Next-Generation Sequencing for Microbial Characterization Of

Next-Generation Sequencing For Microbial Characterization Of Biovermiculations From A Sulfuric Acid Cave In Apulia (Italy) Ilenia M.D’Angeli1, Jo De Waele1, Maria Grazia Ieva1, Stefan Leuko2, Martina Cappelletti3, Mario Parise4, Valme Jurado5, Ana Z. Miller5, Cesareo Saiz-Jimenez5 Afliation: 1Department of Biological, Geological and Environmental Sciences, University of Bologna, Italy, ilenia.dangeli@alice. it; [email protected] 2DLR Deutsches Zentrum für Luf- und Raumfahrt Institute of Aerospace Medicine Radiation Biology, Köln, Germany, [email protected] 3Department of Pharmacy and Biotechnology, University of Bologna, Italy, [email protected] 4Department of Earth and Environmental Sciences, University “Aldo Moro” of Bari, Italy (ex CNR-IRPI Bari, Italy), [email protected] 5Instituto de Recursos Naturales y Agrobiologia, IRNAS-CSIC, Sevilla, Spain, [email protected]; anamiller@ irnas.csic.es; [email protected]

Abstract Sulfuric acid cave systems host abundant microbial communities that can colonize several environments displaying a variety of morphologies, i.e. white flamentous mats foating on the water surface, white creamy moonmilk deposits on the walls, and biovermiculations. Up to date, only few reports have described the microbiological aspects behind biovermiculation geomicrobiology of Italian sulfuric acid caves despite their overall abundance.

Here, we present the frst characterization of biovermiculation microbial populations from the Santa Cesarea Terme (Apulia, Italy) using next-generation sequencing and feld emission scanning electron microscopy (FESEM) approaches. We focused our analysis on biovermiculations from Fetida Cave located along the Adriatic Sea coastline. Tis cave is at sea level, and moving from the entrance to its inner part, it is possible to observe a decrease of marine infuence accompanied by a corresponding increase in the acidic efect of the upwelling waters. Biovermiculations are copiously observed covering the ceiling and walls mainly in the inner and middle environments of the cave, while they are absent near the entrance. Biovermiculations have a widespread range of colors going from grey to dark brown with an overall slimy appearance; dendritic morphologies alternate to very dense wall-covering sheets. Total DNA was extracted from each sample and 16S rDNA sequences were analyzed through Ilumina MiSeq platform. Te main lineages found in all the samples included Gammaproteobacteria, Betaproteobacteria, Alphaproteobacteria, Bacteroidetes, Planctomycetes, Actinobacteria and Acidobacteria. In particular, the samples from the inner part of the cave had the highest abundance of Acidobacteria and showed the presence of Epsilonproteobacteria that can be related to bacterial sulfur- oxidizing activity. FESEM images revealed microbial mats composed of flamentous cells including Actinobacteria and reticulated flaments of unknown origin, as well as prosthecate bacteria.

Tis study shows that the biovermiculation microbial communities from Fetida cave is in accordance with previous data reported in acidic caves from Italy and support the role of Epsilonproteobacteria typically occurring in sulfdic environments.

Keywords: geomicrobiology, Santa Cesarea Terme, hypogenic cave, FESEM, 16S rDNA

1. Introduction caves to the thermal acidic ones, to lava tubes and any other underground systems. In the last decade microbiological studies of hypogean environments have improved the overall knowledge on the pres- In this work we describe biovermiculation microbial commu- ence of extremophile microorganisms able to live in extreme nities from an active sulfuric acid cave, called Fetida, which conditions such as: absence of light, low organic compounds, is located in the southeastern area of the Apulian foreland, high humidity, possible high gases concentration in cave in Santa Cesarea Terme (Italy). Te microbiology of biover- atmosphere, extreme variability in pH and temperature miculations was previously described by Jones et al. (2008) (Tomczyk-Żak and Zielenkiewicz 2015). Hypogean sys- who reported details on these formations from the Frasassi tems are also getting attention as Mars analogues, and cave cave system (Italy). Sulfuric acid cave systems are hypo- biosignature suites have been used for astrobiological stud- gean environments (Klimchouk 2009), characterized by ies (Boston et al. 2004). In these environments, the microbial high H2S concentration that, under oxidizing conditions, organization in collective structures supports the cooperation tends to become H2SO4 (Egemeier 1981). Te interaction into mutualistic relationships favoring the bacterial survival of H2SO4 with the host rocks (mainly carbonates) tends to and resulting in the formation of features that can be observed form gypsum replacements and carbonic acid. Tis oxidiz- within a cave (Grotenhuis et al., 1991). Indeed, microbial ing environment provides an important source of energy for activities are associated with the formation of diferent spe- extremophile microbes (Jones et al. 2008), which can colo- leothems and biomineralization processes (Tisato et al. 2015; nize the sulfuric acid environments by creating diferent types Bontognali et al. 2016). Bacteria can colonize diferent types of microbial mats such as white sof flaments well visible in of subterranean environments from the normal epigenic sulfdic water (Engel et al. 2004), moonmilk deposits (Cañav- eras et al. 2001) and biovermiculations on walls and ceilings

Proceedings of the 17th International Congress of Speleology 377 Figure 1. Location of Santa Cesarea Terme (Apulian Foreland, Italy)

(Hose and Pisarowicz, 1999). In Fetida Cave, the walls and ceilings from the inner and middle halls are copiously covered by slimy microbial materials showing a wide range of colors going from grey to brown- reddish and dark-brown tones, characterized by dendritic morphologies alternating to dense wall-covering sheets. In this study, next generation sequencing and FESEM analyses have been conducted to character- ize the biovermiculation microbiological diversity in three samples collected from diferent sampling areas inside Fetida Cave (Santa Cesarea Terme).

2. Geological setting Santa Cesarea Terme (Lecce, Italy) is located along the Adri- atic coastline (Fig.1), where four active sulfuric acid caves are developed in Cretaceous carbonate rocks (limestones and Figure 2. Samples BV1, CV1 and CV2 were collected in the inner- dolostones) called “Calcari di Altamura” (Azzaroli 1967). middle part of the cave, where sulfdic rising waters reach the surface, Santa Cesarea Terme represents a commercially exploited spa. meet sea water and degas producing an atmosphere rich in H2S. Te relative abundance diferent phyla detected through Illumina MiSeq Te caves open at the sea level and are infuenced by marine platform sequencing is shown as a pie chart. Te group “Others” and upwelling sulfdic waters that meet creating an aggres- includes the phyla < 2% of the total community. sive mixed solution. Tey represent very peculiar environments where it is possible to observe marine (mainly close the seawater, and degassing in the atmosphere. Oxygen pres- to the entrance) and hypogenic evidences (in the inner and ence in this type of environment is reduced. middle zones). Tese caves host a variety of sulfuric acid spe- leogenetic (SAS) minerals, in particular sulfur (that covers Using the Illumina MiSeq next generation sequencing walls and ceiling), gypsum and jarosite deposits. Microbio- approach we assessed the microbial diversity in the three logical materials are abundant; it is possible to see white sof samples. Te produced data were analyzed using publicly flaments in the acidic waters, white moonmilk deposits and available algorithms and analysis pipeline, Mothur, following slimy biovermiculations (Fig.2). the MiSeq standard operating procedure (SOP) (Schloss et al., 2009, SOP accessed 23.5.2016). In short, the paired end Geochemical analyses of waters have been conducted: they reads were joined together, and the produced sequences were can be classifed as Na-Cl-SO4 waters, with a temperature quality checked. Chimeric sequences were identifed and ranging between 25 and 29°C, and pH of 6.7 - 7.2. removed, and the sequences were clustered into OTUs using average neighbour algorithm. Taxonomic assignment was Biovermiculation pH measured with litmus papers ranges performed by querying the sequence reads against a silva SSU between 5 and 5.5. 123 reference database (Quast et al., 2013) and various diversity indices and richness estimates were calculated. 3. Results Te highest number of OTUs was observed for BV1, CV2, Tree samples were collected from the inner-middle part followed by CV1. We identifed 10 and 9 abundant phyla (rep- of Fetida Cave (Fig.2). Sample BV1 was obtained from the resented by more than 2% total bacterial sequences) in BV1 middle zone of the cave on a wall difusely covered by red- and CV1 samples, respectively (Fig. 2). Both samples were brownish biovermiculations. CV1 was sampled on the ceiling dominated by Proteobacteria; the phylum Chlorobi was solely of the inner room and is characterized by brownish biover- detected in BV1 sample. Other commonly abundant phyla in miculations, while CV2 was collected from a wall in the inner BV1 and CV1 were Acidobacteria, Actinobacteria, Plancto- room covered by greyish biovermiculations. In these galleries mycetes, Bacteroidetes and Nitrospirae. In contrast, only fve it is possible to see upwelling sulfdic acidic water occurring in abundant phyla were detected for sample CV2, with Pro-

378 Proceedings of the 17th International Congress of Speleology Figure 3. Relative class abundance of Proteobacteria classes in the three samples (BV1, CV1, CV2). Epsilonproteobacteria were only detected in sample CV2. teobacteria, Acidobacteria and Actinobacteria representing almost 80% of the total microbial population. Tese results were in line with previous studies reporting Proteobacteria as one of the main bacterial phyla present in caves (Tomczyk-Żak and Zielenkiewicz 2015) along with Actinobacteria (Groth et al. 1999). Acidobacteria can be further associated with the acidophilic conditions due to the production of sulfuric acid.

Considering Proteobacteria phylum, the relative abundance of the respective classes is diferent in the three samples (Fig 3); Gammaproteobacteria and Betaproteobacteria were domi- nant in CV1 and BV1 samples, while in CV2 sample Gam- maproteobacteria was followed by Deltaproteobacteria and Alphaproteobacteria. Interestingly, the Epsilonproteobacteria class was retrieved in CV2 sample, but absent in the other two samples. Te great abundance of Acidobacteria in CV2 and the presence of Epsilonproteobacteria can be related to sulfur- Figure 4. FESEM images of grey biovermiculations collected in oxidizing activity. Fetida Cave. A) General view of microbial mats mainly composed of flamentous FESEM images revealed dense microbial mats in all the structures. studied samples, mainly composed of flamentous bacteria B) Prosthecate bacterium, resembling Ancalomicrobium. C) Prosthecate bacterium with long appendages probably related to (Fig. 4A). In addition, the abundance of prosthecate bacteria the genus Prosthecomicrobium (arrow). was noteworthy, particularly in the grey biovermiculations. D)Reticulated flament of unknown bacterial afliation. Figure 4B depicts a bacterial cell with numerous prosthecae extending in all directions, resembling Ancalomicrobium from this lineage are sulfur-oxidizing bacteria), Deltapro- (Alphaproteobacteria). A prosthecate cell with long append- teobacteria (anaerobic sulfate-reducing bacteria), and in ages was also observed (Fig. 4C); this is probably related to the particular of Epsilonproteobacteria (several bacteria from genus Prosthecomicrobium (Alphaproteobacteria) (Oertli et this lineage are thermophiles and can tolerate high sulfur al. 2006). FESEM examinations also revealed intriguing retic- concentration), together with Acidobacteria (typical of sulfur ulated flaments with honey-comb sheaths (Fig. 4D). Tese dominated environments), can be considered endemic of SAS flaments have been previously reported by several authors in environments. In Fetida Cave, we observed that biovermicu- limestone, granite and lava caves (Melim et al. 2008, Miller et lations develop only on walls and ceiling close to sulfuric acid al. 2012), but their taxonomy is still unknown and represents springs. Tis work shows that the microbial communities in a challenge for the ingenuity of microbiologists. biovermiculations from Fetida Cave are in accordance with previous data reported for acidic caves from Italy (Jones et al. 4. Concluding remarks 2008) and support the role of Epsilonproteobacteria (Engel et al., 2004) typically occurring in sulfdic-rich environments. Sulfuric acid (SAS) cave systems host a wide variety of microbial mats with diferent morphologies, colors, and microbial diversity. Analyzing biovermiculations from Fetida Cave, we Acknowledgements found diversifed bacterial communities dominated by Pro- Many thanks Santa Cesarea Terme spa, for the availability teobacteria and Acidobacteria. As suggested by Tomczyk-Żak to visit caves and sample geochemical and microbiological and Zielenkiewicz (2015), the Proteobacteria lineage gener- materials. Tanks to Lee Florea that helped us in monitoring ally, represents the prevalent group in cave environments, geochemical parameters and in samples collection. Finan- from the sulfuric acid to the normal epigenic ones. Te rela- cial support from the Spanish project CGL2016-75590-P is tive abundance of Betaproteobacteria (some microorganisms acknowledged.

Proceedings of the 17th International Congress of Speleology 379 References Klimchouk A, 2009. Morphogenesis of hypogenic caves. Azzaroli A, 1967. Calcare di Altamura. Note illustrative della Geomorphology, 106, 100-117. Carta Geologica d’Italia, Formazioni geologiche, 1, 151-156. Melim LA, Northup DE, Spilde MN, Jones B, Boston P, Bixby Bontognali TRR, D’Angeli IM, Tisato N, Vasconcelos C, R, 2008. Reticulated flaments in cave pool speleothems: Bernasconi SM, Gonzales ERG, De Waele J, 2016. Mush- microbe or mineral? Journal of Cave and Karst Studies, 70, room speleothems: stromatolites that formed in the absence 135-141. of phototrops. Frontiers of Earth Science, 4, art. 49. DOI: Miller AZ, Hernández-Mariné M, Jurado V, Dionísio A, 10.3389/feart00049. Barquinha P, Fortunato E, Afonso MJ, Chaminé HI, Saiz- Boston PJ, Spilde MN, Northup DE, Melim LA, Soroka DS, Jimenez C, 2012. Enigmatic reticulated flaments in subsur- Kleina LG, Lavoie KH, Hose LD, Mallory LM, Dahm CN, face granite. Environmental Microbiology Reports, 4, 596-603. Crossey LJ, Schelble RT, 2004. Cave biosignature suites: Oertli GE, Jenkins C, Ward N, Rainey F, Stackebrant E, microbes, minerals and Mars. Astrobiology 1(1), 25-55. Staley JT, 2006. Te genera Prosthecomicrobium and Ancal- Cañaveras JC, Sanchez-Moral S, Soler V, Saiz-Jimenez C, omicrobium. In: M Dworkin, S Falkow, E Rosenberg, KH 2001. Microorganisms and microbially induced fabrics in Schleifer, E Stackebrandt (Eds.). Te Prokaryotes, Springer, cave walls. Geomicrobiological Journal, 18(3), 223-240. New York, vol. 5, pp. 65-71.

Egemeier SJ, 1981. Cavern development by thermal water. Quast C, Pruesse E, Yilmaz P, Gerken J, Schweer T, Yarza P, National Speleological Society Bulletin, 43, 31- 51. et al, 2013. Te SILVA ribosomal RNA gene database project: improved data processing and web-based tools. Nucleic Acids Engel A, Stern LA, Bennett PC, 2004. Microbial contribu- Res 41:D590-D596. tions to cave formation: new insights into sulphuric acid speleogenesis. Geology, 32(5), 369-372. Schloss PD, Westcott SL, Ryabin T, Hall JR, Hartmann M, Hollister EB, et al. (2009) Introducing mothur: Open-Source, Groth IR, Vettermann B, Schuetze P, Schumann Saiz-Jimenez Platform-Independent, Community-Supported Sofware for C, 1999. Actinomycetes in karstic caves of northern Spain Describing and Comparing Microbial Communities. Applied (Altamira and Tito Bustillo). Journal of Microbiological Environmental Microbiology 75, 7537-7541. Methods, 36, 115-122. Tisato N, Torriani SFF, Monteux S, Sauro F, De Waele J, Grotenhuis JT, Smit M, Plugge CM, Xu YS, Van Lammer AA, Tavagna ML, D’Angeli IM, Chailloux D, Renda M, Eglinton Stams AJ, Zehnder AJ, 1991. Bacteriological composition and TI, Bontognali TRR, 2015. Microbial mediation of complex structure of granular sludge adapted to diferent substrates. subterranean mineral structures, Scientifc Reports, 5, 15525. Applied Environmental Microbiology, 57, 1942-1949. DOI: 10.1038/srep15525.

Hose LD, Pisarowicz JA, 1999. Cueza de Villa Luz, Tabasco, Tomczyk-Żak K, Zielenkiewicz U, 2015. Microbial diversity Mexico: reconnaissance study of an active sulfur spring cave in caves. Geomicrobiology Journal, 33(1), 20-38. and ecosystem. Journal of Cave and Karst Studies, 61, 13-21.

Jones DS, Lyon EH, Macalady JL, 2008. Geomicrobiology of biovermiculations from the Frasassi cave system, Italy. Journal of Cave and Karst Studies, 70(2), 78-93.

380 Proceedings of the 17th International Congress of Speleology