus infernus) [36]. Several chemolithoautotrophs are facultatively heterotrophic, alternatively using organic matter for growth. These two different modes of metabolism were identified for example in representatives of the genera Sulfolobus, Metallosphaera, Acidianus or Pyrobaculum (Table 5).

Table 5.  Energy conservation in heterotrophic hyperthermophiles
Type of metabolismExternal electron acceptorEnergy-yielding reactionGenera
Respiration2[H]+S°→H2SPyrodictium, Thermoproteus, Pyrobaculum, Thermofilum, Desulfurococcus, Stetteria,Thermodiscus, Stygiolobus, Sulfurisphaera
 SO42− (S2O32−; SO32−)8[H]+H2SO4→H2S+4H2OArchaeoglobus
 O22[H]+1/2O2→H2OSulfolobus, Metallosphaera, Acidianus, Pyrobaculum, Aeropyrum
Fermentationpeptides→isovalerate, isobutyrate, butanol, CO2, H2, etc.Pyrodictium, Hyperthermus, Thermococcus, Pyrococcus, Thermoproteus, Thermosphaera, Sulfophobococcus, Desulfurococcus, Staphylothermus
 glucose→l(+)-lactate+acetate+H2+CO2Thermotoga, Thermosipho, Fervidobacterium
 cellobiose, maltose or pyruvate→acetate+alanine+H2+CO2Pyrococcus

A variety of obligate heterotrophs are known which grow by different types of respiration or fermentation. So far, Aeropyrum pernix is the only hyperthermophile gaining energy exclusively by aerobic respiration of complex organic matter [37]. For anaerobic respiration, sulfur, sulfur-containing compounds or nitrate may be used as electron acceptors (Table 5). Depending on the organism, complex organic compounds (e.g. yeast extract, meat extract, water-extracted crude oil, cell homogenates of archaea or bacteria) or defined substrates (e.g. maltose, glucose, arachnic acid, l(+)-lactate, acetate) are used as growth substrates [2,13]. Fermentatively growing hyperthermophiles are present within the archaeal and bacterial domains. During growth on peptides or carbohydrates, different organic acids, carbon dioxide and hydrogen are formed as major products (Table 5) [3]. Molecular hydrogen, a potent inhibitor of growth for members of Thermotoga, Thermococcus or Pyrococcus[38,39], for example, can be removed by gassing with nitrogen or argon. Alternatively, the inhibition can be prevented by the addition of sulfur, cystine or thiosulfate, whereupon H2S is formed instead of hydrogen. Thermotoga maritima may use Fe(III) in place of sulfur as an electron sink to get rid of inhibitory hydrogen during fermentation. Another possibility to eliminate hydrogen is by interspecies hydrogen transfer during co-cultivation of fermentative and hydrogen-consuming hyperthermophiles [40,41]. Thermococcus stetteri, Thermococcus celer, Staphylothermus marinus and Pyrococcus furiosus exhibited excellent growth, when cultivated together with Methanococcus thermolithotrophicus or Methanococcus jannaschii. A further strategy to prevent hydrogen production was identified in P. furiosus[41]. Instead of acetate, alanine is formed as a reduced end product by reductive amination of pyruvate.

7Respiration of arsenate by hyperthermophilic archaea

Within high-temperature environments like Pisciarelli Solfatara, arsenic is one of the most prominent heavy metals (Table 2). This prompted us to look for the possible existence of hyperthermophiles able to use arsenic compounds in their metabolism. Using arsenate as an electron acceptor, we were able to isolate for the first time hyperthermophilic rods, growing chemolithoautotrophically under anaerobic culture conditions with hydrogen as electron donor. The stoichiometric reduction of arsenate to arsenite was directly coupled with exponential growth. Taxonomic investigations revealed that the new isolate PZ6* represents a new species within the genus Pyrobaculum, which we will name ‘Pyrobaculum arsenaticum’ (Fig. 1d) [42].

Under organotrophic culture conditions in the presence of thiosulfate and arsenate, ‘P. arsenaticum’ formed a precipitate during growth, visible as yellow–orange flocs in the culture medium. This precipitate occurred in a crystalline form and was identified as realgar (As2S2) by X-ray diffraction, scanning electron microscopy and energy-dispersive X-ray spectroscopy.

In geothermal systems, the most common As-bearing minerals found include orpigment (As2S3), realgar (As2S2), arsenopyrite (FeAsS), loellingite (FeAs) and native arsenic [43]. To date, formation of arsenic–sulfur compounds in these high-temperature environments has been attributed to chemical processes [43,44]. ‘P. arsenaticum’ is the first example to indicate that precipitation of realgar can be generated biologically. It can precipitate realgar over its temperature range of growth, which is between 68 and 100°C. Interestingly, these results fit well with geological data from Uzon caldera, Kamchatka. Realgar was the only sulfur-bearing mineral in this solfatara field, found in the temperature zone between 70 and 95°C [44]. By dilution in multiple series, we found that arsenate-reducing hyperthermophiles occur in high densities in the natural environment. In Pisciarelli Solfatara, we have determined at least 107 viable cells of arsenate reducers per g sediment. All these results indicate that microbial activities of hyperthermophiles in volcanic areas play a major role in the biogeochemical cycle of arsenic and it is not simply chemistry that determines the speciation of arsenic in high-temperature environments. By their ability to immobilize soluble arsenate by precipitation, they may also be a critical part of a possible combined arsenic and sulfur cycle in hydrothermal areas.


We wish to thank Reinhard Rachel for electron micrographs. This work was financially supported by the Deutsche Forschungsgemeinschaft (Ste 297/10) and by the Fonds der Chemischen Industrie.