Cultivation of Anaerobic and Facultatively Anaerobic Bacteria from Spacecraft-Associated Clean Rooms

Abstract

In the course of this biodiversity study, the cultivable microbial community of European spacecraft-associated clean rooms and the Herschel Space Observatory located therein were analyzed during routine assembly operations. Here, we focused on microorganisms capable of growing without oxygen. Anaerobes play a significant role in planetary protection considerations since extraterrestrial environments like Mars probably do not provide enough oxygen for fully aerobic microbial growth. A broad assortment of anaerobic media was used in our cultivation strategies, which focused on microorganisms with special metabolic skills. The majority of the isolated strains grew on anaerobic, complex, nutrient-rich media. Autotrophic microorganisms or microbes capable of fixing nitrogen were also cultivated. A broad range of facultatively anaerobic bacteria was detected during this study and also, for the first time, some strictly anaerobic bacteria (Clostridium and Propionibacterium) were isolated from spacecraft-associated clean rooms. The multiassay cultivation approach was the basis for the detection of several bacteria that had not been cultivated from these special environments before and also led to the discovery of two novel microbial species of Pseudomonas and Paenibacillus.

The major issue of planetary protection is to prevent the contamination of extraterrestrial environments by terrestrial biomolecules and life forms. Furthermore, reverse contamination of Earth by extraterrestrial material is also a fundamental concern (1). In order not to affect or even to confound future life detection missions on celestial bodies, which are of interest for their chemical and biological evolution, spacecraft are constructed in so-called clean rooms and are subject to severe cleaning processes and microbiological controls before launch (9). Therefore, these clean rooms are considered extreme environments for microorganisms (47).

Detailed planetary protection protocols for missions to Mars were designed for the Viking missions, which were launched in 1975, and about 7,000 samples were taken from the two Viking spacecraft during prelaunch activities in order to determine the cultivable microbial load (37). Besides human-associated bacteria (pathogens and opportunistic pathogens), which were predominant among the microbes detected in these samples, aerobic spore-forming microorganisms (Bacillus) were found frequently on spacecraft and within the facilities.

Spores are the resting states of bacteria and are often highly resistant to heat, desiccation, and other abiotic stresses. These multiresistance properties of such spore-forming microorganisms make them perfect candidates for surviving a space flight, and thus, the main focus of attention has been on them. Furthermore, only the detection of aerobic spore-forming bacteria is currently included in space agencies' planetary protection protocols for the quantitative determination of microbial burden on spacecraft.

The presence of extraordinarily (UV-) resistant spores in spacecraft facilities has been reported (31), but it also has been proven that vegetative microbial cells (e.g., Deinococcus radiodurans and Halobacterium sp. strain NRC-1) can resist very harsh conditions, such as extreme doses of (UV and ionizing) radiation and desiccation (8, 11). Recent culture-based and molecular studies have shown that the microbial diversity on spacecraft and within the clean rooms is extraordinarily high and does include extremotolerant bacteria and even archaea (25, 30).

The atmospheres of most planets and bodies within the reach of human exploration contain only traces of oxygen (Mars contains 0.13%), probably not enough to support terrestrial aerobic life as we know it (26, 44). Even though Mars' surface is highly oxidizing and radiation exposed, the Martian subsurface, as well as those of other planets and bodies (like, e.g., Titan), has been discussed as an anaerobic biotope for possible life (4, 40).

Therefore, the lack of studies of the existence of anaerobically growing microorganisms in spacecraft-associated clean rooms is quite surprising. One possible reason for this discrepancy might be that the cultivation of anaerobes is challenging. Already in 1969, Hungate published a method for the cultivation of strictly anaerobic methanogenic Archaea (20). Although this technique has undergone a few simplifications during past decades, the cultivation of anaerobes requires specialized and expensive equipment (e.g., anaerobic glove boxes and gas stations), practical experience, and skills in specific methodology. Nevertheless, by the application of anaerobic cultivation strategies, many fascinating microorganisms—such as Nanoarchaeum equitans, the first representative of the new archaeal phylum Nanoarchaeota, or Thermotoga maritima, a hyperthermophilic bacterium growing at up to 90°C (17, 18)—have successfully been isolated from diverse and sometimes extreme biotopes.

Generally, there are different types of anaerobic organisms. Facultative anaerobes (like Escherichia coli) are able to adapt their metabolism and can grow under conditions with or without oxygen but prefer aerobic conditions. Aerotolerant anaerobes do not need oxygen for their growth and show no preference, and strict anaerobes (e.g., methanogens) never require oxygen for their reproduction and metabolism. Even more, obligate (strict) anaerobes can be growth inhibited or even killed by oxygen.

The presence of anaerobic microorganisms (enriched using the BD GasPak system) in surface samples from U.S. clean rooms has rarely been reported. Members of the facultatively anaerobic genera Paenibacillus and Staphylococcus have been isolated in the course of a study about extremotolerant microorganisms (25). During molecular surveys of U.S. clean rooms, the 16S rRNA genes from strictly anaerobic microorganisms, such as the spore-forming genus Clostridium, have already been detected (29). Nevertheless, the cultivation of these microbes has not yet been successful.

With the ExoMars mission impending, the European Space Agency (ESA) is organizing and funding a biodiversity study of the ESA's clean rooms and the spacecraft therein. The microbiology of these special environments is characterized in detail by a combination of standard procedures, new cultivation approaches, and molecular methods that shall illuminate the presence of planetary protection-relevant microorganisms in these facilities. At the date of sampling, all the clean rooms harbored the Herschel Space Observatory, a spacecraft to be launched together with the Planck satellite in spring 2009, as of this writing. Herschel will be fitted with the largest mirror ever built for a space mission (3.5 m in diameter), and its main goal will be the exploration of the cold universe, i.e., the formation and evolution of proto-galaxies (35). The Herschel Space Observatory does not demand planetary protection requirements, but all clean rooms were in a fully operating state during the construction work. This gave us the opportunity to sample the microbial diversity in these extreme environments without bioburden control but under strict contamination-controlled conditions, with respect to particulates and molecular contamination.

This paper presents the results from our attempts to isolate anaerobic and facultatively anaerobic microorganisms from samples of spacecraft and surfaces in European spacecraft-associated clean rooms. For this purpose, we have successfully applied Hungate technology for anaerobic culturing and used an assortment of noncommercial media for the cultivation of a broad variety of microorganisms. Besides the capability of anaerobic growth, many of our isolates revealed special physiological capacities (e.g., nitrogen fixation and autotrophic metabolism) that might be relevant for further planetary protection considerations.

MATERIALS AND METHODS

Sampling sites.

Samples were collected from selected surface areas within two European spacecraft-associated clean rooms, as well as from the surface of the Herschel Space Observatory located therein. Clean rooms are classified according to ISO 14644-1. Therefore, the clean room class ISO 5 corresponds to the former clean room class 100 (US FED STD 209E), allowing a maximum of 3.5 particles with a maximum 0.5-μm diameter per liter air. The clean room (ISO 5) at EADS (European Aeronautic Defense and Space Company) in Friedrichshafen, Germany, was sampled in April and November 2007. Another sampling was done in March 2008 within a clean room at ESTEC (European Space Research and Technology Centre), Noordwijk, The Netherlands (ISO 8). At any sampling, the clean rooms were fully operating and harbored the Herschel Space Observatory. The clean rooms in Friedrichshafen (hall 6, room 6101-04) and at ESTEC (hall HYDRA, Fh) were environmentally controlled with regard to humidity (55%), temperature (22°C), and air circulation (whole clean room air was exchanged every 7 h with 80% air from the inside and 20% air from the outside).

Sample collection.

Samples were taken by using either SpongeSicles (Biotrace [3M], St. Paul, MN) or nylon-flocked swabs (MicroRheologics, Copan, Brescia, Italy). The swabs were used for the sampling of spacecraft surfaces (25 cm²), and in this case three swabs were pooled (three samples [25 cm² each] were taken from all over the spacecraft). Before sampling, swabs were premoistened with sterile water. The sampling surface was swabbed in three different directions, while rotating the swab slowly. SpongeSicles were used to sample larger areas (e.g., floor). The SpongeSicles were premoistened with water (10 ml) before sampling. All samples were kept cool (4 to 8°C) and were processed within 24 h of being taken.

Anaerobic sampling, sample extraction, and inoculation of the media. (i) Preparation of the anaerobic sample storage buffer.

Before sampling, anaerobic PBS buffer (including 0.02% Tween 80 [vol/vol]) was freshly prepared. The buffer was flushed with nitrogen and reduced by adding 0.5 g/liter sodium sulfide·9 H₂O. As a redox indicator, sodium resazurin (0.001 g/liter) was added. Under anaerobic conditions, 10-ml (tubes) and 80-ml (bottles) aliquots were prepared for the swabs and the SpongeSicles, respectively.

(ii) Anaerobic sampling.

Swabs and SpongeSicles were premoistened with water, and the surfaces were sampled. Right after sampling, the stopper from the anaerobic tube or bottle was removed and the swab or head of the SpongeSicle was dropped into the sterile anaerobic buffer. The tube or bottle was immediately closed with a new sterile stopper. If the redox indicator turned red, the tube or bottle content was reduced stepwise by adding 0.1 ml of an anaerobic mixture of cysteine-HCl and sodium sulfide (0.5 g cysteine-HCl and 0.5 g Na₂S·9 H₂O in 10 ml of distilled water, pH 7).

(iii) Sample extraction.

Bottles and tubes were kept closed during the extraction procedure. Tubes containing buffer and swabs were placed on a vortex mixer and vortexed at maximum power for 5 to 6 s. Bottles containing buffer and SpongeSicles were shaken vigorously for several seconds and then placed on a vortex mixer for a few seconds.

(iv) Inoculation of the media.

Samples were removed from tubes and bottles using syringes and needles. Anaerobic liquid media (see below) were inoculated in the laboratory; solid media (plates) were inoculated under anaerobic conditions in an anaerobic glove box.

Aerobic sampling, sample extraction, and inoculation of the media.

After sampling, swabs were stored in 2.5-ml aliquots of aerobic, sterile PBS buffer (including 0.02% Tween 80 [vol/vol]). SpongeSicles were dropped back into their plastic bag containers, and 10 ml of sterile PBS was added to each. The plastic bags were closed as described by the manufacturer. Tubes and bags were opened under a sterile hood in the laboratory. Liquid media were inoculated using syringes and needles; for the inoculation of solid media (plates), aliquots of the samples were transferred into an anaerobic glove box.

Blanks and controls.

For each sampling, field blanks were taken. Field blanks were treated like the samples, but the swabs and SpongeSicles were not used for the sampling of surfaces; they were only waved through the air for a few seconds. Furthermore, medium blanks were used. For each inoculated medium, one vial or plate was incubated without inoculation.

Media.

For the cultivation of anaerobic and facultatively anaerobic microbes, the following media were chosen (the recipes are given for 1 liter medium to be prepared with distilled water): TG (thioglycolate liquid medium) {peptone from casein (Becton Dickinson [BD], NJ), 15.0 g; yeast extract (BD), 5.0 g; d-(+)-glucose, 5.5 g; NaCl, 2.5 g; sodium acetate, 3.0 g; cysteine-HCl, 0.5 g; sodium thioglycolate, 0.5 g; sodium resazurin, 0.001 g; gas phase, N₂; pH 7.1}; TGA (thioglycolate agar plates) (TG medium plus agar, 15.0 g); TS (trypticase soy liquid medium) (TS broth [BD], 30.0 g; sodium resazurin, 0.001 g; sodium thioglycolate, 0.5 g; cysteine-HCl, 1.0 g; gas phase, N₂ [80%] and CO₂ [20%]); TSA (trypticase soy agar plates) (TS medium plus agar, 15.0 g); SRA (sulfate reducer agar plates, medium based on DSMZ medium no. 63, modified) [KH₂PO₄, 0.47 g; NH₄Cl, 1.0 g; CaCl₂·2 H₂O, 0.1 g; yeast extract (BD), 1.0 g; Na₂SO₄, 1.0 g; MgSO₄·7 H₂O, 2.0 g; (40%) (wt/vol) l-(+)-lactate, 2.5 ml; FeSO₄·7 H₂O, 0.004 g; agar, 10.0 g; sodium resazurin, 0.001 g; ascorbic acid, 0.2 g; sodium thioglycolate, 0.2 g; pH 7.0]; MM (methanogenic Archaea liquid medium) (NH₄Cl, 0.5 g; KH₂PO₄, 0.4 g; MgCl₂·6 H₂O, 0.15 g; CaCl₂·2 H₂O, 0.05 g; trace element solution [10×], 1 ml; vitamin solution [10×], 1 ml; sodium resazurin, 0.001 g; Na₂S, 0.5 g; cysteine-HCl, 0.5 g; gas phase, H₂ [80%] and CO₂ [20%]); BM (basal medium) (NH₄Cl, 0.5 g; KH₂PO₄, 0.4 g; MgCl₂·6 H₂O, 0.15 g; CaCl₂·2 H₂O, 0.05 g; NaHCO₃, 1.0 g; trace element solution [10×], 1 ml; vitamin solution [10×], 1 ml; sodium resazurin, 0.001 g; Na₂S, 0.25 g; cysteine-HCl, 0.25 g; pH 7.0); ASM (Archaea-supporting liquid medium) (per 20 ml BM, add 0.1% [wt/vol] sterile yeast extract prior to inoculation; add 0.1 ml of an antibiotic mixture [carbenicillin (0.2% [wt/vol]), streptomycin (0.2% [wt/vol]), rifampin (0.4% [wt/vol]), and cephalosporin (0.2% [wt/vol])]; gas phase, N₂; pH 7.0); AHM (autotrophic homoacetogen liquid medium) (per 20 ml BM, add 0.2 ml 2-bromoethanesulfonic acid [2 M]; gas phase, H₂ [80%] and CO₂ [20%]); N₂ fix (Hino and Wilson N₂-free liquid medium as described by Hino and Wilson [15], with modifications) (sucrose, 20.0 g; MgSO₄·7 H₂O, 0.5 g; NaCl, 0.01 g; FeSO₄·7 H₂O, 0.015 g; Na₂MoO₄·2 H₂O, 0.005 g; CaCO₃, 10.0 g; solve ingredients in 1 liter of K₂HPO₄-KH₂PO₄ buffer [0.1 M, pH 7.7]; medium is not reduced; after sterilization, add 5 μg biotin and 10 μg 4-p-aminobenzoic acid per liter; gas phase, N₂); AAM (autotrophic all-rounder liquid medium) (KH₂PO₄, 0.4 g; CaCl₂·2 H₂O, 0.05 g; MgCl₂·6 H₂O, 0.15 g; NaHCO₃, 1.5 g; Fe₂O₃·9 H₂O, 0.25 g; NaNO₃, 0.5 g; Na₂S₂O₃·5 H₂O, 1.56 g; trace element solution [10×], 1 ml; vitamin solution [10×], 1 ml; sodium resazurin, 0.001 g; Na₂S, 0.5 g; gas phase, N₂ [80%] and CO₂ [20%] [for the composition of the trace element solution and vitamin solution, see DSMZ medium 141 (www.dsmz.de)]).

Preparation of the anaerobic media.

For preparation of the anaerobic media (2, 3, 19, 28), heat-stable ingredients (including agar for solid media) were dissolved in flasks and immediately sealed with rubber stoppers and screw caps with holes. The media were flushed with N₂ for at least 30 min, until the redox indicator (resazurin) turned colorless during this procedure. To transfer chemicals, disposable syringes and hypodermic needles were used. The pH was checked and corrected, if necessary. For liquid media, the flask was transferred to the anaerobic glove box; the medium was dispensed into serum bottles (20 ml each), sealed with butyl-rubber stoppers, and clamped. The desired gas mixture was added by flushing the media three times with gas and applying the vacuum alternately. Finally, the gas mixtures were added at two atmospheres of overpressure, and the media were autoclaved. For solid media, media were autoclaved after they were flushed with N₂ and completely reduced. The media were cooled to 50°C and transferred into the anaerobic glove box, where the plates were poured under sterile conditions. The plates were dried in the glove box.

Maintenance of the anaerobic glove box.

The anaerobic glove box (Coy, MI) was operated as instructed by the manufacturer. The gas phase in the glove box contained N₂ and H₂ (95:5 [vol/vol]). Function was checked by an oxygen detector. Catalysts were frequently changed.

Isolation of microorganisms and cultivation conditions.

Grown colonies were picked, and the strains were purified by two consecutive streakouts on agar plates. Microorganisms enriched in liquid cultures were plated on anaerobic or aerobic agar plates (0.2 ml) and afterwards purified as described above. All cultures were incubated at 32°C. Liquid media were shaken (50 rpm) until growth occurred.

Determination of the oxygen requirement of the isolates.

The sensitivity of the anaerobically cultured strains to oxygen was tested by inoculation on aerobic agar plates and cultivation under aerobic conditions. This was performed either by a streakout or by spreading out 0.2 ml of a liquid culture.

Phylogenetic analysis of the isolates.

Colonies from agar plates were picked and used for colony PCR. Bacterial 16S rRNA gene primers were used for the amplification of approximately 1,400-bp DNA fragments (9bF and 1406uR) (12). For the determination and an approximate classification of 16S rRNA genes of different species, the PCR amplicons were analyzed by restriction length polymorphism using the enzymes HinFI and BsuRI (Fermentas GmbH, Germany), according to the manufacturer's instructions (45). Amplicons with an unknown restriction pattern were fully sequenced. For phylogenetic analyses, an alignment of approximately 100,000 homologous full and partial sequences available in public databases was used (the ARB project) (27). The new 16S rRNA gene sequences were added to the 16S rRNA alignment of the SILVA database (36) using the corresponding automated tools of the ARB software package (27). The resulting alignment was checked manually and corrected, if necessary. For tree reconstruction, methods were applied as implemented in the ARB software package.

Nucleotide sequence accession numbers.

The 16S rRNA gene sequences of the isolates were deposited in the NCBI nucleotide sequence database. The accession numbers are given in Fig. 1.

FIG. 1.

Phylogenetic tree. This maximum-parsimony tree is based on 16S rRNA gene sequences of cultivable bacterial strains isolated from the two different spacecraft assembly facilities. Besides the isolates (bold), the closest neighbors are shown. If two or more isolate names are given, this strain was isolated several times. However, the 16S rRNA gene sequence of only one representative was submitted to GenBank. The GenBank accession numbers are given in parentheses. The scale bar shows a 10% estimated difference in nucleotide sequence positions.

RESULTS

Three samplings in two clean rooms took place, and the surfaces of the facilities, as well as the spacecraft itself, were sampled thoroughly. In total, for the whole cultivation study, 31 samples were taken. During sampling, the facilities were in fully operating states and harbored the Herschel Space Observatory. Although Herschel does not demand planetary protection restrictions, the construction is done under strict contamination controls, with respect to particulates and molecular contamination, but without monitoring of the bioburden. As a consequence, the results represent a conservative estimate for the biodiversity present.

Samples were taken and stored under either anaerobic or aerobic conditions in parallel to ensure that sensitive microorganisms were not killed by oxygen when rehydrated. A complete overview of the samples, sampling areas, and characteristics is given in Table 1.

TABLE 1.

Clean room samples, sampling areas, and sample characteristics

Clean rooma	Sample nameb	Location	Sampling area (cm2)	Total sample vol (inoculation vol) (ml)c	Sampling device	Typed
FR1	SS	Spacecraft	3 × 25	10 (0.3)	3 pooled swabs	AN
	WW	Table, wheels	900	80 (2)	SpongeSicle	AN
	WS	Stairs, ladder	2,500	80 (2)	SpongeSicle	AN
	W3	Door knob	400	12 (0.4)	SpongeSicle	AE
	W6	Floor	3,600	12 (0.4)	SpongeSicle	AE
	W4	Entrance, floor	3,600	12 (0.4)	SpongeSicle	AE
FR2	W4	Floor	3,600	12.5 (0.5/0.2)	SpongeSicle	AE
	W5	Stairs	712	12.5 (0.5/0.2)	SpongeSicle	AE
	WA1	Floor	3,600	80 (1/0.2)	SpongeSicle	AN
	WA2	Floor	3,600	80 (1/0.2)	SpongeSicle	AN
	WA3	Floor	3,600	8 (1/0.2)	SpongeSicle	AN
ES	W1	Floor	3,600	11 (0.4/0.2)	SpongeSicle	AE
	W2	Entrance, floor	3,600	7.5 (0.2)	SpongeSicle	AE
	W3	Floor	2,840	9.3 (0.4/0.2)	SpongeSicle	AE
	WA1	Stairs	1,130	80 (1/0.2)	SpongeSicle	AN

^aFR1, Friedrichshafen sampling, April 2007; FR2, Friedrichshafen sampling, November 2007; ES, ESTEC sampling, March 2008.

^bOnly samples with positive anaerobic counts are given.

^cIf two volumes are given in parentheses, the first volume refers to the inoculation of liquid media and the second to that of solid media.

^dAN, anaerobic; AE, aerobic.

Diverse anaerobic media were chosen with different nutrient compositions and gas phases in order to cover the broadest diversity of anaerobic microorganisms present; they ranged from rich media (e.g., TG or TS) containing a complex mixture of carbon sources to media that provided CO₂ as the only carbon source (MM, AHM, and AAM). Other media focused on special physiological capacities of the microbes, e.g., nitrogen fixation or reduction of ferric iron and sulfate. In addition, a medium was selected that supports the enrichment of archaea (ASM), as it contains a mixture of antibiotics, repressing the growth of most bacteria.

This selection of different anaerobic media led to the successful cultivation of a variety of facultative anaerobes, aerotolerant anaerobes, and some strict anaerobes from the spacecraft-associated clean rooms at Friedrichshafen and ESTEC. Overall, 29 strains capable of anaerobic growth were isolated. The greatest number and diversity of organisms were obtained from the ISO 8 clean room at ESTEC, with 13 cultivated species, while 8 species each were obtained from the two samplings at Friedrichshafen. The total facultatively anaerobic and anaerobic microbial diversity obtained from the Herschel Space Observatory and its surrounding clean rooms is summarized in Fig. 1 and Table 2.

TABLE 2.

Cultivable facultatively anaerobic and anaerobic microbial diversity from European spacecraft-associated clean rooms

Clean roomc	Isolate	Representative straina	Source (sample)	Isolation medium	No. of isolates (%)b	Strict anaerobe
FR1	Bacillus thermoamylovorans	FR1_171P	WW	TG/TGA	2 (1.3)	No
	Cellulomonas parahominis	FR1_160	WS	TG	2 (1.3)	No
	Paenibacillus barengoltzii	FR1_164P	WW	TG	1 (0.6)	No
	Paenibacillus pasadenensis	FR1_1	W6	N2 fix	9 (5.7)	No
	Propionibacterium acnes	FR1_165P	WW, W3	TG/TGA	2 (1.3)	Yes
	Staphylococcus haemolyticus	FR1_68	WS, W3, W4, W6	TG/TGA	63 (39.9)	No
	Staphylococcus lugdunensis	FR1_146	WS	TG	1 (0.6)	No
	Staphylococcus pasteuri	FR1_75	SS	TG	6 (3.8)	No
FR2	Arsenicicoccus bolidensis	FR2_MS13	W5	MM	1 (0.8)	No
	Bacillus thuringiensis	FR2_MS1	W4	TS/TSA	2 (1.6)	No
	Dermabacter hominis	FR2_MS10	W5	SRA	1 (0.8)	No
	Propionibacterium acnes	FR2_MS11	WA1, W4, WA2	TS/TSA, SRA	5 (3.9)	Yes
	Pseudomonas luteola	FR2_MS17	W5	N2 fix	1 (0.8)	No
	Staphylococcus haemolyticus	FR2_MS3	W5, WA2	TS	7 (5.5)	No
	Staphylococcus pasteuri	FR2_MS2	WA3	TS	1 (0.8)	No
	Stenotrophomonas maltophilia	FR2_MS4	WA1	N2 fix	44 (34.7)	No
ES	Bacillus licheniformis	ES_MS4	W3	TS/TSA	4 (1.9)	No
	Bacillus subtilis	ES_MS5	W3	SRA	2 (0.9)	No
	Clostridium perfringens	ES_MS22	W1, W3	SRA	3 (1.4)	Yes
	Corynebacterium pseudogenitalium	ES_MS37	WA1	TSA, SRA	3 (1.4)	Yes
	Enterococcus faecalis	ES_MS6	W1	TS/TSA	2 (0.9)	No
	Enterococcus faecium	ES_MS1	WA1, W3	TS, ASM	2 (0.9)	No
	Paenibacillus ginsengisoli	ES_MS40	W1	AHM	1 (0.5)	No
	Paenibacillus sp.	ES_MS17	W3	TSA	1 (0.5)	No
	Pseudomonas sp.	ES_MS8	W3	N2 fix	1 (0.5)	No
	Propionibacterium acnes	ES_MS23	W1	SRA	2 (0.9)	Yes
	Propionibacterium avidum	ES_MS27	W3	SRA	2 (0.9)	Yes
	Staphylococcus haemolyticus	ES_MS10	W1, W2, W3, WA1	TS/TSA, SRA	9 (4.2)	No
	Staphylococcus pasteuri	ES_MS25	W1, W3	SRA	9 (4.2)	No

^aFor each isolate one representative strain is given.

^bTotal numbers of all isolates were as follows: for FR1, 158; FR2, 125; ES, 216 (data not shown); these totals also include all isolates of this species on aerobic media (for facultative anaerobes).

^cFR1, Friedrichshafen sampling, April 2007; FR2, Friedrichshafen sampling, November 2007; ES, ESTEC sampling, March 2008.

All control samples, including field blanks and negative controls, showed no growth of microorganisms. The strains isolated from the samples were identified as members of the bacterial phyla Firmicutes, Actinobacteria, and Gammaproteobacteria. The most frequently represented phylum, Firmicutes, included the genera Staphylococcus, Bacillus, Clostridium, Paenibacillus, and Enterococcus.

As shown in Fig. 2, the three sampled facilities showed major differences in their microbial diversities, with only a few common organisms detected in all three samplings. Interestingly, the three common species (Staphylococcus haemolyticus, Staphylococcus pasteuri, and Propionibacterium acnes) are known to be human-associated microorganisms, as was the overwhelming majority (85%) of all the facultatively anaerobic and anaerobic strains isolated during this study. Most identified species are detected generally in association with humans and/or are opportunistic pathogens (e.g., Propionibacterium acnes and Clostridium perfringens); only a minor portion of the isolates can be considered environmental microorganisms (e.g., Arsenicicoccus bolidensis).

FIG. 2.

Origin of organisms. Schematic drawing showing the location from which the different microorganisms were isolated. FR1, first sampling in Friedrichshafen (April 2007); FR2, second sampling (November 2007); ES, ESTEC sampling (March 2008). The middle box (gray) summarizes the organisms that were detected at all three samplings.

All strictly anaerobic strains isolated during this study were opportunistic pathogens. Most of them were obtained from the sampling at ESTEC (Clostridium perfringens, Propionibacterium avidum, Propionibacterium acnes, and Corynebacterium pseudogenitalium). However, within the overall, cultivable microbial burden, as determined in the frame of the whole biodiversity study (data not shown), the obligate anaerobic organisms represented only a minority of the isolated cultivable diversity (1 to 4%) (Fig. 3). In most cases, anaerobically enriched species were identified to be facultative or aerotolerant anaerobes, comprising 23 to 53% of all microbes isolated in the course of the entire study (Fig. 3).

FIG. 3.

Quantitative diagram of isolates and their relationship to oxygen. This diagram shows the abundances of microorganisms with different oxygen requirements after enrichment and cultivation on media with different oxygen content. Anaerobes were strict anaerobes, growing only on oxygen-free medium. Facultative anaerobes were isolated on anaerobic and aerobic media and were capable of growing under both conditions. Aerobic isolates were obtained only on aerobic media, and whether they also had the ability to grow under anaerobic conditions was not tested (data not shown.).

The majority of the isolated strains were obtained on anaerobic, complex, nutrient-rich media, indicating that they possess a fermentative metabolism or respire anaerobically without oxygen (Table 2). Nevertheless, several autotrophic organisms, which grew with CO₂ as the only carbon compound in the medium, were cultivated. Arsenicicoccus bolidensis was isolated on MM, originally designed to enrich methanogenic archaea. Paenibacillus ginsengisoli was isolated on AHM for homoacetogenic microorganisms. Four strains capable of fixing nitrogen were identified; Paenibacillus pasadenensis, Pseudomonas luteola, Stenotrophomonas maltophilia, and a Pseudomonas sp. were successfully isolated on medium lacking organic nitrogen. Also, an antibiotic-resistant strain was cultivated: Enterococcus faecium was grown on ASM which was supplemented with four different antibiotics intended to repress bacterial microbes. Although some media were specifically designed for the selective cultivation of archaea (e.g., methanogens), so far the cultivation of those organisms has not been successful.

Except for strain FR1_75 (Staphylococcus pasteuri), which was isolated from a surface sample of the Space Observatory, all other strains were obtained from samples from the clean room itself (floor, stairs, or other surfaces). The redox status of the sample seemed not to have a lot of influence on the cultivable bacterial diversity. Nevertheless, Corynebacterium pseudogenitalium (strain ES_MS37) was obtained only from an anaerobic sample (ESTEC, WA1). All other anaerobic or facultatively anaerobic strains were obtained from either aerobic or anaerobic samples. Most isolates were obtained from samples WW (table wheels) and WS (stairs) during the first sampling (FR1) and from sample W5 (stairs) during the second sampling (FR2). The majority of strains from ESTEC were observed in cultures of sample W3 (floor).

Colony counts obtained from samples spread on TSA or TGA plates allowed the estimation of the order of magnitude of the anaerobic, microbial load in the space-associated facilities. The counts in the samples from two facilities ranged from 1.7 × 10² to 7 × 10³ per m² (Table 3). The average colony count per m² was 1.4 × 10³ cells capable of growing on TGA or TSA under anaerobic conditions.

TABLE 3.

Number of anaerobes per m² sampling area

Clean rooma	Sample	Location	Anaerobic counts per m2b
FR1	WW	Table, wheels	1.3 × 103
	WS	Stairs, ladder	3.2 × 102
	W3	Door knob	7.5 × 102
FR2	W4	Floor	1.7 × 102
	WA1	Floor	1.1 × 103
ES	W1	Floor	5.5 × 102
	W2	Entrance, floor	2.1 × 102
	W3	Floor	6.5 × 102
	WA1	Stairs	7.1 × 103

^aFR1, Friedrichshafen sampling, April 2007; FR2, Friedrichshafen sampling, November 2007; ES, ESTEC sampling, March 2008.

^bColony counts on anaerobic TGA/TSA plates were determined for the given samples. All other anaerobic cultures grown on liquid medium or SRA could not be considered for this calculation.

Spore-forming species, belonging to the phylum Firmicutes, were successfully cultivated from samples of each facility. Besides the genus Bacillus, a well-known spore-forming organism frequently isolated from spacecraft-associated clean rooms, representatives of the genera Clostridium and Paenibacillus were found among our isolates.

Interestingly, the 16S rRNA gene sequences of the strains ES_MS8 (Pseudomonas sp.) and ES_MS17 (Paenibacillus sp.) showed a sequence similarity to the type strains of their nearest evolutionary neighbors (Pseudomonas luteola and Paenibacillus campinasensis, respectively) of less than 97.5%, and the strains can therefore be considered novel (41).

DISCUSSION

Anaerobes are considered one group of microorganisms that might be able to survive or even thrive in extraterrestrial environments, since the atmospheres of most of the planets and bodies within human reach contain only traces of oxygen. Even if the oxidizing, cold, and radiation-exposed surface of Mars is quite hostile, there are some regions on Mars (e.g., the subsurface) that could enable the existence of life (4). Recent experiments have shown that strictly anaerobic methanogenic archaea are able to grow on Mars soil simulant (21). Furthermore, some Martian regions resemble permafrost biotopes on Earth, which are the habitats for a broad variety of microbes. Besides psychrophilic, aerobic, and spore-forming species, anaerobes (e.g., methanogens, sulfate-reducing bacteria, and Clostridium spp.) have also been isolated from permafrost samples (42).

On Earth, anaerobes are generally widespread and can be found in, e.g., oxic soil, aerobic desert soils, or other habitats, such as the human body (22, 34, 43). In particular, the last makes them potential contaminants of spacecraft assembly facilities through the human workforce. For planetary protection concerns, a possible transfer of terrestrial life onto a planet's surface or an introduction of terrestrial life into Mars' subsurface during the landing and the impact of a spacecraft has to be considered. Facing ESA's ExoMars mission, we have screened two different European spacecraft-associated clean rooms in three samplings which concentrated on the cultivable anaerobic microbial community.

In general, two types of anaerobes are distinguished; they can grow either chemolithotrophically or chemoheterotrophically and use inorganic or organic molecules as electron donors, respectively. For that reason, different anaerobic media were employed for the enrichment of microorganisms from samples from spacecraft and their housings. Most of the media aiming to enrich chemolithotrophs were based on the chemical conditions that could be expected on Mars. The Martian atmosphere consists mostly of CO₂ and smaller amounts of N₂, H₂, O₂, CH₄, CO, and H₂O (48). The Martian soil contains mainly sulfates, manganese, and iron compounds (13, 46). Nevertheless, since these media are appropriate for metabolic specialists only, we have also used nutrient-rich media to cover anaerobically growing and fermenting microorganisms.

Media rich in inorganic compounds (TSA and TGA) were used for the estimation of the quantitative anaerobic microbial load. On average, about 1.4 × 10³ cells per m² clean room surface were identified to be able to grow on these media. Nevertheless, the overall cultivable microbial bioburden usually ranges from about 10⁴ to 10⁷ cells per m² (25).

Using our multiassay cultivation approach, a broad diversity of facultative anaerobes and aerotolerant anaerobes was detected in these specific environments. Some strictly anaerobic cultures were also obtained (Clostridium and Propionibacterium sp.), representing the first reported isolation of strictly anaerobic microorganisms from space-related environments.

Clostridium is a spore-forming microorganism and widespread in nature, but it is also known as a human-associated or even pathogenic microbe. Despite being considered a strict anaerobe, it has been reported that some Clostridium strains can tolerate oxygen under certain circumstances (14). Even though many propionibacteria were described to be aerotolerant anaerobes, our isolates turned out to be strict anaerobes, since no growth on aerobic media was observed. Interestingly, isolate ES_MS37 (Corynebacterium pseudogenitalium) also could not be grown under aerobic conditions, although corynebacteria have been described to be facultative anaerobes (6). Currently the reason for this observation is unclear.

All strictly anaerobic strains were isolated on complex media, rich in organic material, and thus a fermentative metabolism of these microbes is very likely. To our knowledge, Mars has no reserves of such organic material, and therefore, chemolithotrophic autotrophs could serve as pioneers, providing the basis of a food chain (44). For example, the capabilities to fix nitrogen from the gaseous atmosphere or to grow autotrophically on CO₂ are important properties of such primary producers. The activities of these microbes are the prerequisites for other microorganisms to colonize new nutrient-poor environments (44). Therefore, the assortment of anaerobic media tested during this study also included media selective for chemolithoautotrophs, e.g., those that provide CO₂ as the only carbon source as well as media containing sulfate as the possible electron acceptor. Media selective for nitrogen fixation were also applied, and several nitrogen-fixing microorganisms were detected in the course of this study. Paenibacillus pasadenensis, Pseudomonas luteola, Stenotrophomonas maltophilia, and a Pseudomonas sp. were isolated on liquid media, providing N₂ in the gas phase only. For all genera, this capability has already been described in other studies (5, 10, 38), whereas the autotrophic capabilities of Paenibacillus and Arsenicicoccus have not been reported yet. Members of these two genera have been isolated from samples collected in Friedrichshafen and at ESTEC (Paenibacillus ginsengisoli and Arsenicicoccus bolidensis) on media selective for autotrophic microorganisms. Arsenicicoccus bolidensis was described as a facultatively anaerobic environmental organism living mostly in contaminated lakes (7), whereas Paenibacillus seems to be widespread in nature, but it is also a common contaminant in spacecraft facilities (25).

Some of our isolates were originally enriched on medium containing sulfate as a possible electron acceptor under anaerobic conditions. Nevertheless, none of our isolates appeared to be a sulfate-reducing organism. This is because the colonies on the agar plates (containing iron compounds) did not turn black during growth, indicating sulfate reduction and the precipitation of FeS. It can therefore be argued that these isolates metabolized the organic compounds provided.

In this study, the facultative anaerobes dominated the bacterial community in Friedrichshafen (first sampling) and were also very prevalent in the other samples (45% and 22.6%). This might be due to the fact that the overwhelming majority of all isolates can be attributed to the mostly facultatively anaerobic human microbiota. Even for the Viking mission in 1975, a predominance of human-associated bacteria was reported and Staphylococcus strains were most frequently isolated (37). This is still the case for screenings of spacecraft-associated clean rooms nowadays (25).

Although many anaerobes are non-spore-forming microorganisms (e.g., methanogens or homoacetogens), some genera are known to be capable of forming spores. In particular, spores of the facultatively anaerobic genus Bacillus are a main consideration of planetary protection. Bacillus spores and their resistances have been studied in detail. For example, spores of Bacillus subtilis have been reactivated after exposure to the space environment for over 6 years onboard the Long Duration Exposure Facility (LDEF) (16). Besides bacilli, clostridia and paenibacilli have been detected during our study. Clostridium spores have been described to be very hardy. They can resist boiling for several hours (32, 39) and are resistant to desiccation, certain chemicals, and UV radiation (33). However, like bacilli, clostridia require complex carbon sources for growth. On the other hand, spore-forming paenibacilli seem to be able to deal with nutrient-poor environments, as shown in the present study, in which obtained isolates were able to fix nitrogen or to grow autotrophically. While very limited information on the resistance of spores from paenibacilli is available in literature, our preliminary studies hint at an extraordinary heat resistance of some species (data not shown).

Previous cultivation studies of microbial communities from spacecraft-associated clean rooms focused on the cultivation of aerobic and spore-forming members of the Bacteria. Thus, they used mainly one nutrient-rich, aerobic medium, namely, TS/TSA (24, 47). One recent study reported the detection of a very broad, even extremotolerant microbial community in these facilities through various commercially available R2A media in pH, salt, or other conditions (25).

Here, the usage of noncommercial media and the application of the Hungate anaerobic technology led to the discovery of an even broader diversity. Members of the bacterial genera Clostridium, Propionibacterium, Arsenicicoccus, Dermabacter, and Enterococcus were isolated from spacecraft-associated environments for the first time, although Clostridium, Propionibacterium, and Enterococcus have already been detected via molecular diagnostic methods in former studies (23, 29). In addition, two strains representing novel species (Paenibacillus sp. and Pseudomonas sp.) were obtained.

Our results indicate that the facultatively anaerobic and anaerobic microbial community is large and maybe even dominant in spacecraft assembly facilities. So far, the importance of anaerobes can only be estimated, and it is unclear if the organisms identified here can survive a space flight or even thrive in extraterrestrial environments. Nevertheless, their metabolic activity and versatility, as well as ability to grow and proliferate, have been demonstrated, and their presence in these critical environments must not be ignored.