Exploring the microbial diversity featuring the geochemical complexity of the quartz-sandstone cave Imawarì Yeuta, Auyan Tepui, Venezuela

Abstract

In the last three years, one of the longest quartz-sandstone caves, Imawarì Yeuta, has been explored in the Precambrian rocks of the Auyan Tepui massif in Venezuela. The uniqueness of this quartz-sandstone cave resides in its great age (estimated over 30 Ma) and its complete isolation from anthropogenic activities. Therefore, subsurface ecosystems might have been preserved from contamination and possible subsequent alteration. The cave contains a high level of microbial activity as demonstrated by the presence of silica stromatolite-like speleothems, biologically mediated sulphate-phosphate deposits and lakes covered by patinas of violaceous or other colourful biofilms. The high diversity of the environments, in terms of mineralogical substrate and geochemistry of the waters, suggests that niche-differentiation of the microbial communities could be mainly controlled by the specific chemical characteristics of each site rather than by light attenuation, i.e., the distance from the entrances of the cave toward the dark zones. To investigate this possibility, we characterised the microbial community composition in 17 samples collected during the expeditions of 2014 and 2016 to the cave system. In particular, the next-generation sequencing (NGS) approach was performed by using Illumina MiSeq technology targeting the 16S rRNA gene. The acquired data have then been correlated with geochemical (water pH, EC, major and minor elements) and mineralogical (predominant and secondary minerals) information from each sampling site. The results showed that the cave hosts three main types of microbial communities, each of these characteristic of a specific environment dominated by silica (crystalline or amorphous), iron hydroxides or hydrated sulphate-phosphates.
This work presents a first set of results on the microbial characterisation of each sampling site and proposes a preliminary hypothesis on the functional biological factors supporting the biodiversity distribution in relation with the different geochemical environments described in this exceptional cave.

Full text

Proceedings of the 17th International Congress of Speleology 385
Exploring the microbial diversity featuring the geochemical complexity of the quartz-
sandstone cave Imawarì Yeuta, Auyan Tepui, Venezuela
Daniele Ghezzi1, Francesco Sauro2, 3, Hosam Mamoon Zowawi4, Pei-Ying Hong5, Martina Cappelletti1,
Leonardo Piccini3,6, Davide Zannoni1, Freddy Vergara3,7, Jo De Waele2,3
Aliation: 1Department of Pharmacy and BioTechnology, University of Bologna, Via Irnerio 42, Bologna, Italy
2Department of Biological, Geological and Environmental Sciences, Italian Institute of Speleology, Bologna University,
Via Zamboni 67, 40126, Bologna, Italy
3La Venta Geographic Explorations Association, Via Priamo Tron 35/F, 31100, Treviso, Italy
4e University of Queensland, UQ Centre for Clinical Research, Building 71/918 Royal Brisbane Hospital, Herston,
QLD 4006, Australia
5Division of Biological and Environmental Science and Engineering, King Abdullah University of Science and
Technology (KAUST), 4700 King Abdullah Boulevard, uwal 23955-6900, Saudi Arabia
6Department of Earth Sciences, Florence University, Via La Pira 4, 50121, Florence (Italy)
7eraphosa Exploring Team, Puerto Ordaz (Venezuela)
Abstract
In the last three years, one of the longest quartz-sandstone caves, Imawarì Yeuta, has been explored in the Precambrian rocks
of the Auyan Tepui massif in Venezuela. e uniqueness of this quartz-sandstone cave resides in its great age (estimated over
30 Ma) and its complete isolation from anthropogenic activities. erefore, subsurface ecosystems might have been preserved
from contamination and possible subsequent alteration. e cave contains a high level of microbial activity as demonstrated
by the presence of silica stromatolite-like speleothems, biologically mediated sulphate-phosphate deposits and lakes covered by
patinas of violaceous or other colourful biolms. e high diversity of the environments, in terms of mineralogical substrate
and geochemistry of the waters, suggests that niche-dierentiation of the microbial communities could be mainly controlled
by the specic chemical characteristics of each site rather than by light attenuation, i.e., the distance from the entrances of
the cave toward the dark zones. To investigate this possibility, we characterised the microbial community composition in 17
samples collected during the expeditions of 2014 and 2016 to the cave system. In particular, the next-generation sequencing
(NGS) approach was performed by using Illumina MiSeq technology targeting the 16S rRNA gene. e acquired data have then
been correlated with geochemical (water pH, EC, major and minor elements) and mineralogical (predominant and secondary
minerals) information from each sampling site. e results showed that the cave hosts three main types of microbial communities,
each of these characteristic of a specic environment dominated by silica (crystalline or amorphous), iron hydroxides or hydrated
sulphate-phosphates.
is work presents a rst set of results on the microbial characterisation of each sampling site and proposes a preliminary
hypothesis on the functional biological factors supporting the biodiversity distribution in relation with the dierent geochemical
environments described in this exceptional cave.
Keywords: microbial diversity; silica; sulphates; biospeleothems; tepui.
1. Introduction
Deep subsurface and deep-sea environments are the two last
largely unexplored habitats on earth. Since caves are natural
openings into the deep surface they oer the unique possibil-
ity to explore and investigate this important ecosystem. Glob-
ally only 10% of all caves have been discovered and only a
small fraction of these has been biologically explored (Eavis
2010; Engel, 2010, 2011). Up to now only ~60 caves have
been investigated using molecular biological methods. So far
results showed that subsurface microbes might be unique and
genetically divergent from surface organisms, with only about
half of bacterial and archaeal phyla somehow identiable (Lee
et al. 2012).
e identication of cave specic species and ecology opens
interesting questions related to their interaction with the
mineral substratum and with peculiar weathering and min-
erogenetic processes. Also the evolutionary mechanisms,
which allowed microbes to proliferate in the extreme envi-
ronment of deep surface where usually nutrients are low, are
still largely unknown. Recently it was hypothesised that older
caves might function as long-term reservoirs of largely non
described microorganisms, similar to deep-sea hydrothermal
vents that are used as model systems to better understand
the origin and evolution of life on Earth or on other planets
(Northup and Lavoie 2001; Engel and Northup 2008; Engel
2010; Lee et al. 2012).
Most previous research on this topic dealt with classic carbon-
ate caves while very little attention has been given to pecu-
liar environments like quartz-sandstone and quartzite caves,
where silica is by far the dominant element. About twenty
years ago cavers and karst scientists believed that speleogen-
esis of caves in quartz-sandstones was related to exceptional
conditions and only of local importance. On the contrary,
since 2000, several huge horizontal cave systems have been
explored in the tepui massifs of Venezuela. ere is still an
ongoing debate in the scientic community concerning the
processes responsible for the formation of such extensive
caves in this extremely hard and barely soluble lithology. e

386 Proceedings of the 17th International Congress of Speleology
formation of caves and karst features in quartz-rich rocks was
considered exceptional given the low solubility and dissolu-
tion rates of quartz (Wray 1997). Based on the slow dissolu-
tion rate of SiO2, on geochemical analysis of dissolved silica
in cave water and on the geomorphic history of the region,
many authors agree that the formation of these caves could be
reasonably dated back at least to ~20-30 Ma (Piccini and Mec-
chia 2009). In addition, all these caves present several types
of silica speleothems (opal and amorphous silica) that have
been proposed as biologically mediated or even as true silica
stromatolites (Aubrecht et al. 2011). e discovery of new
(Galli et al. 2014) or extremely rare sulphates and sulphate-
phosphate minerals (Sauro et al. 2014) in these underground
environments, with clear presence of EPS (extracellular poly-
meric substances) has also opened a debate on whether or
not microbes are involved in their genesis. e uniqueness of
the tepui caves resides in their great age and isolation from
anthropogenic activities. erefore subsurface ecosystems
might have been preserved from contamination and possible
subsequent alteration in these caves.
In 2013 a new giant cave, named Imawarì Yeuta, was discov-
ered by a joint Italian-Venezuelan expedition on the Auyan
Tepui in the Canaima National Park (Venezuela). Besides its
dimensions (this is the longest quartzite cave in the world),
the scientic interest of this karst system is very high ranging
from the processes of weathering that lead to the cave forma-
tion, to the exceptional secondary minerals and speleothems
Figure 1. Dierent geochemical environments in Imawarì Yeuta: A) Eroded and polished quartzite walls along a stream active area (HSOR);
B) Stromatolite-like speleothems made of amorphous silica (HSOR); C) A lake satured in SiO2 and covered by a violet patina (HSOW); D) A
sulphate-phosphate deposit is accumulated below a quartzite boulder (SP); E) Mould speleothems made by philaments encrusted by amorphous
silica (MS); F) A deposit of iron-hydroxides (HI). (Photos by R. Shone, R. De Luca, A. Romeo, V. Crobu – La Venta).

Proceedings of the 17th International Congress of Speleology 387
(Sauro 2014). A set of 17 samples has been collected from
this cave in order to start a research project on the micro-
bial diversity and on the functional role of microorganisms
in the speleogenesis and minerogenesis in silica environment.
In this abstract we describe the rst results of the microbial
diversity next-generation sequencing analysis (NGS) and
their relation with dierent geochemical environments (high
silica, high sulphates, high iron) within the cave system.
2. Geochemical environments and samples
Imawarì Yeuta is actually one of the most developed cave sys-
tems in quartzite environment of the world. e rock hosting
the cave is 85 to 95 % quartz, with small amounts of phyllosili-
cates (kaolinite and pyrophyllite) and iron hydroxides. Quartz
has an extremely low solubility at the local T (15° C) and pH
(3.4 to 5), thus stream waters crossing the cave are usually
undersaturated and with extremely low EC values, almost
approximating to that of distilled water. Nonetheless in the
cave there are locally peculiar biogeochemical niches, like
sulphate substrata, iron hydroxide deposits and rare opal and
amorphous silica biostromatolites (Fig. 1). In general there is
a clear transition between hydrologically active zones, where
the quartzite rock is polished by water erosion and secondary
mineral deposits are absent, to hydrologically inactive fossil
zones where amorphous silica crusts, silica speleothems, and
secondary minerals are abundant. In these areas the quartzite
rock is more weathered and weaker and there is frequently the
evidence of biological activity in the form of biolms. Also
here cave waters in still-standing pools have dierent chemi-
cal characteristics, with higher levels of pH (6) and signicant
levels of dissolved SiO2 (most of them saturated at 8.5 mg L-1).
Fossil areas are environments with extremely low nutrient
sources, where the only exchange with the surface is through
air currents.
Most of the samples come from these fossil areas of the cave,
being the most isolated spots where the biological activity has
HSOR HSOW MS IH SP
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
Ay301 Ay302 Ay304 Ay305 Ay306 Ay312 Ay313 Ay317 Ay303 Ay314 Ay325 Ay311 Ay323 Ay316 Ay307 Ay308 Ay315
Others
Archea
WPS-2 incertae sedis
Firmicutes
Bacteroidetes
Nitrospira
Planctomycetes
Actinobacteria
Acidobacteria
Deltaproteobacteria
Gammaproteobacteria
Betaproteobacteria
Alphaproteobacteria
Unclassified Proteobacteria
A
B
Figure 2. Taxonomy classication of sequences (reads adjusted for the copy number) for the samples under analysis. Panel A) shows the percent-
age of taxonomically classied (black) and unclassied (white) sequences found in each sample. Panel B) displays the relative abundance of the
bacterial phyla and archea found in the classied fraction of sequences in each sample. “Others” comprises all the classied bacterial phyla present
<2% in all samples. e corresponding sample group names dening the geological settings reported in Table 1 are reported below the histograms

388 Proceedings of the 17th International Congress of Speleology
lasted for longer times in more or less stable environmental
conditions, possibly for millions of years. e main group of
samples is related to environments with high silica (HSOR)
either on rocky substratum or in waters. In this group, sam-
ples have been collected on clean quartzite walls (Ay317),
amorphous silica speleothems (Ay302-306), silica coralloids
(Ay313), and quartz sands (Ay312). A peculiar subset of sam-
ples also related to unusual high silica content is that of silica-
saturated waters and violet blue lakes in fossil areas (HSOW).
Sulphate and phosphate-sulphate substrata represent another
geochemically peculiar set of samples collected mainly in
one chamber, where minerals like sanjuanite, rossiantonite,
gypsum and alunite have been found in form of mounds
(SP). One other set, with only one sample, is from giant iron
hydroxide deposits forming owstones and stalagmites (IH),
where iron is surely the dominant element. Finally another set
(MS) comes from a peculiar mould biospeleothem made of a
mixture of amorphous silica, manganese and iron hydroxides,
representing one of the most enigmatic features of this cave
covering the cave roofs for hundreds of square metres.
3. Methods
Samples have been taken with a sterilized spoon directly from
the rocky substratum, water ponds, or mineral aggregates and
stored in sterilized 4 ml eppendorf tubes lled with a solution
of LifeGuard RNA. e transport from the site to the lab was
carried out in a portable fridge and then samples were stored
at -90° C.
Genomic DNA was extracted from each sample using the
UltraClean Soil DNA Isolation Kit (MoBio, Carlsbad, CA,
USA) with slight modications to the manufacturer’s protocol
(Cappelletti et al. 2016).
To provide amplicon for Illumina MiSeq analysis, the total
DNA was amplied for V4-V5 region of 16S rRNA gene
with universal forward 515F (5’-Illumina overhang-GTGY-
CAGCMGCCGCGGTA-3’) and reverse 907R (5’- Illumina
overhang-CCCCGYCAATTCMTTTRAGT-3’) primers. One
µL of total DNA was added to a 50 µL (nal volume) PCR
reaction mixture containing 25 µL of Premix F (Epicentre
Biotechnologies, WI, USA), 200 mM (each) forward and
reverse primers, and 0.5 U of Ex Taq DNA polymerase (Takara
Bio, Japan). Amplication reactions were carried out under
the following thermocycling conditions: 95°C for 3 min, 30
cycles of 95°C for 30 s, 55°C for 30 s, 72°C for 30 s, with a nal
extension at 72°C for 5 min.
PCR amplicons were conrmed by electrophoresis with a
1% (w/v) agarose gel and then puried by AMPure XP beads
(Beckman Coulter) prior to the index PCR. Nextera XT Index
was incorporated into each of the individual samples during
PCR. e thermal cycling program included a rst denatur-
ation step at 95°C for 3 min, followed by 8 cycles of denatur-
ation at 95°C for 30 s, annealing at 55°C for 30 s, elongation at
72°C for 30 s, with a nal extension at 72°C for 5 min. Puried
amplicons were submitted to KAUST Genomic Core Lab for
unidirectional sequencing reads on an Illumina MiSeq plat-
form. Chimeric sequences were detected using UCHIME37
and deleted.
To annotate the 16S rRNA gene sequences obtained from
high-throughput sequencing, the Ribosomal Database Proj-
ect (RDP) Classier was used for taxonomical assignments at
a 95% condence level. Aer annotation, the relative abun-
dance of each bacterial genus was calculated, collated and the
normalized data were square-root transformed (Ansari et al.
2015). e transformed dataset was then computed for their
Bray-Curtis similarities and represented graphically for spa-
tial distribution in a non-metric threshold multidimensional
scaling (nMDS) plot using Primer E version 7.
4. Results and Discussion
e analysis of Illumina MiSeq sequencing revealed a total
number of 56,622 sequences (reads corrected for the taxon
copy number) of which around 40% were not taxonomically
dened by the RDP program (Fig. 2A). is high number
of unclassied reads is probably related to the lack of previ-
ous microbiological studies in these pristine environments
showing the interesting potential of this cave for biodiversity
studies, particularly in the search of novel bacteria and novel
metabolic activities (e.g. production of antibiotics, secondary
metabolites generation). e classied sequences were mainly
dened as Eubacteria and, in 12 out of 17 samples, sequences
belonging to Archaea were also present within an abundance
range of 0.1-5% (Fig. 2B).
e multi-dimensional scaling (MDS) plot in Fig. 3 compares
the 17 samples on the basis of the relative abundance of clas-
sied and unclassied bacteria and archaea revealed by next-
generation sequencing. In the MDS, the samples belonging
to the geochemical groups (HS-R and HS-W) featured by the
presence of high concentration of silica generally clustered
together. On the other hand, the samples characterised by
sulphate and phosphate presence tended to co-localize in the
MDS plot and clustered apart from the HS-W sample group.
us, dierences in the geochemical settings seem to reect
the microbial community structure diversication.
In all the samples, the classied bacterial sequences belonged
to the 8 most abundant phyla (representing >2% in at least
one sample) i.e., Proteobacteria, Acidobacteria, Actinobacte-
ria, Nitrospirae, Bacteroidetes, Firmicutes, Planctomycetes, and
the taxonomically undened lineage WPS-2 (Fig. 2B). Pro-
teobacteria, Acidobacteria and Actinobacteria were dominant,
comprising 75–100% of the classied bacterial sequences in
all the samples. Nitrospirae, Bacteroidetes and Firmicutes gen-
erally represented less abundant phyla; only in Ay323 of the
mold-like speleothem (MS-R) sample group, Bacteroidetes
and Nitrospirae amounted for 16% and 4%, respectively, while
they were <2% in the other samples. Firmicutes was >2% only
in Ay315 of the SP-R sample group. e Planctomycetes com-
munity was identied in almost all samples with an occur-
rence lower than 3%.
Alphaproteobacteria and Gammaproteobacteria were the
dominant classes within Proteobacteria. Interestingly, in one
of the samples (Ay303) collected from a pond with saturated
content of silica and an unusual sulphate concentration
(HSOW group), Betaproteobacteria dominated over Alpha-
and Gammaproteobacteria being mainly composed by Jan-
thinobacterium genus. ese bacteria have been previously
isolated from cave samples for their Mn(II)-oxidizing abili-
ties and violacein-producing capacity (Rodrigues et al. 2012;
Carmichael et al. 2013). is last feature might be related to
the violet-like appearance of the water pond where Ay303

Proceedings of the 17th International Congress of Speleology 389
was collected. Alphaproteobacteria and Rhizobiales covered
more than 50% of the Proteobacteria found in the micro-
bial community in some HS-R samples (quartzite rock) and
one HS-W sample (violet slime on water pond). A number
of novel Rhizobiales species have been described from caves
and they are well represented in subterranean environments.
ey are known to x nitrogen and to play key roles in bio-
deterioration processes (Diaz-Herraiz et al. 2014). Among
Gammaproteobacteria, Pseudomonadales, Xanthomonadales,
Alteromonadales and Enterobacterales represented the most
abundant orders in dierent sample groups being associated
with various natural and cave fauna (e.g. bats or oil birds)
carbon organic sources. A representative group of Deltapro-
teobacteria in the HS-R samples was the Myxococcales order
whose abundance was higher in sample Ay317 collected from
pure quartzite rock walls. Previously, Myxobacteria have been
found within caves in areas apparently untouched by man
before (Menne and Rückert 1988). Some Myxobacteria spe-
cies are known to induce bioprecipitation of calcium carbon-
ate and to tolerate high concentrations of heavy metals (Tisato
et al. 2015).
e Actinobacteria represents the second most abundant
phylum aer Proteobacteria found in this study. More than
half of the samples tested showed Actinobacteria abundance
>20% of the classied bacterial community. In particular, the
sample groups collected from water environments showed
lower percentage (<4%) of this phylum. Actinobacteria have
been described as one of the prevalent lineages in freshwater,
but are also a small component in saltwater (Jensen and Lauro
2008).
Acidobacteria represent the third most abundant phylum
found in our tepui samples. eir distribution is variable
among samples, with the highest abundance being detected
in the HS-R group. Acidobacteria are ubiquitous in dierent
types of subterranean environments; in particular, Acido-
bacteria subgroups 1, 3, 4, and 6 are prevalent in terrestrial
environments, including some cave samples (Sáiz-Jiménez
2015). In HS-R samples subgroups 1, 2, 3 and 13 were the
most abundant, while in MS-R samples only Acidobacteria
subgroup 1 was detected. Metabolic associations have been
described between the dierent Acidobacteria subgroups and
other bacteria-like methanotrophs, phototrophs, Gamma-
and Epsilonproteobacteria (Sáiz-Jiménez, 2015).
In all the samples, considering only the classied sequences,
the maximum Archaea representation was 5% (Ay312). au-
marchaeota comprised a signicant portion of the classied
archaeal community detected in this study. e second most
abundant phylum was Crenarchaeota, while Euryarchaeota
phylum was identied exclusively in sample Ay307 (SP-R),
where Crenarchaeota and aumarchaeota were absent. Pre-
viously, the presence of aumarchaeota and Euryarchaeota
in caves has been related to oxygen concentration and to
nitrogen source production for microbial community devel-
opment.
Non-metric MDS
Transform: Square root
Resemblance: S17 Bray-Curtis similarity
Group
Ay301
Ay302
Ay303
Ay304
Ay305
Ay306
Ay307
Ay308
Ay309
Ay311
Ay312
Ay313
Ay314
Ay315
Ay316
Ay317
Ay323
Ay325
2D Stress: 0.11
HS-R
HS-W
MS-R
IH-W
SP-R
Figure 3. Multi-dimensional scaling (MDS) plot for the microbial community in tepui samples described in Table 1. e MDS plot displays the
community structure distribution of the 17 samples under analysis on the basis of the relative abundance of classied and unclassied bacterial
and archeal sequences detected in each sample through Illumina Miseq analysis. Symbols with the same shape and colour represent samples of the
same geochemistry-based grouping (Table 1). e acronyms shown in the legend are explained in Table 1.

390 Proceedings of the 17th International Congress of Speleology
In general these rst data show that the biodiversity of each
is group of samples seems to be controlled by the specic
geochemical and environmental characteristics, with some
variations in between group samples depending also on the
substratum (rock or water). Further genomic studies will
probably enable identication of functional metabolic asso-
ciations that might induce silica bioweathering or have a
control on the genesis of certain minerals, thus revealing the
mutual inuence between microbial communities and geo-
chemical processes.
Acknowledgements
is research has been funded by the Rolex Award for Enter-
prise within the project Tepui of La Venta Geographic Explo-
rations Association. Permit and support for this research has
been provided by the Gobernación de Estado Bolivar (Ven-
ezuela). is study has also beneted from a nancial support
by the Rectorate of the University of Bologna.
References
Ansari MI, Harb M, Jones B, Hong P-Y, 2015. Molecular-
based approaches to characterize coastal microbial commu-
nity and their potential relation to the trophic state of Red
Sea. Scientic Reports, 5, 9001.
Aubrecht R, Brewer-Carías C, Šmída B, Audy M, Kováčik Ľ,
2008. Anatomy of biologically mediated opal speleothems
in the world’s largest sandstone cave Cueva Charles Brewer,
Chimanta Plateau, Venezuela. Sedimentary Geology, 203,
181-195.
Cappelletti, M., Ghezzi, D., Zannoni, D., Capaccioni, B., &
Fedi, S. (2016). Diversity of methane-oxidizing bacteria in
soils from “Hot Lands of Medolla” (Italy) featured by anoma-
lous high-temperatures and biogenic co2 emission. Microbes
and Environments, 31, 369–377.
Carmichael MJ, Carmichael SK, Santelli CM, Strom A,
Brauer SL, 2013. Mn(II)-oxidizing bacteria are abundant and
environmentally relevant members of the ferromanganese
deposits in caves of the upper Tennessee River Basin. Geomi-
crobiology Journal, 30, 779-800.
Diaz-Herraiz M, Jurado V, Cuezva S, Laiz L, Pallecchi P,
Tiano P, Sanchez-Moral S., Saiz-Jimenez C, 2014. Deteriora-
tion of an Etruscan tomb by bacteria from the order Rhizo-
biales. Scientic Reports, 4, 1–7.
Eavis A, 2009. An up to date report of cave exploration
around the world. In: WB White (Ed.). Proceedings of the
15th International Congress of Speleology, Kerrville, Texas, pp.
21-25.
Engel AS, 2010. Microbial diversity of cave ecosystems.
In: LL Barton, M Mandl, A Loy (Eds.). Geomicrobiology:
Molecular and Environmental Perspective. Springer, Amster-
dam, pp. 219-238.
Engel AS, 2011. Karst microbial ecosystems. In: J Reitner, V
iel (Eds.). Encyclopedia of Geobiology. Springer Encyclope-
dia of Earth Sciences Series (LESS, formerly Kluwer Edition),
Berlin, pp. 521-531.
Engel AS, Northup D.E., 2008. Caves and karst as model
systems for advancing the microbial sciences. In: J Martin
and WB White (Eds.), Frontiers in Karst Research. Karst
Waters Institute Special Publication, 13, Leesburg, Virginia,
pp. 37-48.
Galli E, Brigatti MF, Malferrari D, Sauro F, De Waele J, 2013.
Rossiantonite, Al3(PO4)(SO4)2(OH)2(H2O)10· 4H2O, a new
hydrated aluminum phosphatesulfate mineral from Chi-
manta massif, Venezuela: description and crystal structure.
American Mineralogist, 98, 1906-1913.
Jensen PR, Lauro FM, 2008. An assessment of actinobacterial
diversity in the marine environment. Antonie Van Leeuwen-
hoek, 94, 51–62.
Lee NM, Meisinger DB, Aubrecht R, Kovacik L, Saiz-Jimenez
C, Baskar S, Baskar R, Liebl W, Porter ML, Engel AS, 2012.
Caves and Karst Environments. In: EM Bell (Ed.) Life at
Extremes: Environments Organisms and Strategies for Sur-
vival. CABI, Wallingford, UK, pp. 320-344.
Menne B, Rückert G, 1988. Myxobakterien (Myxobacte-
rales) in Höhlensedimenten des Hagengebirges (Nördliche
Kalkalpen). Die Höhle, 39(4), 120-131.
Northup DE, Lavoie KH, 2001. Geomicrobiology of caves: a
review. Geomicrobiology Journal, 18(3), 199-222.
Piccini L, Mecchia M, 2009. Solution weathering rate and
origin of karst landforms and caves in the quartzite of
Auyan-tepui (Gran Sabana, Venezuela). Geomorphology,
106, 15-25.
Rodrigues AL, Gocke Y, Bolten C, Brock NL, Dickschat JS,
Wittmann C, 2012. Microbial production of the drugs viola-
cein and deoxyviolacein: analytical development and strain
comparison. Biotechnological Letters, 34, 717–20.
Sáiz-Jiménez C, 2015. e microbiology of show caves,
mines, tunnels, and tombs: implications for management
and conservation. In: A Summers Engel (Ed.). Microbial
Life of Cave Systems. Berlin, München, Boston, Walter de
Gruyter, pp. 231-262.
Sauro F, Tisato N, Waele J, Bernasconi SM, Bontognali TRR,
Galli E, 2014. Source and genesis of sulphate and phosphate-
sulphate minerals in a quartz-sandstone cave environment.
Sedimentology, 61, 1433-1451.
Tisato N, Torriani SFF, Monteux S, Sauro F, De Waele J,
Tavagna ML, D’Angeli IM, Chailloux D, Renda M, Eglinton
TI, Bontognali TRR, 2015. Microbial mediation of complex
subterranean mineral structures. Scientic Reports, 5, 15525.
Wray RAL, 1997. A global review of solutional weathering
forms on quartz sandstones. Earth-Science Reviews, 42, 137-
160.

Proceedings of the 17th International Congress of Speleology 391
Table 1. Description of the geological settings and chemical parameters featuring the tepui samples analysed in this study.
Sample Description Type of
substratum pH EC Si SO4
(mg/L) PO4 FeOH
(μg/L)
Al
(μg/L)
T
(°C)
Water
activity
High Silica on Rocks (HS-R)
Ay301 Whitish dots of amor-
phous silica
Quartzite
oor / / High No No No Low 15,2
Wet, close
to small
stream
Ay302 White paste of amor-
phous silica
Quartzite
oor / / High No No No Low 15,2
Wet, close
to small
stream
Ay304 White spots resembling
lichens
Quartzite
oor / / High No No No Low 15,2
Wet, along
small
stream
Ay305 White shiny on the
oor
Quartzite
oor / / High No No No Low 15,2
Wet, along
small
stream
Ay306 Black amorphous silica
corallloid
Quartzite
oor / / High No No No Low 15,2
Wet, along
small
stream
Ay312 Yellow dots and
agglomerates Quartz sand / / High No No No Low 15,2 Wet
satured
Ay313 Black encrustation of
amorphous Si Quartz sand / / High No No No Low 15,2 Wet, not
satured
Ay317 Quartz sandstone walls Quartzite
rock wall / / High No No No Low 15,2
Wet, close
to small
stream
High Silica on Waterpond (HS-W)
Ay303 Brown slime with
yellow dots Water 5,1 6,9 8,87* 0,66 No 19,7 No 14,7 Stagnant
waters
Ay314 Violet slime on water
pond Water 5,1 6,9 8,87* 0,66 No 19,7 No 14,7 Stagnant
waters
Ay325 Irridescent patina on
water Water 6,1 11 8,6* 0,63 No No No 14,8 Stagnant
waters
Mold-Like Speleothem (MS-R)
Ay311 Violet patina on the
roof
Quartzite
oor / / High No No Low No 15-17 Dry and
windy
Ay323 Mold on silica coral-
loids on roof
Amorphous
silica / / High No No Low No 15-17 Dry and
windy
Iron Hydroxides (IH-W)
Ay316 Black gours of Goethite Water 5 9,8 6,3 0,59 Low 10,1 57,0 14,4 Percolating
water
Sulphates And Phosphates (SP-R)
Ay307 So deposit of sul-
phates in the oor Sulphates / / Low High High Low Low 15,2 Dry on
oor
Ay308 Powder of sulphates
and amorphous silica
Sulphates and
silica / / Med High High Low High 15,2 Dry on
oor
Ay315 Sanjuanite and Gypsum
deposit
Sulphate
powder / / Low High High Low High 15,2 Dry
* = saturate