1   MICROORGANISMS, ORGANIC CARBON, AND THEIR RELATIONSHIP WITH 2   OXIDANT ACTIVITY IN HYPER-ARID MARS-LIKE SOILS: IMPLICATIONS FOR 3   SOIL HABITABILITY 4   JULIO E. VALDIVIA-SILVA1,2,*, FATHI KAROUIA3,4, RAFAEL NAVARRO-GONZALEZ2, 5   CHRISTOPHER MCKAY1 6   1Space Science & Astrobiology Division, NASA Ames Research Center, Moffett Field, CA 94035, 7   USA, [email protected]; 2Laboratorio de Química de Plasmas y Estudios 8   Planetarios, Instituto de Ciencias Nucleares, Universidad Nacional Autónoma de México, 9   Circuito Exterior, Ciudad Universitaria México D.F., 04510, México; 3Exobiology Branch, 10   NASA Ames Research Center, Moffett Field, CA 94035, USA; 4Department of Pharmaceutical 11   Chemistry, University of California San Francisco, San Francisco, CA 94158, USA. 12   13   *Corresponding Author: Julio E. Valdivia-Silva M.D., Ph.D., NASA Ames Research Center, 14   Space Science & Astrobiology Division, M/S: 245-3, Moffett Field, CA 94035, USA, 15   Telephone: +1-650-604-1136, Fax: +1-650-604-6779, E-mail: 16   [email protected] 17   18   RRH: OXIDANT ACTIVITY AND SOIL HABITABILITY IN MARS-LIKE SOILS 19   LRH: VALDIVIA-SILVA ET AL. 20   Keywords: Mars-like soils, Pampas de La Joya, soil habitability, oxidant activity 21   22   ABSTRACT: 23   Soil samples from the hyper-arid region in the Atacama Desert in Southern Peru (La Joya 24   Desert) were analyzed for total and labile organic carbon (TOC & LOC), phospholipid fatty 25   acids analysis (PLFA), quantitative real time polymerase chain reaction (qRT-PCR), 4’,6- 26   diamidino-2-phenylindole (DAPI)-fluorescent microscopy, culturable microorganisms, and 27   oxidant activity, in order to understand the relationship between the presence of organic matter 28   and microorganisms in these types of soils. TOC content levels were similar to the labile pool of 29   carbon suggesting the absence of recalcitrant carbon in these soils. The range of LOC was from 2 30   to 60 µg/g of soil. PLFA analysis indicated a maximum of 2.3 x 105 cell equivalents/g. Culturing 31   of soil extracts yielded 1.1 x 102–3.7 x 103 CFU/g. qRT-PCR showed between 1.0 x 102 and 8 x 32   103 cells/g; and DAPI fluorescent staining indicated bacteria counts up to 5 x 104 cells/g. Arid 33   and semiarid samples (controls) showed values between 107 and 1011 cells/g with all of the 34   methods used. Importantly, the concentration of microorganisms in hyper-arid soils did not show 35   any correlation with the organic carbon content; however, there was a significant dependence on 36   the oxidant activity present in these soil samples evaluated as the capacity to decompose sodium 37   formate in 10 hours. We suggest that the analysis of oxidant activity could be a useful indicator 38   of the microbial habitability in hyper-arid soils, obviating the need to measure water activity over 39   time. This approach could be useful in astrobiological studies on other worlds. 40   41   INTRODUCTION 42   43   Studies of Mars-like soils on Earth provide an important approach for better understanding the 44   physical, geochemical, and microbiological processes that occur, or could have occurred, on 45   Mars (Navarro-González et al., 2006; Mahaffy, 2008; Peters et al., 2008). The value of Mars-like 46   areas relies on the similarity of the analogue to its target, either in terms of their mineralogical or 47   geochemical context, or current physical or chemical environmental conditions. They illustrate 48   preservation mechanisms that could guide the search for fossil and biological remnants of 49   microbial life, which could be extrapolated to the ancient or current Mars (Preston and Dartnell, 50   2014). Although, Mars today is a cold dry desert world with surface conditions that are not 51   habitable even for the hardiest “known” life forms from Earth, there is ample evidence of past 52   water activity and the presence of interesting niches for life, such as subsurface or/and evaporitic 53   minerals, that make the Red Planet in a prime target for looking for extraterrestrial 54   microorganisms (Davila et al., 2010; McKay, 2010). 55   In this regard, a Martian soil analogue extensively studied and of great scientific interest is the 56   Atacama Desert, located in northern Chile and southern Peru. This desert lies on the west slopes 57   of Central Andes between 15°S and 30°S (Houston and Hartley, 2003; Hartley et al., 2005) and 58   is considered one of the oldest and driest desert on Earth. Previous research has identified the 59   Atacama as a key analogue model for life in dry Mars-like conditions (McKay et al., 2003; 60   Navarro-González et al., 2003). Hyper-arid soils from the Chilean and Peruvian region have 61   represented an ideal test bed for constraining the limits of life or its preserved remnants (McKay 62   et al., 2003; Drees et al., 2006). Previous studies have shown the presence of very low levels of 63   organic carbon (20-40 ppm), non-biological oxidants and exotic evaporitic minerals, which all 64   together are common characteristics expected on Mars (Connon et al., 2007; Fletcher et al., 65   2012). Importantly, detailed multidisciplinary studies comparing those characteristics from the 66   Pampas de La Joya, our sampling site located in the Peruvian hyper-arid region, and Mars have 67   demonstrated that this area represents a valuable Martian analogue for studies of oxidative 68   processes that may occur on Mars, and can be used for the testing of instruments designed for 69   future Martian life detection missions (Valdivia-Silva et al., 2009; 2011; 2012a; 2012b). These 70   studies showed different types of abiotic oxidants with strong chemical activity on the 71   accelerated destruction of organics under several experimental conditions (Quinn et al., 2007; 72   Valdivia-Silva et al., 2012c), and were extrapolated to Mars trying to explain the low levels of 73   organics on its surface (Ponnamperuma et al., 1977; Oyama and Berdahl, 1979; Quinn and Zent, 74   1999; Quinn et al., 2007; Quinn et al., 2013). Since the return of the Viking data several 75   hypotheses have been presented to explain oxidative activity on Martian surface (Quinn et al., 76   2007). Hydrogen peroxide, superoxides, UV radiation, peroxide-modified titanium dioxide, 77   peroxinitrites, radiolysis products of perchlorates, etc., are possible candidates (McKay et al., 78   1998; Zent et al., 2008; Quinn et al., 2011, 2013). Similarly, Pampas de La Joya and other 79   samples from the Atacama have shown the presence of non-chirally specific and as-yet- 80   unidentified oxidants – indicating a chemical oxidation not biological consumption of amino 81   acids and sugars (Navarro-Gonzalez et al., 2003; Valdivia-Silva et al., 2012c). Although this 82   region is characterized by large amounts of deposited salt (Michalski et al., 2004) and contains 83   highly oxidative species, including iodates (IO3-), chromates (CrO 2-), perchlorates (ClO -) and 4 4 84   probably persulfates (S O -2 ) (Ewing et al., 2006); these candidates do not completely explain 2 8 85   our results, nor experiments similar to those made by the Viking Lander on Mars (Navarro- 86   Gonzalez et al., 2003; Quinn et al., 2007; Valdivia-Silva 2009; 2011). The Quinn et al. (2013) 87   proposal based on radiolysis of perchlorates may explain the Viking results but would not 88   provide a complete explanation for the Atacama oxidants. 89   On the other hand, an important and contradictory issue to resolve in this type of hyper-arid soils 90   is that the levels of organic content have no shown a direct correlation with the number of 91   microorganisms or biomass inside the dry core region (Navarro-Gonzalez et al 2003; Valdivia- 92   Silva et al 2011). Although the microbiological studies have shown good correlation curves 93   between organics and microorganisms in precipitation gradients throughout semiarid, arid, and 94   hyper-arid areas and/or from the Pacific coast to the core of the desert (Dress et al 2006; Fletcher 95   et al., 2011); most analyses in the hyper arid core have shown conflicting results with regard to 96   number of microorganisms as discussed later (Orlando et al., 2010; Fletcher et al., 2011). 97   It is known that in extreme arid deserts, the low water activity severely limits microbial growth, 98   abundance, and diversity (Kieft, 2002; Warren-Rhodes et al., 2006; Crits-Christoph et al., 2013). 99   However, these environmental conditions do not completely explain the high variability of 100   microorganisms forming patches in the hyper-arid core even few meters apart. Indeed, although 101   previous reports of near sterile Atacama Desert soils (~10 Colony Forming Unit (CFU)/g) have 102   led to the suggestion that the dry limit for microbial life had been crossed (McKay et al., 2003; 103   Navarro-González et al., 2003), other investigations in the region produced results ranging 104   values near to ~106 cells/g (Glavin et al., 2004) or even more (Drees et al., 2006; Orlando et al., 105   2010). Moreover, in a comprehensive survey of the Yungay regions, Bagaley (2006) found 106   variations in CFU by orders of magnitude over distances of few hundred meters. 107   In this regard, this high heterogeneity in the presence of microorganisms despite having almost 108   uniform, albeit low, levels of water in the area, we hypothesize that it is the level of oxidant 109   activity, rather than the amount of water, which determines the survival of microorganisms and 110   maybe organic matter in hyper-arid soils. Indeed, although the presence of water could be 111   beneficial for habitability, on the other hand it could also be detrimental to the survival of 112   microorganisms and organic compounds in these soils since all mineral oxidants require water to 113   trigger chemical reactions. 114   In this work, we present the results of a multi-component investigation involving different 115   microbiological and geochemical analyses of soil samples collected in the hyper-arid soils from 116   Pampa de la Joya in southern Peru (Valdivia-Silva et al., 2011; Preston and Dartnell, 2014), and 117   in two points along a latitudinal moisture gradient to the Pacific coast used as controls (from 0.5 118   mm to 120 mm/y rainfall) (Fig. 1) in order to evaluate: 1) the biomass and organic carbon 119   distribution in the core of the hyper-arid region, and 2) the relationship between these 120   measurements and the oxidizing processes in the area which could be limiting the growth of 121   microorganisms. 122   123   METHODOLOGY 124   Soil samples 125   Soil samples used in this study were collected from 2008 to 2012 in the hyper-arid area of 126   Pampas de La Joya, or alternatively called “La Joya Desert”, located about 70 km from the city 127   of Arequipa, Peru, along the South Pacific coast, between latitudes of 16°S–17°S, longitudes of 128   71.5°W–72.5°W, and approximately between 1000 and 2000 m.a.s.l. The entire area is part of 129   the Atacama Desert, inside of the region considered as “hyper-arid” (Houston and Hartley, 130   2003). This region is considered hyper-arid because the Aridity Index (AI) calculated as the ratio 131   P/PET (evapotranspiration / precipitation) is less than 0.05 (Thornwaite, 1948; UNEP, 1997). 132   Our specific area of analyses encompasses an area of 96 km2 (12 x 8 km) referred in this paper as 133   the “quadrangle of interest” (diagonal coordinates: 16°38.386’S – 72°2.679’W and 16°44.986’S 134   – 71°58.279’W) (Fig. 1). Importantly, this quadrangle contains interesting locations of evaporitic 135   minerals, quartz rocks, and soils with high oxidant activity and very low levels of organic matter 136   (Valdivia-Silva et al., 2009; 2011; 2012a). 137   Between 50 g and 100 g, representing a composite of 5 individual nearby sites (~1.5 m in radius) 138   of each sampling site, were collected from the surface to a depth of 5 – 10 cm using sterile 139   scoops and stored in sterile polyethylene (Whirlpak TM) bags for transport until analysis. The 140   samples were homogenized to obtain a representative sample for each time to compensate for 141   any small scale spatial heterogeneity (Girvan et al., 2003). Immediately after collection, the 142   samples were stored in a cool, dry, and dark location before being shipped to NASA ARC. The 143   transport required 4 days, and after arrival, the samples were stored in a -20°C refrigerator until 144   the analyses of TOC, LOC, PLFA, qRT-PCR, DAPI, and oxidation activity. Culture dependent 145   assays were initiated 14 days after sample collection. Importantly, separate bags of soil were 146   used for microbiology analysis than those used for chemical studies. Although, storage 147   conditions could change structure and function of communities, several studies have shown that 148   microorganisms in hyper-arid soils are in a dormant state due to very low water activity, and the 149   storage in similar conditions (mainly low humidity) did not show significantly change the 150   microbial community (Navarro-Gonzalez et al., 2003; Glavin et al., 2004; Drees et al., 2006; 151   Connon et al., 2007; Lester et al., 2007; Orlando et al., 2010; Fletcher et al., 2011; Crits- 152   Christoph et al., 2013). Furthermore, in this study we track changes in total microbial biomass 153   and not microbial community structure. 154   Because of the high spatial heterogeneity reported in these soils (Peeters et al., 2009; Valdivia- 155   Silva et al., 2012b) the sampling was a systematic “grid” type (Webster and Oliver, 1990) (Fig. 156   1). For labile organic carbon (LOC) and fluorescence microscopy (DAPI-FM) 35 soil samples 157   spaced 2 km apart were analyzed; for total organic carbon (TOC), quantitative polymerase chain 158   reaction (qRT-PCR), and cultures 12 samples spaced 4 km apart were evaluated; and finally for 159   phospholipid fatty acid analysis (PLFA) 5 samples from the vertices and at the midpoint of the 160   quadrangle were analyzed. Additionally, two more samples were collected (positive controls) 161   following a short precipitation gradient from an arid (P/PET > 0.05; coord.16°43.945’S – 162   72°18.645’W) and a semiarid (P/PET > 0.2; coord.16°50.31’S – 72°17.16’W) area (Fig. 1). 163   Importantly, LOC and DAPI analyses were used to first evaluate the presence of a correlation 164   between both variables (Fig. 2) and this was followed by the application of the other methods 165   used in this study to confirm and validate the number of bacteria obtained in the DAPI results. 166   LOC and DAPI are faster than the other methods we use and can to complete many samples over 167   large areas in order to build a preliminary map of variations and correlations on which to base 168   further specific analysis (Fletcher et al., 2011; 2014). 169   170   Organic carbon analysis 171   The labile organic carbon (LOC) content was evaluated by permanganate titration in acid media 172   as our group has reported before (Valdivia-Silva et al., 2011; Fletcher et al., 2012). This 173   technique has proven to be simple, accurate, sensitive, and reproducible for the quantification of 174   labile organic carbon in hyper-arid soils (Fletcher et al., 2012; 2014). Since, these soils have 175   shown negligible levels, or almost absence, of recalcitrant carbon (or passive pool), the labile 176   form (or active pool) is the most abundant in these soils and includes molecules with biological 177   importance such as amino acids, nucleotides, lipids, sugars, aliphatic and aromatic hydrocarbons, 178   etc (Skelley et al., 2007; Ewing et al., 2008) . In order to demonstrate the low levels of other 179   forms of organic carbon we evaluated the total organic carbon (TOC) using an Elemental 180   Analyzer model EA1108 (Fisions, Loughborough, U.K.) at 1200°C. The inorganic carbon from 181   carbonates and bicarbonates was removed by acid treatment as has been reported before 182   (Navarro-Gonzalez et al., 2006). 183   Negative controls were made by heating some soil samples to 500°C for 8 hours. 184   185   Microbiological analyses 186   The quantification of microorganisms in hyper-arid soils has shown to be a real challenge for 187   microbiologists since each method has specific limitations. Because of this, in this work we 188   analyze the number of microorganisms using different independent and dependent culture-based 189   techniques in order to improve our approach to the number of microorganisms present in the soil. 190   A more comprehensive comparison between molecular enumeration techniques used in this type 191   of dry soils was published for our group (see Table 3 in Fletcher et al., 2011). 192   193   DAPI staining by fluorescence microscopy 194   Soil samples from La Joya were analyzed under the fluorescence microscope by means of 4’,6- 195   diamidino-2-phenylindole (DAPI) stain protocol, which is a fluorescent stain that binds strongly 196   to double-stranded DNA. For fluorescence microscopy, DAPI is excited with ultraviolet light 197   (absorption at 358 nm) emitting blue light at 461 nm. Despite being less specific for DNA than 198   previously thought, DAPI has been used as the bacterial stain of choice for a wide range of 199   sample types and is particularly useful for quantifying the total number of nonviable and viable 200   bacteria in natural samples (Kepner and Pratt, 1994). The protocol of this technique has been 201   previously reported for hyper-arid soils demonstrating good sensibility and reproducibility 202   (Glavin et al., 2004; Fletcher et al., 2011). The analysis of the images was done by the program 203   NIH Image-J, which enabled the calculation of bacterial biomass in terms of pixels (Kemp, 1993; 204   Posch et al., 2001). 205 206   Phospholipid fatty acid analysis (PLFA) 207   PLFA analyses on 50 g soil samples were performed by Microbial Insights (Rockford, TN). The 208   limit of detection for PLFA analysis is ~50 picomoles of total PLFA and the limit of 209   quantification is ~150 picomoles of total PLFA (104 cell/g). Microbial biomass was determined 210   from the total concentration of ester-linked phospholipid fatty acids in the samples (White and 211   Ringelberg, 1997). Importantly, the PLFA analysis could detect phospholipids from dead 212   microorganisms which can remain for long time due to very low, if any, bacterial metabolism 213   and the lack of chemical activity in the dry soils. Preliminary results of the structural community 214   profiles were generated for each sample by community-level PLFA analyses based on six major 215   PLFA structural groups, each of which corresponds to a broad phylogenetic group of 216   microorganisms (White, 1979; Lehman et al., 1995; White et al., 1996). The complete protocol 217   of this technique is published on the Microbial Insights website (http://www.microbe.com/). 218   219   Quantitative real time polymerase chain reaction (qRT-PCR) 220   The DNA extraction and purification of soil samples was done using the Ultra-Clean Mega Soil 221   DNA extraction kit (MoBio Laboratories Inc.) following the recommended instructions. 222   Amplifications were performed using a Cepheid, Inc., Smart Cycler automated real-time PCR 223   system with 25 µL reaction tubes. The necessary enzymes were added by using the OmniMix 224   PCR Beads (Cepheid, Inc.) as they are a premade enzyme mix containing reagents for the PCR 225   reaction that are stable in a field environment. The 16S rRNA genes were amplified using the 226   universal primers 8F (5'-AGA GTT TGATCM TGG CTC AG-3'), and 1492R (5'-GGY TAC 227   CTT GTT ACG ACT T-3'). A calibration curve was generated by running a serial dilution series 228   using purified Bacillus globigii DNA at known concentrations, from a starting concentration of