Biomarkers and Metabolic Patterns in the Sediments of Evolving Glacial Lakes As a Proxy for Planetary Lake Exploration

ASTROBIOLOGY Volume 18, Number 5, 2018 ª Mary Ann Liebert, Inc. DOI: 10.1089/ast.2015.1342

Biomarkers and Metabolic Patterns in the Sediments of Evolving Glacial Lakes as a Proxy for Planetary Lake Exploration

Vıćtor Parro,1 Yolanda Blanco,1 Fernando Puente-Sańchez,1 Luis A. Rivas,1 Mercedes Moreno-Paz,1 Alex Echeverrıá,2 Guillermo Chong-Dıáz,2 Cecilia Demergasso,2 and Nathalie A. Cabrol3

Abstract

Oligotrophic glacial lakes in the Andes Mountains serve as models to study the effects of climate change on natural biological systems. The persistent high UV regime and evolution of the lake biota due to deglaciation make Andean lake ecosystems potential analogues in the search for life on other planetary bodies. Our objective was to identify microbial biomarkers and metabolic patterns that represent time points in the evolutionary history of Andean glacial lakes, as these may be used in long-term studies as microscale indicators of climate change processes. We investigated a variety of microbial markers in shallow sediments from Laguna Negra and Lo Encanãdo lakes (Regioń Metropolitana, Chile). An on-site immunoassay-based Life Detector Chip (LDChip) revealed the presence of sulfate-reducing bacteria, methanogenic archaea, and exopolymeric substances from Gammaproteobacteria. Bacterial and archaeal 16S rRNA gene sequences obtained from field samples confirmed the results from the immunoassays and also revealed the presence of Alpha-, Beta-, Gamma-, and Deltaproteo- bacteria, as well as cyanobacteria and methanogenic archaea. The complementary immunoassay and phylogenetic results indicate a rich microbial diversity with active sulfate reduction and methanogenic activities along the shoreline and in shallow sediments. Sulfate inputs from the surrounding volcanic terrains during deglaciation may explain the observed microbial biomarker and metabolic patterns, which differ with depth and between the two lakes. A switch from aerobic and heterotrophic metabolisms to anaerobic ones such as sulfate reduction and methanogenesis in the shallow shores likely reflects the natural evolution of the lake sediments due to deglaciation. Hydrodynamic deposition of sediments creates compartmentalization (e.g., sediments with different structure and composition surrounded by oligotrophic water) that favors metabolic transitions. Similar phenomena would be expected to occur on other planetary lakes, such as those of Titan, where watery niches fed by depositional events would be surrounded by a ‘‘sea’’ of hydrocarbons. Key Words: Glacier lakes—Sedimentation—Prokaryotic metabolisms and biomarkers—Deglaciation—Life detection—Planetary exploration. Astrobiology 18, 586–606.

1. Introduction of the world’s largest sources of glacial water, which includes mountain glaciers, ice, and snowfields that are receding rap- ith global temperature rising, ice worldwide re- idly (Haeberli et al., 2002; Le Quesne and Acunã, 2003; Downloaded by Consejo Superior De Investigaciones Cientificas CSIC from www.liebertpub.com at 07/22/18. For personal use only. Wtreats and thins. It is projected that many low-altitude Coudrain et al., 2005; Bradley et al., 2006; Rivera et al., 2007; glaciers could disappear within 20 years (IPCC, 2007). Long- Vuille et al., 2008; Le Quesne et al., 2009). Glacial lakes and term, multiproxy studies in regions between 33S and 36Sin their sediments are highly sensitive temporal markers of en- Chile and Argentina have shown a mean frontal retreat of vironmental variability, which in turn affects their biota. between -50 and -9my-1, thinning rates between 0.76 and Microbiological studies of oligotrophic Andean lakes have 0.56 m y-1, and a mean ice area reduction of 3% since 1955 shown that changes in the water column occur in association (Roig et al., 2000; Le Quesne and Acunã, 2003; Lara et al., with fluctuations in water transparency or turbidity during de- 2005; Vuille, 2006; Le Quesne et al., 2009). The IPCC (2007) glaciation (Modenutti et al., 2012). During early deglaciation lists the Central and Southern Andean countries as particu- (phase 1), the high silt content of water protects microorgan- larly vulnerable (Painter, 2007). Among those, Chile has one isms from the high UV that occurs at the higher altitudes of

1Department of Molecular Evolution, Centro de Astrobiologı´a (INTA-CSIC), Madrid, Spain. 2Centro de Biotecnologı´a ‘‘Profesor Alberto Ruiz,’’ Universidad Cato´lica del Norte, Antofagasta, Chile. 3The SETI Institute, Carl Sagan Center, Mountain View, California, and NASA Ames Research Center, Moffett Field, California, USA.

586 BIOMARKERS IN EVOLVING GLACIAL LAKES 587

glacial lakes. With time, however, as glaciers continue to re- past 10 years and has been implemented—along with the cede, the meltwater discharge and sediment load into the lakes Signs of Life Detector (SOLID) instrument—for in situ detec- decrease (phase 2), and the transparency and UV levels of the tion of biomarkers (Parro et al., 2008, 2011b). The LDChip has water column increase. A comparative study of six Andean been used for in situ detection of prokaryotes and biomarker ultra-oligotrophic lakes characterized by low-phosphorus profiling in different extreme environments that include the concentrations revealed that high photosynthetically active acidic iron-rich sediments of the Rıó Tinto in Spain (Parro et al., radiation (PAR), UVA, and UVB forced planktonic organisms 2008, 2011c; Puente-Sańchez et al., 2014), subsurface sedi- into deeper layers of the water column (Callieri et al., 2007). In ments (down to 5 m depth) cored in the hypersaline Atacama contrast, the microbial diversity (i.e., picocyanobacterial as- Desert (Parro et al., 2011a; Fernańdez-Remolar et al., 2013), semblages) of such lakes may remain high, a finding attributed and the surface and permafrost (down to 4.2 m) sediments to habitat fragmentation generated by geographic barriers, drilled on Deception Island in Antarctica (Blanco et al., 2012). which resulted in rapid speciation (Caravati et al., 2010). The LDChip results reported here reveal a rich geochemistry in The combination of environmental and climate factors can Andean lakes that is capable of sustaining an active anaerobic also affect the biodiversity of microbial communities in gla- metabolism and broad microbial diversity. Such data also cial lake sediments. Leoń et al. (2012) reported differences in provides a baseline for further monitoring to understand the the metabolic activity and structural and functional compo- evolution of deglaciation as recorded in glacial lake sediments. sition of bacterial communities between the sediments of three Patagonian Chilean oligotrophic lakes of quaternary 2. Geological Setting glacial origin. They attributed these differences to the unique geomorphological pattern of each lake due to both local (e.g., Laguna Negra and Lo Encanãdo are located on the south slope volcanic activity) and global (climate change) disturbances. of the Echaurren glacier watershed in the Central Andes of Chile Such examples illustrate how the microbial communities (33.65S, 70.13W; Fig. 1). Monitoring these lakes over time that live in the water column and sediments of glacial lakes and would allow for characterization of their prokaryotic diversity, reservoirs reflect their environmental setting and geochemical physical processes, and spatiotemporal changes. The two lakes input. Although lower and more acidic precipitation, increased are part of a complex of freshwater resources in the Santiago area temperature, and snow melting that accompanies deglaciation that includes El Yeso lake to the east, which is damned (on its strongly affect the dissolved organic carbon concentration of southern end). Though Laguna Negra and Lo Encanãdo were glacial lakes (Beniston et al., 1997; Sala et al., 2000; Wil- connected by a human-made overflow tunnel, the lake level of liamson et al., 2009), microbes ultimately control the carbon Laguna Negra has decreased enough that the two lakes have cycle of glacial lakes through metabolic processes, such as been isolated from each other for at least the past 5 years. fermentation of complex organic matter or the production of Both lakes are located in the same catchment area (up to methane (Wadham et al.,2008). 4600 m in elevation), yet they are fed by two distinct stream Oligotrophic water, characterized by low organic carbon systems and have contrasting physical characteristics: Laguna and microbial biomass, is a general feature of glacial Andean Negra is a large (6.1 · 1.7 km) and deep (276 m in 2013) lake lakes. Although some works have reported on the microbial located at 2700 m above sea level, whereas Lo Encanãdo is content, photosynthetic activity, and geochemistry of the water a smaller, 980 · 635 m wide, and 45 m deep lake located at column of these lakes, very little is actually known about their 2492 m elevation. sediments, which may reveal key aspects of the evolution of The watershed area of the lakes is composed primarily of these important ecosystems with regard to ecology, water re- volcanic constructs and basalt and andesite deposits that sources (e.g., for human consumption), and the evolution of were covered by modern glacial deposits from the receding glacier lakes at planetary scales. Determining whether there are Echaurren glacier, which now lies above the lakes at 3500 m predictable microbial successions or key biomarkers that might elevation. Lahar deposits mark recent to modern interactions be indicative of the progressive evolution of glacial lake sys- between volcanic and glacial activity on the west shore of tems, and whether those biomarkers are preserved in their Laguna Negra. Residual frontal moraines enclose the basin sediments, could aid in our understanding of the evolutionary on its south and southeast shore and separate Laguna Negra history of these lakes and reservoirs. from Lo Encanãdo to the south and El Yeso to the east. Here, we report the results of a geomicrobiological study Moraine deposits, as well as large glacial erratics at the that included biomarker profiling via the use of an on site bottom of the lake, consist of residual granodiorite that was Downloaded by Consejo Superior De Investigaciones Cientificas CSIC from www.liebertpub.com at 07/22/18. For personal use only. multiplex immunological assay and additional laboratory- excavated by glacial erosion from the basement of the basin. based analyses of sediments from two oligotrophic Andean Both lakes are monomictic and oligotrophic and in dif- lakes. Immunological techniques are advantageous for field ferent phases of the deglaciation process: Laguna Negra is study because sample preparation is relatively easy and can highly transparent (i.e., phase 2), while the latter is turbid as be performed in the field. Conventional molecular ecology a consequence of the larger biomass content (i.e., phase 3), techniques used for massive DNA sequencing of environ- which translates into distinct orbital spectral signatures in mental samples also require time-intensive and specialized visible (black and green, respectively) spectra. expertise, which is not required for immunological assays. The Life Detector Chip (LDChip) is an antibody microarray- 3. Materials and Methods based biosensor designed for in situ life-detection studies of 3.1. Sampling and environmental parameters analog deposits in preparation for planetary exploration and for monitoring microbial diversity and metabolisms in ex- Sediments were sampled from the shoreline of Laguna treme environments (Rivas et al., 2008; Parro et al., 2008, Negra (LNS, coordinates 3366061S, 7012161W) and Lo 2011a, 2011b). It has been developed over the course of the Encanãdo (LES, coordinates 33674168S, 701278W) and 588 PARRO ET AL.

FIG. 1. ASTER image showing (1) Laguna Negra, (2) Lo Encanãdo, and (3) El Yeso, and their respective spectral orbital signatures in the visible. El Yeso is the richest in glacial flour, which gives its aqua color. Topographical profiles A–B and C–D are shown.

at 4 m water depth in Laguna Negra (LN4) during a field and oxalate) of interest. All samples were run on a Metrohm campaign in 2011. Sample LN4 was collected by diving to 861 Advanced Compact IC (Metrohm AG, Herisau, Switzer- 4 m depth below the water surface some meters away from the land) as described previously (Parro et al.,2011a). shore. Sediments along the shorelines of both lakes appeared Two types of samples were extracted for geochemical similar, that is, highly granulated and low in organic material, analysis via the ion chromatograph (IC): the interstitial fluids whereas the sediment collected from the lake bottom con- of the sediments and supernatants prepared by physical agi- tained significantly more decomposing matter. All the sedi- tation of the sediments after they had been separated from the Downloaded by Consejo Superior De Investigaciones Cientificas CSIC from www.liebertpub.com at 07/22/18. For personal use only. ments were collected in 50 mL sterile Falcon tubes and kept interstitial fluids. To separate the interstitial water (IW) from refrigerated during transport until they reached the laboratory the sediments, 2 g of wet sediment from each sample was and were analyzed. Conductivity, pH, temperature, and redox centrifuged at 2000 g, and 1 mL of the IW was removed from potential were measured at the time of sample collection with the tube with the sediment ‘‘pellet.’’ The 1 mL of IW was then a multiparametric portable instrument (WTW, Germany). diluted in 10 or 20 mL of distilled water (analytical grade) to The sediment samples were analyzed in the field by the im- let the IC values fit into the calibration curves and to have munological biosensor LDChip (see below) and then in the different IC measurements. To extract the small organic laboratory by ion chromatography and DNA sequencing. molecules and anions of interest that were associated with the surfaces of the sediment materials (minerals and particulate organic matter), any remaining IW supernatant was removed 3.2. Geochemical analysis from the centrifuged sediment samples so that the sediments Ion chromatography was used to measure the concentration could be resuspended in 10 mL of sterile distilled water, - = - - = of inorganic anions (Cl ,SO4 ,NO2 ,NO3 ,PO4 ) and low- mixed in a vortex, and then physically agitated for 1 h. After molecular-weight organic acids (acetate, propionate, formate, the solid particles were removed by centrifugation at 2000 g BIOMARKERS IN EVOLVING GLACIAL LAKES 589

for 10 min, the supernatants of these samples were then To prepare the samples for sandwich microarray immuno- analyzed by ion chromatography to measure the small or- assay analysis, approximately 0.5 g of sediment was collected ganic molecules and inorganic anions in the same manner as from the field, resuspended in 2 mL of TBSTRR, and sonicated the IW, as noted above. (3 · 1 min cycle) with a handheld Ultrasonic Processor (UP50H, Hielscher, Germany) at the field site. The sediment was decanted 3.3. Antibody microarrays: printing LDChip for 2–5 min, depending the clay content, and 50 lL of the crude and fluorescent labeling of antibodies environmental extracts were injected into each of the nine incubation chambers and incubated for 1 h with the LDChip at In this work, the LDChip antibody microarray contained 193 ambient temperature. After a wash with TBSTRR, all chambers antibodies (see Appendix Table 1 on page 000) and was designed were flooded with the fluorescently labeled antibody mixture for to enable the identification of bacterial strains from each main 1 h. The slides were then washed, dried, and scanned for fluo- phylogenetic group of bacteria and archaea; a variety of uni- rescence at 635 nm in a GenePix4100A scanner (in a nearby versal proteins, peptides, and biological polymers (nucleic acids, laboratory). Parallel immunoassays were performed as a blank lipopolysaccharides, exopolysaccharides) used in modern (and control by using only buffer as a negative control and treated presumably ancient) metabolisms (e.g., nitrogen fixation, sulfate with the same fluorescent antibody mixture. All sets of newly reduction, iron homeostasis, and nitrate metabolism); see also printed LDChips are tested in the laboratory with randomly se- Rivas et al. (2008, 2011) and Parro et al. (2008, 2011a, 2011c). lected antibody-immunogen pairs. We have demonstrated that The immunoglobulin (IgG) fraction of each of the 193 protein A- the LDChip is fully functional for at least 9 months of storage at purified antibodies was printed in a spot pattern in duplicate on ambient temperature (de Diego-Castilla et al., 2011). Ad- the surface of epoxy-activated glass slides (Arrayit, CA, USA) ditionally, as a positive control sample to be used in the field, we with a MicroGrid II TAS arrayer (Biorobotics, Genomic Solu- spiked the buffer with the human hepatitis B antigen and ran a tions, UK) as reported previously (Rivas et al., 2008; Parro et al., test to verify that the antibodies were still functional. The scan- 2011a). For this particular study, 141 out of the 193 antibodies in nedimageswereanalyzedinthefieldwithGenePixProsoftware Appendix Table 1 (52 antibodies produced against RıóTinto (Genomic Solutions) installed on a laptop computer. The final natural samples were excluded to avoid bias in the results) were fluorescence intensity was quantified as previously reported fluorescently labeled with Alexa 647 (Molecular Probes) as (Parro et al., 2011a; Rivas et al., 2011). tracers for fluorescent immunoassay detection. To reveal the immunoreactions on the antibody micro- 3.5. DNA extraction and sequencing array, we produced a homogenate that consisted of 141 fluorescent antibodies (final concentration was 100 lg/mL). The DNA was extracted from 10 g of sediments with the MoBio concentration of the individual antibodies in the mixture DNA extraction kit according to the manufacturer’s instruc- (inferred via titration, data not shown) averaged 0.7 lg/mL. tions. The 16S rRNA gene from Bacteria and Archaea was This mixture was frozen and then lyophilized for transport PCR amplified, cloned, and sequenced as described previously and use in the field at a 1/10 final working dilution. (Parro et al., 2011a). Additionally, the amplicons were sub- Six new antibodies have been used in this work for the first jected to high throughput sequencing by the 454 Roche pyro- time (Appendix Table 1, bottom): one genus-specific anti- sequencing system (Lifesequencing S.L., Valencia, Spain). body produced to recognize Polaromonas sp., two strain- For pyrosequencing, the V3–V5 region of the 16S rRNA gene specific antibodies produced to recognize Planococcus was amplified using key-tagged eubacterial primers (Life- spp. (kindly supplied by Dr. Lyle Whyte, McGill University, sequencing S.L.) based on the design of Sim et al. (2012). PCR Montreal, Canada), and three antibodies produced as de- reactions were performed with 20 ng of metagenomic DNA, scribed previously (Rivas et al., 2008) that reacted positively 200 lM of each of the four deoxynucleoside triphosphates, to environmental extracts of samples collected from (a) 2 m 400 nM of each primer, 2.5 U of FastStart HiFi Polymerase and depth in the Atacama Desert, (b) a permafrost sample from the appropriate buffer with MgCl2 supplied by the manufac- Deception Island, Antarctica, and (c) a biofilm growing on a turer (Roche, Germany), 4% of 20 g/mL BSA (Sigma, Dorset, concrete wall on a building structure next to Centro de As- UK), and 0.5 M betaine (Sigma). Thermal cycling consisted of trobiologıá (Madrid, Spain). The titration and detailed per- an initial denaturation at 94C for 2 min followed by 35 cycles formance of these six antibodies will be reported elsewhere. of denaturation at 94C for 20 s, annealing at 50C for 30 s, and extension at 72C for 5 min. Amplicons were combined in a

Downloaded by Consejo Superior De Investigaciones Cientificas CSIC from www.liebertpub.com at 07/22/18. For personal use only. 3.4. Sandwich microarray immunoassays single tube in equimolar concentrations. The pooled amplicon mixture was puriﬁed twice (AMPure XP kit, Agencourt, Sandwich-type microarray immunoassays were performed Takeley, UK), and the cleaned pool was requantiﬁed by using as described previously (Parro et al., 2011a; Blanco et al., the PicoGreen assay (Quant-iT, PicoGreen DNA assay, In- 2012). In summary, printed microscope slides with the anti- vitrogen). Subsequently, an amplicon was submitted to the body microarray (LDChip with 193 antibodies) were blocked pyrosequencing services provided by Lifesequencing S.L. with 0.5 M Tris-HCl in 5% BSA for 5 min. Slides were then (Valencia, Spain) where EmPCR was performed. Subse- immersed in 0.5 M Tris-HCl with 2% BSA and gently agitated quently, unidirectional pyrosequencing was carried out on a for 30 min. After washing with TBSTRR buffer (0.4 M Tris- 454 Life Sciences GS FLX+ instrument (Roche) following the HCl pH 8, 0.3 M NaCl, 0.1% Tween 20) and drying the chip by Roche Amplicon Lib-L protocol. quick centrifugation, the slides were mounted in a portable multi-array analysis module (MAAM) device (Rivas et al., 3.6. 16S rRNA gene phylogenetic analysis 2008; Parro, 2010) composed of nine incubation chambers that isolated and sealed each antibody microarray and allowed Partial-length bacterial and archaeal 16S sequences were for the processing of all samples simultaneously. trimmed, assembled, and screened for vector contamination 590 PARRO ET AL.

with the CodonCode Aligner (CodonCode Corporation, being associated with the solid material and the latter with Centerville, MA) software package. Assemblies longer than IW. The concentration of nitrite was lower than the limit of 795 bp were used as input files for the mothur version 1.21.1 detection of the technique used to analyze all the sediment (Schloss et al., 2009). Sequences were aligned to the SILVA samples. Relatively high concentrations of bromide were (Pruesse et al., 2007) reference databases provided by mothur detected but only in the IW in the LN4 sediments, where in order to scan them for chimeras with the use of the chimera visible decomposing macrophyte algal remains were ob- slayer tool. The alignment was then used to calculate distance served. The ratio of Cl-/Br- ions was low (23.95), though the matrices, and the sequences were clustered into operative accumulation of Br- was found to be much higher than that taxonomic units (OTUs) by using the average neighbor al- of many freshwater lakes (Davis et al., 1998). gorithm. Rarefaction curves, richness estimators (Ace, The presence of relatively high concentrations of acetate Chao1), and community diversity indexes (Simpson Index) and formate, indicative of microbial fermentation processes, implemented in mothur were also obtained. For each sample, supported the growth of sulfate-reducing bacteria and/or one representative of each 0.03 distance OTU was selected for acetoclastic methanogenic archaea. phylogenetic tree construction. Sequences were aligned with the SILVA reference alignment using SINA (Pruesse et al., 4.2. Biomarker profiling with the LDChip 2012) and imported into the ARB phylogenetic package immunosensor in the field (Ludwig et al., 2004). ARB parsimony algorithm was used to Samples LNS, LES, and LN4 were analyzed in the add the sequences to the SSUref_NR98 reference tree, ap- field with the LDChip antibody microarray immunosensor plying a filter to exclude the most variable positions. (Materials and Methods). Sediment samples (ca. 0.5 g) were The 16S rRNA gene sequences obtained by 454 pyro- homogenized by a handheld ultrasonicator, filtered, and sequencing were analyzed as follows: Raw GS FLX +454 incubated with the LDChip. Positive immunoreactions were reads were analyzed with mothur version 1.31.2 as rec- revealed with a fluorescent antibody mixture and scanned ommended by Schloss et al. (2011). Sequences were de- for fluorescence, and the digital data was plotted (Fig. 2). multiplexed, denoized using mothur’s implementation of the Although one might expect that the immunograms from the PyroNoise algorithm (Quince et al., 2009), aligned to a sediments collected along the shoreline of both lakes (LNS combination of silva.archaea and silva.bacteria databases, and LES) would be more similar, they showed meaningful screened for chimeras using UCHIME (Edgar et al., 2011), differences. The immunogram of LNS had a lower number and clustered by using the average neighbor algorithm. Since of positive immunoreactions, while that of LES exhibited a various diversity metrics are sensitive to the procedure used richer inmunoprofile (Fig. 2 and Table 2). For example, for sampling, 4979 sequences for each sampling site were spots/bars 11 and 24, which correspond to antibodies pro- randomly selected for further analysis. Rarefaction curves, duced to Desulfotalea psychrophila and Desulfosporosinus richness estimators, community diversity indexes, and com- meridiei (two sulfate-reducing bacteria)orspots/barspro- munity structure were obtained as previously described. duced to Shewanella gelidimarina and Geobacter spp. (metal reducers), were absent in LNS. Notwithstanding, positive 4. Results reactions with antibodies produced to extracellular polymeric substances and whole cell extracts from Gammapro- 4.1. Lake geochemistry teobacteria were obtained along the shoreline of both lakes. Differences in the concentrations of low-molecular- Interestingly, the immunopattern obtained with the LN4 weight organic acids and anions involved in microbial me- sample was more similar to that of LES than to LNS. An- tabolism were observed between both lakes and, in the case tibodies produced to a psychrophilic sulfate-reducing bac- of Laguna Negra, with depth (Table 1). On average, the terium (D. psychrophila), other sulfate-reducing bacteria anion concentration was higher in the IW, except for nitrate, (Desulfovibrio vulgaris, Geobacter sulfurreducens,orD. which was preferably attached to the coarse material. meridiei), and methanogenic archaea (Methanosarcina ma- Acetate, formate, chloride, phosphate, and sulfate were es- zeii, Methanobacterium formicicum) showed clear positive pecially abundant in LN4 IW and coarse material. These reactions. Similarly, antibodies produced to heterotrophic values were lower in the shore sediments, whose visual bacteria, such as Burkholderia fungorum, Acidiphilium sp., or inspection showed less decomposing organic matter. Pro- Acidocella aminolytica also showed positives in both LN4

Downloaded by Consejo Superior De Investigaciones Cientificas CSIC from www.liebertpub.com at 07/22/18. For personal use only. pionate and oxalate were only detected in LN4, the former and LES samples. Somewhat unexpectedly, the LN4 sample

Table 1. Geochemical Analysis of the Laguna Negra and Lo Encan˜ ado Sediments Showing Low-Molecular-Weight Organic Acids and the Main Inorganic Anions (lg/mL) - - - - = = Sample Propionate Acetate Formate Tartrate Oxalate F Cl Br NO3 PO4 SO4 LES_w 0 3.23 0.45 2.82 0 0 31.77 0 0.05 0 2.53 LES_ex 0 0.02 0.02 0 0 0 7 0 1.79 0.42 1.34 LNS_w 0 0.26 0.23 0 0 0 81.38 0 0.38 0 117.05 LNS_ex 0 0.31 0 0 0 0 9.61 0 1.7 0 7.62 LN4_w 0 655.45 174.62 4.34 1.2 0.24 504.23 21.05 0 62.52 263.45 LN4_ex 7.73 137.43 37.89 3.36 0.36 0 91.06 0 1.75 8.86 47.89

LES, Lo Encanãdo shore sediments; LNS, Laguna Negra shore sediments; LN4, Laguna Negra sediments from 4 m below the surface water; _w, interstitial water; _ex, extract from coarse material from the centrifuged sediments, with ultrapure water. Table 2. List of the Antibodies Showing Positive Immunoreactions after On-Site Analysis of the Sediment Samples from Laguna Negra (LNS and LN4) and Lo Encan˜ ado (LES) Phylogenetic group Ab to Ab name* LNS LN4 LES References Alfaproteobacteria Acidocella aminolytica IVG1C1 Rivas et al., 2011 Acidiphillium spp. IVG2C1 Parro et al., 2011a Betaproteobacteria, Polaromonas IVF15C1 This paper Burkholderiales Burkholderia fungorum IVI4C2 Rivas et al., 2008 B. fungorum IVI4S100 Rivas et al., 2008 B. fungorum IVI4S2 Rivas et al., 2008 Gammaproteobacteria Acidithiobacillus ferrooxidans A183 Parro et al., 2005 A. ferrooxidans IVE3C1 Rivas et al., 2008 A. ferrooxidans IVE3S100 Rivas et al., 2008 A. thiooxidans A184 Rivas et al., 2008 A. thiooxidans IVE4C1 Rivas et al., 2008 A. thiooxidans IVE4C2 Rivas et al., 2008 A. thiooxidans IVE4S100 Rivas et al., 2008 A. albertensis IVE5C1 Rivas et al., 2008 A. albertensis IVE5C2 Rivas et al., 2008 A. albertensis IVE5S100 Rivas et al., 2008 A. caldus IVE6S100 Rivas et al., 2008 Halothiobacillus neapolitanus IVE7C1 Rivas et al., 2011 Methylomicrobium capsulatum IVI15C1 Rivas et al., 2008 Pseudomonas putida IVI1C2 Rivas et al., 2008 Shewanella gelidimarina IVF2S2 Rivas et al., 2008 Deltaproteobacteria Desulfotalea psychrophila IVF18C1 Rivas et al., 2011 Desulfovibrio vulgaris IVI10C1 Rivas et al., 2008 Geobacter sulfurreducens IVI11C1 Rivas et al., 2008 G. metallireducens IVI12C1 Rivas et al., 2008 Actinobacteria Acidimicrobium ferrooxidans IVE8C1 Parro et al., 2011a A. ferrooxidans IVE8S2 Parro et al., 2011a Cryobacterium psychrophilum IVF6S1 Rivas et al., 2008 Firmicutes Desulfosporosinus meridiei IVI19C1 Parro et al., 2011a Bacillus spp. (environ. isol.) IVI2S2 Rivas et al., 2008 Bacillus subtilis 3610 IVI8C1 Rivas et al., 2008 Planococcus or2 IVF31C1 This paper Planococcus IVF34S2 This paper Nitrospiraceae Leptospirillum ferrooxidans A139 Parro et al., 2005 L. ferroxidans A186 Rivas et al., 2008 L. pherrifilum (LPH2) IVE1S100 Rivas et al., 2008 L. pherrifilum spp. IVE2S1 Rivas et al., 2008 Bacteroidetes Psychroserpens burtonensis IVF4S2 Rivas et al., 2008 Salinibacter ruber PR1 IVI21C1 Rivas et al., 2011 Verrumicrobia Verrucomicrobium spinosum IVI14C1 Rivas et al., 2008 Euryarchaeota Haloferax mediterranei IVJ1C1 Rivas et al., 2008 Methanobacterium formicicum IVJ4C1 Rivas et al., 2008 Methanosarcina mazeii IVJ5C1 Rivas et al., 2008 Halorubrum spp. IVJ8C1 Parro et al., 2011a Proteins Glutathione-S-transferase A-GST Sigma-Aldrich (G7781) Cellular extracts (C1) Atacama Extract VID1C1 This paper and extracellular Solar saltern EPS A-EPS_SP Parro et al., 2011a substances (S2) Biomass from concrete VIID3BF This paper from environmental Rıó Tinto (3.2 water dam) IA1S1 Rivas et al., 2008 samples Rıó Tinto IA2C1 Rivas et al., 2008 Rıó Tinto IA2S1 Rivas et al., 2008 Rıó Tinto (Arroyo 3.1) IA3C1 Rivas et al., 2008 Rıó Tinto (Arroyo 3.1) IA3S1 Rivas et al., 2008 Rıó Tinto (3.2 water dam) IA3C1 Rivas et al., 2008 Rıó Tinto (3.0 Stream) A140 Rivas et al., 2008 Rıó Tinto (‘‘Nacimiento’’) A138 Rivas et al., 2008 Downloaded by Consejo Superior De Investigaciones Cientificas CSIC from www.liebertpub.com at 07/22/18. For personal use only. Rıó Tinto (Playa 3.1) IC1C1 Rivas et al., 2008 Rıó Tinto (Playa 3.2) IC1S1 Rivas et al., 2008 Rıó Tinto (Arroyo 3.1) IC2S2 Rivas et al., 2008 Rıó Tinto (3.1 water dam) IC2C3 Rivas et al., 2008 Rıó Tinto (‘‘Nacimiento’’) IC3S2 Rivas et al., 2008 Rıó Tinto (3.2 water dam) IC4C1 Rivas et al., 2008 Rıó Tinto (Playa 3.1) IC6C1 Rivas et al., 2008 Rıó Tinto (3.2 water dam) IC7C1 Rivas et al., 2008 Rıó Tinto (3.1 stream bank) IC8C1 Rivas et al., 2008 Rıó Tinto (3.1 stream bank) IC8S1 Rivas et al., 2008 Rıó Tinto (3.1 mine ruins) IC9C1 Rivas et al., 2008 Penã de Hierro (93m deep) ID18S2 Rivas et al., 2008 Penã de Hierro (154m deep) ID4S2 Rivas et al., 2008 Antarctic (DI) permafrost IIIC3C1 This paper

Shaded boxes indicate positive detection, that is, the presence of the corresponding strain/compound or a highly similar one. *See Appendix Table 1 for Ab details.

591 592 PARRO ET AL.

FIG. 2. On-site microbial biomarker detection with the LDChip, an antibody microarray-based life-detector chip. (A) LDChip fluorescent image corresponding to the on-site analysis of the Laguna Negra 4 m deep sediment (LN4) sample (top figure) and the negative control where only buffer was used (lower figure). (B) The image in (A) and those corresponding to the other samples (not shown) were quantified and the relative fluorescence units plotted: Laguna Negra shore sample (LNS), Laguna Negra 4 m deep (LN4), and Lo Encanãdo shore sample (LES). The antibodies showing positive im- munodetection were clustered (a–f) and numbered as follows: a, antibodies to acidic environmental extracts enriched in Gammaproteobacteria (e.g., 1, IC4C1, a cellular extract from a biofilm); b, bacterial cell extracts from acidic environment, mostly Gammaproteobacteria and Nitrospira (2, Leptospirillum ferrooxidans; 3, L. pherrifilum; 4, Acidithiobacillus ferrooxidans; 5, A. thiooxidans; 6, A. albertensis; 7, A. caldus; 8, Halothiobacillus neapolitanus; and 9, Acidimicrobium ferrooxidans); c, psychrophiles (10, Polaromonas spp.; 11, Desulfotalea psychrophila; 12, Shewanella gelidimarina; 13, Planococcus; 14, Planococcus spp.; 15, Psychroserpens meridei; 16, Cryobacterium psychrophilum); 17, Acidocella aminolytica; 18, Acidiphillium spp.; 19, Desulfovibrio vulgaris; 20, Geobacter sulfurreducens; 21, G. metallireducens; 22, Verrucomicrobium spinosum; 23, Methylomicrobium capsulatum; 24, Desulfoporosinus meridiei; 25, Pseudomonas putida; 26, Salinibacter ruber PR1; 27, Bacillus spp.; 28, Burkholderia fungorum; 29, Bacillus subtilis 3610; d, Archaea (30, Haloferax mediterranei; 31, Methanobacterium formicicum; 32, Methanosarcina mazeii; 33, Halorubrum spp.); e, peptides Downloaded by Consejo Superior De Investigaciones Cientificas CSIC from www.liebertpub.com at 07/22/18. For personal use only. and proteins; f, nitroaromatic compounds and nucleotide derivatives; 34, extracts from a biofilm on a concrete wall pavement; 35, whole extracts from 3–4 m deep cores of permafrost (Deception Island, Antarctica); 36, extracts from a 2.5 m depth sulfate- and halite-rich core sample from the Atacama Desert; 37, EPS from a solar saltern.

gave positive reactions to halophilic archaea (Haloferax subjected to massive sequencing by 454-pyrosequencing. mediterranii and Halorubrum sp.) and other halophilic bac- More than 4900 bacterial sequences were retrieved from each teria (Salinibacter ruber). The number of positive signals sample, and they showed nearly saturated rarefaction curves from aromatic compounds and nucleotide derivatives was (Fig. 3), while very few archaeal sequences were obtained. higher in sediments from LES than those in LNS or LN4. Regarding the bacterial sequences, the LNS sediments from Laguna Negra showed a lower number of OTUs (827 from LNS and 884 from LN4) than that of LES (1549) for a similar 4.3. Prokaryotic diversity by 16S rRNA gene sequencing number of sequences. The Simpson index for an OTU defini- Total environmental DNA was extracted from all the tion of 0.03 was 18.58, 41.62, and 145.58 for LNS, LN4, and samples, and the universal 16S rRNA gene was amplified and LE, respectively, confirming that the bacterial diversity was BIOMARKERS IN EVOLVING GLACIAL LAKES 593

were analyzed for accurate phylogenetic afﬁliations (Fig. 5). Again, no amplicons were obtained with archaeal primers from LNS. Of the 72 clones retrieved from LN4 and 78 from LES, 70 clones (in each case) corresponded to the Metha- nosarcinales order (mostly Methanosarcina genus) in LN4 sample and Methanosaeta = Methanothrix in LES sample. The phylogeny of bacterial clones (46 for LNS, 46 from LN4, and 18 from LES) was generally in agreement with massive sequencing, although with a different proportion between the groups. For example, Proteobacteria were ma- jor components in the libraries (Fig. 5) from the shoreline of both lakes (LNS and LES), and in particular, Betaproteo- bacteria representatives from the Gallionellaceae, Hydro- genophilaceae, and Comamonadaceae families dominated in the LNS sample. Further, we obtained more clones in this sample that belong to Deltaproteobacteria (one uncultured Desulfuromonadales, one Desulfobacteraceae bacterium, two Geobacter strains) than in the other two samples, and a Gammaproteobacteria (Thiofaba tediphila) of the Halothiobacillaceae family was also detected. Different Deltaproteobacteria, from the order Myxococcales (Ha- liangium spp), as well as bacteria from the order Gemmati- monadetes and the phylum Acidobacteria were found in the LES sample. In the LN4 sample, a different Deltaproteo- bacteria was identiﬁed, corresponding to the Syntrophaceae family, while the Firmicutes group (mostly Clostridia) ac- counted for more than 20% of the sequences in LN4, which is in close agreement with the massive sequencing data. The only cloned sequence within the phylum Cyanobacteria was found in the LN4 sample. Archaeal sequences were obtained from LN4 and LES, mostly corresponding to the Methano- sarcina Methanosaeta FIG. 3. Rarefaction curves for bacterial diversity by 454- genus in the former and the = pyrosequencing in the three analyzed samples (a) and for the Methanothrix genus in the latter (Fig. 5), which is in agree- diversity of Archaea by gene cloning and sequencing in LES ment with the pyrosequencing data (see above). No archaeal (b) and LN4 (c). See main text for explanation. clones and sequences were obtained after several trials from the LNS sample.

much higher in the sediments from Lo Encanãdo than in the 5. Discussion sediments from Laguna Negra. Members of the main groups of 5.1. Biomarker immunoprofiling Proteobacteria (Alpha-, Beta-, Gamma- and Deltaproteo- for monitoring deglaciation bacteria) were present in all samples (Fig. 4), with a significant increase in the number of Deltaproteobacteria in the We previously demonstrated that the LDChip is a useful LN4 and LES samples. Two of the most abundant phylo- tool for environmental monitoring (Rivas et al., 2008; Parro genetic groups, Actinobacteria and Acidobacteria, were et al.,2011a;Blancoet al., 2012, 2014; Fernańdez-Remolar preferentially abundant in the shoreline sediments of both et al., 2014; Puente-Sańchez et al., 2014). Biomarker and lakes (Fig. 4), while the groups of Firmicutes, Planctomy- microbial profiles in Laguna Negra and Lo Encanãdo were cetes, Chloroflexi, Bacteroidetes, and Chlorobi dominated obtained by sandwich antibody microarray immunoassays and showed similar patterns to those of the LN4 and LES with an LDChip (Fig. 2 and Table 2) with 193 immobilized Downloaded by Consejo Superior De Investigaciones Cientificas CSIC from www.liebertpub.com at 07/22/18. For personal use only. samples, with very few representatives in LNS. Members of antibodies and a mixture of 141 fluorescent ones as tracers of the Deinococcus-Thermus group were only found in LNS the immunoreactions (Materials and Methods). We did not and, by contrast, Cyanobacteria were absent in this sample include the 52 antibodies developed to recognize cells and though they appeared in LN4 and LES. No 16S rRNA gene polymeric materials for Rıó Tinto samples to avoid any po- amplicons related to the domain Archaea were obtained in tential bias on the LDChip results. Positive immunological LNS with the universal primers used. However, most of the interactions were identified with several antibodies, such as 13 archaeal sequences obtained by pyrosequencing from the those from sulfate-reducer biomarkers detected in the LN4 other two samples corresponded to members of the metha- sample, which is consistent with a higher amount of sulfate in nogenic Archaea (10 sequences from Methanosaeta = this sample. Exopolymeric material from Gammaproteo- Methanothrix genusinLN4and1fromMethanosarcina in bacteria, such as exopolymeric substances (EPS) and whole LES), while two sequences from Halobacteria were re- cell extracts, was detected in all samples, but preferentially in trieved from LN4. the LN4 and LES samples where the largest number of DNA Additionally, a full-length 16S rRNA gene was amplified sequences were retrieved for this phylum. Although individ- by PCR, cloned, and sequenced, and the retrieved sequences ual compound and derivative confirmation would be needed, 594 PARRO ET AL.

FIG. 4. Prokaryotic diversity (A, archaea and B, bacteria) in the sediments of Lo Encan˜ado (LES) and Laguna Negra (LNS) shores and at 4 m below the water surface in Laguna Negra (LN4). Percentage of the retrieved 16 S rRNA gene sequences ascribed to the corresponding phylogenetic taxon. The numbers above the bars indicate the number of sequences used in the analysis. Sequences from bacteria were obtained by massive sequencing by 454 pyrosequencing and those from archaea by gene cloning and Sanger-type sequencing (asterisks). Bars represent the percentage of the retrieved 16 S rRNA gene sequences ascribed to each phylogenetic group (legend to the right) from the total number of sequences analyzed.

positive signals from nitro-aromatic compounds and nucleo- geochemical and molecular ecology results, we propose a tide derivatives were higher in the LES sediments. geomicrobiological model for the shallow sediments, which The immunological profiles (Fig. 2) indicate that the LN4 applies to the ecosystem at the shoreline and 4 m under and LES samples were closer to each other than those from water at Laguna Negra and Lo Encanãdo (Fig. 7). Primary the shorelines (LNS and LES), suggesting that local envi- producers, such as microalgae, macrophytes, and the cya- ronmental conditions (nutrients, geochemistry) control the nobacteria that grow along the shorelines, produce organic microbial diversity regardless of the lake or the depth. The matter that is transformed by fermenters in shallow anaer- LDChip detected positive immunoreactions with antibodies obic niches. Key products from fermentation are the small- to prokaryotes capable of anaerobic metabolisms (e.g., molecular-weight organic acids such as acetate, formate, sulfate-reducing bacteria) and to methanogens in the LN4 propionate, and oxalate, all of them detected by ion chro- and LES samples, which is consistent with the DNA se- matography (Table 1). These compounds are excellent en- quencing data. ergy sources for anaerobic microbial metabolisms such as sulfate reduction and methanogenesis, which is consistent with our detection of sulfate-reducing bacteria and metha- Downloaded by Consejo Superior De Investigaciones Cientificas CSIC from www.liebertpub.com at 07/22/18. For personal use only. 5.2. A geomicrobiological model of the Laguna nogenic archaea with the immunosensor LDChip (Fig. 2) Negra and Lo Encanãdo sediments and 16S rRNA gene sequencing (Fig. 4). The presence of Geochemical and immunological studies combined with immunoreactive material from halophilic archaea (with 16S rRNA gene cloning and massive sequencing identified anti-Haloferax and anti-Halorubrum antibodies) and bacte- rich prokaryotic communities and allowed us to infer the ria (Salinibacter, from Bacteroidetes group) in LN4 may diverse metabolisms that can operate in two Andean oli- be a consequence of the relatively high accumulation gotrophic lakes. We identified bacteria capable of sulfate (504 ppm) of chloride at the sampling site (Table 1), which reduction, sulfide oxidation, nitrogen oxidation, anaerobic allows proliferation of these microbes. DNA sequences ammonia oxidation, nitrate reduction, and methane oxidation from Halobacteria and Bacteroidetes were actually retrieved (methanotrophic), photosynthesis, anaerobic phototrophy, and from this sample. At this stage, further sampling would be methanogenic archaea (Fig. 6). The fermentative and anaerobic necessary to conclude whether this chloride accumulation is (mainly reduction) metabolisms seemed to dominate in the the result of a local phenomenon (e.g., seepage from the LN4 and LES samples, while heterotrophic and oxidative lake floor) or if it is widespread and occurs across the entire processes were significant in the LNS sample. Based on the bottom of the lake basin. Downloaded by Consejo Superior De Investigaciones Cientificas CSIC from www.liebertpub.com at 07/22/18. For personal use only.

FIG. 5. Panel a (Continued).

595 Downloaded by Consejo Superior De Investigaciones Cientificas CSIC from www.liebertpub.com at 07/22/18. For personal use only.

FIG. 5. Panel b (Continued).

596 BIOMARKERS IN EVOLVING GLACIAL LAKES 597

FIG. 5. Maximum parsimony tree based on 16S rRNA sequences showing the relationships of the OTUs retrieved from samples LNS (a), LN4 (b), and LES (c) with their closest phylogenetic neighbors. The number of clones retrieved for each OTU is indicated in parentheses. Bar: substitutions per nucleotide.

The bacterial diversity was roughly similar in sediments 0.055, 0.004, and 11.11 mM in the LES, LNS, and LN4 Downloaded by Consejo Superior De Investigaciones Cientificas CSIC from www.liebertpub.com at 07/22/18. For personal use only. from the shorelines of both lakes (LNS and LES), with samples, respectively (Table 1). These values rendered ac- representatives of Alpha-, Beta-, and Deltaproteobacteria etate/sulfate ratios of 2.08, 0.004, and 4.05 for the LES, (Fig. 4). However, in the LN4 sample the bacterial pattern LNS, and LN4 samples, respectively. It has been shown that changed, and the larger number of sequences corresponded sulfate-reducing bacteria compete with methanogenic ar- to the Gram-positive Firmicutes group, particularly from the chaea for hydrogen and acetate (Lovley and Klug, 1983) at Clostridiales order. No archaebacterial representatives were in situ sulfate concentrations of 0.06–0.105 mM. In both detected in the LNS sample, while in the LES sample most Laguna Negra sediments, the sulfate concentration is higher of the retrieved archaeal DNA sequences corresponded (see above and Table 1). The absence of archaea in the LNS to the methanogenic Methanosarcina genus and to the sample can be explained by the very low acetate/sulfate Methanosaeta = Methanothrix genus that were observed in ratio, which indicates that they were completely out- the LN4 sample. These differences might be due to the competed by sulfate-reducing bacteria. However, the ace- speciﬁc geochemical features in each site. The sulfate con- tate/sulfate ratio is much higher in the LES and LN4 centration was 0.026, 1.22, and 2.74 mM in the LES, LNS, samples, which might compensate the sulfate concentration and LN4 samples, respectively, while that of acetate was effect. The genus Methanosarcina is a highly versatile, 598 PARRO ET AL.

metabolic generalist (Thauer et al., 1989) capable of using several substrates, which include methanol, H2, methyl- amines, and acetate. The presence of Methanosarcina in the LES samples indicates that it most likely proliferated preferentially under low organic acid content (e.g., acetate) and low sulfate concentration. By contrast, the Methanosaeta = Methanothrix strains, which are the most nutritionally re- strictive specialist with a high affinity for acetate (Patel and Sprott, 1990), were found to grow at the high concentration of acetate detected in the LN4 sample. At high sulfate concentration, sulfate reducers compete for acetate and in- hibit acetoclastic methanogens (Ward and Winfrey, 1985). Therefore, our finding of methanogenic species such as those from the Methanosaeta = Methanothrix genus, which have a high affinity for acetate, were found to proliferate and coexist with sulfate reducers in the LN4 sample. The sequence data and the number of bacterial OTUs FIG. 6. Potential metabolisms in Laguna Negra and Lo obtained in all the samples indicate that the sediments of Encanãdo lakes. Based on the prokaryotic diversity from both lakes are rich and diverse ecosystems (Fig. 3), in spite massive 16 S rRNA gene sequence analysis, we inferred the of the oligotrophic characteristics of the water. In fact, the potential metabolisms associated with the detected micro- geochemical analysis (Table 1) showed the presence of organisms. The chart represents the percentage of the re- relatively high concentrations of organic acids from fer- trieved sequences that can be ascribed to microbes capable mentation or acetogenic processes, which can be used as of the corresponding metabolism (ordinates). Sample origin carbon and energy sources by microbes, while oxidized is LN4 (Laguna Negra 4 m deep), LES (Lo Encanãdo shore), compounds such as nitrate and sulfate can be used as elec- and LNS (Laguna Negra shore). tron acceptors for respiration. In contrast to Bacteria, the archaeal diversity was found to be very low, with rarefaction curves next to saturation with less than 100 clones and a Downloaded by Consejo Superior De Investigaciones Cientificas CSIC from www.liebertpub.com at 07/22/18. For personal use only.

FIG. 7. Geomicrobiological processes operating in the shallow sediments of the oligotrophic lakes Laguna Negra (LN) and Lo Encanãdo (LE). Scheme showing the geochemistry and microbial metabolisms as deduced from the geochemical and biodiversity analysis (immunological and DNA sequencing) from the sediment samples collected at the shore of LN (LNS), at 4 m below the water surface (LN4), and at the shore of Lo Encanãdo (LES). Basically, in LN4 the high concentration of small organic acids (such as acetate and formate), obtained from the fermentation of organic matter (OM), and sulfate are the main drivers for sulfate reduction by sulfate-reducing bacteria (SRB). In this scenario, small organic acids are mainly consumed by SRB, and only some acetoclastic methanogens, such as Methanosaeta = Methanothrix strains that have high affinity for acetate, may compete for these resources. Contrarily, in LES sediments, where sulfate concentration is much less, other more versatile methanogens, such as Methanosarcina strains, can proliferate. While anaerobic metabolisms dominated in the LN4 and LES samples, the heterotrophic and oxidative ones, including methane oxidation (methanotrophy), dominate in the shore of LNS (see text for further discussions). Yellow arrows indicate the difference in the amount of sulfate supplied by the water stream. SOB, sulfur oxiding bacteria. BIOMARKERS IN EVOLVING GLACIAL LAKES 599

clear predominance of two methanosarcinales genera, Me- 6. Conclusions thanosarcina Methanothrix and (Fig. 3). Understanding the evolution of high-altitude oligotrophic lakes requires knowledge of their geochemical and geomi- 5.3. Deglaciation evolution determines the microbial crobiological characteristics. For Laguna Negra and Lo metabolisms in the sediments Encanãdo, we have presented critical geochemical features and the prokaryotic community compositions of each lake, Lakes are considered sentinels of climate change and inferred their operating microbial metabolisms in shal- et al. (Williamson , 2009) that record environmental fluctu- low sediments. Although the waters are oligotrophic, the ations in their sediments. As their habitats change, it is sediments from the shore of Lo Encanãdo and from 4 m expected that their microbial ecosystems will be modified. below the water surface in Laguna Negra are rich in nutri- Therefore, understanding the link between microbial com- ents and contain a great bacterial diversity, with sulfate re- munities and their metabolic traits, climate history, and duction and methanogenesis as key anaerobic metabolisms. physicochemical evolution may reveal microbial markers Although attempting to associate alteration of the lakes to associated with different phases of deglaciation. climate change would be a complex task that would need This work provides a baseline to monitor the geomicro- time-course monitoring, our data constitute a reference point biological evolution of these two oligotrophic Andean lakes for future studies. Additionally, we demonstrated that on site in response to climate change. More data will be necessary immunosensing techniques such as the LDChip for bio- along with time-course monitoring of the microbial com- markers and microbial detection are useful tools for such munities associated with the types of sediments studied; studies. Ongoing geochemical and microbiological work in however, regardless of whether they evolve toward eutroph- the water column at different depths and deeper sediment ication or oligotrophication with time, the geochemistry and samples will help clarify the effect of deglaciation phases on microbiology of the sediment will play critical roles in the the prokaryotic communities. evolution of these communities. Important variables such as the nutrient inputs from the water catchment, precipitation Acknowledgments regime, temperature, lake stratification, and the proliferation of primary producers will affect the evolutionary pathway of We thank Aguas Andinas Company (Santiago, Chile) for the microbial communities as a function of time. their help and issuing of the permits to access to the field sites, Figure 7 illustrates the metabolic processes that we found to Campoalto Operaciones for its support in the field, and Miriam operate in the Laguna Negra and Lo Encanãdo sediments along Garcıá-Villadangos for technical assistance. This work was with a schematic of our hypothesis for the behavior of lake funded by the Spanish ‘‘SecretarıádeEstadodeInvestigacioń sediments in an advanced deglaciation scenario. In this case, Desarrollo e Innovacioń’’ from the Economy and Competi- recently deglaciated soils might be colonized by a diverse tiveness Ministry (MINECO) grants No. AYA2011-24803 and community of cyanobacteria and other microbes during the first ESP2014-58494-R ‘‘Detection of Biomolecules in Planetary years following the glacial retreat (Schmidt et al.,2008).In Exploration,’’ and NASA’s ASTEP grant No. 10-ASTEP10- these soils, carbon and nitrogen fixation would dominate and 0011 ‘‘Planetary Lake Lander.’’ F.P. is a JAE-pre fellow from increase the production of organic matter in the watershed. the Consejo Superior de Investigaciones Cientı´ficas (CSIC). Such organic material would feed the glacial lakes through DNA sequences have been deposited in Genebank under runoff, and depending on the pluviometric regime and glacial accession number PRJNA256272. melting, produce an increase in the internal primary production that would lead to greater sedimentation of organic matter. Author Disclosure Statement Under aerobic conditions, microbial heterotrophs would dominate organic matter oxidation, as is the case of LNS No competing financial interests exist. (Fig. 7). In deeper and anaerobic sediments where sulfate from References the catchment accumulates, fermenters would oxidize the complex organic material and produce small-size organic acids Beniston, M., Diaz, H.F., and Bradley, R.S. (1997) Climatic to fuel sulfate reduction and specialized acetoclastic metha- change at high elevation sites: an overview. Clim Change nogenesis as in LN4 (Fig. 7). However, in lakes at a later de- 36:233–251. glatiation stage, with higher organic content and sedimentation, Blanco, Y., Prieto-Ballesteros, O., Go´mez, M.J., Moreno-Paz, Downloaded by Consejo Superior De Investigaciones Cientificas CSIC from www.liebertpub.com at 07/22/18. For personal use only. the anaerobic metabolisms already dominate even in the shal- M., Garcıá-Villadangos, M., Rodrı´guez-Manfredi, J.A., Cruz- low sediments of the shore as in LES (Fig. 7 right): fermenta- Gil, P., Sańchez-Romań, M., Rivas, L.A., and Parro, V. tion, anaerobic ammonia oxidation, sulfate reduction, and the (2012) Prokaryotic communities and operating metabolisms in the surface and the permafrost of Deception Island (Ant- versatile multi-substrate-powered methanogenesis. In fact, the arctica). Environ Microbiol 14:2495–2510. metabolic profiles from LES and the deeper LN4 sediments are Blanco, Y., Rivas, L.A., Garcıá-Moyano, A., Aguirre, A., Cruz- more similar between them than with the LNS (in agreement Gil, P., Palacıń, A., van Heerden, E., and Parro, V. (2014) De- with the LDChip results), with the difference that, among the ciphering the prokaryotic community and metabolisms in South methanogens, only the highly specialized acetoclastic Metha- African deep-mine biofilms through antibody microarrays and nosaeta spp. are detected in LN4. Whether similar processes graph theory. PloS One 9, doi:10.1371/journal.pone.0114180. ocurred on ancient Mars is unknown; however, deglaciation Bradley, R.S., Vuille, M., Diaz, H.F., and Vergara, W. (2006) phenomena might have generated new sedimentary environ- Climate change. Threats to water supplies in the tropical ments where a rich microbial population flourished. Remnants Andes. Science 312:1755–1756. of such ancient environments are candidate targets for the Callieri, C., Modenutti, B., Queimalinõs, C., Bertoni, R., and search for remains of life on future planetary missions. Balseiro, E. (2007) Production and biomass of picophyto- 600 PARRO ET AL.

plankton and larger autotrophs in Andean ultraoligotrophic Ludwig, W., Strunk, O., Westram, R., Richter, L., Meier, H., lakes: differences in light harvesting efficiency in deep layers. Yadhukumar, Buchner, A., Lai, T., Steppi, S., Jacob, G., Aquat Ecol 41:511–523. Fo¨rster, W., Brettske, I., Gerber, S., Ginhart, A.W., Gross, O., Caravati, E., Callieri, C., Modenutti, B., Corno, G., Balseiro, E., Grumann, S., Hermann, S., Jost, R., Ko¨nig, A., Liss, T., Bertoni, R., and Michaud, L. (2010) Picocyanobacterial as- Lu¨ßbmann, R., May, M., Nonhoff, B., Reichel, B., Strehlow, semblages in ultraoligotrophic Andean lakes reveal high re- R., Stamatakis, A., Stuckmann, N., Vilbig, A., Lenke, M., gional microdiversity. J Plankton Res 32:357–366. Ludwig, T., Bode, A., and Schleifer, K.H. (2004) ARB: a Coudrain, A., Francou, B., and Kundzewicz, Z.W. (2005) software environment for sequence data. Nucleic Acids Res Glacier shrinkage in the Andes and consequences for water 32:1363–1371. resources. Hydrological Sciences Journal 50:925–932. Modenutti, B., Balseiro, E., Bastidas Navarro, M., Laspouma- Davis, S.N., Whittemore, D.O., and Fabryka-Martin, J. (1998) deres, C., Souza, M.S., and Cuassolo, F. (2012) Environ- Uses of chloride/bromide ratios in studies of potable water. mental changes affecting light climate in oligotrophic Ground Water 36:338–350. mountain lakes: the deep chlorophyll maxima as a sensitive Edgar, R.C., Haas, B.J., Clemente, J.C., Quince, C., and Knight, variable. Aquat Sci 75:361–371. R. (2011) UCHIME improves sensitivity and speed of chi- Painter, J. (2007) Deglaciation in the Andean Region, Human mera detection. Bioinformatics 27:2194–2200. Development Report Office occasional paper, United Nations Fernańdez-Calvo P., Na¨ke C., Rivas L.A., Garcıá-Villadangos Development Programme, New York. M., Go´mez-Elvira J., and Parro V. (2006) A multi-array Parro, V. (2010) Antibody microarray for environmental moni- competitive immunoassay for the detection of broad-range toring. In Handbook of Hydrocarbon and Lipid Microbiology, molecular size organic compounds relevant for astrobiology. edited by K.N. Timmis, Springer-Verlag, Berlin, pp 2699–2710. Planet Space Sci 54:1612–1621. Parro, V., Rodrı´guez-Manfredi, J.A., Briones, C., Compostizo, Fernańdez-Remolar D.C., Chong-Dıáz G., Ruı´z-Bermejo M., C., Herrero, P.L., Vez, E., Sebastiań, E., Moreno-Paz, M., Harir M., Schmitt-Kopplin P., Tziotis D., Go˜mez-Ortı´z D., Garcıá-Villadangos, M., Fernańdez-Calvo, P., Gonza´lez- Garcıá-Villadangos M., Martıń-Redondo M.P., Go˜mez F., Toril, E., Pe´rez-Mercader, J., Fernańdez-Remolar, D., and Rodrı´guez-Manfredi J.A., Moreno-Paz M., De Diego-Castilla Go´mez-Elvira, J. (2005) Instrument development to search G., Echeverrıá A., Urtuvia V.N., Blanco Y., Rivas L., Izawa for biomarkers on Mars: terrestrial acidophile iron-powered M.R.M., Banerjee N.R., Demergasso C., and Parro V. (2013) chemolithoautotrophic communities as model systems. Pla- Molecular preservation in halite- and perchlorate-rich hyper- net Space Sci 53:729–737. saline subsurface deposits in the Salar Grande basin (Atacama Parro, V., Fernańdez-Calvo, P., Rodrı´guez Manfredi, J.A., Desert, Chile): Implications for the search for molecular bio- Moreno-Paz, M., Rivas, L.A., Garcıá-Villadangos, M., Bo- markers on Mars. J Geophys Res Biogeosci 118:922–939. naccorsi, R., Gonza´lez-Pastor, J.E., Prieto-Ballesteros, O., Haeberli, W., Maisch, M., and Paul, F. (2002) Mountain gla- Schuerger, A.C., Davidson, M., Go´mez-Elvira, J., and Stoker, ciers in global climate-related observation networks. World C.R. (2008) SOLID2: an antibody array-based life-detector Meterological Organization Bulletin 51:18–25. instrument in a Mars Drilling Simulation Experiment IPCC. (2007) Climate Change 2007: The Physical Science (MARTE). Astrobiology 8:987–999. Basis, contribution of Working Group 1 to the Fourth As- Parro, V., De Diego-Castilla, G., Moreno-Paz, M., Blanco, Y., sessment Report of the Intergovernmental Panel on Climate Cruz-Gil, P., Rodrı´guez-Manfredi, J.A., Fernańdez-Remolar, D., Change, Cambridge University Press, Cambridge, UK. Go´mez, F., Go´mez, M.J., Rivas, L.A., Demergasso, C., Eche- Lara, A., Wolodarsky-Franke, A., Aravena, J.C., Villalba, R., So- verrıá, A., Urtuvia, V.N., Ruiz-Bermejo, M., Garcıá-Villadangos, lari, M.E., Pezoa, L., Rivera, A., and Le Quesne, C. (2005) Cli- M., Postigo, M., Sańchez-Romań, M., Chong-Dıáz, G., and mate fluctuationsderived fromtree-ringsand other proxy-records Go´mez-Elvira, J. (2011a) A microbial oasis in the hypersaline in the Chilean Andes: state of the art and future prospects. In Atacama subsurface discovered by a life detector chip: implica- Global Change and Mountain Regions: An Overview of Current tions for the search of life on Mars. Astrobiology 11:969–996. Knowledge, edited by U.M. Huber, H.K. Bugmann, and M.A. Parro, V., De Diego-Castilla, G., Rodrı´guez-Manfredi, J.A., Reasoner, Springer, Dordrecht, The Netherlands, pp 145–156. Rivas, L.A., Blanco-Lo´pez, Y., Sebastiań, E., Romeral, J., Le Quesne, C. and Acunã, C. (2003) Fluctuaciones histo´ricas Compostizo, C., Herrero, P.L., Garcıá-Marıń, A., Moreno- del ventisquero Cipreses y su relatioń con el registro de an- Paz, M., Garcıá-Villadangos, M., Cruz-Gil, P., Peinado, V., illos de Austrocedrus chilensis en Chile Central Glacier Martıń-Soler, J., Pe´rez-Mercader, J., and Go´mez-Elvira, J. (3433¢S7022¢W). In Symposium on Global Change: To- (2011b) SOLID3 a multiplex antibody microarray-based op- wards a Systemic View, held at the meeting of the IGBP tical sensor instrument for in situ life detection in planetary Downloaded by Consejo Superior De Investigaciones Cientificas CSIC from www.liebertpub.com at 07/22/18. For personal use only. Scientific Steering Committee, Punta Arenas, Chile. exploration. Astrobiology 11:15–28. Le Quesne, C., Acunã, C., Boninsegna, J.A., Rivera, A., and Parro, V., Fernańdez-Remolar, D., Rodrı´guez-Manfredi, J.A., Barichivich, J. (2009) Long-term glacier variations in the Cruz-Gil, P., Rivas, L.A., Ruiz-Bermejo, M., Moreno-Paz, Central Andes of Argentina and Chile, inferred from histor- M., Garcıá-Villadangos, M., Go´mez-Ortiz, D., Blanco-Lo´pez, ical records and tree-ring reconstructed precipitation. Pa- Y., Menor-Salvań, C., Prieto-Ballesteros, O., and Go´mez- laeogeogr Palaeoclimatol Palaeoecol 281:334–344. Elvira, J. (2011c) Classification of modern and old Rıó Tinto Leoń, C., Campos, V., Urrutia, R., and Mondaca, M.A. (2012) sedimentary deposits through the biomolecular record using a Metabolic and molecular characterization of bacterial com- life marker biochip: implications for detecting life on Mars. munity associated to Patagonian Chilean oligotrophic-lakes Astrobiology 11:29–44. of quaternary glacial origin. World J Microbiol Biotechnol Patel, G.B. and Sprott, G.D. (1990) Methanosaeta concilii gen. nov., 28:1511–1521. sp.nov.(‘Methanothrix concilii’) and Methanosaeta thermo- Lovley, D.R. and Klug, M.J. (1983) Sulfate reducers can out- acetophila nom. rev., comb. nov. Int J Syst Bacteriol 40:79–82. compete methanogens at freshwater sulfate concentrations. Pruesse, E., Quast, C., Knittel, K., Fuchs, B.M., Ludwig, W., Appl Environ Microbiol 45:187–192. Peplies, J., and Glo¨ckner, F.O. (2007) SILVA: a compre- BIOMARKERS IN EVOLVING GLACIAL LAKES 601

hensive online resource for quality checked and aligned ri- Schmidt S.K., Reed S.C., Nemergut D.R., Grandy A.S., Cle- bosomal RNA sequence data compatible with ARB. Nucleic veland C.C., Weintraub M.N., Hill A.W., Costello E.K., Acids Res 35:7188–7196. Meyer A.F., Neff J.C., and Martin A.M. (2008) The earliest Pruesse, E., Peplies, J., and Glo¨ckner, F.O. (2012) SINA: ac- stages of ecosystem succession in high-elevation (5000 me- curate high-throughput multiple sequence alignment of ribo- tres above sea level), recently deglaciated soils. Proc R Soc B somal RNA genes. Bioinformatics 28:1823–1829. Biol Sci 275:2793–2802. Puente-Sańchez, F., Moreno-Paz, M., Rivas, L.A., Cruz-Gil, P., Sim, K., Cox, M.J., Wopereis, H., Martin, R., Knol, J., Li, M.S., Garcıá-Villadangos, M., Go´mez, M.J., Postigo, M., Garrido, P., Cookson, W.O.C.M., Moffatt, M.F., and Kroll, J.S. (2012) Gonza´lez-Toril, E., Briones, C., Fernańdez-Remolar, D., Stoker, Improved detection of bifidobacteria with optimised 16S C., Amils, R., and Parro, V. (2014) Deep subsurface sulfate rRNA-gene based pyrosequencing. PloS One 7, doi:10.1371/ reduction and methanogenesis in the Iberian Pyrite Belt (IPB), journal.pone.0032543. revealed through geochemistry and molecular biomarkers. Thauer, R.K., Zinkhan-Moller, D., and Spormann, A.M. (1989) Geobiology 12:34–47. Biochemistry of acetate catabolism in anaerobic chemo- Quince, C., Lanzeń, A., Curtis, T.P., Davenport, R.J., Hall, N., trophic bacteria. Annu Rev Microbiol 43:43–67. Head, I.M., Read, L.F., and Sloan, W.T. (2009) Accurate Vuille, M. (2006) Climate change in the tropical Andes—ob- determination of microbial diversity from 454 pyrosequen- servations, models, and simulated future impacts on glaciers cing data. Nat Methods 6:639–641. and streamflow. Mt Res Dev 26:3. Rivas, L.A., Garcıá-Villadangos, M., Moreno-Paz, M., Cruz- Vuille, M., Francou, B., Wagnon, P., Juen, I., Kaser, G., Mark, Gil, P., Go´mez-Elvira, J., and Parro, V. (2008) A 200- B.G., and Bradley, R.S. (2008) Climate change and tropical antibody microarray biochip for environmental monitoring: Andean glaciers: past, present, and future. Earth-Science searching for universal microbial biomarkers through im- Reviews 89:76–96. munoprofiling. Anal Chem 80:7970–7979. Wadham, J.L., Tranter, M., Tulaczyk, S., and Sharp, M. (2008) Rivas, L.A, Aguirre, J., Blanco, Y., Gonza´lez-Toril, E., and Parro, Subglacial methanogenesis: a potential climatic amplifier? V. (2011) Graph-based deconvolution analysis of multiplex Global Biogeochem Cycles 22, doi:10.1029/2007GB002951. sandwich microarray immunoassays: applications for environ- Ward, D.M. and Winfrey, M.R. (1985) Interactions between mental monitoring. Environ Microbiol 13:1421–1432. methanogenic and sulfate-reducing bacteria in sediments. Rivera, A., Benham, T., Casassa, G., Bamber, J., and Dow- Advances in Aquatic Microbiology 3:141–179. deswell, J.A. (2007) Ice elevation and areal changes of gla- Williamson, C.E., Saros, J.E., and Schindler, D.W. (2009) ciers from the Northern Patagonian Icefield, Chile. Glob Sentinels of change. Science 323:887–888. Planet Change 59:126–137. Roig, F.A., Le-Quesne, C., Boninsegna, J.A., Briffa, K.R., Lara, Address correspondence to: A., Grudd, H., Jones, P.D., and Villagrań, C. (2000) Climate Victor Parro variability 50,000 years ago in mid-latitude Chile as re- Department of Molecular Evolution constructed from tree rings. Nature 410:567–570. Centro de Astrobiologıá (INTA-CSIC) Sala, O.E., Chapin, F.S., III, Armesto, J.J., Berlow, E., Bloomfield, Carretera de Ajalvir km 4 J., Dirzo, R., Huber-Sanwald, E., Huenneke, L.F., Jackson, R.B., Torrejoń de Ardoz, Madrid Kinzig, A., Leemans, R., Lodge, D.M., Mooney, H.A., Oes- Spain terheld, M., Poff, N.L., Sykes, M.T., Walker, B.H., Walker, M., and Wall, D.H. (2000) Global biodiversity scenarios for the year E-mail: [email protected] 2100. Science 287:1770–1774. Schloss, P.D., Westcott, S.L., Ryabin, T., Hall, J.R., Hartmann, Submitted 24 April 2015 M., Hollister, E.B., Lesniewski, R.A., Oakley, B.B., Parks, Accepted 5 October 2015 D.H., Robinson, C.J., Sahl, J.W., Stres, B., Thallinger, G.G., Van Horn, D.J., and Weber, C.F. (2009) Introducing mothur: open-source, platform-independent, community-supported Abbreviations Used software for describing and comparing microbial communi- EPS ¼ exopolymeric substances ties. Appl Environ Microbiol 75:7537–7541. IC ¼ ion chromatograph Schloss, P.D., Gevers, D., and Westcott, S.L. (2011) Reducing IW ¼ interstitial water the effects of PCR amplification and sequencing artifacts LDChip ¼ Life Detector Chip on 16S rRNA-based studies. PloS One 6, 10.1371/journal OTUs ¼ operational taxonomic units Downloaded by Consejo Superior De Investigaciones Cientificas CSIC from www.liebertpub.com at 07/22/18. For personal use only. .pone.0027310.

(Appendix Table 1 follows) Downloaded by Consejo Superior De Investigaciones Cientificas CSIC from www.liebertpub.com at 07/22/18. For personal use only.

Appendix Table 1. List of the Antibodies Printed on the LDChip for This Study No Ab name Source/Strain Sample/Culture conditions Immunogen/Fraction References 1 IA1C1 Rıó Tinto (3.2 water dam) Water Cellular fraction Rivas et al., 2008 2 IA1S1 Rıó Tinto (3.2 water dam) Water Supernatant Rivas et al., 2008 3 IA1S2 Rıó Tinto (3.2 water dam) Water Supernatant from EDTA wash Rivas et al., 2008 4 IA1C2 Rıó Tinto (3.2 water dam) Water Insoluble cell pellet from S100 Rivas et al., 2008 5 IA1S100 Rıó Tinto (3.2 water dam) Water Soluble cellular fraction S100 Rivas et al., 2008 6 IA1C3 Rıó Tinto (3.2 water dam) Water EDTA washed cells (sonicated) Rivas et al., 2008 7 IA2C1 Rıó Tinto (3.1 stream) Green filaments Cellular fraction Rivas et al., 2008 8 IA2S1 Rıó Tinto (3.1 stream) Green filaments Supernatant Rivas et al., 2008 9 IA3C1 Rıó Tinto (3.1 stream) Black filaments Cellular fraction Rivas et al., 2008 10 IA3S1 Rıó Tinto (3.1 stream) Black filaments Supernatant Rivas et al., 2008 11 IC1C1 Rıó Tinto (Playa 3.2) Sediments Cellular fraction Rivas et al., 2008 12 IC1S1 Rıó Tinto (Playa 3.2) Sediments Supernatant Rivas et al., 2008 13 IC1S2 Rıó Tinto (Playa 3.2) Sediments Supernatant from EDTA wash Rivas et al., 2008 14 IC2C1 Rıó Tinto (3.1 water dam) Dark red sediments Cellular fraction Rivas et al., 2008 15 IC2S2 Rıó Tinto (3.1 water dam) Dark red sediments Supernatant from EDTA wash Rivas et al., 2008 16 IC2C3 Rıó Tinto (3.1 water dam) Dark red sediments EDTA washed cells (sonicated) Rivas et al., 2008 17 IC3C1 Rıó Tinto (Main spring) Brown filaments Cellular fraction Rivas et al., 2008 18 IC3S2 Rıó Tinto (Main spring) Brown filaments Supernatant from EDTA wash Rivas et al., 2008 19 IC3C3 Rıó Tinto (Main spring) Brown filaments EDTA washed cells (sonicated) Rivas et al., 2008 20 IC4C1 Rıó Tinto (3.2 water dam) Brown filaments Cellular fraction Rivas et al., 2008 602 21 IC4S2 Rıó Tinto (3.2 water dam) Brown filaments Supernatant from EDTA wash Rivas et al., 2008 22 IC5C1 Rıó Tinto (Playa 3.1) Red crusts Cellular fraction Rivas et al., 2008 23 IC5S1 Rıó Tinto (Playa 3.1) Red crusts Supernatant Rivas et al., 2008 24 IC6C1 Rıó Tinto (Playa 3.1) Red sediment 1–2 cm under crust Cellular fraction Rivas et al., 2008 25 IC6S1 Rıó Tinto (Playa 3.1) Red sediment 1–2 cm under crust Supernatant Rivas et al., 2008 26 IC7C1 Rıó Tinto (3.2 water dam) Dried wall sediments Cellular fraction Rivas et al., 2008 27 IC8C1 Rıó Tinto (3.1 stream’s banks) Green-orange sediments Cellular fraction Rivas et al., 2008 28 IC8S1 Rıó Tinto (3.1 stream’s banks) Green-orange sediments Supernatant Rivas et al., 2008 29 IC9C1 Rıó Tinto (3.1 mine’s ruins) Red-gray sulfate-rich precipitates Cellular fraction Rivas et al., 2008 30 A138 Rıó Tinto (‘‘Nacimiento’’) Yellow mats (central stream) Whole Gu/HCl extraction Rivas et al., 2008 31 A140 Rıó Tinto (3.0 stream) Pink superficial layer Cellular fraction Rivas et al., 2008 32 A141 Rıó Tinto (3.0 stream) Pink superficial layer Supernatant Rivas et al., 2008 33 A143 Rıó Tinto (3.2 water dam) Wall sediments. Lithified overgrowth Whole Gu/HCl extraction Rivas et al., 2008 34 A152 Rıó Tinto (3.0 stream) Pink superficial layer Whole Gu/HCl extraction Rivas et al., 2008 35 ID1C1 Rıó Tinto (Main spring) Iron-sulfate-rich precipitates Cellular fraction Rivas et al., 2008 36 ID1S1 Rıó Tinto (Main spring) Iron-sulfate-rich precipitates Supernatant Rivas et al., 2008 37 ID1S2 Rıó Tinto (Main spring) Iron-sulfate-rich precipitates Supernatant from EDTA wash Rivas et al., 2008 38 ID1C3 Rıó Tinto (Main spring) Iron-sulfate-rich precipitates EDTA washed cells (sonicated) Rivas et al., 2008 39 ID2S2 Penã de Hierro (148 m deep) 4-59c sample (MARTE project) Supernatant from EDTA wash Rivas et al., 2008 40 ID3S2 Penã de Hierro (96 m deep) 4-39c sample (MARTE project) Supernatant from EDTA wash Rivas et al., 2008 41 ID4S2 Penã de Hierro (154 m deep) 4-61a sample (MARTE project) Supernatant from EDTA wash Rivas et al., 2008 42 ID5S2 Penã de Hierro (141 m deep) 4-56c sample (MARTE project) Supernatant from EDTA wash Rivas et al., 2008 43 ID7S2 Penã de Hierro (84–97 m deep) 8-42b-46bc (MARTE project) Supernatant from EDTA/guanidinium Rivas et al., 2008 (continued) Downloaded by Consejo Superior De Investigaciones Cientificas CSIC from www.liebertpub.com at 07/22/18. For personal use only.

Appendix Table 1. (Continued) No Ab name Source/Strain Sample/Culture conditions Immunogen/Fraction References 44 ID10S2 Penã de Hierro (119–127 m deep) 8-54a+54c+56c (MARTE project) Supernatant from EDTA/guanidinium Rivas et al., 2008 45 ID11S2 Penã de Hierro (138 m deep) 8-60b+60c (MARTE project) Supernatant from EDTA/guanidinium Rivas et al., 2008 46 ID12S2 Penã de Hierro (147–152 m deep) 8-63b-65b (MARTE project) Supernatant from EDTA/guanidinium Rivas et al., 2008 47 ID13S2 Penã de Hierro (2–3 m deep) 8-2a (MARTE project) Supernatant from EDTA/guanidinium Rivas et al., 2008 48 ID18S2 Penã de Hierro (93 m deep) 8-45b (MARTE project) Supernatant from EDTA/guanidinium Rivas et al., 2008 49 ID14S2 Penã de Hierro (105 m deep) 8-49b (MARTE project) Supernatant from EDTA/guanidinium Rivas et al., 2008 50 ID16S2 Penã de Hierro (152 m deep) 8-65b (MARTE project) Supernatant from EDTA/guanidinium Rivas et al., 2008 51 ID17C1 Rıó Tinto (3.2 water dam) Yellow Fe-S-rich salt precipitates Cellular fraction Rivas et al., 2008 52 ID17S1 Rıó Tinto (3.2 water dam) Yellow Fe-S-rich salt precipitates Supernatant Rivas et al., 2008 53 A185 Acidiphilium spp. Batch culture Whole cells (intact + sonicated) Parro et al., 2005 54 A139 Leptospirillum ferrooxidans Batch (N2 fixing) Sonicated cells Parro et al., 2005 55 A186 L. ferrooxidans Batch culture Whole cells (intact + sonicated) Parro et al., 2005 56 IVE1C1 L. pherrifilum (LPH2) Fermenter Whole cells Rivas et al., 2008 57 IVE1S1 L. pherrifilum (LPH2) Fermenter Culture supernatant Rivas et al., 2008 58 IVE1C2 L. pherrifilum (LPH2) Fermenter Insoluble cell pellet from S100 Rivas et al., 2008 59 IVE1S100 L. pherrifilum (LPH2) Fermenter Soluble cellular fraction S100 Rivas et al., 2008 60 IVE1BF L. pherrifilum (LPH2) Fermenter Biofilm Rivas et al., 2008 61 IVE2C1 L. pherrifilum spp. Batch + Fe2+ Whole cells Rivas et al., 2008 62 IVE2S1 L. pherrifilum spp. Batch + Fe2+ Culture supernatant Rivas et al., 2008 2+

603 63 IVE2S100 L. pherriﬁlum spp. Batch + Fe Soluble cellular fraction S100 Rivas et al., 2008 64 A183 Acidithiobacillus ferrooxidans Batch + Fe2+ Whole cells (sonicated) Parro et al., 2005 65 IVE3C1 At. ferrooxidans Batch + Fe2+ Whole cells Rivas et al., 2008 66 IVE3S1 At. ferrooxidans Batch + Fe2+ Culture supernatant Rivas et al., 2008 67 IVE3C2 At. ferrooxidans Batch + Fe2+ Insoluble cell pellet from S100 Rivas et al., 2008 68 IVE3S100 At. ferrooxidans Batch + Fe2+ Soluble cellular fraction S100 Rivas et al., 2008 69 A184 At. thiooxidans Batch + S Whole cells (intact + sonicated) Parro et al., 2005 70 IVE4C1 At. thiooxidans Batch + S Whole cells Rivas et al., 2008 71 IVE4C2 At. thiooxidans Batch + S Insoluble cell pellet from S100 Rivas et al., 2008 72 IVE4S100 At. thiooxidans Batch + S Soluble cellular fraction S100 Rivas et al., 2008 73 IVE5C1 At. albertensis Batch + S Whole cells Rivas et al., 2008 74 IVE5C2 At. albertensis Batch + S Insoluble cell pellet from S100 Rivas et al., 2008 75 IVE5S100 At. albertensis Batch + S Soluble cellular fraction S100 Rivas et al., 2008 76 IVE6C1 At. caldus Batch + S Whole cells Rivas et al., 2008 77 IVE6S2 At. caldus Batch + S Supernatant from EDTA wash Rivas et al., 2008 78 IVE6C2 At. caldus Batch + S Insoluble cell pellet from S100 Rivas et al., 2008 79 IVE6S100 At. caldus Batch + S Soluble cellular fraction S100 Rivas et al., 2008 80 IVE7C1 Halothiobacillus neapolitanus Batch + S Whole cells Parro et al., 2011a 81 IVE8C1 Acidimicrobium ferrooxidans Biomass from DSM N10331 Whole cells Parro et al., 2011a 82 IVE8S2 Acidimicrobium ferrooxidans Biomass from DSM N10331 Supernatant from EDTA wash Parro et al., 2011a 83 IVF1S1 Shewanella gelidimarina Batch (marine broth 15C) Culture supernatant Rivas et al., 2008 84 IVF2C1 S. gelidimarina Batch (marine broth 4C) Whole cells Rivas et al., 2008 (continued) Downloaded by Consejo Superior De Investigaciones Cientificas CSIC from www.liebertpub.com at 07/22/18. For personal use only.

Appendix Table 1. (Continued) No Ab name Source/Strain Sample/Culture conditions Immunogen/Fraction References 85 IVF2S2a S. gelidimarina Batch (marine broth 4C) Supernatant from EDTA wash Rivas et al., 2008 86 IVF2S100 S. gelidimarina Batch (marine broth 4C) Soluble cellular fraction S100 Rivas et al., 2008 87 IVF3C2 Psychroserpens burtonensis Batch (marine broth 15C) Insoluble cell pellet from S100 Rivas et al., 2008 88 IVF4C1 P. burtonensis Batch (marine broth 4C) Whole cells Rivas et al., 2008 89 IVF4S1 P. burtonensis Batch (marine broth 4C) Culture supernatant Rivas et al., 2008 90 IVF4S100 P. burtonensis Batch (marine broth 4C) Soluble cellular fraction S100 Rivas et al., 2008 91 IVF5C1 Psychrobacter frigidicola Batch (Harpo’s medium 15C) Whole cells Rivas et al., 2008 92 IVF5S1 Ps. frigidicola Batch (Harpo’s medium 15C) Culture supernatant Rivas et al., 2008 93 IVF5C2 Ps. frigidicola Batch (Harpo’s medium 15C) Insoluble cell pellet from S100 Rivas et al., 2008 94 IVF5S100 Ps. frigidicola Batch (Harpo’s medium 15C) Soluble cellular fraction S100 Rivas et al., 2008 95 IVF6C1 Cryobacterium psychrophilum Batch (TSA) Whole cells Rivas et al., 2008 96 IVF6S1 C. psychrophilum Batch (TSA) Culture supernatant Rivas et al., 2008 97 IVF6S2 C. psychrophilum Batch (TSA) Supernatant from EDTA wash Rivas et al., 2008 98 IVF6C2 C. psychrophilum Batch (TSA) Insoluble cell pellet from S100 Rivas et al., 2008 99 IVF6S100 C. psychrophilum Batch (TSA) Soluble cellular fraction S100 Rivas et al., 2008 100 IVF7C1 Colwellia psychrerythraea Bath culture (marine broth) Whole cells Parro et al., 2011a 101 IVF7S1 C. psychrerythraea Bath culture (marine broth) Culture supernatant Parro et al., 2011a 102 IVF7S2 C. psychrerythraea Bath culture (marine broth) Supernatant from EDTA wash Parro et al., 2011a 103 IVG1C1 Acidocella aminolytica Batch (DSMZ N 269) Whole cells Parro et al., 2011a DSM 11237 604 104 IVG2C_185 Acidiphillium spp. Batch (DSMZ N 269) Whole cells Parro et al., 2011a 105 IVG2C1 Acidiphillium sp. Batch (DSMZ N 269) Whole cells Parro et al., 2011a 106 IVG3C1 Acidobacterium capsulatum Batch (DSMZ N 269) Whole cells Parro et al., 2011a DSM 11244 107 IVG4C1 Thermus scotoductus Batch (TYG) Whole cells Parro et al., 2011a 108 IVG4C2 T. scotoductus Batch (TYG) Insoluble cell pellet from S100 Parro et al., 2011a 109 IVG5C1 Sulfobacillus acidophilus Biomass DSMZ No 10332 Whole cells Parro et al., 2011a 110 IVG6C1 T. thermophilus Batch (TYG) Whole cells Parro et al., 2011a 111 IVH1C1 Bacillus subtilis (spores) Batch (Schaeffer medium) Whole spores Ferna´ndez-Calvo et al., 2006 112 IVI1C1 Pseudomonas putida Batch (LB) Whole cells Rivas et al., 2008 113 IVI1C2 P. putida Batch (LB) Insoluble cell pellet from S100 Rivas et al., 2008 114 IVI1S100 P. putida Batch (LB) Soluble cellular fraction S100 Rivas et al., 2008 115 IVI1RB P. putida Batch (LB) Ribosome fraction Rivas et al., 2008 116 IVI2C1 Bacillus spp. (environ. isol.)* Batch (LB) Whole cells Rivas et al., 2008 117 IVI2S2 Bacillus spp. (environ. isol.)* Batch (LB) Supernatant from EDTA wash Rivas et al., 2008 118 IVI2C2 Bacillus spp. (environ. isol.)* Batch (LB) Insoluble cell pellet from S100 Rivas et al., 2008 119 IVI2S100 Bacillus spp. (environ. isol.)* Batch (LB) Soluble cellular fraction S100 Rivas et al., 2008 120 IVI3C1 Shewanella oneidensis Batch (LB) Whole cells Rivas et al., 2008 121 IVI3S2 S. oneidensis Batch (LB) Supernatant from EDTA wash Rivas et al., 2008 122 IVI3C2 S. oneidensis Batch (LB) Insoluble cell pellet from S100 Rivas et al., 2008 123 IVI3S100 S. oneidensis Batch (LB) Soluble cellular fraction S100 Rivas et al., 2008 (continued) Downloaded by Consejo Superior De Investigaciones Cientificas CSIC from www.liebertpub.com at 07/22/18. For personal use only.

Appendix Table 1. (Continued) No Ab name Source/Strain Sample/Culture conditions Immunogen/Fraction References 124 IVI4C1 Burkholderia fungorum Batch (LB) Whole cells Rivas et al., 2008 125 IVI4S2 B. fungorum Batch (LB) Supernatant from EDTA wash Rivas et al., 2008 126 IVI4C2 B. fungorum Batch (LB) Insoluble cell pellet from S100 Rivas et al., 2008 127 IVI4S100 B. fungorum Batch (LB) Soluble cellular fraction S100 Rivas et al., 2008 128 IVI5C1 S. oneidensis Anaerobic (fumarate) Whole cells Rivas et al., 2008 129 IVI5S1 S. oneidensis Anaerobic (fumarate) Culture supernatant Rivas et al., 2008 130 IVI5C2 S. oneidensis Anaerobic (fumarate) Insoluble cell pellet from S100 Rivas et al., 2008 131 IVI5S100 S. oneidensis Anaerobic (fumarate) Soluble cellular fraction S100 Rivas et al., 2008 132 IVI6C3 Azotobacter vinelandii Batch culture (LB) EDTA washed cells (sonicated) Rivas et al., 2008 133 IVI7C1 B. subtilis 168 Batch culture (LB) vegetative cells Whole cells Rivas et al., 2008 134 IVI8C1 B. subtilis 3610 Biofilm Whole cells Rivas et al., 2008 135 IVI8S1 B. subtilis 3610 Biofilm Culture supernatant Rivas et al., 2008 136 IVI9C1 Deinococcus radiodurans Biomass DSMZ No 20539 Whole cells Rivas et al., 2008 137 IVI10C1 Desulfovibrio vulgaris (vulgaris) Biomass DSMZ No 644 Whole cells Rivas et al., 2008 138 IVI11C1 Geobacter sulfurreducens Biomass DSMZ No 12127 Whole cells Rivas et al., 2008 139 IVI12C1 Geobacter metallireducens Biomass DSMZ No 7210 Whole cells Rivas et al., 2008 140 IVI13C1 Thermotoga maritima Biomass DSMZ No 3109 Whole cells Rivas et al., 2008 141 IVI14C1 Verrucomicrobium spinosum Biomass DSMZ No 4136 Whole cells Rivas et al., 2008 142 IVI15C1 Methylomicrobium capsulatum Biomass DSMZ No 6130 Whole cells Rivas et al., 2008 143 IVI16C1 Planctomyces limnophilus Biomass DSMZ No 3776 Whole cells Rivas et al., 2008 605 144 IVI17C1 Hydrogenobacter thermophilus Biomass DSMZ No 6534 Whole cells Rivas et al., 2008 145 IVI19C1 Desulfosporosinus meridiei Biomass DSMZ No 13257 Whole cells Parro et al., 2011a 146 IVI20C1 Salinibacter ruber M8 Batch (SW25% marine salt) Whole cells Parro et al., 2011a 147 IVI21C1 S. ruber PR1 Batch (SW25% marine salt) Whole cells Parro et al., 2011a 148 IVI21C2 S. ruber PR1 Batch (SW25% marine salt) Insoluble cell pellet from S100 Parro et al., 2011a 149 IVI21S1 S. ruber PR1 Batch (SW25% marine salt) Culture supernatant Parro et al., 2011a 150 IVJ1C1 Haloferax mediterranei Batch (SW25% marine salt) Whole cells Rivas et al., 2008 151 IVJ2C1 Methanococcoides burtonii Biomass DSMZ No 6242 Whole cells Rivas et al., 2008 152 IVJ3C1 Thermoplasma acidophilum Biomass DSMZ No 1728 Whole cells Rivas et al., 2008 153 IVJ4C1 Methanobacterium formicicum Biomass DSMZ No 1535 Whole cells Rivas et al., 2008 154 IVJ5C1 Methanosarcina mazeii Biomass DSMZ No 3647 Whole cells Rivas et al., 2008 155 IVJ6C1 Pyrococcus furiosus Biomass DSM No 3638 Whole cells Parro et al., 2011a 156 IVJ8C1 Halorubrum sp. Batch (SW25% marine salt) Whole cells Parro et al., 2011a 157 IVJ9C1 Halobacterium sp. Batch (SW25% marine salt) Whole cells Parro et al., 2011a 158 A-Mycob Mycobacterium genus Batch culture M. tuberculosis Genus-specific antigens extract BIODESIGN (B47827R) 159 A-Paer P. aeruginosa Batch culture (P. aeruginosa) Outer membrane protein extract BIODESIGN (B47578G) 160 A-GroEL GroEL (E. coli) Recombinant GroEL (purified) Sigma-Aldrich (G6532) 161 A-Hsp70 HSP-70 Batch culture (M. tuberculosis) HSP-70 (purified protein) BIODESIGN (H86313M) (continued) Downloaded by Consejo Superior De Investigaciones Cientificas CSIC from www.liebertpub.com at 07/22/18. For personal use only.

Appendix Table 1. (Continued) No Ab name Source/Strain Sample/Culture conditions Immunogen/Fraction References 162 A-Hfer Human ferritin From human liver Ferritin (purified) BIODESIGN (H53715) 163 A-HBVAg Hepatitis B virus surface Ag Hepatitis B virus surface antigen Highly purified hepatitis B antigen Abcam (ab9216) 164 A-ASB_11362 Archaeoglobus fulgidus ATP synthase, subunit B (Archaea) Purified recombinant polypeptide Parro et al., 2011a 165 A-ASF1_11355 Thermotoga maritima ATP synthase, Alpha (Bacteria) Purified recombinant polypeptide Parro et al., 2011a 166 A-BaFER Bacterio ferritin Bacterio ferritin Purified recombinant polypeptide Parro et al., 2011a 167 A-CrReTs_977 T. scotoductus Chromate reductase (membrane) Purified recombinant polypeptide Parro et al., 2011a 168 A-DsrA-11365 DsrA (Archaeoglobus fulgidus) Dissimilatory sulfite reductase Purified recombinant polypeptide Parro et al., 2011a 169 A-DsrB_11368 DsrB (Archaeoglobus fulgidus) Dissimilatory sulfite reductase Purified recombinant polypeptide Parro et al., 2011a 170 A-EFG_11359 Thermotoga maritima Elongation factor G Purified recombinant polypeptide Parro et al., 2011a 171 A-FeReTs_983 Iron reductase (T. scotoductus) Iron reductase Purified recombinant polypeptide Parro et al., 2011a 172 A-NifD_11466 NifD (G. metallireducens) NifD protein Purified recombinant polypeptide Parro et al., 2011a 173 A-NirS_11369 NirS (P. aeruginosa) Nitrite reductase Purified recombinant polypeptide Parro et al., 2011a 174 A-NOR1_11375 NOR1 (N. hamburgensis) Nitrite oxidoreductase Beta subunit Purified recombinant polypeptide Parro et al., 2011a 175 A-NRA-11912 NRA (G. metallireducens) Nitrate reductase subunit Alpha Purified recombinant polypeptide Parro et al., 2011a 176 A-ABCtrans ABC transporter (T. scotoductus) ABC-transporter protein Purified recombinant polypeptide Parro et al., 2011a 177 A-ApsA-11754 ApsA (Desulfovibrio desulfuricans) HMMLREMREGRGPIYC Conjugate Parro et al., 2011a

606 178 A-BfR BFR (D. desulfuricans) CAENFAERIKELFFEP Conjugate Parro et al., 2011a 179 A-Groel GroEL (G. metallireducens) ETEMKEKKARVEDALC Conjugate Parro et al., 2011a 180 A-Ktrans_11750 K+ -transporter (G. metallireducens) LAMSLGRKGEGGTIVC Conjugate Parro et al., 2011a 181 A-Stv Streptavidin Culture (Streptomyces avidinii) Streptavidin (purified) Sigma-Aldrich (S6390) 182 A-EPS_SP Exopolysaccharides Solar saltern (Alicante, Spain) EPS extract Parro et al., 2011a 183 A-LPS_N LPS (Pseudomonas sp.) Lipopolysaccharide LPS-BSA Parro et al., 2011a 184 A-GST Glutathione S-transferase Recombinant (S. japonicum) GST (purified) Sigma-Aldrich (G7781) 185 A-dinitroph. Dinitrophenol Dinitrophenol Dinitrophenol-BSA Sigma-Aldrich (D 9656) 186 A-Tc Tetracycline Tetracycline Tetracycline-BSA USBiological (T2965-05) 187 A-Trp Tryptophan Tryptophan Trp-glutaraldehyde-poly-lysine USBiological (T9013-05) 188 IVF15C1 Polaromonas sp. Batch (TSA) Whole cells This work 189 IVF31C1 Planococcus sp1 Batch (TSA) Whole cells This work 190 IVF34C1 Planococcus sp3 Batch (TSA) Whole cells This work 191 VID1C1 Atacama subsurface (2 m) Crude extract Cellular and EPS fractions This work 192 IIIC3C1 Deception Island permafrost Crude extract Cellular and EPS fractions This work 193 VIID3BF Biofilm from a concrete wall Crude extract Cellular and EPS fractions This work

From number 53 to the bottom are those ﬂuorescently labeled antibodies used as tracers for sandwich immunoassay (see Materials and Methods).