SEARCH

SEARCH BY CITATION

Keywords:

  • 16S rDNA;
  • biodeterioration;
  • epilithic biofilm;
  • ITS region;
  • microbial diversity;
  • phylogenetic analysis;
  • restriction analyses of amplified ribosomal DNA

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Aims:  The aim of the present study was to reveal the microbial genetic diversity of epilithic biofilms using a DNA-based procedure.

Methods and Results:  A DNA extraction protocol was first selected to obtain PCR-amplifiable metagenomic DNA from a limestone biofilm. Extracted DNA was used to amplify either 16S rRNA genes or ITS regions from prokaryotic and eukaryotic genomes, respectively. Amplified DNAs were subsequently cloned, amplified by colony PCR and screened by restriction analysis [restriction analyses of amplified ribosomal DNA (ARDRA)] for DNA sequencing. Phylogenetic analysis using 16S rDNA sequences showed that predominating bacteria were Alphaproteobacteria belonging to the genera Sphingomonas, Erythrobacter, Porphyrobacter, Rhodopila and Jannashia; Cyanobacteria and Actinobacteria were also identified. Analysis of ITS rDNA sequences revealed the presence of algae of the Chlorophyceae family and fungi related either to Rhinocladiella or to a melanized ascomycete. Statistical analysis showed that the specific richness evidenced was representative of the original sampled biofilm.

Conclusions:  The molecular methodology developed here constitutes a valuable tool to investigate the genetic diversity of microbial biofilms from building stone.

Significance and Impact of the Study:  The easy-to-run molecular method described here has practical importance to establish microbiological diagnosis and to define strategies for protection and restoration of stone surfaces.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Stone monuments and buildings are continuously exposed to weathering agents such as air, sunlight and rain. Although ineluctable, evolution of stone decay depends on its mineral composition and environmental conditions, mainly influenced by climate and human activities (Warscheid and Braams 2000). Moreover, besides physical and chemical factors, biological agents also play an important role in the deterioration of stone material. Knowledge of such processes is crucial to define strategies for restoration and conservation of historic monuments, and thus, is an economic issue (Wakefield and Jones 1998).

Micro-organisms classically found on building surfaces are autotrophic and heterotrophic bacteria, fungi, algae and lichens (Gorbushina 2007). This consortium of heterogeneous microbial species forms a biofilm whose internal cohesion and adhesion to underlying surface are ensured by extracellular polymeric substances. Main detrimental effects of microbial biofilms are: (i) discolouration of stone surface because of the presence of photosynthetic or protective pigments, (ii) physical alteration of the material structure by penetration of bacterial and fungal hyphae and by differential mechanical pressure imposed by shrinking and swelling cycles of the adhesive biofilms, and (iii) chemical modification of the mineral support by acidolytic and oxidoreductive corrosion processes generated by products of the microbial metabolism (Warscheid and Braams 2000).

The characterization of microbial communities colonizing stone surfaces is a prerequisite to monitor deteriogenic micro-organisms and to set up suitable treatments to circumvent their undesirable effects. Because culture-dependent methods conduce inevitably to underestimate the microbial diversity (Ward et al. 1990), DNA-based typing methods have been widely used to study the microbial diversity of communities from various ecological niches. However, it should be noted that molecular techniques share some limitations that are mainly caused by inherent biases of the PCR technology, as primer specificity or formation of chimeric sequences (Nocker et al. 2007). As yet, only few reports mentioned their use in the genotyping of biofilms present at the surface of building stones (Gonzàlez and Saiz-Jiménez 2005). Moreover, all previous molecular studies were focused on specific components of stone biofilms – for example, bacteria (Zimmermann et al. 2005; McNamara et al. 2006; Imperi et al. 2007) or fungi (Möhlenhoff et al. 2001) – but to our knowledge, studies concerning the entire consortium remained very scarce. The objective of this study is to develop an easy-to-run molecular procedure suitable to assess the genetic diversity of the overall microbial community from building stone biofilms.

Materials and methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Sampling

Sampled microbial biofilms appeared as rosy areas on the face-wall of ‘Villa Isabel’, a private house located in the centre of Biarritz city (France). Samples were taken by scraping off limestone material with sterile scalpels and microtubes, maintained at 4°C during the transport to the laboratory and stored at −80°C until use.

DNA extraction and purification

DNA extractions were carried out using two commercial kits: the Easy-DNA™ kit (Invitrogen, Carlsbad, CA, USA) and the UltraClean™ Soil DNA Isolation kit (MoBio Laboratories Inc, Solana Beach, CA, USA), following furnisher’s instructions. The two conditions of cell lysis proposed by the furnisher of the later DNA extraction kit were essayed, consisting in submitting cells either to bead beating for 5 min at room temperature or to four cycles of heating (65°C, 5 min) and brief (5 s) bead beating (soft lysis). In any cases, DNA extractions started by resuspending 100 mg of sample material in lysis solution.

rDNA amplification

The 16S rDNA was amplified by PCR using primers 8F (5′-AGAGTTTGATCCTGGCTCAG-3′) and 1489R (5′-TACCTTGTTACGACTTCA-3′), previously designed by Weisburg et al. (1991). For PCR amplification of the ITS region, primers ITS1 (5′-TCCGTAGGTGAACCTGCGG-3′) and ITS4 (5′-TCCTCCGCTTATTGATATGC-3′) were used (White et al. 1990). PCRs were carried out in 50 μl final volume containing 20–50 ng of extracted DNA, 20 pmol of each primer, 100 μmol l−1 of each deoxyribonucleotide triphosphate, 5 μl of 10× buffer (500 mmol l−1 KCl, 15 mmol l−1 MgCl2, 100 mmol l−1 Tris–HCl pH 8·3) and 2·5 U of Taq DNA polymerase (5 PRIME Inc., Gaithersburg, MD, USA). PCR was performed in a robocycler (Themocycler PTC-100; MJ Research, Waltham, MA) with the following thermocycling programme: 5 min denaturation at 94°C, followed by 35 cycles at 94°C for 1 min, 48°C for 1 min and 72°C for 1 min. 10 min at 72°C was used as a final extension step. Amplified DNA was separated by electrophoresis in TEA 1% w/v agarose gel, cut out and purified on column using the GFX™ PCR and DNA Gel Band Purification kit (GE Healthcare Life Sciences, Uppsala, Sweden). 16S rDNA PCR amplifications gave an amplimer of 1500 bp. ITS rDNA PCR amplifications yielded two amplimers of 650 and 750 bp, which were grouped before purification.

rDNA cloning

Four microlitres of purified PCR products was cloned with the pCR 2·1-TOPO vector (Invitrogen), following the manufacturer’s protocol. The ligation mixture was transformed into One Shot® TOP10 Electrocomp™E. coli cells (Invitrogen), and recombinant clones were selected on LB solid medium containing ampicillin (100 μg ml−1) and X-Gal (0·1 mmol l−1).

Inserts amplification and purification

White colonies randomly picked were screened for inserts by performing colony PCR with the vector specific primers M13 F (5′-GTAAAACGACGGCCAG-3′) and M13 R (5′-CAGGAAACAGCTATGAC-3′). PCRs were carried out in 25 μl final volume containing 20 pmol of each primer, 100 μmol l−1 of each deoxyribonucleotide triphosphate, 2.5 μl of 10× buffer, 1·25 U of Taq DNA polymerase (5 PRIME Inc.). Thermal cycling conditions were as follows: 10 min denaturation at 94°C, followed by 35 cycles at 94°C for 45 s, 56°C for 45 s and 72°C for 1 min. A final extension step of 10 min at 72°C was applied. The amplimer of all clones containing inserts of the correct size was purified using GFXTM PCR and Gel Band Purification kit (GE Healthcare Life Sciences), following the manufacturer’s protocol.

Restriction analyses of amplified ribosomal DNA (ARDRA)

Purified amplimers were separately digested with restriction enzymes HinfI and HaeIII (10U, New England Biolabs) for 4 h at 37°C. The restricted DNA fragments were separated on a 1·5% w/v agarose gel containing ethidium bromide (0·5 μg ml−1). After migration in TBE buffer at 25 V/h for 1 h 30 min, DNA fragments were visualized under UV illumination at 312 nm. The restriction profiles were captured as JPEG format with a CDD camera and compared manually for grouping clones into ribotypes.

Sequencing and phylogenetic analysis

rDNA inserts representative of each distinct ARDRA pattern were selected and sequenced with Big Dye Terminator ver. 1·1 cycle sequencing kit (Applied Biosystems, Foster City, CA). Concretely, one DNA clone was sequenced when the number of clones grouped under the same ARDRA pattern ranged between 1 and 3, two when it ranged between 4 and 10, three when it ranged between 10 and 20, four when it ranged between 20 and 30 and five when it encompassed 30. Sequencing reactions were analysed with ABI 330X sequencer at the Genotyping and Sequencing facility of Bordeaux (France). The presence of possible chimeric DNA sequences was investigated by using the chimera_check program of the Ribosomal Database Project II (Cole et al. 2003). Clones containing chimeric sequences were excluded from further analysis. All DNA sequences were compared with sequences in the Genbank/EMBL/DDBJ database using BlastN (Basic Local Alignment Search Tool, http://blast.ncbi.nih.gov/Blast.cgi). 16S or ITS rDNA clones were clustered into phylotypes at an overlap percentage identity cut-off of 97 using ClustalW2 program (http://www.ebi.ac.uk/Tools/clustalw2/index.html?). Evolutionary distances were calculated by the Kimura-2-parameter algorithm, and the phylogenetic trees were constructed by the neighbour-joining method, using Mega3 program, ver. 3·0 (Kumar et al. 2004). Confidence in tree topology was determined by bootstrap analysis using 100 resamplings of the sequence. Sequences reported in this study can be accessed under numbers FJ028664 to FJ028676 and FJ028701 to FJ028712.

Statistical analysis of phylotype richness from DNA clone libraries

To assess whether DNA clone libraries were large enough to be representative of phylotype richness from environmental samples, we used the web interface constructed by Kemp and Aller (2004) and located at http://www.aslo.org/lomethods/free/2004/0114a.html. Author instructions were followed for calculations using the abundance-based richness estimator SChao1 and Good’s C index of coverage (Kemp and Aller 2004).

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Whole community DNA extraction from epilithic biofilms

Comparison of DNA preparations obtained from the different procedures (Table 1) showed that DNA extraction with the UltraClean™ Soil DNA Isolation kit using a soft cell lysis procedure yielded pure and stable DNA of high molecular weight from stone microbial biofilms.

Table 1.   DNA extraction protocol comparison
DNA extraction protocolPurity (A260/A280)DNA fragment length Stability*
  1. *Estimated by electrophoresis analysis after a storage for 6 days at 4°C.

Easy-DNA™ kit1·7–1·8>23 kbpNo
UltraClean™ Soil DNA Isolation kit1·7–1·8∼20 kbpYes
UltraClean™ Soil DNA Isolation kit with soft lysis1·7–1·8>23 kbpYes

Phylogenetic analysis of 16S and ITS rDNA

Analysis of ARDRA patterns from cloned 16S rRNA genes (SLB clones) and ITS regions (ILB clones) allowed to group 67 SLB clones into eight ribotypes and 76 ILB clones into three ribotypes (Figs 1 and 2). In order to further characterize microbial communities from building stone biofilms under study, DNA clones with identical ARDRA patterns were randomly selected and sequenced. A total of 27 clones were partially sequenced, two of them contained chimeric sequences and were discarded from phylogenetic analysis. Pairwise alignments of DNA clone sequences using 97% similarity as the cut-off allowed to discriminate 11 phylotypes: SLB clones clustered into eight phylotypes and ILB clones into 3 (Figs 1 and 2).

image

Figure 1.  Phylogenetic tree of partial 16S rDNA gene sequences (613 bp) of SLB clones isolated from microbial community of a limestone biofilm. Percentages of 100 bootstrap resamplings that support branching points above 50% confidence are indicated. GenBank accession numbers are given in parentheses. The number of clones having the same ribotype is indicated within grey boxes. The number of clones grouped into the same phylotype is given within black boxes. Scale bar represents 0·02 nucleotide changes per position.

Download figure to PowerPoint

image

Figure 2.  Phylogenetic tree of partial ITS region sequences (513 bp) of ILB clones isolated from microbial community of a limestone biofilm. Percentages of 100 bootstrap resamplings that support branching points above 50% confidence are indicated. GenBank accession numbers are given in parentheses. The number of clones having the same ribotype is indicated within grey boxes. The number of clones grouped into the same phylotype is given within black boxes. Scale bar represents 0·05 nucleotide changes per position.

Download figure to PowerPoint

Phylogenetic analysis of SLB clones (16S rRNA genes from limestone biofilm, Fig. 1) revealed that 40% of clones were distantly related to chloroplast rDNA sequences of an unknown alga (79% identity with chloroplastic rDNA sequence of Monomorphina pyrum). Other clones belonged to Alphaproteobacteria division (55%), Cyanobacteria (3%) and Actinobacteria (2%). The two most important groups of sequences that fell into the Alphaproteobacteria division were related to Sphingomonas sp. KIN169 (20 clones, 93–95% identity) and to Erythrobacter sp. R14 (11 clones, 96–100% identity), two species of the Sphingomonadaceae family. Among minor groups of sequences also belonging to Alphaproteobacteria division, one is similar (97%) to an uncultured bacterium belonging to the genus Porphyrobacter, a second to an uncultured bacterium of the Rhodopila genus (95%) and a third to Jannaschia sp. RP11 (99% identity). The clone SLB-73, affiliated to Actinobacteria (High GC gram-positive bacteria), is closely related (98% identity) to an uncultured bacterium of the genus Rubrobacter already identified on limestone monuments (Ortega-Morales et al. 2004). The 16S rDNA sequences of two clones matched (98% identity) with that of an uncultured cyanobaterium.

DNA sequencing of eukaryotic rDNA sequences (ITS regions) isolated from limestone biofilm allowed to separate ILB clones into three phylotypes (Fig. 2). The most abundant comprised 37 clones that were affiliated to an uncultured chlorobiont of the Chlorophyceae family (94% identity). The two other phylotypes contained fungal ITS sequences related either to Rhinocladiella sp. (24 clones, 86–95% identity) or to a melanized limestone ascomycete (15 clones, 95% identity).

Reliability of phylotype richness

Phylotype richness (SChao1 estimator) was compared against sampling effort (library size) using the free software tool developed by Kemp and Aller (2004). The obtained curves (Fig. 3) showed that in all cases, richness estimates reached a stable asymptotic value indicating that the number of clones analysed were sufficient to yield a reliable estimate of phylotype richness. Richness estimates (9·9 and 3 for SLB and ILB clone library, respectively) were identical or slightly greater than the actual number of phylotypes (9 and 3 for SLB and ILB clone library, respectively). Calculations of Good’s C indexes (Fig. 3) confirmed that clone libraries were large enough to capture most of the diversity of microbial biofilms sampled. Coverage was 0·91 and 1 for SLB and ILB clone library, respectively.

image

Figure 3.  Estimated phylotype richness (left panel) and coverage (right panel) vs library size for the 16S rDNA clone library (SLB clones, filled symbols) and the ITS rDNA clone library (ILB clones, open symbols).

Download figure to PowerPoint

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

The goal of this study was to develop a molecular methodology, based on a single DNA extraction procedure and to assess the genetic diversity of all members (prokaryotes and eukaryotes) of microbial communities living at the surface of building stones. As previously reported for samples derived from soil or concrete (Yeates and Gillings 1998; Giannantonio et al. 2009), the bead beating DNA extraction procedure we used in this study produced high quality DNA suitable for PCR amplification and cloning of either prokaryotic 16S rRNA genes or eukaryotic ITS regions. It appeared efficient to isolated DNA from any micro-organism present at the surface of building stone, including DNA from organisms that are hard to lyse, such as Actinobacteria (Kauffmann et al. 2004) or thick-walled Cyanobacteria. Although Cyanobacteria were poorly represented in the studied biofilm, this could not be because of a lack of the DNA extraction method because its use to characterize a sandstone biofilm containing 60% of Cyanobacteria was conclusive (data not shown).

Phylogenetic analysis of partial 16S rRNA gene sequences showed that Alphaproteobacteria are predominant in the studied limestone biofilm, as in Altamira cave biofilms within which they have been considered as the most metabolically active group (Portillo et al. 2009). Most of them are affiliated either to the halotolerant Sphingomonas sp. (Pinhassi and Berman 2003) or to the marine Erythrobacter sp. (Yurkov and Beatty 1998), indicating that most of bacteria present within the investigated biofilm are well adapted to the coastal environment where they originate from. Phototrophs such as Cyanobacteria and algae where also identified. They are supposed to be the first colonizers, and so, to support the growth of subsequent heterotrophs (bacteria and fungi). Analysis of ITS region of rRNA genes confirmed the presence of algae and allowed to identify two ascomycetous fungi previously described as typical inhabitants of stone surfaces (Chertov et al. 2004; Sert et al. 2007). On the whole, the combined 16S and ITS DNA analysis revealed a relatively low microbial diversity (11 phylotypes). However, this result is not surprising, as it could be assumed that stone surfaces are oligotrophic substrata. Comparable specific richnesses were reported following DNA analyses of microbial biofilms from frescoes or concrete surfaces (Imperi et al. 2007; Giannantonio et al. 2009). Statistical evaluation of phylotype richness of 16S and ITS rDNA clone libraries confirmed that microbial diversity evidenced in our study was representative of the original stone biofilm. As mentioned by Kemp and Aller (2004), such an evaluation is missing in the most of the published works dealing with the assessment of microbial diversity.

Concerning the methodology used in this study, our results showed that clones grouped under a same ARDRA pattern were closely related to each other, and thus, confirmed that ARDRA profiling is a good screening tool to identify clones of interest for sequencing (Nocker et al. 2007). This method of screening has often been used to investigate the microbial diversity living at the surface of stone material (Imperi et al. 2007; Giannantonio et al. 2009), but DGGE has frequently been used too (Möhlenhoff et al. 2001; Imperi et al. 2007); besides this, some studies have been performed by direct sequencing of rDNA cloned sequences (Zimmermann et al. 2005; McNamara et al. 2006). Such a practice may become more and more spread as costs for DNA sequencing are decreasing.

In conclusion, whether a screening step is used or not, the development of any molecular procedure suitable to assess the genetic diversity of microbial communities needs to ensure that DNA isolation is efficient for the totality of members who make up the consortium and that the number of analysed clones is sufficient to yield a genetic diversity representative from the original sample. As it was the case for the molecular methodology described here, we adopted it for further studies. Its use in preliminary studies designed to follow microbial colonization of building stone revealed that it was convenient to identify micro-organisms even on poorly colonized stones (data not shown).

Acknowledgements

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

This study was carried out with financial support from the Conseil des Elus du Pays Basque. MB thanks the Région Aquitaine and the Agence Nationale de la Recherche et de la Technologie for PhD fellowship. The authors are grateful to J. Valero (Inasmet, Spain) for valuable discussions and help in sampling.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  • Chertov, O., Gorbushina, A.A. and Deventer, B. (2004) A model for microcolonial fungi growth on rock surfaces. Ecol Modell 177, 415426.
  • Cole, J.R., Chai, B., Marsch, T.L., Farris, R.J., Wang, Q., Kulam, S.A., Chandra, S., McGarrell, D.M. et al. (2003) The ribosomal database project (RDP-II): previewing a new autoaligner that allows regular updates and the new prokaryotic taxonomy. Nucleic Acids Res 31, 442443.
  • Giannantonio, D.J., Kurth, J.C., Kimberly, E.K. and Sobecky, P.A. (2009) Molecular characterizations of microbial communities fouling painted and unpainted concrete structures. Int Biodeterior Biodegrad 63, 3040.
  • Gonzàlez, J.M. and Saiz-Jiménez, C. (2005) Application of molecular nucleic acid-based techniques for the study of microbial communities in monuments and artworks. Int Microbiol 8, 189194.
  • Gorbushina, A. (2007) Life on rocks. Environ Microbiol 9, 16131631.
  • Imperi, F., Caneva, G., Cancellieri, L., Ricci, M.A., Sodo, A. and Visca, P. (2007) The bacterial aetiology of rosy discolouration of ancient wall paintings. Environ Microbiol 9, 28942902.
  • Kauffmann, I.M., Schmitt, J. and Schmid, R.D. (2004) DNA isolation from soil samples for cloning in different hosts. Appl Microbiol Biotechnol 64, 665670.
  • Kemp, P.F. and Aller, J.Y. (2004) Estimating prokaryotic diversity: when are 16S rDNA libraries large enough? Limnol Oceanogr Methods 2, 114125.
  • Kumar, S., Tamura, K. and Nei, M. (2004) MEGA3: integrated software for molecular evolutionary genetics analysis and sequence alignment. Brief Bioinform 5, 150163.
  • McNamara, C.J., Perry, T.D. IV, Bearce, K.A., Hernandez-Duque, G. and Mitchell, R. (2006) Epilithic and endolithic bacterial communities in limestone from a Maya archaeological site. Microbial Ecol 51, 5164.
  • Möhlenhoff, P., Müller, L., Gorbushina, A. and Petersen, K. (2001) Molecular approach to the characterisation of fungal communities: methods for DNA extraction, PCR amplification and DGGE analysis of painted art objects. FEMS Microbiol Lett 195, 169173.
  • Nocker, A., Burr, M. and Camper, A.K. (2007) Genotypic microbial community profiling: a critical technical review. Microb Ecol 54, 276289.
  • Ortega-Morales, B.O., Narvàez-Zapata, J.A., Schmalenberger, A., Sosa-Lopez, A. and Tebbe, C.C. (2004) Biofilms fouling ancient limestone Mayan monuments in Uxmal, Mexico: a cultivation-independent analysis. Biofilms 1, 7990.
  • Pinhassi, J. and Berman, T. (2003) Differential growth response of colony-forming α- and γ-Proteobacteria in dilution culture and nutrient addition experiments from lake Kinneret (Israel), the eastern mediterranean sea, and the gulf of Eilat. Appl Environ Microbiol 69, 199211.
  • Portillo, M.C., Saiz-Jimenez, C. and Gonzalez, J.M. (2009) Molecular characterization of total and metabolically active bacterial communities of “white colonizations” in the Altamira cave, Spain. Res Microbiol 160, 4147.
  • Sert, H.B., Sumbul, H. and Sterflinger, K. (2007) Microcolonial fungi from antique marbles in Perge/Side/Termessos (Antalya/Turkey). Antonie Van Leeuwenhoek 91(3), 217227.
  • Wakefield, R. and Jones, M. (1998) An introduction to stone colonizing micro-organisms and biodeterioration of building stone. Q J Eng Geol 31, 301313.
  • Ward, D.M., Weller, R. and Bateson, M.M. (1990) 16S rRNA sequences reveal numerous uncultured microorganisms in a natural community. Nature (London) 345, 6365.
  • Warscheid, T. and Braams, J. (2000) Biodeterioration of stone: a review. Int Biodeterior Biodegrad 46, 343363.
  • Weisburg, W.G., Barns, S.M., Pelletier, D.A. and Lane, D.J. (1991) 16S ribosomal DNA amplification for phylogenetic study. J Bacteriol 173, 697703.
  • White, T.J., Bruns, T.D., Lee, S. and Taylor, J. (1990) Analysis of phylogenetic relationships by amplification and direct sequencing of ribosomal RNA genes. In PCR Protocols: a Guide to Methods and Applications ed. Innis, M.A., Gelfand, D.H., Sninsky, J.J. and White, T.J. pp. 315322. New York: Academic Press.
  • Yeates, C. and Gillings, M.R. (1998) Rapid purification of DNA from soil for molecular biodiversity analysis. Lett Appl Microbiol 27, 4953.
  • Yurkov, V.V. and Beatty, J.T. (1998) Aerobic anoxygenic phototrophic bacteria. Microbiol Mol Biol Rev 62, 695724.
  • Zimmermann, J., Gonzàlez, J.M. and Saiz-Jiménez, C. (2005) Epilithic biofilms in Saint Callixtus Catacombs (Rome) harbour a broad spectrum of Acidobacteria. Antonie Van Leeuwenhoek 89, 203208.