Pristine but metal-rich Río Sucio (Dirty River) is dominated by Gallionella and other iron-sulfur oxidizing microbes

Abstract

Whether the extreme conditions of acidity and heavy metal pollution of streams and rivers originating in pyritic formations are caused primarily by mining activities or by natural activities of metal-oxidizing microbes living within the geological formations is a subject of considerable controversy. Most microbiological studies of such waters have so far focused on acid mine drainage sites, which are heavily human-impacted environments, so it has been problematic to eliminate the human factor in the question of the origin of the key metal compounds. We have studied the physico-chemistry and microbiology of the Río Sucio in the Braulio Carrillo National Park of Costa Rica, 22 km from its volcanic rock origin. Neither the remote origin, nor the length of the river to the sampling site, have experienced human activity and are thus pristine. The river water had a characteristic brownish-yellow color due to high iron-dominated minerals, was slightly acidic, and rich in chemolithoautotrophic iron- and sulfur-oxidizing bacteria, dominated by Gallionella spp. Río Sucio is thus a natural acid-rock drainage system whose metal-containing components are derived primarily from microbial activities.

Full text

1 3
Extremophiles
DOI 10.1007/s00792-016-0898-7
ORIGINAL PAPER
Pristine but metal‑rich Río Sucio (Dirty River) is dominated
by Gallionella and other iron‑sulfur oxidizing microbes
Alejandro Arce‑Rodríguez1,2 · Fernando Puente‑Sánchez3 · Roberto Avendaño4 ·
Eduardo Libby5 · Leonardo Rojas5 · Juan Carlos Cambronero6,7 ·
Dietmar H. Pieper2 · Kenneth N. Timmis1 · Max Chavarría4,5,6
Received: 28 June 2016 / Accepted: 7 November 2016
© Springer Japan 2016
high iron-dominated minerals, was slightly acidic, and rich
in chemolithoautotrophic iron- and sulfur-oxidizing bac-
teria, dominated by Gallionella spp. Río Sucio is thus a
natural acid-rock drainage system whose metal-containing
components are derived primarily from microbial activities.
Keywords Río Sucio · Braulio Carrillo National Park ·
Costa Rica · Acid-rock drainage · Gallionella spp. ·
“Ferrovum” spp.
Introduction
Acid-rock drainage (ARD) ecosystems based on acidic,
heavily metal-polluted water streams originating in
both natural (volcanic) and anthropogenically impacted
(mines) sulfidic geological formations occurs worldwide
(Fernández-Remolar 2003; Johnson and Hallberg 2005).
ARD environments are characterized by the presence
of large concentrations of metal-sulfide deposits (John-
son 1998; Baker and Banfield 2003), where chemical and
Abstract Whether the extreme conditions of acidity and
heavy metal pollution of streams and rivers originating in
pyritic formations are caused primarily by mining activities
or by natural activities of metal-oxidizing microbes living
within the geological formations is a subject of consider-
able controversy. Most microbiological studies of such
waters have so far focused on acid mine drainage sites,
which are heavily human-impacted environments, so it has
been problematic to eliminate the human factor in the ques-
tion of the origin of the key metal compounds. We have
studied the physico-chemistry and microbiology of the Río
Sucio in the Braulio Carrillo National Park of Costa Rica,
22 km from its volcanic rock origin. Neither the remote ori-
gin, nor the length of the river to the sampling site, have
experienced human activity and are thus pristine. The river
water had a characteristic brownish-yellow color due to
Communicated by M. da Costa.
Electronic supplementary material The online version of this
article (doi:10.1007/s00792-016-0898-7) contains supplementary
material, which is available to authorized users.
* Max Chavarría
max.chavarria@ucr.ac.cr
1 Institute of Microbiology, Technical University
of Braunschweig, 38106 Brunswick, Germany
2 Microbial Interactions and Processes Research Group,
Helmholtz Centre for Infection Research, 38124 Brunswick,
Germany
3 Department of Molecular Evolution, Centro de Astrobiología
(INTA-CSIC), Carretera de Ajalvir Km 4, Torrejón de Ardoz,
28850 Madrid, Spain
4 Centro Nacional de Innovaciones Biotecnológicas
(CENIBiot), CeNAT-CONARE, San José 1174-1200, Costa
Rica
5 Escuela de Química & Centro de Investigaciones en
Productos Naturales (CIPRONA), Universidad de Costa
Rica, Sede Central, San Pedro de Montes de Oca, San
José 11501-2060, Costa Rica
6 Centro de Investigaciones en Productos Naturales
(CIPRONA), Universidad de Costa Rica, San
José 11501-2060, Costa Rica
7 Centro de Investigación en Biología Celular y Molecular
(CIBCM), Universidad de Costa Rica, San José 11501-2060,
Costa Rica

Extremophiles
1 3
microbiological oxidation produce waters with low pH,
hydrated iron(III) oxides and sulfate, often in substantial
quantities (Sánchez-Andrea et al. 2011).
Examples of studied ARD environments generated by
human activity include the Summitville Mine (Colorado,
USA) (Farrand 1997), Iron Mountain Mine (California,
USA) (Edwards et al. 1999) and others (Bruneel et al. 2006;
Fabisch et al. 2013; Jones et al. 2015). However, only a few
studies have focused on natural ARD environments, such as
the Pastoruri Glacier area in Huascarán National Park (Perú)
(González-Toril et al. 2015) and specific sites in the Antarc-
tic landmass (Dold et al. 2013). One of the most extensively
studied ARD environments is the Río Tinto system that orig-
inates in the Iberian Pyritic Belt (IPB) of Spain (González-
Toril et al. 2003; López-Archilla et al. 2004; García-Moy-
ano et al. 2012; Sánchez-Andrea et al. 2012). Because the
IPB has been extensively mined for more than 4500 years,
there is controversy about the causes of the extreme acidity
and heavy metal pollution of the river, with some authori-
ties claiming that they are due to human activity (Olías and
Nieto 2015) and others arguing that they basically result
from natural activities within the IPB (Fernández-Remolar
2003; Amils et al. 2014). Due to the extensive anthropo-
genic impacts in most ARD systems that have been studied,
with the consequent changes in geology and hydrology, it is
difficult to separate natural from engineered causes of pollu-
tion. It is, therefore, critical to study ARD systems with no
history of anthropogenic influence.
Costa Rica has four mountain ranges, which are, from
NW to SE, Guanacaste, Tilarán, Central Volcanic and Tala-
manca, with most volcanoes in the Guanacaste and Central
volcanic ranges (Castellón et al. 2013). The Braulio Car-
rillo National Park, having an area of 47,000 ha, is located
in the latter range. A brownish-yellow river known as Río
Sucio (Dirty River) flows through the heart of the park’s
cloud and rain forests (Fig. 1a, b). This river has its source
at the cliff of an unstable landslide produced by the Río
Sucio fault just northwest of the Irazú volcano (Fig. 1b).
Río Sucio is thus accepted to have a volcanic origin: i.e,
the river conditions are not due to anthropogenic activity
(Dr. Rolando Mora, personal communication). To date, due
to the highly inaccessible terrain and danger of major land-
slides, it has been impractical to access the exact point at
which the river begins. The first safe and accessible sam-
pling point is roughly 22 km from its source (Fig. 1b),
just before it joins the Hondura River. From its source, the
river maintains its peculiar color all the way to the sam-
pling point and along its entire course large amounts of
brownish-yellow materials have precipitated on the banks
of the river (Fig. 1c–f). In this work, we investigated for
the first time the physico-chemical and microbiological
characteristics of Río Sucio. Results indicate that Río Sucio
is a natural acid-rock drainage environment, the chemical
composition of which is biologically driven by iron and
sulfur oxidizing bacteria.
Materials and methods
Sampling and field measurements
All necessary permits for sampling water were obtained
from the National System of Conservation Areas (SINAC)
of the Ministry of Environment and Energy (MINAE) of
Costa Rica (Resolution VS-052). On 24 February 2014,
samples of water were collected 22 km from the source of
Río Sucio (Fig. 1b), just before it joins the Hondura River
(10.147299, −83.947448). We sampled the river in the
middle of Costa Rica’s dry season to minimize the effect
of runoff water. The temperature in the field was measured
with a partial immersion thermometer; pH was measured
with a pH meter (Scholar 425 pH meter, Corning, Inc.,
Corning, NY) in the laboratory. Water samples for chemi-
cal analysis were collected in clean glass bottles, chilled on
ice, and stored at 4 °C until analysis. Samples for analysis
of microbial communities were collected in clean and ster-
ile glass bottles, chilled on ice, stored at 4 °C and processed
within less than 24 h.
Chemical analysis
The water samples were analyzed for major anionic and
cationic species with an ion-exchange chromatograph (IC)
and an inductively coupled-plasma mass spectrometer
(ICP-MS). Samples were filtered with polycarbonate mem-
brane filters (0.45 µm) before analysis. The anions were
determined with an ion chromatograph (MIC-II, Metrohm
Co., Switzerland) equipped with an anionic exchange
resin (Metrosep A Supp 5–100/4.0). Operating conditions
were a mobile phase at 33 °C, Na2CO3 (3.2 mM)/NaHCO3
(1.0 mM) and flow rate of 0.7 mL/min. The anions were
identified and quantified relative to certified commercial
solution standards (Certipur ®Anion multi-element stand-
ard I, II, Merck, Germany). For ICP-MS analysis (Agilent
7500 instrument, Agilent Technologies, Tokyo, Japan), the
optimized operating conditions are listed in Table S1. A
certified multi-element stock solution (Perkin-Elmer Pure
Plus standard, product number 9300233) was used to pre-
pare standard solutions. All determinations were performed
in triplicates.
Scanning electron microscope and electron‑dispersive
spectrometer (SEM–EDS) experiments
SEM–EDS experiments were performed on the suspended
material recovered by filtration of Río Sucio waters. The

Extremophiles
1 3
brownish-yellow solid was dried in the laboratory at 23 °C,
recovered with a spatula and analyzed with a scanning
electron microscope with energy-dispersive X-ray spectra
(Hitachi S-570, Japan).
X‑ray difraction (XRD) analysis
Powder X-ray diffraction patterns were recorded with a
D8 Advance diffractometer (Bruker AXS, Madison, USA,
LynxEye detector) with Cu radiation (λ = 0.154 nm) from
2θ 15º to 80º and step size 0.018º. Identification was made
with ICDD PDF-2 2007 Database (ICDD 2007).
Total DNA isolation, construction of 16S rRNA gene
libraries and Illumina sequencing
Three water samples (1L each) taken at different points
along the river width were filtered through a vacuum filtra-
tion system under sterile conditions using a membrane filter
(pore size 0.22 μm; Millipore, GV CAT No GVWP04700).
To prevent rupture, another filter membrane (pore size
0.45 μm; Phenex, Nylon Part No AF0-0504) was placed
below. The upper filter was collected and stored at −80 °C
until processing. The DNA was extracted from aseptically
cut pieces of the filter with a PowerSoil® DNA Isolation
Fig. 1 Río Sucio at Braulio
Carrillo National Park, Costa
Rica. a Río Sucio is located in
the heart of Braulio Carrillo
National Park. This park is
located on the Central Mountain
Range between San José and
Limón in the Caribbean coast.
Map taken from OpenStreet-
Maps (http://www.openstreet-
map.org/). b The map shows
the distance between the Irazú
Volcano crater and the sampling
point, in addition to several non-
colored rivers flowing into Río
Sucio (Montero et al. 1998).
c, d The river has a particular
brownish-yellow color that is
clearly seen over a distance of
about 30 km. e Along the river
can be seen much brownish-yel-
low material precipitated on the
bottom and banks of the river. f
The brownish-yellow material
deposited on rocks that dry-out
becomes red

Extremophiles
1 3
Kit (MoBio, Carlsbad, CA, USA). Cell lysis was accom-
plished by two steps of bead beating (FastPrep-24, MP Bio-
medicals, Santa Ana, CA, USA) for 30 s at 5.5 m s−1. For
the construction of microbial 16S rRNA amplicon libraries,
the V5–V6 hypervariable regions were PCR-amplified with
universal primers 807F and 1050R (Bohorquez et al. 2012).
The barcoding of the DNA amplicons and the addition of
Illumina adaptors were conducted by PCR as described pre-
viously (Camarinha-Silva et al. 2014; Burbach et al. 2016).
The PCR-generated amplicon libraries were subjected to
250 nt paired-end sequencing on an Illumina MiSeq plat-
form (Illumina, San Diego, CA, USA).
Bioinformatic and phylogenetic analysis of 16S rDNA
amplicon data
Raw MiSeq sequences were quality-filtered (moira.py
script; Puente-Sánchez et al. 2016). Filtered sequences
were subsequently analyzed (mothur version 1.31.2;
Schloss et al. 2009), as recommended by Kozich et al.
(2013). Briefly, the sequences were aligned to a combina-
tion of silva.archaea and silva.bacteria databases (Quast
et al. 2013), screened for chimeras (UCHIME; Edgar et al.
2011) and clustered at 97% similarity with the average-
neighbor algorithm. The most abundant taxa (i.e, OTUs
with relative abundance >0.05% of the total reads sam-
pled) were retrieved and classified against the Ribosomal
Database Project (RDP) reference using the Classify tool
(version 4.3.3; Wang et al. 2007). The results from this ini-
tial classification were individually checked and curated
manually using the RDP Seqmatch tool. Computational
resources were provided by the Data Intensive Academic
Grid (http://diagcomputing.org) and by the MINP group at
Helmholtz-HZI.
For phylogenetic analysis, the closest 16S rRNA
sequences from validly described microbial type strains
and isolates were retrieved using the RDP SeqMatch tool
and also by blastn against the curated 16S rRNA database
of NCBI. In the case of representative 16S rRNA sequences
of candidate division OD1 (Parcubacteria), the sequences
from the closest uncultured organisms were used. The
downloaded sequences and the Río Sucio OTU representa-
tives were subsequently aligned with the SINA web-based
tool (Pruesse et al. 2012). These alignments enabled phy-
logenetic reconstruction with MEGA6 software (Tamura
et al. 2013) and the maximum-likelihood method based on
the general time-reversible model. In total, 100 bootstrap
replications were calculated to ensure the robustness of the
results.
Results and discussion
Physicochemical analysis of Río Sucio waters
The pH of the Río Sucio samples was ca. 5.0 (Table 1).
However, as the sample point is located 22 km from the
source of Río Sucio, the Irazú volcano, it is likely that con-
siderable mixing with non-acidic tributaries occurred, with
resulting dilution and rising of the pH from that at the river
source. According to topographic scale maps (1:50,000),
the river flows 3 km along the bottom of a landslide pro-
duced by the Río Sucio fault, before it meets the first of
seven major tributaries that dilute its deep-yellow water
(Fig. 1b; Montero et al. 1998). The inaccessible and too
dangerous terrain makes the collection of samples in areas
nearer the source of the Río Sucio impracticable. At the
sampling point, brownish-yellow suspended matter was
observed, which constantly adheres to the river bed and
rocks (Fig. 1e). This sedimented material eventually turned
reddish upon drying (Fig. 1f). In the laboratory, water
samples held at 23 °C also showed complete precipitation
within 24 h (Fig. 2a–c).
The chemical analysis of filtered samples (Table 1)
revealed the presence of sulfate, calcium, magnesium, alu-
minum and iron at concentrations much greater than of
Table 1 Physical properties
and chemical composition of
Río Sucio
Property/element/ion Value/concentration
(mg L−1)
Property/element/ion Value/concentration (mg L−1)
Temperature 22.0 ± 0.1 °C Aluminum 17.1 ± 1.9
pH 5.0 ± 0.1 Chromium 0.003195 ± 0.000071
Sulfate 502 ± 29 Arsenic 0.00049 ± 0.00015
Nitrate 0.50 ± 0.21 Iron 5.20 ± 0.11
Chloride 43.1 ± 2.8 Nickel 0.05924 ± 0.00043
Fluoride 2.23 ± 0.25 Copper 0.03305 ± 0.00031
Sodium 22.901 ± 0.066 Zinc 0.1482 ± 0.0026
Potassium 6.379 ± 0.093 Cadmium 0.000279 ± 0.000029
Calcium 96.6 ± 2.2 Manganese 1.407 ± 0.014
Magnesium 27.36 ± 0.16 Lead 0.001080 ± 0.000085

Extremophiles
1 3
typical freshwater rivers; the sulfate level was particularly
high at ~0.5 g/L, which is characteristic of ARD environ-
ments. Heavy metals like arsenic, chromium, cadmium,
zinc, copper, nickel and lead were detected at minute
concentrations. The solid material from the water column
(Fig. 2c) was further characterized with SEM–EDS and
XRD (Figs. 3 and S1A). The SEM micrographs (Fig. 3a,
b) showed that the sediment consists of lumps of mainly
amorphous material of nanometer size. The EDS spectrum
(Fig. 3c) of the particulate material demonstrates the pres-
ence of oxygen (51.1%), iron (32.6%), silicon (5.3%), sul-
fur (4.9%) and aluminum (3.0%) as the principal elemental
components. The EDS analysis suggests that the sediment
is formed almost completely of iron oxides with alumino-
silicates at only a few percent by mass. The XRD pattern
of the solid material (Fig. S1A) pointed to the presence
of quartz and clays but the amorphous iron oxides, which
are the main constituents of the sample, produced only
broad diffraction signals. As the solid material was weakly
attracted to a strong rare-earth magnet, we applied this
property to isolate the iron-oxide particles. Taping the mag-
net to the side of a bottle with suspended matter allowed
us to pipette out a small amount of adhered particles after
sedimentation was complete and thus eventually to collect
Fig. 2 Water samples of Río
Sucio. a, b Laboratory samples
showed the complete precipita-
tion of the solid material within
24 h near 23 °C. c Material
recovered from the water sam-
ples after filtering
Fig. 3 SEM-EDS analysis
of Río Sucio sediments. a, b
SEM micrographs to 500 μm
(20.0 kV 7.6 mm × 60)
and 20 μm (20 kV
7.1 mm × 2.70 k), respectively,
of the brownish-yellow powder
sedimented at the bottom of Río
Sucio. Micrographs indicate
that the sediment consists of
lumps of mainly amorphous
material of nanometer size. c
EDS analysis of the brownish-
yellow powder showed that it
is composed of mainly iron and
oxygen. Other species such as
silicon, sulfur, aluminum and
potassium were found in the
sediment in minor proportions

Extremophiles
1 3
enough adhered material to obtain improved infrared (Fig.
S2) and XRD data (Fig. S1B) on the isolated powder. The
infrared spectrum clearly shows hydrated iron oxide and
sulfate lines at these wavenumbers and relative intensi-
ties: 498w, 619w, 704 m, 1124 s (SO42−), 1636 m (H2O),
3393 s (H2O) (Fig. S2). Yellow iron oxides that contain sul-
fate are formed when precipitation occurs in a sulfuric acid
medium, as in the recently characterized iron-oxyhydroxy-
sulfate Schwertmannite (Fernández-Martínez et al. 2010).
If chloride is present in the acidic medium, Akaganeite
forms. The XRD pattern of our sample (Fig. S1B) is iden-
tical to that of Schwertmannite and also of sulfate-substi-
tuted Akaganeite (Bakoyannakis et al. 2003). In summary,
the physico-chemical data are consistent with the composi-
tion of an ARD environment where the high sulfate concen-
trations, presence of iron minerals and acidity most likely
originate from the oxidation of sulfidic compounds.
Analysis of microbial communities thriving in Río
Sucio ecosystem
In total 64 phylotypes were obtained and grouped into nine
phyla from both bacterial and archaeal domains (Fig. 4a;
Supplementary Table S2). The largest number of sequence
reads (~89.39%) was assigned to the phylum Proteobac-
teria, in particular to class Betaproteobacteria (~ 80.16%).
Members of the Gamma—(~7.78%), Alpha—(~0.49%)
Delta—(~0.08%) and Epsilon—(~0.08%) proteobacteria
were found to lesser extents. The remaining bacterial phyla
identified correspond to Nitrospirae, Actinobacteria, Par-
cubacteria (OD1), Bacteroidetes, Spirochaetes, Elusimi-
crobia and Firmicutes, which accounted, respectively, for
~5.54, ~1.58, ~1.45, ~0.23, ~0.17, ~0.09 and ~0.08% of the
sequence reads.
Further analysis of the microbial composition of Río
Sucio showed a community dominated by microorganisms
commonly found in ARD environments like iron-, sulfur-,
sulfide- and thiosulfate-oxidizing bacteria (Figs. 4, 5). As
shown in Fig. 4b, the betaproteobacterial families Gal-
lionellaceae and “Ferrovaceae” were particularly abundant
in Río Sucio. Moreover, a single Gallionella phylotype
(OTU RS001) comprised 43.89% of total 16S rRNA gene
reads, this organism being by far the most abundant in the
river. Members of family Gallionellaceae has been found in
a variety of different habitats including freshwater ecosys-
tems at neutral pH (Haaijer et al. 2008) and ARD-related
ecosystems such as mine tailings (Bruneel et al. 2006; He
et al. 2007), plants to treat mine water (Heinzel et al. 2009)
and acidic subterranean waters (Kimura et al. 2011). Specif-
ically, the taxonomic breadth of OTU RS001 indicates that
it is closely affiliated to Gallionella sp. JA52 (Fig. 5). This
bacterium was isolated from a mine water treatment plant
for the biological oxidation of iron(II), in which members
from the genera ‘Ferrovum’ and Gallionella were highly
represented (Heinzel et al. 2009; Tischler et al. 2013). In
accordance with the results from this mine water treatment
plant and from other ARD environments (Kimura et al.
2011; Fabisch et al. 2013; Tischler et al. 2013; Jones et al.
Fig. 4 Taxonomic composition
of Río Sucio waters. Relative
abundance of bacterial and
archaeal organisms to phylum
(a) and family (b) taxonomic
levels. The classification of
the OTUs was performed by
uploading their V5–V6 DNA
sequence to the Classify and
Seqmatch tools from the Ribo-
somal Database Project (RDP),
as described in the methods

Extremophiles
1 3
2015), several ‘Ferrovum myxofaciens’-like members were
also highly abundant in Río Sucio (32.6% of total reads).
Both ‘F. myxofaciens’ and members of genus Gallionella
are autotrophic iron oxidizers, which catalyze the oxidation
of iron(II). This oxidation might result in the precipitation
of Fe3+ in minerals such as iron(III) oxyhydroxides and
oxyhydroxysulfates (e.g., Schwertmannite) (Bigham et al.
1990; Straub et al. 2000; Hedrich et al. 2011). In addition
to these two highly represented genera, we identified the
presence of other microorganisms closely related to aci-
dophilic, iron-oxidizing bacteria such as Acidithiobacillus
spp. and Leptospirillum spp. (representing ~6 and 5.54% of
total reads, respectively). Other bacteria related to common
iron oxidizers in ARD environments from the genera Acidi-
ferrobacter (A. thiooxydans) or Ferritrophicum (F. radici-
cola) were detected in smaller amounts.
Fig. 5 Phylogenetic classification of bacteria found in Río Sucio
waters. The evolutionary history of bacterial OTU was inferred with
a maximum-likelihood method based on the general time-reversible
model using MEGA. Bootstrap values (in total 100 bootstrap replica-
tions were calculated) are represented for the main branches of the
phylogenetic tree. Phylotypes belonging to Archaea were excluded
from this analysis

Extremophiles
1 3
Besides bacteria associated with iron transformations,
many microorganisms involved in sulfur metabolism were
identified. The lithotrophic bacterium Gallionella ferrug-
inea is also capable of utilizing reduced sulfur compounds
(sulfide and thiosulfate) as electron donors and energy
sources (Lütters-Czekalla 1990). The high abundance of
the Gallionella OTU RS001 described above indicates
that sulfur metabolism is also important for the bioen-
ergetics of the Río Sucio microbial community. Another
organism related with sulfur metabolism found in Río
Sucio waters was Acidithiobacillus ferrooxidans. This
chemolithotrophic bacterium is also capable of oxidizing
sulfide to sulfate, coupling this reaction to iron(III) (under
anoxic conditions) or oxygen reduction (under oxic con-
ditions) (Suzuki et al. 1990; Pronk et al. 1992). Thus, part
of iron(III) could be reduced, coupled to the oxidation of
S2− under anoxic conditions by A. ferrooxidans, or cou-
pled to the oxidation of organic matter by heterotrophic
acidophiles such as Acidiphilium spp. (e.g., A. multivo-
rum, A. rubrum) or Acidocella spp. (Fig. 5; Supplemen-
tary Table S2). These results suggest that both Gallionella
spp. and A. ferrooxidans are key bacteria in the Río Sucio
ecosystem as they participate in both iron and sulfur
metabolism.
According to Bohorquez et al. (2012), the primers used
for the amplification of V5–V6 regions of 16S rRNA gene
could theoretically amplify >97% of Archaeal reads from
the RDP database. We found only five archaeal representa-
tives that belong to phylum Euryarcheota (order Thermo-
plasmatales), representing 1.12% of the total microbial
community (Supplementary Table S2); a similar pattern
is found in other ARD systems, such as in the sediments
of the extremely acidic Río Tinto (Sánchez-Andrea et al.
2011). We were unable to examine further the classifica-
tion of this archaeal OTUs as their DNA sequences are not
related to known cultured specimens.
Conclusions
This work corresponds to the first report where the physico-
chemical and microbiological composition of Río Sucio is
described. The geographic location of this river suggests
that it corresponds to a natural ARD system, which results
from hydrothermal springs along the cliff of the Río Sucio
fault (Dr. Rolando Mora, personal communication). Our
data are consistent with the chemistry of the river being
largely created and maintained by microorganisms, i.e, the
extreme conditions of its waters are biologically driven.
Río Sucio is one of few ARD environments in which condi-
tions are due solely to natural processes, unlike most ARD
environments around the world that have been influenced
by human activity, in particular mining.
Acknowledgements We are grateful to Ricardo Amils from the
Astrobiology Centre (CSIC-INTA) for critical comments and helpful
suggestions on this manuscript, and Raul Mora and Carlos Ramírez
of the Costa Rican National Seismological Network (RSN) for inspir-
ing conversations. We also thank to Rolando Mora from Escuela Cen-
troamericana de Geología (Universidad de Costa Rica) by providing
information about the origin of Río Sucio and Solange Voysest for
help in the design of some figures.
Compliance with ethical standards
Funding This work was supported by resources of the Vice-rectory of
Research of Universidad de Costa Rica (809-B4-282) and by the ERC
grant IPBSL (ERC250350-IPBSL) to Ricardo Amils and Kenneth N.
Timmis. Computational resources were provided by the Data Inten-
sive Academic Grid, which is supported by the USA National Science
Foundation (0959894). F.P.S. was supported by a JAE-pre fellowship
from the Spanish Consejo Superior de Investigaciones Científicas
(CSIC) and the European Union FP7 programme Grant Agreement
(607346).
References
Amils R, Fernández-Remolar D, The IPBSL Team (2014) Río Tinto:
a geochemical and mineralogical terrestrial analogue of Mars.
Life 4:511–534
Baker BJ, Banfield JF (2003) Microbial communities in acid mine
drainage. FEMS Microbiol Ecol 44:139–152
Bakoyannakis DN, Deliyanni EA, Zouboulis AI et al (2003) Akaga-
neite and goethite-type nanocrystals: synthesis and characteriza-
tion. Microporous Mesoporous Mater 59:35–42
Bigham JM, Schwertmann U, Carlson L et al (1990) A poorly crys-
tallized oxyhydroxysulfate of iron formed by bacterial oxida-
tion of Fe(II) in acid mine waters. Geochim Cosmochim Acta
54:2743–2758
Bohorquez LC, Delgado-Serrano L, López G et al (2012) In-depth
characterization via complementing culture-independent
approaches of the microbial community in an acidic hot spring
of the Colombian Andes. Microb Ecol 63:103–115
Bruneel O, Duran R, Casiot C et al (2006) Diversity of microorgan-
isms in Fe–As-rich acid mine drainage waters of Carnoulès,
France. Appl Environ Microbiol 72:551–556
Burbach K, Seifert J, Pieper DH et al (2016) Evaluation of DNA
extraction kits and phylogenetic diversity of the porcine gastroin-
testinal tract based on Illumina sequencing of two hypervariable
regions. Microbiologyopen 5:70–82
Camarinha-Silva A, Jáuregui R, Chaves-Moreno D et al (2014) Com-
paring the anterior nare bacterial community of two discrete
human populations using Illumina amplicon sequencing. Envi-
ron Microbiol 16:2939–2952
Castellón E, Martínez M, Madrigal-Carballo S et al (2013) Scattering
of light by colloidal aluminosilicate particles produces the unu-
sual sky-blue color of Río Celeste (Tenorio Volcano complex,
Costa Rica). PLoS One 8:e75165
Dold B, González-Toril E, Aguilera A et al (2013) Acid rock drain-
age and rock weathering in antarctica: important sources for iron
cycling in the southern ocean. Environ Sci Technol 47:6129–6136
Edgar RC, Haas BJ, Clemente JC et al (2011) UCHIME improves
sensitivity and speed of chimera detection. Bioinformatics
27:2194–2200
Edwards KJ, Rodgers TM, Schrenk MO et al (1999) Geomicrobiol-
ogy of pyrite (FeS2) dissolution: case study at Iron Mountain,
California. Geomicrobiol J 16:155–179

Extremophiles
1 3
Fabisch M, Beulig F, Akob DM et al (2013) Surprising abundance of
Gallionella-related iron oxidizers in creek sediments at pH 4.4 or
at high heavy metal concentrations. Front Microbiol 4:390
Farrand WH (1997) Identification and mapping of ferric oxide and
oxyhydroxide minerals in imaging spectrometer data of Summit-
ville, Colorado, USA, and the surrounding San Juan Mountains.
Int J Remote Sens 18:1543–1552
Fernández-Martínez A, Timon V, Romaman-Ross G et al (2010) The
structure of schwertmannite, a nanocrystalline iron oxyhydroxy-
sulfate. Am Mineral 95:1312–1322
Fernández-Remolar DC (2003) Geological record of an acidic envi-
ronment driven by iron hydrochemistry: the Tinto River system.
J Geophys Res 108:1–15
García-Moyano A, González-Toril E, Ángeles Aguilera et al (2012)
Comparative microbial ecology study of the sediments and the
water column of the Río Tinto, an extreme acidic environment.
FEMS Microbiol Ecol 81:303–314
González-Toril E, Llobet-Brossa E, Casamayor EO et al (2003)
Microbial ecology of an extreme acidic environment, the Tinto
River. Appl Environ Microbiol 69:4853–4865
González-Toril E, Santofimia E, Blanco Y et al (2015) Pyrosequenc-
ing-based assessment of the microbial community structure of
Pastoruri glacier area (Huascarán National Park, Perú), a natural
extreme acidic environment. Microb Ecol 70:936–947
Haaijer SCM, Harhangi HR, Meijerink BB et al (2008) Bacteria asso-
ciated with iron seeps in a sulfur-rich, neutral pH, freshwater
ecosystem. ISME J 2:1231–1242
He Z, Xiao S, Xie X et al (2007) Molecular diversity of microbial
community in acid mine drainages of Yunfu sulfide mine. Extre-
mophiles 11:305–314
Hedrich S, Lunsdorf H, Kleeberg R et al (2011) Schwertmannite for-
mation adjacent to bacterial cells in a mine water treatment plant
and in pure cultures of Ferrovum myxofaciens. Environ Sci Tech-
nol 45:7685–7692
Heinzel E, Hedrich S, Janneck E et al (2009) Bacterial diversity in a
mine water treatment plant. Appl Environ Microbiol 75:858–861
Johnson DB (1998) Biodiversity and ecology of acidophilic microor-
ganisms. FEMS Microbiol Ecol 27:307–317
Johnson DB, Hallberg KB (2005) Acid mine drainage remediation
options: a review. Sci Total Environ 338:3–14
Jones DS, Kohl C, Grettenberger C et al (2015) Geochemical niches
of iron-oxidizing acidophiles in acidic coal mine drainage. Appl
Environ Microbiol 81:1242–1250
Kimura S, Bryan CG, Hallberg KB et al (2011) Biodiversity and geo-
chemistry of an extremely acidic, low-temperature subterranean
environment sustained by chemolithotrophy. Environ Microbiol
13:2092–2104
Kozich JJ, Westcott SL, Baxter NT et al (2013) Development of a
dual-index sequencing strategy and curation pipeline for analyz-
ing amplicon sequence data on the miseq illumina sequencing
platform. Appl Environ Microbiol 79:5112–5120
López-Archilla AI, Gérard E, Moreira D et al (2004) Macrofilamen-
tous microbial communities in the metal-rich and acidic River
Tinto, Spain. FEMS Microbiol Lett 235:221–228
Lütters-Czekalla S (1990) Lithoautotrophic growth of the iron bacte-
rium Gallionella ferruginea with thiosulfate or sulfide as energy
source. Arch Microbiol 154:417–421
Montero W, Denyer P, Barquero R et al. (1998) Map of Quaternary
Faults and Folds of Costa Rica. United States Geological Survey,
Open-file report 98–481. http://pubs.usgs.gov/of/1998/ofr-98-
0481/. Accessed 04 Dec 2015
Olías M, Nieto JM (2015) Background conditions and mining pollu-
tion throughout history in the Río Tinto (SW Spain). Environ-
ments 2:295–316
Pronk JT, De Bruyn JC, Bos P et al (1992) Anaerobic growth of Thio-
bacillus ferrooxidans. Appl Environ Microbiol 58:2227–2230
Pruesse E, Peplies J, Glöckner FO (2012) SINA: accurate high-
throughput multiple sequence alignment of ribosomal RNA
genes. Bioinformatics 28:1823–1829
Puente-Sánchez F, Aguirre J, Parro V (2016) A novel conceptual
approach to read-filtering in high-throughput amplicon sequenc-
ing studies. Nucleic Acids Res 44:e40
Quast C, Pruesse E, Yilmaz P et al (2013) The SILVA ribosomal RNA
gene database project: improved data processing and web-based
tools. Nucleic Acids Res 41:D590–D596
Sánchez-Andrea I, Rodríguez N, Amils R et al (2011) Microbial
diversity in anaerobic sediments at Río Tinto, a naturally acidic
environment with a high heavy metal content. Appl Environ
Microbiol 77:6085–6093
Sánchez-Andrea I, Rojas-Ojeda P, Amils R et al (2012) Screening of
anaerobic activities in sediments of an acidic environment: tinto
River. Extremophiles 16:829–839
Schloss PD, Westcott SL, Ryabin T et al (2009) Introducing mothur:
open-source, platform-independent, community-supported soft-
ware for describing and comparing microbial communities. Appl
Environ Microbiol 75:7537–7541
Straub KL, Benz M, Schink B (2000) Iron metabolism in anoxic envi-
ronments at near neutral pH. FEMS Microbiol Ecol 34:181–186
Suzuki I, Takeuchi TL, Yuthasastrakosol TD et al (1990) Ferrous iron
and sulfur oxidation and ferric iron reduction activities of Thio-
bacillus ferrooxidans are affected by growth on ferrous iron, sul-
fur, or a sulfide ore. Appl Environ Microbiol 56:1620–1626
Tamura K, Stecher G, Peterson D et al (2013) MEGA6: molecu-
lar evolutionary genetics analysis version 6.0. Mol Biol Evol
30:2725–2729
Tischler JS, Jaffer R, Gelhaar N et al (2013) New cultivation medium
for “Ferrovum” and Gallionella-related strains. J Microbiol
Methods 95:138–144
Wang Q, Garrity GM, Tiedje JM et al (2007) Naive Bayesian classi-
fier for rapid assignment of rRNA sequences into the new bacte-
rial taxonomy. Appl Environ Microbiol 73:5261–5267