Sage Journals: Discover world-class research

Abstract

Microbe-mineral interactions can produce unique composite materials, which can preserve biosignatures. Geological evidence suggests that iron sulfide (Fe-S) minerals are abundant in the subsurface of Mars. On Earth, the formation of Fe-S minerals is driven by sulfate-reducing microorganisms (SRM) that produce reactive sulfide. Moreover, SRM metabolites, as well as intact cells, can influence the morphology, particle size, aggregation, and composition of biogenic Fe-S minerals. In this work, we evaluated how simple and complex organic molecules—hexoses and amino acid/peptide mixtures, respectively—influence the formation of Fe-S minerals (simulated prebiotic conditions), and whether the observed patterns mimic the biological influence of SRM. To this end, organo-mineral aggregates were characterized with X-ray diffraction, scanning electron microscopy, and scanning transmission X-ray microscopy coupled to near-edge X-ray absorption fine structure spectroscopy. Overall, Fe-S minerals were found to have a strong affinity for proteinaceous organic matter. Fe-S minerals precipitated at simulated prebiotic conditions yielded organic carbon distributions that were more homogeneous than treatments with whole SRM cells. In prebiotic experiments, spectroscopy detected potential organic transformations during Fe-S mineral formation, including conversion of hexoses to sugar acids and polymerization of amino acids/peptides into larger peptides/proteins. In addition, prebiotic mineral-carbon assemblages produced nanometer-scaled filamentous aggregated morphologies. On the contrary, in biotic treatments with cells, organic carbon in minerals displayed a more heterogeneous distribution. Notably, “hot spots” of organic carbon and oxygen-containing functional groups, with the size, shape, and composition of microbial cells, were preserved in mineral aggregates. We propose a list of characteristics that could be used to help distinguish biogenic from prebiotic/abiotic Fe-S minerals and help refine the search of extant or extinct microbial life in the martian subsurface.

1. Introduction

It is now established that liquid water has been present at the surface of early Mars for periods of time that are long enough for microbial life to have emerged and proliferated (McKay et al., 1992; Squyres et al., 2004; Cockell, 2014; Grotzinger et al., 2014). Based on the hypothesis that martian life would be carbon-based as is life on Earth, the next step in the search of life on the red planet is the investigation of organic signatures (Westall et al., 2015; Hays et al., 2017; Brack, 2018). The presence of strong oxidants at the surface of Mars precludes the accumulation of organic molecules (Freissinet et al., 2015; Lasne et al., 2016; Eigenbrode et al., 2018) and casts doubt about the results of early in situ biological experiments at the surface of Mars by the Viking landers (Klein et al., 1976; Levin and Straat, 1977, 2016; Mazur et al., 1978; Margulis et al., 1979). However, organic molecules could potentially be well preserved in the subsurface of Mars, where they would be shielded from ultraviolet and ionizing radiation and be in a more reducing environment than at the surface (Cockell, 2014; Westall et al., 2015). On Earth, the most important mechanism that preserves organic carbon over geological timescales is the protection by minerals (Keil et al., 1994; Keil and Mayer, 2014; Hemingway et al., 2019). Therefore, organo-mineral aggregates precipitated in reducing subsurface environments represent a promising target to look for traces of microbial life on Mars (Hays et al., 2017).

Iron sulfide (Fe-S) minerals have been proposed to be abundant in the subsurface of Mars (Burns and Fisher, 1990a), and therefore represent a promising system for the search of life on Mars. Fe-S minerals on Earth provide the justification for this supposition, and an opportunity to further explore the relationship between Fe-S minerals and biosignatures. On Earth, Fe-S minerals precipitate in anoxic sedimentary environments from the reaction between ferrous iron and sulfide, with the latter being produced by sulfate-reducing microorganisms (SRM). The process of microbial sulfate reduction has been driving the sulfur cycle on Earth for potentially as long as the past 3.4 Gy (Shen and Buick, 2004; Shen et al., 2009; Jørgensen et al., 2019). The SRM not only provide nearly all sulfide for the formation of iron sulfide minerals, but they can also influence the physical properties and the composition of Fe-S minerals (Picard et al., 2018, 2019; Mansor et al., 2019). In addition, SRM can impart large sulfur isotopic fractionation between sulfate and sulfide that can be preserved in S-bearing solid phases, including Fe-S minerals, on geological timescales (Canfield, 2001; Fike et al., 2015). Finally, Fe-S minerals are associated with biomass-derived organic molecules when precipitated in the presence of SRM and can potentially preserve organic signatures for long periods (Picard et al., 2019).

If life emerged on Mars, it has likely happened in its early history, when environmental conditions were more similar to those on Earth (Nisbet and Sleep, 2001; Zahnle et al., 2007). Life on Mars would likely have been anaerobic and chemotrophic (Cockell, 2014; Westall et al., 2015), and microbial sulfate reduction is a plausible metabolic pathway for martian microorganisms based on available reactants (Banfield et al., 2001; McMahon et al., 2018). Sulfur is abundant and ubiquitous on Mars. On the surface of modern Mars, it is mostly found in the oxidized state (S⁶⁺) in a variety of sulfate minerals, whereas reduced sulfur species (S²⁻, S¹⁻) have been proposed to be abundant in the martian subsurface (Franz et al., 2018). With such a range of oxidation states, there is potential for a biologically mediated sulfur cycle on Mars (King and McLennan, 2010). Early anaerobic life could have thrived in the reducing subsurface, and/or could have retreated to the subsurface when the surface dried out and became oxidizing (Cockell, 2014). Abundant sulfate salts could have been used by SRM if liquid water was available. Abiotic hydrocarbons or H₂ have been proposed as potential energy sources, and carbon dioxide or magmatic carbon as a potential carbon source for building biomass (Michalski et al., 2013; Cockell, 2014). As Fe is also abundant on Mars, iron sulfide minerals have been proposed to be abundant in the subsurface. This is supported by meteorite data and orbital and in situ measurements (Burns and Fisher, 1990a, 1990b; Lorand et al., 2005; Zolotov and Shock, 2005; Vaniman et al., 2014; Franz et al., 2018). Massive sulfide ores mainly composed of pyrrhotite and pentlandite could have formed on Mars as a result of magmatic activity (Burns and Fisher, 1990a), whereas pyrite deposition could have occurred through hydrothermal activity induced by impacts (Zolotov and Shock, 2005). If microbial sulfate reduction evolved on Mars, iron sulfide minerals formed from microbially reduced sulfur may also be present (Zolotov and Shock, 2005; Parnell et al., 2010; Frank et al., 2015).

Although microbial biomass can associate with Fe-S minerals and influence their properties, it is unknown whether individual organic molecules or organic mixtures could influence the formation of iron sulfide minerals in a way that could mimic a biological signature. In other words, differentiating the processes of iron sulfide mineral formation requires that biotic and abiotic processes produce discernably distinct minerals. Iron sulfide minerals can record a biological imprint in their morphology, element and isotopic composition, but they can also preserve organic signatures. On Mars, organic molecules received in significant amounts from space or produced abiotically by past hydrothermalism (Steele et al., 2012, 2016, 2018; Franz et al., 2020) could have influenced the properties of iron sulfide minerals in the subsurface of Mars without the presence of life.

In this work, we aimed at identifying the differences between iron sulfide minerals precipitated in the presence of simple molecules (glucose or mannose) and complex organic mixtures (tryptone or yeast extract) at simulated prebiotic conditions, and iron sulfide minerals precipitated at biotic conditions (i.e., fully grown cultures of Desulfovibrio hydrothermalis that contained biotic sulfide and microbial cells). In the latter case, minerals were precipitated away from cells, as Fe(II) was added after sulfide had been produced, leaving no opportunity for Fe(II) to bind to the cells (Picard et al., 2018). We compare our results with minerals precipitated with microbial cells grown in the presence of Fe(II), in which case minerals precipitate at the surface of cells and strongly associate with organic molecules (Picard et al., 2018, 2019). We used X-ray diffraction (XRD), scanning electron microscopy (SEM), and scanning transmission X-ray microscopy (STXM) coupled to near-edge X-ray absorption fine structure (NEXAFS) spectroscopy to characterize the mineralogy, morphology, and chemistry of iron sulfide minerals and to determine whether we can distinguish prebiotic from biotic signatures.

2. Materials and Methods

2.1. Glassware, medium and solutions

Glassware used in this study was combusted at 550°C, washed with 1 N HCl, rinsed with ultrapure water, and autoclaved. Rubber stoppers were boiled three times in ultrapure water and autoclaved. The medium used for iron sulfide mineral precipitation experiments was an anoxic marine mineral medium (DSMZ 195c medium, omitting the addition of sulfide and resazurin) that is typically used for the cultivation of sulfate-reducing bacteria. The complete medium consisted of five sterile solutions that were mixed together in the anaerobic chamber: anoxic marine mineral solution (870 mL), anoxic sodium bicarbonate buffer (100 mL), anoxic sodium l-lactate solution (10 mL), ATCC^® vitamin solution (MD-VS, 10 mL), and ATCC trace element solution (MD-TES, 10 mL) (see medium recipe in Supplementary Table S1). The medium was dispensed in 100-mL serum vials in the anaerobic chamber (50 mL of medium per serum vial). Serum vials were sealed with butyl rubber stoppers and aluminum caps. Concentrated solutions of glucose (C content: 40%) and mannose (C content: 40%) were each prepared at 90 g/L, which corresponds to a C concentration of 36 g C/L. Concentrated solutions of tryptone (C content: 44%) and yeast extract (C content: 39%) were each prepared at 50 g/L, which corresponds to 22 and 19.5 g C/L, respectively. These concentrated solutions were prepared with ultrapure water, degassed with nitrogen gas, and autoclaved in serum vials sealed with butyl rubber stoppers and aluminum caps. Anoxic sterile water was prepared by boiling ultrapure water and then cooling it down under a nitrogen gas flow to allow degassing of oxygen, and it was then made sterile by autoclaving. An iron (II) chloride solution was prepared at a concentration of 1 M by using iron chloride powder (FeCl₂•4 H₂O, >99%; Sigma-Aldrich) that was stored in the anaerobic chamber. It was briefly taken out of the anaerobic chamber to be weighted and was then dissolved in anoxic sterile water in the anaerobic chamber. The FeCl₂ solution was filtered (0.22 μm) in the anaerobic chamber into a sterile serum vial and sealed with a butyl rubber stopper and an aluminum cap. No sign of oxidation was observed in the FeCl₂ solution over the course of the project. A sodium sulfide solution was prepared by dissolving large crystals of sodium sulfide (Na₂S•9 H₂O, Spectrum), previously washed and weighted, in anoxic sterile water to reach a concentration of about 900 mM. The sodium sulfide solution was transferred to a serum vial in the anaerobic chamber and sealed with a butyl rubber stopper and an aluminum cap.

2.2. Iron sulfide mineralization experiments

All experiments were prepared in an anaerobic chamber (Coy Laboratory Products, Inc.). The sulfate-reducing bacterium D. hydrothermalis AM13^T was purchased from the German culture collection DSMZ (DSM 14728). Strain AM13 grows optimally at 35°C and is an incomplete lactate oxidizer (Alazard et al., 2003). The strain was inoculated in the complete medium at a concentration of ∼10⁶ cells/mL and incubated for at least 5 days at 35°C until the cells reached stationary phase (∼10⁸ cells/mL). Two hundred microliters of the 1 M FeCl₂ solution was added to the fully grown culture (50 mL) to precipitate iron sulfide minerals with biotic H₂S in the presence of cells. The final concentration of Fe(II) was about 4 mM, and the Fe(II):sulfide ratio was ∼1:2. Cultures immediately turned black and opaque upon the addition of sodium sulfide, indicating precipitation of iron sulfide minerals. At these conditions, iron sulfide minerals precipitate away from cells, as cells do not serve as templates for the nucleation or growth of minerals, in contrast to when cells are grown in the medium containing Fe(II) (Picard et al., 2018). After addition of Fe(II), cultures were placed back in the incubator at 35°C and incubated for several months until analysis (Table 1). Carbon concentrations and initial C/Fe molar ratios were calculated by using the cell concentration reached at stationary phase in these experiments (∼10⁸ cells/mL) and a reported biomass yield of 10.3 g/mol sulfate for Desulfovibrio grown with lactate (Voordouw, 2002) (Table 1).

Table 1.

Summary of Iron Sulfide Mineralization Experiments Discussed in This Article

Experimental conditions	Organic carbon concentration (mmol C/L)	C/Fe molar ratios	Sampling time XRD	Sampling time SEM	Sampling time STXM	Sampling time NEXAFS
Abiotic conditions
Abiotic H₂S+Fe(II)	NA	NA	5 Months^a	1 Week^a	NA	Fe (19 months),^b S (19 months)
Prebiotic conditions (abiotic conditions+organic molecules)
Abiotic H₂S+glucose+Fe(II)	60.0	15.0	6 Months	ND	Fe/C (1 week)	Fe/C/O (1 week), S (16 months)
Abiotic H₂S+mannose+Fe(II)	60.0	15.0	6 Months	ND	Fe/C (1 week, 16 months)	Fe/C/O (1 week)
Abiotic H₂S+tryptone+Fe(II)	36.7	9.2	6 Months	ND	Fe/C (1 week, 16 months)	Fe (4 months), C/O (1 week, 4 months), S (16 months)
Abiotic H₂S+yeast extract+Fe(II)	32.5	8.1	3, 6 Months	3 Months	Fe/C (1 week, 4 months)	Fe/C/O (1 week, 4 months)
Biotic conditions
Biotic H₂S+cells+Fe(II) added after full growth of cells	∼4.2	1.1	5 Months^a	11 Months	Fe/C/O (9 months)	Fe/C/O 9 months, S 24 months
Biotic H₂S produced by live cells grown with Fe(II)	∼4.2	1.1	1–11 Months^a	1 Week^a	Fe/C (3, 9, 24 months)^b,c	3–24 months^b
Dead cells+Fe(II)+abiotic H₂S	∼4.2	1.1	1–5 Months^a	ND	Fe/C (3.5, 5, 16 months)^b,c	3–16 months^b

The organic carbon concentration in experiments is calculated from the amount of organic molecules added to prebiotic experiments or estimated from the cell concentrations in biotic experiments. The Fe(II) concentration is ∼4 mM in all experiments. XRD, SEM, STXM, and NEXAFS spectroscopy were used to characterize mineral aggregates.

Picard et al. (2018).

Picard et al. (2019).

Data analysis made for this study.

NEXAFS, near-edge X-ray absorption fine structure; SEM, scanning electron microscopy; STXM, scanning transmission X-ray microscopy; XRD, X-ray diffraction.

Iron sulfide minerals were also precipitated in the complete medium in the presence of organic molecules to simulate prebiotic conditions. First, 1 mL of a concentrated solution (glucose, mannose, tryptone, yeast extract) was added to 50 mL of complete medium (see Table 1 for C concentrations and initial C/Fe molar ratios). Then, 200 μL of the 1 M FeCl₂ solution was added to the medium to reach a final concentration of Fe(II) of about 4 mM. Finally, sodium sulfide was added to the bottles to reach an Fe(II):sulfide ratio of ∼1:2, as in the biotic experiments. The precipitation of iron sulfide minerals occurred on addition of sodium sulfide. Experiments were incubated at 35°C for several months.

Solid phases were recovered by centrifugation, washed with sterile anoxic water, left to dry on sample holders in the anaerobic vinyl chamber (Coy Laboratory Products, Inc.), and transported to analytical instruments in air-tight containers to prevent contact with atmospheric oxygen. A summary of experiments and analytical measurements is given in Table 1.

2.3. X-ray diffraction

Solid phases were analyzed by powder XRD to determine their mineralogy. Solid phases were recovered from whole experiments (50 mL) by centrifugation at ≈3000 g in 50-mL centrifuge tubes, rinsed with anoxic water, and dried in small tubes that were left open in the anaerobic chamber for about 48 h. The samples were transported in an air-tight jar to the instrument at the X-ray laboratory of the Department of Chemistry and Chemical Biology at Harvard University. Immediately before measurements, samples were homogenized with 200-proof ethanol in an agate mortar and mounted on a zero-diffraction silicon plate (MTI Corporation). Powder XRD data were acquired by using Cu-Kα radiation (30 mV, 10 mA) from a Bruker^™ D2 Phaser. X-ray diffractograms were acquired at 2θ angles from 10° to 60° with increment steps of 0.05° and a scan speed of 2 s. A background subtraction was performed on the XRD data by using the DIFFRAC algorithm. Mineral phase identification was performed by comparing peak positions and relative intensities against powder diffraction databases with the DIFFRAC.EVA module of the DIFFRAC.SUITE by Bruker. The reference pattern for mackinawite displayed in this study is PDF 86-0389 (Lennie et al., 1995). Diffraction peaks in XRD patterns were fitted with Lorentzian functions to determine their position and their full width at half maximum (FWHM). The average size of crystalline domains was estimated by using the Scherrer equation as described in the work of Wolthers et al. (2003): $L = K λ ∕ (β c o s θ)$

where L is the average size of the crystalline domain, K is the Scherrer constant (0.9), λ is the wavelength of the applied X-rays (0.15418 nm for Cu Kα), β is the FWHM of the peak, and θ is the Theta-angle of the position of the peak. The Scherrer equation was applied to the (001) and (200) reflections of mackinawite to determine the length of the crystalline domain along the c-axis and along the a/b-axes, respectively.

2.4. Scanning electron microscopy

SEM was used to image the texture of organo-mineral aggregates. Solid phases were recovered from 1 mL of suspension by centrifugation in small centrifuge tubes (14,000 rpm, 15 min), rinsed, and resuspended in anoxic water. A drop of suspension was then deposited and dried on a small chip broken off an Si wafer in the anaerobic chamber. The Si chip was fixed onto an aluminum stub with double-sided carbon tape. Samples were imaged with a Supra 55VP field emission SEM (Zeiss) operating at the Harvard Center for Nanoscale Systems (CNS). Secondary electron images were obtained at a voltage of 10 kV and a working distance of 3–4 mm with an Everhart-Thornley secondary electron detector or an InLens detector.

2.5. STXM and NEXAFS spectroscopy

STXM and NEXAFS spectroscopy were carried out at the 10ID-1 beamline of the Canadian Light Source (Saskatoon, Canada) and at the 11.0.2 beamline of the Advanced Light Source (Berkeley, CA). Technical details about the beamlines can be found elsewhere (Bluhm et al., 2006; Kaznatcheev et al., 2007). Mineral preparation for STXM was done in the anaerobic chamber. Minerals were recovered by centrifugation (1-mL samples, ≈11,000 g, 20 min), washed, and resuspended in anoxic sterile water. For each sample, a small drop of mineral suspension was deposited on a Cu 200-mesh formvar-coated transmission electron microscopy grid (Electron Microscopy Sciences) that had been previously attached with double-sided tape to a sample holder adapted to the beamlines' microscopes. Samples were left to dry in the anaerobic chamber. Sample holders were transported to the beamlines in air-tight jars, limiting the contact of minerals with air to less than 30 s during the transfer to the microscope chamber. The air in the microscope chamber was evacuated and replaced by an He atmosphere for measurements. This experimental procedure has been previously established to prevent oxidation of iron sulfide minerals (Picard et al., 2019). Image sequences (stacks) were acquired on mineral aggregates across the Fe L_3,2-edges (700–740 eV) and the S L-edge (155–185 eV) to characterize the Fe and S chemistry and distribution, respectively; across the C K-edge (280–320 eV) and the O K-edge (525–560 eV) to characterize organic molecules and functional groups present in the organic carbon associated with iron sulfide minerals. Measurements at the O K-edge also served to verify that anoxic conditions were maintained during the preparation and the transport of samples to the synchrotrons, as spectral features of iron oxides at the O K-edge are very specific (Chan et al., 2011; Picard et al., 2016b). A spectral resolution of 0.1 eV was used for regions of interest in the absorption spectra. The C K-edge NEXAFS spectra of standard organic compounds (saturated lipid, albumin, alginate, and glucose) presented in Fig. 3 are from other studies (Lawrence et al., 2003; Dynes et al., 2006; Hitchcock et al., 2009).

2.6. STXM and NEXAFS data processing

Data were processed with the aXis2000 software package and standard procedures (Hitchcock, 2019). Images were converted from transmission scale to optical density (OD) scale by using established procedures (Jacobsen et al., 2000). The OD maps of elements (Fe, C, and O) were generated by subtracting pre-edge images from images that showed the highest absorption (708.7, 288.2–288.6, and 538 eV), respectively. Maximum OD values and average OD values (after removing 0 values) were obtained from the OD images with aXis2000. Average NEXAFS spectra were extracted from whole sequence images or from selected structures (hot spots or extracellular material). Spectra were normalized to unity with the software Athena (Ravel and Newville, 2005). Thicknesses of Fe in FeS and C in organic material were estimated in mineral aggregates by using average OD values extracted from Fe(II) and organics maps and the following scaling factors (SF): 45.6 nm/OD unit for FeS, 324.3 nm/OD unit for glucose and mannose, 327.5 nm/OD unit for tryptone, 453.5 nm/OD unit for yeast extract, and 368.5 nm/OD unit for microbial biomass. The SF were calculated as follows based on published recommendations (Wan et al., 2007): $S F = R E \times μ \times f \times ρ$

where RE is the resonance enhancement factor (unitless), μ is the difference between the mass absorption coefficient above and below the absorption edge of the element considered (cm²/g), f is the mass fraction of the element in the solid phase, and ρ is the density of the material (g/cm³). The RE factors are 5 and 1 for Fe and C, respectively. The mass absorption coefficients (μ) are 15,800 and 51,400 cm²/g for Fe and C, respectively, at their respective absorption edge energy. The mass fractions (f) of the elements within the solid phases were the following: 0.635 (in g of Fe per g of FeS) and 0.4 for glucose/mannose, 0.44 for tryptone, 0.39 for yeast extract, 0.48 for biomass (in g of C per g of organic molecules/biomass). The densities (ρ) used for FeS, glucose/mannose, and bacterial biomass were 4.37, 1.5, and 1.1 g/cm³, respectively_. As tryptone is an enzymatic digest of proteins, the density used for our calculations is based on the average density of proteins of 1.35 g/cm³ (Fischer et al., 2004). For yeast extract, which is a water-soluble extract of autolyzed yeast cells, we use the density of bacterial biomass of 1.1 g/cm³ _. The C/Fe thickness ratios were calculated from the estimated thicknesses.

2.7. Image analysis

The correlation between Fe and C, and between C and O, in mineral aggregates was investigated by using the ScatterJ plugin (Zeitvogel et al., 2016), available in the free software ImageJ (Rueden et al., 2017) (https://imagej.net/ImageJ). The analysis was performed as in our previous study to enable comparisons of the results (Picard et al., 2019). Input images were Fe(II) (equivalent to FeS maps) and amide C/carboxylic C (organic C) OD maps of the same area that had been acquired with the same size and pixel resolution and aligned to one another. Masks were created to remove pixels in the background of both images. For pixels at the same position in both Fe and C maps, datapoints were extracted and plotted in a scatterplot matrix to visualize the relationship between Fe and C (x = Fe, y = C). A major axis linear regression analysis was performed on the whole dataset by using the major axis regression model in the ScatterJ plugin (Zeitvogel et al., 2016). The plugin calculates the Pearson's correlation factor, the slope, the y-intercept, and the root-mean square error. A deviation map of perpendicular distances from the major axis regression line was generated to visualize the areas in iron-sulfide mineral aggregates that were enriched or depleted in organic carbon. Regions that appear red in the deviation map indicate hot spots of organic carbon, whereas dark regions indicate the absence of OC. Regions of similar colors indicate similar deviations of the pixels from the regression line, suggesting compositional variations within the samples (Zeitvogel et al., 2016).

3. Results

3.1. Characterization of the mineral phase of iron sulfide mineral aggregates precipitated in prebiotic and biotic conditions

The XRD patterns of minerals precipitated in prebiotic conditions, that is, abiotic conditions in the presence of simple organic molecules (glucose and mannose) and complex organic mixtures (tryptone and yeast extract), displayed intense and broad Bragg peaks at angles of 17.6°, 30.1°, 39.0°, 49.6°, and 50.5°, and smaller Bragg peaks at angles of 34.5° and 53.0° (Fig. 1A). These peaks were assigned to mackinawite, which was the only crystalline iron sulfide mineral detected over the course of these experiments. Traces of siderite were detected in some experimental conditions, as indicated by the small Bragg peak appearing at an angle of 32.0° (Fig. 1A). The XRD patterns collected in this study were comparable to those of minerals precipitated in abiotic conditions in the mineral medium without organic molecules and minerals precipitated with biotic sulfide away from cells (redrawn in Fig. 1A for comparison) (Picard et al., 2018).

FIG. 1.

(A) XRD patterns of iron sulfide minerals precipitated in the presence of organic molecules (i.e., prebiotic conditions): tryptone (T), yeast extract (YE), mannose (Man), glucose (Glu) and incubated for up to 6 months at 35°C. XRD patterns labeled with * are of abiotic minerals precipitated in the culture medium and of minerals precipitated by adding Fe(II) to a fully grown culture of Desulfovibrio hydrothermalis AM13 containing biotic sulfide, and are redrawn from Picard et al. (2018). PDF 86-0389 of mackinawite is presented in the bottom panel (Lennie et al., 1995). In all experiments, mackinawite (M) is the only crystalline iron sulfide mineral. Traces of siderite (S) were detected in some experimental conditions. (B) SEM images of abiotic iron sulfide minerals precipitated with yeast extract (left) and biotic sulfide minerals precipitated with biotic sulfide and microbial cells (right). In both conditions, minerals display a very fine texture. Scale bar = 10 μm. SEM, scanning electron microscopy; XRD, X-ray diffraction. Color images are available online.

The calculated mineral particle size of mackinawite precipitated with simple and complex organic molecules (Table 2) was in the range of abiotic mackinawite (Wolthers et al., 2003; Picard et al., 2018; Mansor et al., 2019) and of mackinawite precipitated with biotic sulfide away from cells, but much smaller than mackinawite precipitated at the surface of live and dead sulfate-reducing bacteria (Picard et al., 2018). Iron sulfide minerals precipitated with yeast extract and those precipitated in the presence of biotic sulfide and cells displayed a fine texture (Fig. 1B). A fine texture observed in SEM indicates limited aggregation and is consistent with a small particle size (Picard et al., 2018).

Table 2.

Particle Size of Iron Sulfide Minerals Precipitated in the Mineral Medium in the Presence of Simple and Complex Organic Molecules (This Study), Compared with Abiotic Minerals and Biotic Minerals from Other Studies (Wolthers et al., 2003; Picard et al., 2018; Mansor et al., 2019)

Experimental conditions	Mineral particle size, nm (a/b axis, c axis)	References
Abiotic conditions
Abiotic H₂S+Fe(II)	Mackinawite: 9.7 ± 2.9 nm, 7.1 ± 2.1 nm; Mackinawite: ND, 4–6 nm; Mackinawite: 2.2–7.4 nm, 1.7–2.9 nm	Picard et al. (2018); Mansor et al. (2019); Wolthers et al. (2003)
Prebiotic conditions
Abiotic H₂S+simple molecules (glucose/mannose)+Fe(II)	Mackinawite: 9.0 ± 0.3 nm, 3.4 ± 0.1 nm	This study
Abiotic H₂S+complex molecules (tryptone/yeast extract)+Fe(II)	Mackinawite: 7.6 ± 0.6 nm, 8.5 ± 3.8 nm	This study
Biotic conditions
Biotic H₂S+cells+Fe(II) added after full growth of cells	Mackinawite: 6.4 ± 2.1 nm, 3.8 ± 0.9 nm	Picard et al. (2018)
Biotic H₂S produced by live cells grown with Fe(II)	Mackinawite: 20.1 ± 5.0 nm, 13.8 ± 4.8 nm; Greigite: 19.2 ± 2.6 nm, NA Mackinawite: ND, 7–9 nm	Picard et al. (2018); Mansor et al. (2019)
Dead cells+Fe(II)+abiotic H₂S	Mackinawite: 21.5 ± 3.2 nm, 18.5 ± 7.7 nm	Picard et al. (2018)

In all studies, the mineral particle size was calculated by using the Scherrer equation.

NEXAFS spectroscopy was used at the Fe L_3,2-edges and at the S L-edge to gain insights into the chemistry of Fe and S in iron sulfide mineral aggregates, respectively. Average NEXAFS spectra at the Fe L_3,2-edges collected from mineral aggregates were all similar to the spectrum of abiotic mackinawite precipitated in the culture medium (Fig. 2A). These spectral features are characteristic of the tetragonal coordination of Fe(II) in FeS (Todd et al., 2003). NEXAFS spectra at the S L-edge were acquired for a subset of conditions and all displayed spectral characteristics of abiotic mackinawite (Fig. 2B). However, the limited intensity of the pre-edge spectral features of FeS prevents further interpretation of the spectra (Farrell et al., 2002). Overall, the Fe and S chemistry was homogeneous in mineral aggregates precipitated in prebiotic and biotic conditions.

FIG. 2.

Fe L_3,2-edges and S L-edge NEXAFS spectroscopy of Fe and S in iron sulfide minerals precipitated in abiotic conditions, iron sulfide minerals precipitated in prebiotic conditions in the presence of tryptone (T), yeast extract (YE), mannose (Man), and glucose (Glu), and iron sulfide minerals precipitated in the presence of biotic H₂S and cells. (A) Spectra acquired at the Fe L_3,2-edges are indicative of Fe(II) in tetrahedral coordination (Todd et al., 2003). The spectrum of abiotic minerals labeled with # is redrawn from Picard et al. (2019). (B) NEXAFS spectra recorded at the sulfur L-edge. Dashed lines indicate the two main pre-edge features associated with reduced sulfur in iron sulfide minerals (Farrell et al., 2002). NEXAFS, near-edge X-ray absorption fine structure. Color images are available online.

3.2. Characterization of the organic phase of iron sulfide mineral aggregates precipitated in prebiotic and biotic conditions

NEXAFS spectroscopy was used at the C K-edge to characterize the composition of organic carbon associated with minerals (Fig. 3). The spectra of monosaccharides typically exhibit a peak between 289.1 and 289.6 eV (289.3 eV for the glucose spectrum in Fig. 3A), which is attributed to the C1s-3p/σ* transition of C-O in O-alkyl groups (C-OH) (Solomon et al., 2009). The C K-edge spectroscopic signature of minerals precipitated with glucose and mannose was similar to one another but did not match the spectrum of glucose (Fig. 3A). A main broad and undefined feature appeared between 288.3 and 288.7 eV. These values are in the range of the C1s-3p/π* transition of C = O in -COOH groups (288.4–288.7 eV). If the presence of carboxylic functional groups can be confirmed, these spectral modifications indicate a potential transformation of glucose/mannose into sugar acids during the formation of iron sulfide minerals. Polymerization of these sugar acids into acidic polysaccharides is not excluded; the C K-edge spectrum of acidic polysaccharides (see alginate spectrum in Fig. 3B) also presents the main feature attributed to carboxylic groups at 288.7 eV (Chan et al., 2004). In addition, a new feature appeared at ∼285.2 eV in the spectra of minerals precipitated with glucose and mannose (Fig. 3A). This indicates the presence of stable aromatic ring structures. This supports the hypothesis that iron sulfide minerals promote transformations and changes in the structure of monosaccharides. The spectra of tryptone and yeast extract were similar to one another and displayed a peak at 285.2 eV, which corresponds to the C1s-3p/π* transition of C = C in aromatic ring structures and a peak at 288.7 eV which corresponds to the C1s-3p/π* transition of C = O in carboxyl-carbonyl (COOH-COO⁻) structures in amino acids and small peptides (Fig. 3A) (Stewart-Ornstein et al., 2007; Solomon et al., 2009). Although the exact composition of tryptone and yeast extract is unknown, their C K-edge spectra are consistent with these organic mixtures being produced by enzymatic hydrolysis of large proteins (tryptone) or autolytic digestion of yeast cells (yeast extract), respectively. The C K-edge spectra of minerals precipitated with tryptone and yeast extract were similar to one another and retained a shape similar to these of the original molecules. However, the main peak of the spectra was slightly shifted toward lower energies (288.2–288.3 eV) in comparison to the tryptone and yeast extract spectra (Fig. 3A). The spectral feature at 288.2 eV is from the C1s-3p/π* transition of C = O in the amide bond, as seen in the spectrum of albumin (Fig. 3B). Such an energy shift is commonly attributed to the formation of amide bonds (Stewart-Ornstein et al., 2007). This observation indicates that small peptides and/or amino acids might have undergone polymerization into larger peptides/proteins during the precipitation of iron sulfide minerals. The spectral feature at 285.2 eV is still present in the spectra of minerals and indicates the presence of aromatic amino acids. In addition, the C K-edge spectra of minerals precipitated with tryptone and yeast extract were very similar to the spectra of minerals precipitated with biotic H₂S and cells, described hereafter.

FIG. 3.

C K-edge NEXAFS spectroscopy analysis of organic carbon in iron sulfide minerals. The main features of standard spectra and their energies (in eV) are represented by dashed lines. The explanation of these features is in the text. (A) Average C K-edge NEXAFS spectra of iron sulfide mineral aggregates precipitated in prebiotic conditions in the presence of tryptone (T), yeast extract (YE), mannose (Man), and glucose (Glu), and C K-edge spectra of organic molecules used in these experiments (tryptone, yeast extract, and glucose). (B) C K-edge NEXAFS spectra of mineral aggregates precipitated in the presence of biotic H₂S away from cells, and C K-edge spectra of organic molecules representative of major microbial components (albumin = protein, alginate = polysaccharides, saturated lipid = cell membrane). Color images are available online.

Average C K-edge spectra acquired from three mineral aggregates precipitated with biotic sulfide in the presence of cells (but away from them) are represented in Fig. 3B. Their features were most similar to the C K-edge spectroscopic signature of microbial cells reported in other studies (see, e.g., Miot et al., 2009a; Cosmidis et al., 2015), as well as of iron sulfide minerals precipitated near microbial cells (Picard et al., 2019). The spectra can be deconvoluted into a mixture of spectra of the three main biomolecules composing cells (proteins, polysaccharides, and lipids) (Picard et al., 2019). Indeed, the main features of the spectra of microbial cells are those of proteins (288.2 eV, C1s-3p/π* transition of C = O in the amide bond, and 285.2 eV, C1s-3p/π* transition of C = C in aromatic ring structures), polysaccharides (contribution in the range 288.4–288.7 eV, C1s-3p/π* transition of C = O in -COOH groups), and saturated lipids (shoulder at 287.3 eV, C1s-3p/σ* and C1s-3p/π* Rydberg-like excitations from C-H and in -CH and -CH₂ groups). Spectra of albumin, alginate, and saturated lipids are presented in Fig. 3B to illustrate the main spectral features described earlier.

NEXAFS spectroscopy at the O K-edge confirmed that the sample preparation was done in strict anoxic conditions. Features of Fe oxides and oxyhydroxides are clearly visible in O K-edge NEXAFS spectra (Chan et al., 2011; Picard et al., 2016b) and were not observed in our spectra. When used in samples precipitated in the absence of oxygen, spectroscopy at the O K-edge allows the probing of oxygen in functional groups of organic carbon (Picard et al., 2019). The main peak observed at ∼532.1 eV is indicative of the O1s → π* transition of C = O that occurs in functional groups of biomolecules (Fig. 4) (Chan et al., 2011; Picard et al., 2016b). The broader feature at 538–540 eV is assigned to the O1s → σ* transition of C-O in hydroxyl groups (Zubavichus et al., 2005). Although the broad feature is present in the O K-edge spectra of minerals precipitated with glucose and mannose and indicates the presence of some oxygen-containing functional groups, the first feature is absent and the spectra are very noisy. Overall, organic carbon (and therefore associated functional groups) might be present at low concentrations in these minerals, indicating a low affinity of Fe-S minerals with hexoses. These results correlate well with those of NEXAFS spectroscopy at the C K-edge.

FIG. 4.

O K-edge NEXAFS spectroscopy analysis of oxygen-containing functional groups in iron sulfide minerals precipitated in the presence of organic molecules (Glu, glucose; Man, mannose; T, tryptone; YE, yeast extract) and precipitated in the presence of biotic sulfide and cells. The dashed line represents the energy in eV of the main feature of the O K-edge spectra of organic molecules. Color images are available online.

3.3. Organic carbon distribution in prebiotic iron sulfide mineral aggregates

STXM revealed that organic carbon (C_org) was homogeneously distributed in prebiotic minerals precipitated with simple and complex organic molecules after 1 week of incubation and after up to several months of incubation (Fig. 5). Not all conditions could be measured at all times because of beamtime availability, but the sampling procedure gives an overview of the general evolution of minerals precipitated in prebiotic conditions. A positive linear correlation between Fe and C_org was observed in prebiotic mineral aggregates (Fig. 5C). The Pearson's correlation coefficient calculated for the prebiotic dataset indicated a decreasing correlation between Fe and C in the following order: Prebiotic tryptone > Prebiotic glucose/mannose >> Prebiotic yeast extract (Table 3). The Pearson's correlation coefficient also decreased as a function of incubation time (Table 3). In the Fe-C scatter plot for minerals precipitated with yeast extract (Fig. 5, 1 week incubation), we observed a deviation of the C_org content to lower levels at high Fe levels (measured in OD units), which could be explained by the thickness of mineral aggregates. When the Fe content is high, mineral aggregates are sometimes too thick for C measurements to be performed reliably. Therefore, in this case, the C_org content in areas of high-Fe contents could have been underestimated.

FIG. 5.

STXM high-resolution imaging of Fe and C in iron sulfide minerals precipitated with glucose (Glu), mannose (Man), tryptone (T), and yeast extract (YE) and incubated at 35°C for up to 16 months. (A) Transmission images at 708.7 eV, which is the energy at which Fe absorbs the most in samples. (B) Species-specific maps (OD maps) of Fe(II) and C_org. As Fe is only present as Fe(II) in mackinawite, Fe maps represent iron sulfide mineral aggregates. C_org maps represent the maps of carboxylic C in minerals precipitated with glucose and mannose, and amide C in minerals precipitated with tryptone and yeast extract. Legends below the maps indicate the maximum OD values. (C) Scatter plots were generated with FeS and C_org maps to evaluate the correlation between both elements. Major axis regression lines (black dashed lines) are represented on the scatter plots. (D) Deviation maps show the perpendicular distances of data points from the major axis regression lines in the scatter plots. The scale on the right indicates the deviation in OD values. A red color in the deviation map indicates an enrichment in organic carbon, whereas a black color indicates the absence of organic carbon. C_org, organic carbon; OD, optical density; STXM, scanning transmission X-ray microscopy. Color images are available online.

Table 3.

Analysis of Scanning Transmission X-Ray Microscopy Datasets from This Study and from Picard et al. (2019)^#

Experimental conditions and sampling time	OD maximum (Fe maps)	OD maximum (C maps)	OD average (Fe maps)	Fe thickness average (nm)	OD average (C maps)	C thickness average (nm)	C/Fe ratios (thickness)	Pearson's correlation coefficient
Prebiotic conditions
Abiotic H₂S+glucose+Fe(II)
1 Week	2.10	0.59	0.47 ± 0.28	21.5	0.15 ± 0.05	48.7	2.3	0.75868
Abiotic H₂S+mannose+Fe(II)
1 Week	1.30	0.53	0.31 ± 0.20	14.0	0.15 ± 0.04	49.6	3.5	0.7701
16 Months	2.57	0.37	0.73 ± 0.32	33.4	0.13 ± 0.05	41.2	1.2	0.5443
Abiotic H₂S+tryptone+Fe(II)
1 Week	0.65	0.57	0.17 ± 0.10	7.9	0.14 ± 0.08	45.9	5.8	0.89222
1 Week	1.23	0.93	0.30 ± 0.22	13.6	0.37 ± 0.14	119.9	8.8	0.9277
16 Months	1.64	0.49	0.77 ± 0.31	35.2	0.12 ± 0.05	40.3	1.1	0.75737
Abiotic H₂S+yeast extract+Fe(II)
1 Week	2.38	1.77	0.96 ± 0.55	43.9	0.96 ± 0.57	434.9	9.9	0.68489
4 Months	2.02	10.40	0.55 ± 0.39	25.0	0.15 ± 0.05	65.8	2.6	0.48963
Biotic conditions
Biotic H₂S+cells+Fe(II) added after full growth of cells
9 Months	1.94	1.59	0.48 ± 0.31	21.9	0.30 ± 0.23	109.4	5.0	0.44327
	1.71	1.82	0.50 ± 0.29	22.8	0.38 ± 0.26	140.0	6.1	0.25403
	1.54	1.64	0.50 ± 0.30	22.7	0.38 ± 0.27	138.6	6.1	0.47415
Biotic H₂S produced by live cells grown with Fe(II)^#
3 Months	2.28	1.88	0.88 ± 0.64	40.0	0.53 ± 0.33	194.9	4.9	0.17165
	2.10	2.07	1.01 ± 0.64	46.1	0.88 ± 0.49	324.3	7.0	−0.44025
	2.59	1.67	0.73 ± 0.65	33.1	0.43 ± 0.30	157.0	4.7	0.25602
9 Months	2.05	1.41	0.67 ± 0.45	30.0	0.41 ± 0.26	139.7	4.7	−0.07891
24 Months	3.71	1.81	1.47 ± 0.89	67.0	0.44 ± 0.36	162.1	2.4	−0.07356
Dead cells+Fe(II)+abiotic H₂S^#
3.5 Months (autoclaved)	0.92	0.56	0.24 ± 0.14	10.9	0.15 ± 0.04	55.6	5.1	0.73175
5 Months (γ-irradiated)	2.21	1.35	0.76 ± 0.52	34.6	0.26 ± 0.24	97.3	2.8	0.13766
5 Months (γ-irradiated)	2.26	1.26	0.65 ± 0.40	29.5	0.49 ± 0.27	179.5	6.1	−0.16069
16 Months (γ-irradiated)	2.31	1.41	0.74 ± 0.51	33.6	0.29 ± 0.24	105.4	3.1	0.30512

Maximum and average values of OD were collected from Fe and C maps acquired of the same mineral aggregates. Fe and C thicknesses are estimated from average OD values (see Section 2).

# indicates that the dataset is from Picard et al. 2019.

OD, optical density.

3.4. Filamentous structures in prebiotic iron sulfide minerals

Filamentous structures were observed at a high resolution in STXM images of minerals precipitated with organic molecules (Fig. 6). Most observations occurred in minerals precipitated in the presence of glucose and yeast extract, although some hook-like structures were also observed in minerals precipitated in the presence of mannose (See Fe map of 16-month mineral aggregate in Fig. 5). The structures contained organic material (as evidenced by maps of C_org and/or O) (Fig. 6). Those structures were not observed in SEM images, probably because mineral aggregates were more diluted and dispersed in samples prepared for STXM than for SEM.

FIG. 6.

STXM transmission and OD maps of minerals precipitated in the presence of glucose and yeast extract (after 1 week of incubation at 35°C). Ambiguous filamentous structures are indicated by arrows. The first column represents transmission images at 708.7 eV; the other images are either Fe, C or O maps, as indicated. Color images are available online.

3.5. Organic carbon distribution in biotic iron sulfide minerals

In comparison to prebiotic minerals, mineral aggregates precipitated with biotic sulfide in the presence of cells displayed a more heterogeneous C_org distribution (Fig. 7A). C_org was distributed throughout the minerals, which is shown by a clear relationship between Fe and C levels in the scatterplots. However, features with high organic carbon content with the shape and the size of microbial cells could be observed in mineral aggregates (pointed by arrows in C_org maps in Fig. 7A). This is indicated in the scatterplot by a cloud of data points with high C and low Fe deviating from the major axis regression line (Fig. 7A). The distribution of O from oxygen-containing functional groups was closely linked to that of C_org (Fig. 7B), as indicated by a Pearson's correlation coefficient of 0.95464. C K-edge and O K-edge spectra were extracted from the hot spots in the C and O maps and compared with those of extracellular regions in the maps (Fig. 7C). The C K-edge and O K-edge spectra of hot spots displayed an increase in intensity of the main peak at 288.2 and 532.1 eV, respectively, in comparison to the spectra of extracellular material.

FIG. 7.

STXM analysis of iron sulfide minerals precipitated with biotic H₂S away from cells and incubated for 9 months at 35°C. (A) Iron (Fe) and C_org distribution analysis in two mineral aggregates. Transmission images were acquired at 708.7 eV, which is the energy at which Fe absorbs the most in samples. Fe(II) (mackinawite) and C_org (amide C) maps are used to produce a scatter plot to identify the correlation between both elements. Yellow arrows in C maps point to hot spots of carbon that represent cells. Major axis regression lines (black dashed lines) are represented on the scatter plots. For each dataset, a deviation map was generated to display perpendicular distances of data points from the major axis regression lines in the scatter plots. The scale on the right indicates the deviation in OD values. A red color in the deviation map indicates an enrichment in organic carbon, whereas a black color indicates the absence of organic carbon. (B) C_org and oxygen (O) distribution analysis in the second mineral aggregate. Amid C maps and O maps were used to generate a scatter plot to identify the correlation between both elements. The major axis regression line (black dashed line) is represented on the scatter plot. (C) C K-edge and O K-edge NEXAFS spectra of the hot spot structures (cells in C and O maps in C) and of the extracellular material. The peaks at 288.2 and 532.1 eV are more intense in the hot spots. Color images are available online.

3.6. Affinity of iron sulfide minerals for organic molecules

The evolution of the affinity of iron sulfide minerals for C_org was evaluated as a function of incubation time (Fig. 8). The C/Fe ratios in mineral aggregates were calculated by using thicknesses of Fe and C in solid phases estimated from STXM data (Table 3). We also estimated thickness ratios in iron sulfide minerals precipitated in other biotic conditions by using the STXM dataset from the work of Picard et al. (2019). All experiments had an excess of C to Fe in the solution, as indicated by C/Fe molar ratios >1; however, prebiotic experiments had C/Fe molar ratios roughly eight to fifteen times higher than biotic experiments (Table 1). Overall, the C/Fe thickness ratios in mineral aggregates decreased with increasing incubation times in all experimental conditions (Fig. 8). The sharpest decrease in C/Fe thickness ratios was observed in prebiotic minerals precipitated with yeast extract and tryptone. Although the exponential decay curve predicts the C/Fe thickness ratio in minerals precipitated with yeast extract to reach zero, it is more likely that the C/Fe thickness ratios in those minerals will reach levels similar to this measured in minerals precipitated with tryptone. The C/Fe thickness ratios decreased at a similar rate in minerals precipitated with mannose and in minerals from biotic experiments. After 3 months of incubation, C/Fe thickness ratios were always higher in all biotic experiments than in prebiotic experiments. Although we do not have decrease rates for C/Fe thickness ratios of minerals precipitated with glucose and minerals precipitated away from cells [biotic experiments to which Fe(II) was added], the data available fall in the range of minerals with mannose, and minerals from other biotic experiments, respectively. Overall, iron sulfide minerals appear to have a strong affinity with organic molecules from microbial biomass, and interactions between Fe-S minerals and these microbial organic molecules seem more stable over time than interactions occurring in prebiotic conditions.

FIG. 8.

Average C/Fe thickness ratios in mineral aggregates plotted as a function of time. C/Fe thickness ratios were calculated by using average OD values extracted from OD maps (Table 3). When several datasets are available for one time point, the average C/Fe thickness ratio is represented. Exponential decay curves were used to fit data, except for minerals precipitated with glucose and minerals precipitated away from cells [Biotic H₂S+cells+Fe(II)], for which only one time point is available. C/Fe thickness ratios in mineral aggregates decrease as a function of time. C/Fe thickness ratios appear to be more stable in mineral aggregates precipitated in biotic conditions (with live or dead cells) than in mineral aggregates precipitated in prebiotic conditions. Color images are available online.

4. Discussion

The success of astrobiological research hinges strongly on our ability to look for both extinct and extant life. The upcoming NASA Mars2020 and ESA ExoMars missions are prioritizing life detection and include instruments that will characterize organic molecules in the subsurface of Mars and help determining whether these could constitute the building blocks of martian life or record past microbial activity (Vago et al., 2017; Williford et al., 2018). Minerals have the potential to preserve organic carbon for geologic timescales (Hemingway et al., 2019). A number of their properties—for example, mineralogy, morphology, particle size, aggregation pattern, organic carbon content and composition, and isotopic signatures—can be characterized at various spatial resolutions to determine whether living organisms influenced their formation (Banfield et al., 2001; Hays et al., 2017; Benzerara et al., 2019). With their biosignature suite, minerals therefore represent a promising target for the search of life on Mars. In this work, we used an experimental approach to constrain the formation pathways of organo-mineral interactions of interest for the search of life in the Mars subsurface.

If life has evolved on Mars, it would likely be microbial. In prokaryotes, there are only a few examples of well-controlled mineralization that produce intracellular minerals: for example, magnetite and greigite in magnetotactic bacteria (Lefevre and Bazylinski, 2013; Posfai et al., 2013), elemental sulfur globules in sulfide-oxidizing bacteria (Maki, 2013), and Ca-carbonate minerals in Cyanobacteria (Benzerara et al., 2014). As such, most prokaryotic influence on mineral formation occurs extracellularly as a result of their metabolic activity, as well as by providing cell surfaces and exopolymers as templates for mineralization (so-called biologically influenced mineralization) (Beveridge, 1989; Benzerara et al., 2019). Where oxygen prevails on Earth, iron-oxidizing microorganisms confer unique morphologies and composition to extracellular Fe(III)-oxyhydroxides, preserving unique signatures of microbial activity under low oxygen concentrations (Chan et al., 2011; Krepski et al., 2013). Early life on Earth evolved in the absence of oxygen; therefore, if life evolved on Mars, it would very likely be anaerobic (Westall et al., 2013). In this work, we focused on the potential for minerals produced in anoxic environments to preserve signatures of microbial life.

On Earth, iron sulfide minerals precipitate in anoxic sedimentary environments where iron is available and the organic carbon supply is high enough to fuel microbial sulfate reduction (Berner, 1970, 1984; Rickard et al., 2017). In these environments, SRM produce most of the sulfide that is required to precipitate iron sulfide minerals. Recent studies have demonstrated that the influence of SRM on the formation of metastable iron sulfide minerals (e.g., mackinawite FeS and greigite Fe₃S₄) goes beyond providing sulfide (Picard et al., 2016a, 2018, 2019; Stanley and Southam, 2018; Mansor et al., 2019). However, the biological influence is more complex than it seems. The availability of microbial cell surfaces for the nucleation and growth of iron sulfide minerals appears to be the main driver in determining particle size of iron sulfide minerals. In conditions that promote the formation of mackinawite (FeS) (summarized in Fig. 9), minerals precipitated in the presence of live and dead cells of the bacterium D. hydrothermalis have a larger particle size and aggregate more than minerals precipitated in abiotic conditions (Picard et al., 2018). In these experiments, cells are in contact with Fe(II) in the medium before the production of sulfide starts. Live and dead cells, therefore, serve as templates for the nucleation of mackinawite and promote its growth. In the process, live and dead cells also become encrusted in iron sulfide minerals (Picard et al., 2018). A larger mackinawite particle size was also observed in the presence of live cells of the bacterium Desulfovibrio vulgaris than in abiotic conditions; however, the magnitude of size difference was not as large as in our study (Mansor et al., 2019). On the contrary, when Fe(II) is added in experiments after sulfide has been produced by live cells, the increase in particle size and aggregation does not occur, and cells do not become encrusted. Iron sulfide mineral precipitation is very rapid when Fe(II) and sulfide come in contact with one another and there is no possibility for Fe(II) to bind to cells (Picard et al., 2018). In the current study, we showed that the particle size of mackinawite precipitated in the presence of simple and complex organic molecules is in the range of abiotic minerals and minerals precipitated away from cells (Table 2).

FIG. 9.

Schematic diagram of biosignatures that can be preserved in iron sulfide minerals. This is based on data collected for (A) minerals precipitated in the presence of microbial cells in conditions that promote cell encrustation or (B) in conditions that do not promote cell encrustation (Picard et al., 2018, 2019, this study), and for (C) minerals precipitated at simulated prebiotic conditions (i.e., in the presence of simple sugars: glucose or mannose, and in the presence of complex mixtures: tryptone or yeast extract). Color images are available online.

Mineral aggregation and texture reflect particle size in iron sulfide minerals. In SEM images, mackinawite precipitated in abiotic, prebiotic, and biotic (with minerals away from cells) appears fine-textured, whereas mackinawite/greigite precipitated near live cells appears bulky and aggregated (Fig. 1B) (Picard et al., 2018). The surface of microbial cells that become encrusted in iron sulfide minerals appears similar to that of non-encrusted cells when observed with SEM (Picard et al., 2018). Although crusts can be detected around cells when using STXM (Picard et al., 2019), the detailed study of cell encrustation in iron sulfide minerals requires delicate preparation and ultramicrotomy to perform transmission electron microscopy (Picard et al., 2018). Prebiotic conditions promote the formation of iron sulfide minerals with thin filamentous aggregated morphologies, which could potentially be mistaken for extracellular appendages that some microorganisms produce (Figs. 5 and 6). For example, Miot et al. (2009b) reported extracellular filaments of similar size in cultures of Fe(II)-oxidizing bacteria (Miot et al., 2009b). In our study, abiotic filamentous structures could be observed with STXM but not with SEM, which might be due to the way samples are prepared. In the STXM preparation, mineral suspensions are much more diluted, which can influence aggregation. One has to be cautious when interpreting the origin of organo-mineral structures based solely on the morphology. A variety of biomorphic and ambiguous organo-minerals structures have been shown to form as a result of abiotic sulfide oxidation in the presence of organic molecules (Cosmidis and Templeton, 2016; Cosmidis et al., 2019). Nanoparticle aggregation occurs in a variety of abiotic organic and inorganic systems (Lin et al., 1989). Oriented aggregation can lead to the formation of various morphologies, including the ones observed in this study (Penn and Soltis, 2014). In the biomedical field, the use of proteins or peptides is an established procedure to control the morphology of nanoparticle aggregates (Yang et al., 2017). The morphology of nanoparticle aggregates is controlled by the mechanisms of aggregation, which can be varied by changing pH or ionic strength in solution, or by drying nanoparticles (Gilbert et al., 2009). Indeed, the morphology of mackinawite nanoparticle aggregates can be controlled by changing temperature (Lai and Chen, 2009) or pH (Matamoros-Veloza et al., 2018b). In our experiments, we could observe abiotic filamentous morphologies in the presence of glucose and yeast extract, and abiotic hook-like structures in the presence of mannose. Experimental conditions were kept stable during incubations (pH and temperature); however, sample preparation included washing and drying minerals, which also influence aggregation. What exactly controls the morphology of the iron sulfide filaments in our study is unknown, and in situ studies would be required to explore the aggregation mechanisms of iron sulfide minerals at simulated prebiotic conditions.

Although scaffolding is important for the physical properties of minerals, microbial activity appears to drive mineral transformations. However, only live encrusted cells promote the transformation of mackinawite to greigite (Fig. 9) (Picard et al., 2018). A transformation of mackinawite to greigite happens after 5–6 months in cultures where live cells are grown with Fe(II) (Picard et al., 2018; Mansor et al., 2019). In contrast, in the presence of dead cells, in experiments where Fe(II) is added to fully grown cultures, in simulated prebiotic conditions, and in abiotic conditions, mackinawite remains stable (Picard et al., 2018).

Minerals precipitated with live and dead cells in the conditions that promote encrustation and mineral growth incorporate organic molecules derived from microbial biomass. Interactions between microbes and minerals were shown to last for at least 2 years (Picard et al., 2019). The current study follows up on these observations and reports the associations between organic molecules and minerals precipitated at simulated prebiotic conditions (Fig. 5), as well as between microbial biomass and minerals precipitated in conditions that do not promote encrustation and mineral growth (Fig. 7). In all these conditions, the chemistry and distribution of Fe and S appears homogeneous, based on NEXAFS spectroscopy at the Fe L-edge and S L-edge (Fig. 2) (Picard et al., 2019). On the contrary, the chemistry and distribution of C and O appear as important factors to help distinguishing between prebiotic and biotic conditions. While C was distributed homogeneously in prebiotic minerals (Fig. 5), hot spots of C appear in mineral aggregates that were precipitated with live and dead cells (whether conditions promoted encrustation or not) (Fig. 7) (Picard et al., 2019). Interestingly, differences were observed between autoclaved cells and gamma-irradiated cells (Picard et al., 2019). Although in both cases, cells were preserved (as indicated by the presence of hot spots), the correlation between Fe and C in minerals precipitated with autoclaved cells was much stronger than with gamma-irradiated cells and the Fe-C scatter plot of autoclaved cells resembled that of prebiotic minerals (Picard et al., 2019). This could be explained by the release of intracellular material and lysis of many cells during autoclaving. We assume that, in natural environments, dead cells would mostly remain intact, similarly to gamma-irradiated cells, and that the distribution of C in mineral aggregates containing dead cells would preserve cell shape and size. The distribution of O (from O-containing functional groups) shadowed the distribution of C (Fig. 7). In 1-week incubations in prebiotic conditions, iron sulfide minerals displayed more affinity to proteinaceous material (tryptone and yeast extract) than for simple sugars (as indicated by lower C/Fe thickness ratios in the latter case) (Table 3 and Fig. 8), although the concentration of simple sugars in solution was higher than this of complex molecules. This could be explained by the different functional groups available in organic molecules to bind to minerals. Chemical bonds between N-atoms in amine groups and Fe-atoms in mackinawite and pyrite seem to play an important role for the interactions between oligonucleotides and iron sulfides minerals (Plekan et al., 2007; Hatton and Rickard, 2008; Mateo-Martí et al., 2008), and between methylamine and iron sulfide minerals (Dzade et al., 2013). Similarly, in biogenic zinc sulfide aggregates, proteins have been found to be the binding component between nanoparticles (Moreau et al., 2007). After 16 months of incubation, C/Fe thickness ratios in prebiotic mineral aggregates precipitated with mannose and tryptone reached the same level. The C/Fe thickness ratios in mineral aggregates decreased as a function of time in all experimental conditions; however, they remained twice as high in biotic minerals than in prebiotic minerals after >1 year of incubation (Fig. 8 and Table 3), despite higher C/Fe molar ratios in prebiotic experiments than in biotic experiments. This indicates that biotic conditions produce more stable organo-mineral aggregates than prebiotic conditions. Also, C/Fe thickness ratios are in the same range in all experiments that contain cells (live or dead). It should be noted that C concentrations in our experiments were higher than natural dissolved organic carbon (DOC) concentrations in sediments, which can vary greatly (Burdige and Komada, 2015). However, DOC concentrations in anoxic marine sediments of organic-rich coastal areas, where iron sulfide minerals precipitate, can peak up to 10 mM (Alperin et al., 1994). Similarly, in organic-rich sediments of Guaymas Basin, dissolved organic concentration can reach up to 8.8 mM (Zhuang et al., 2019). The decrease in C/Fe thickness ratios in iron sulfide minerals over time points toward a loss of organic carbon originally contained in the mineral structure and/or bound to the surface of minerals. The release of organic carbon in the medium could happen due to weakening interactions between organic molecules and minerals, which would imply either changes in the structure of the minerals and/or changes in the physical properties of the mineral surface. Neither spectroscopy nor XRD data indicate obvious changes in the chemistry of prebiotic or biotic minerals as a function of time, which could explain the decrease in C/Fe thickness ratios of minerals. The crystallinity or particle size did not significantly change as a function of time in minerals (Picard et al., 2018). Analytical methods such as X-ray absorption spectroscopy at the Fe K-edge and S K-edge, as well as surface characterization methods such as X-ray photoelectron spectroscopy, could help determine the binding mechanisms between organic molecules and Fe-S minerals and explain the carbon loss over time.

Shifts in the spectroscopic signatures of organic carbon associated with prebiotic minerals were observed (Fig. 3), suggesting potential changes in the composition of organic molecules during iron sulfide mineral precipitation. We observed the apparent formation of sugar acids from hexoses in one case (simple prebiotic conditions), whereas a potential polymerization of peptides and/or amino acids produced large peptides/proteins in the other case (complex prebiotic conditions). It has been proposed that pyrite could have played a role in the emergence of life, with pyrite forming thin layers similar to membranes and playing a catalytic role in the formations of early organic molecules (Russell et al., 1993, 1994; Mielke et al., 2011; White et al., 2015). In the formation of mackinawite in abiotic systems, a highly reactive nanoparticulate solid-phase precursor has been identified (Matamoros-Veloza et al., 2018a). Based on these results, the catalytic power of metastable iron sulfide minerals should be further explored.

5. Conclusions

Microbial sulfate reduction is an ancient metabolism on Earth that has been proposed to be a candidate metabolism for early life on Mars. Given the proposed abundance of iron sulfides in the subsurface of Mars, our results suggest that, if microbial life evolved on Mars, iron sulfides could potentially retain a suite of biosignatures. Therefore, understanding the formation of iron sulfide minerals in laboratory experiments is essential to constrain potential biosignatures that could be expected in the Mars subsurface. Overall, the organo-iron-sulfide-mineral system is complex. However, we have identified parameters that would help identify whether microbial life was involved in the precipitation of iron sulfide minerals (Fig. 9). Although some parameters could be remotely assessed (mineralogy and particle size using XRD, bulk elemental composition, bulk isotopic composition), high-resolution analyses would be required to fully characterize the biosignature suite contained in iron sulfide minerals. Morphology and texture analysis using several types of microscopy would help distinguish cell-like elements from abiotic structures, and elemental analysis and speciation at high resolution would help identify C/O hotspots associated with preserved cells. Bulk and high-resolution isotopic analysis of S and C would complement these measurements, as both these elements would record the metabolic activity of SRM (Kaplan and Rittenberg, 1964; Canfield, 2001; Londry and Des Marais, 2003; Londry et al., 2004). Additional high-resolution analyses would include trace metal analysis that could potentially be sequestered by cells and/or by iron sulfide minerals (Large et al., 2014; Glass et al., 2018; Gregory et al., 2019). All these parameters would increase the chances of correctly identifying the precipitation pathways of iron sulfide minerals (Westall et al., 2015; Hays et al., 2017; Neveu et al., 2018). Proving that life existed on another planet is the biggest challenge in astrobiology; however, biominerals in the subsurface appear as a promising target for the search of life on Mars.

Footnotes

Acknowledgments

This work was performed in part at the CNS, a member of the National Nanotechnology Coordinated Infrustructure Network (NNCI). CNS is part of Harvard University. The authors thank Shao-Liang Zheng for support at the X-ray laboratory of the Department of Chemistry and Chemical Biology, Harvard University. This research used resources of the Advanced Light Source, which is a DOE Office of Science User Facility under contract no. DE-AC02-05CH11231. The STXM measurements were also performed at the Canadian Light Source, which is supported by the Canada Foundation for Innovation, Natural Sciences and Engineering Research Council of Canada, the University of Saskatchewan, the Government of Saskatchewan, Western Economic Diversification Canada, the National Research Council Canada, and the Canadian Institutes of Health Research. The authors thank the staff of beamlines and analytical facilities for their support. A.P. would like to thank Caroline Peacock, Julie Cosmidis and Besty Swanner for inspiring discussions.

Author Disclosure Statement

No competing financial interests exist.

Funding Information

This work was supported by the National Science Foundation under Grant Numbers NSF-1344241 and NSF-1542506, as well as by NASA under grant numbers 80NSSC19K1427 and 80NSSC18K1140 awarded to P.R.G. This work benefited from the support of a Deep Energy award to A.P. from the Deep Carbon Observatory.

Supplementary Material

Supplementary Table S1

Abbreviations Used

References

Alazard

, Dukan

, Urios

, et al. (2003) Desulfovibrio hydrothermalis sp. nov., a novel sulfate-reducing bacterium isolated from hydrothermal vents. Int J Syst Evol Micr, 53:173–178.

Alperin

, Albert

, and Martens

(1994) Seasonal variations in production and consumption rates of dissolved organic carbon in an organic-rich coastal sediment. Geochim Cosmochim Acta, 58:4909–4930.

Banfield

, Moreau

, Chan

, et al. (2001) Mineralogical biosignatures and the search for life on Mars. Astrobiology, 1:447–465.

Benzerara

, Skouri-Panet

, Li

, et al. (2014) Intracellular Ca-carbonate biomineralization is widespread in cyanobacteria. Proc Natl Acad Sci U S A, 111:10933–10938.

Benzerara

, Bernard

, and Miot

(2019) Mineralogical identification of traces of life. In Biosignatures for Astrobiology, edited by B Cavalazzi and F Westalls, Springer Nature, Switzerland, pp 123–144.

Berner

(1970) Sedimentary pyrite formation. Am J Sci, 268:1–23.

Berner

(1984) Sedimentary pyrite formation: an update. Geochim Cosmochim Acta, 48:605–615.

Beveridge

(1989) Role of cellular design in metal accumulation and mineralization. Annu Rev Microbiol, 43:147–171.

Bluhm

, Andersson

, Araki

, et al. (2006) Soft X-ray microscopy and spectroscopy at the molecular environmental science beamline at the Advanced Light Source. J Electron Spectrosc, 150:86–104.

10.

Brack

(2018) An origin of life on Mars? In From Habitability to Life on Mars, edited by NA Cabrol and EA Grin, Elsevier, pp 13–35.

11.

Burdige

and Komada

(2015) Sediment pore waters. In Biogeochemistry of Marine Dissolved Organic Matter, edited by DA Hansell and CA Carlson, Elsevier, pp 535–577.

12.

Burns

and Fisher

(1990a) Evolution of sulfide mineralization on Mars. J Geophys Res Solid Earth, 95:14169–14173.

13.

Burns

and Fisher

(1990b) Iron-sulfur mineralogy of Mars: magmatic evolution and chemical weathering products. J Geophys Res Solid Earth, 95:14415–14421.

14.

Canfield

(2001) Biogeochemistry of sulfur isotopes. Rev Mineral Geochem, 43:607–636.

15.

Chan

, De Stasio

, Welch

, et al. (2004) Microbial polysaccharides template assembly of nanocrystal fibers. Science, 303:1656–1658.

16.

Chan

, Fakra

, Emerson

, et al. (2011) Lithotrophic iron-oxidizing bacteria produce organic stalks to control mineral growth: implications for biosignature formation. ISME J, 5:717–727.

17.

Cockell

(2014) Trajectories of Martian habitability. Astrobiology, 14:182–203.

18.

Cosmidis

and Templeton

(2016) Self-assembly of biomorphic carbon/sulfur microstructures in sulfidic environments. Nat Commun, 7:12812.

19.

Cosmidis

, Benzerara

, Nassif

, et al. (2015) Characterization of Ca-phosphate biological materials by scanning transmission X-ray microscopy (STXM) at the Ca L_2,3-, P L_2,3- and C K-edges. Acta Biomater, 12:260–269.

20.

Cosmidis

, Nims

, Diercks

, et al. (2019) Formation and stabilization of elemental sulfur through organomineralization. Geochim Cosmochim Acta, 247:59–82.

21.

Dynes

, Lawrence

, Korber

, et al. (2006) Quantitative mapping of chlorhexidine in natural river biofilms. Sci Total Environ, 369:369–383.

22.

Dzade

, Roldan

, and de Leeuw

(2013) Adsorption of methylamine on mackinawite (FES) surfaces: a density functional theory study. J Chem Phys, 139:124708.

23.

Eigenbrode

, Summons

, Steele

et al. (2018) Organic matter preserved in 3-billion-year-old mudstones at Gale crater, Mars. Science, 360:1096–1101.

24.

Farrell

, Fleet

, Stekhin

, et al. (2002) Evolution of local electronic structure in alabandite and niningerite solid solutions [(Mn, Fe) S,(Mg, Mn) S,(Mg, Fe) S] using sulfur K-and L-edge XANES spectroscopy. Am Mineral, 87:1321–1332.

25.

Fike

, Bradley

, and Rose

(2015) Rethinking the ancient sulfur cycle. Annu Rev Earth Planet Sci, 43:593–622.

26.

Fischer

, Polikarpov

, and Craievich

(2004) Average protein density is a molecular-weight-dependent function. Protein Sci, 13:2825–2828.

27.

Frank

, Rogers

, et al. (2015) Key factors influencing rates of heterotrophic sulfate reduction in active seafloor hydrothermal massive sulfide deposits. Front Microbiol, 6:1449.

28.

Franz

, King

, and Gaillard

(2018) Sulfur on Mars from the atmosphere to the core. In Volatiles in the Martian Crust, edited by J Filiberto and SP Schwemzers, Elsevier, Saint Louis, MO, pp 119–183.

29.

Franz

, Mahaffy

, Webster

, et al. (2020) Indigenous and exogenous organics and surface–atmosphere cycling inferred from carbon and oxygen isotopes at Gale crater. Nat Astron, 4:526–532.

30.

Freissinet

, Glavin

, Mahaffy

, et al. (2015) Organic molecules in the sheepbed mudstone, gale crater, mars. J Geophys Res Planets, 120:495–514.

31.

Gilbert

, Ono

, Ching

, et al. (2009) The effects of nanoparticle aggregation processes on aggregate structure and metal uptake. J Colloid Interface Sci, 339:285–295.

32.

Glass

, Chen

, Dawson

, et al. (2018) Trace metal imaging of sulfate-reducing bacteria and methanogenic archaea at single-cell resolution by synchrotron X-ray fluorescence imaging. Geomicrobiol J, 35:81–89.

33.

Gregory

, Mukherjee

, Olson

, et al. (2019) The formation mechanisms of sedimentary pyrite nodules determined by trace element and sulfur isotope microanalysis. Geochim Cosmochim Acta, 259:53–68.

34.

Grotzinger

, Sumner

, Kah

, et al. (2014) A habitable fluvio-lacustrine environment at Yellowknife Bay, Gale Crater, Mars., Science, 343: 124, 2777.

35.

Hatton

and Rickard

(2008) Nucleic acids bind to nanoparticulate iron(II) monosulphide in aqueous solutions. Orig Life Evol B, 38:257–270.

36.

Hays

, Graham

, Des Marais

, et al. (2017) Biosignature preservation and detection in Mars analog environments. Astrobiology, 17:363–400.

37.

Hemingway

, Rothman

, Grant

, et al. (2019) Mineral protection regulates long-term global preservation of natural organic carbon. Nature, 570:228–231.

38.

Hitchcock

(2019) aXis2000 Is Written in Interactive Data Language (IDL). Available online at http://unicorn.macmaster.ca/axis2000.html, accessed January 22, 2021 .

39.

Hitchcock

, Dynes

, Lawrence

, et al. (2009) Soft X-ray spectromicroscopy of nickel sorption in a natural river biofilm. Geobiology, 7:432–453.

40.

Jacobsen

, Wirick

, Flynn

, et al. (2000) Soft X-ray spectroscopy from image sequences with sub-100 nm spatial resolution. J Microsc, 197:173–184.

41.

Jørgensen

, Findlay

, and Pellerin

(2019) The biogeochemical sulfur cycle of marine sediments. Front Microbiol, 10:Article 849.

42.

Kaplan

and Rittenberg

(1964) Carbon isotope fractionation during metabolism of lactate by Desulfovibrio desulfuricans . Microbiology, 34:213–217.

43.

Kaznatcheev

, Karunakaran

, Lanke

, et al. (2007) Soft X-ray spectromicroscopy beamline at the CLS: commissioning results. Nucl Instrum Meth A, 582:96–99.

44.

Keil

and Mayer

(2014) Mineral matrices and organic matter. In: Treatise on Geochemistry, 2nd edition, edited by HD Holland and KK Turekian, Elsevier, pp 337–359.

45.

Keil

, Montluçon

, Prahl

, et al. (1994) Sorptive preservation of labile organic matter in marine sediments. Nature, 370:549.

46.

King

and McLennan

(2010) Sulfur on mars. Elements, 6:107–112.

47.

Klein

, Lederberg

, Rich

, et al. (1976) The Viking Mission search for life on Mars. Nature, 262:24–27.

48.

Krepski

, Emerson

, Hredzak-Showalter

, et al. (2013) Morphology of biogenic iron oxides records microbial physiology and environmental conditions: toward interpreting iron microfossils. Geobiology, 11:457–471.

49.

Lai

H-Y

and Chen

C-J

(2009) Shape-controlled synthesis of iron sulfide nanostructures via oriented attachment mechanism. J Cryst Growth, 311:4698–4703.

50.

Large

, Halpin

, Danyushevsky

, et al. (2014) Trace element content of sedimentary pyrite as a new proxy for deep-time ocean-atmosphere evolution. Earth Planet Sci Lett, 389:209–220.

51.

Lasne

, Noblet

, Szopa

, et al. (2016) Oxidants at the surface of mars: a review in light of recent exploration results. Astrobiology, 16:977–996.

52.

Lawrence

, Swerhone

GDW

, Leppard

, et al. (2003) Scanning transmission X-ray, laser scanning, and transmission electron microscopy mapping of the exopolymeric matrix of microbial biofilms. Appl Environ Microbiol, 69:5543–5554.

53.

Lefevre

and Bazylinski

(2013) Ecology, diversity, and evolution of magnetotactic bacteria. Microbiol Mol Biol R, 77:497–526.

54.

Lennie

, Redfern

SAT

, Schofield

, et al. (1995) Synthesis and Rietveld crystal structure refinement of mackinawite, tetragonal FeS. Mineral Mag, 59:677–683.

55.

Levin

and Straat

(1977) Life on Mars? The Viking labeled release experiment. Biosystems, 9:165–174.

56.

Levin

and Straat

(2016) The case for extant life on Mars and its possible detection by the Viking labeled release experiment. Astrobiology, 16:798–810.

57.

Lin

, Lindsay

, Weitz

, et al. (1989) Universality in colloid aggregation. Nature, 339:360–362.

58.

Londry

and Des Marais

(2003) Stable carbon isotope fractionation by sulfate-reducing bacteria. Appl Environ Microbiol, 69:2942–2949.

59.

Londry

, Jahnke

, and Des Marais

(2004) Stable carbon isotope ratios of lipid biomarkers of sulfate-reducing bacteria. Appl Environ Microbiol, 70:745–751.

60.

Lorand

, Chevrier

, and Sautter

(2005) Sulfide mineralogy and redox conditions in some shergottites. Meteorit Planet Sci, 40:1257–1272.

61.

Maki

(2013) Bacterial intracellular sulfur globules: structure and function. J Mol Microbiol Biotechnol, 23:270–280.

62.

Mansor

, Berti

, Hochella Jr

, et al. (2019) Phase, morphology, elemental composition, and formation mechanisms of biogenic and abiogenic Fe-Cu-sulfide nanoparticles: a comparative study on their occurrences under anoxic conditions. Am Mineral, 104:703–717.

63.

Margulis

, Mazur

, Barghoorn

, et al. (1979) The Viking Mission: implications for life on Mars. J Mol Evol, 14:223–232.

64.

Matamoros-Veloza

, Cespedes

, Johnson

BRG

, et al. (2018a) A highly reactive precursor in the iron sulfide system. Nat Commun, 9:3125.

65.

Matamoros-Veloza

, Stawski

, and Benning

(2018b) Nanoparticle assembly leads to mackinawite formation. Cryst Growth Des, 18:6757–6764.

66.

Mateo-Martí

, Briones

, Rogero

, et al. (2008) Nucleic acid interactions with pyrite surfaces. Chem Phys, 352:11–18.

67.

Mazur

, Barghoorn

, Halvorson

, et al. (1978) Biological implications of the Viking mission to Mars. Space Sci Rev, 22:3–34.

68.

McKay

, Friedman

, Wharton

, et al. (1992) History of water on Mars: a biological perspective. Adv Space Res, 12:231–238.

69.

McMahon

, Bosak

, Grotzinger

, et al. (2018) A field guide to finding fossils on Mars. J Geophys Res Planets, 123:1012–1040.

70.

Michalski

, Cuadros

, Niles

, et al. (2013) Groundwater activity on Mars and implications for a deep biosphere. Nat Geosci, 6:133–138.

71.

Mielke

, Robinson

, White

, et al. (2011) Iron-sulfide-bearing chimneys as potential catalytic energy traps at life's emergence. Astrobiology, 11:933–950.

72.

Miot

, Benzerara

, Morin

, et al. (2009a) Iron biomineralization by anaerobic neutrophilic iron-oxidizing bacteria. Geochim Cosmochim Acta, 73:696–711.

73.

Miot

, Benzerara

, Obst

, et al. (2009b) Extracellular iron biomineralization by photoautotrophic iron-oxidizing bacteria. Appl Environ Microbiol, 75:5586–5591.

74.

Moreau

, Weber

, Martin

, et al. (2007) Extracellular proteins limit the dispersal of biogenic nanoparticles. Science, 316:1600–1603.

75.

Neveu

, Hays

, Voytek

, et al. (2018) The ladder of life detection. Astrobiology, 18:1375–1402.

76.

Nisbet

and Sleep

(2001) The habitat and nature of early life. Nature, 409:1083–1091.

77.

Parnell

, Boyce

, Thackrey

, et al. (2010) Sulfur isotope signatures for rapid colonization of an impact crater by thermophilic microbes. Geology, 38:271–274.

78.

Penn

and Soltis

(2014) Characterizing crystal growth by oriented aggregation. Cryst Eng Comm, 16:1409–1418.

79.

Picard

, Gartman

, and Girguis

(2016a) What do we really know about the role of microorganisms in iron sulfide mineral formation?. Front Earth Sci, 4:68.

80.

Picard

, Obst

, Schmid

, et al. (2016b) Limited influence of Si on the preservation of Fe mineral-encrusted microbial cells during experimental diagenesis. Geobiology, 14:276–292.

81.

Picard

, Gartman

, Clarke

, et al. (2018) Sulfate-reducing bacteria influence the nucleation and growth of mackinawite and greigite. Geochim Cosmochim Acta, 220:367–384.

82.

Picard

, Gartman

, Cosmidis

, et al. (2019) Authigenic metastable iron sulfide minerals preserve microbial organic carbon preservation in anoxic environments. Chem Geol, 530:119343.

83.

Plekan

, Feyer

, Šutara

, et al. (2007) The adsorption of adenine on mineral surfaces: iron pyrite and silicon dioxide. Surf Sci, 601:1973–1980.

84.

Posfai

, Lefevre

, Trubitsyn

, et al. (2013) Phylogenetic significance of composition and crystal morphology of magnetosome minerals. Front Microbiol, 4:Article 344.

85.

Ravel

and Newville

(2005) ATHENA, ARTEMIS, HEPHAESTUS: data analysis for X-ray absorption spectroscopy using IFEFFIT. J Synchrotron Radiat, 12:537–541.

86.

Rickard

, Mussman

, and Steadman

(2017) Sedimentary sulfides. Elements, 13:117–122.

87.

Rueden

, Schindelin

, Hiner

, et al. (2017) ImageJ2: ImageJ for the next generation of scientific image data. BMC Bioinformatics, 18:529.

88.

Russell

, Daniel

, and Hall

(1993) On the emergence of life via catalytic iron-sulphide membranes. Terra Nova, 5:343–347.

89.

Russell

, Daniel

, Hall

, et al. (1994) A hydrothermally precipitated catalytic iron sulphide membrane as a first step toward life. J Mol Evol, 39:231–243.

90.

Shen

and Buick

(2004) The antiquity of microbial sulfate reduction. Earth Sci Rev, 64:243–272.

91.

Shen

, Farquhar

, Masterson

, et al. (2009) Evaluating the role of microbial sulfate reduction in the early Archean using quadruple isotope systematics. Earth Planet Sci Lett, 279:383–391.

92.

Solomon

, Lehmann

, Kinyangi

, et al. (2009) Carbon (1s) NEXAFS spectroscopy of biogeochemically relevant reference organic compounds. Soil Sci Soc Am J, 73:1817–1830.

93.

Squyres

, Grotzinger

, Arvidson

et al. (2004) In situ evidence for an ancient aqueous environment at Meridiani Planum, Mars. Science, 306:1709–1714.

94.

Stanley

and Southam

(2018) The effect of gram-positive (Desulfosporosinus orientis) and gram-negative (Desulfovibrio desulfuricans) sulfate-reducing bacteria on iron sulfide mineral precipitation. Can J Microbiol, 64:629–637.

95.

Steele

, McCubbin

, Fries

, et al. (2012) A reduced organic carbon component in martian basalts. Science, 337:212–215.

96.

Steele

, McCubbin

, and Fries

(2016) The provenance, formation, and implications of reduced carbon phases in Martian meteorites. Meteorit Planet Sci, 51:2203–2225.

97.

Steele

, Benning

, Wirth

, et al. (2018) Organic synthesis on Mars by electrochemical reduction of CO₂ . Sci Adv, 4:eaat5118.

98.

Stewart-Ornstein

, Hitchcock

, Hernández Cruz

, et al. (2007) Using intrinsic X-ray absorption spectral differences to identify and map peptides and proteins. J Phys Chem B, 111:7691–7699.

99.

Todd

, Sherman

, and Purton

(2003) Surface oxidation of chalcopyrite (CuFeS₂) under ambient atmospheric and aqueous (pH 2-10) conditions: Cu, Fe L-and O K-edge X-ray spectroscopy. Geochim Cosmochim Acta, 67:2137–2146.

100.

Vago

, Westall

, Coates

, et al. (2017) Habitability on early Mars and the search for biosignatures with the ExoMars Rover. Astrobiology, 17:471–510.

101.

Vaniman

, Bish

, Ming

, et al. (2014) Mineralogy of a mudstone at Yellowknife bay, Gale crater, Mars., Science, 343: 124, 3480.

102.

Voordouw

(2002) Carbon monoxide cycling by Desulfovibrio vulgaris Hildenborough. J Bacteriol, 184:5903–5911.

103.

Wan

, Tyliszczak

, and Tokunaga

(2007) Organic carbon distribution, speciation, and elemental correlations within soil microaggregates: applications of STXM and NEXAFS spectroscopy. Geochim Cosmochim Acta, 71:5439–5449.

104.

Westall

, Loizeau

, Foucher

, et al. (2013) Habitability on Mars from a microbial point of view. Astrobiology, 13:887–897.

105.

Westall

, Foucher

, Bost

, et al. (2015) Biosignatures on Mars: what, where, and how? Implications for the search for martian life. Astrobiology, 15:998–1029.

106.

White

, Bhartia

, Stucky

, et al. (2015) Mackinawite and greigite in ancient alkaline hydrothermal chimneys: identifying potential key catalysts for emergent life. Earth Planet Sci Lett, 430:105–114.

107.

Williford

, Farley

, Stack

, et al. (2018) The NASA Mars 2020 rover mission and the search for extraterrestrial life. In: From Habitability to Life on Mars, edited by NA

Cabrol

and EA Grin

Elsevier

, pp 275–308.

108.

Wolthers

, Van der Gaast

, and Rickard

(2003) The structure of disordered mackinawite. Am Mineral, 88:2007–2015.

109.

Yang

, Guo

, Chang

, et al. (2017) Protein/peptide-templated biomimetic synthesis of inorganic nanoparticles for biomedical applications. J Mater Chem B, 5:401–417.

110.

Zahnle

, Arndt

, Cockell

, et al. (2007) Emergence of a habitable planet. Space Sci Rev, 129:35–78.

111.

Zeitvogel

, Schmid

, Hao

, et al. (2016) ScatterJ: an ImageJ plugin for the evaluation of analytical microscopy datasets. J Microsc, 261:148–156.

112.

Zhuang

, Montgomery

, Samarkin

, et al. (2019) Generation and utilization of volatile fatty acids and alcohols in hydrothermally altered sediments in the Guaymas Basin, Gulf of California. Geophys Res Lett, 46:2637–2646.

113.

Zolotov

and Shock

(2005) Formation of jarosite-bearing deposits through aqueous oxidation of pyrite at Meridiani Planum, Mars. Geophys Res Lett, 32:L21203.

114.

Zubavichus

, Shaporenko

, Grunze

, et al. (2005) Innershell absorption spectroscopy of amino acids at all relevant absorption edges. J Phys Chem A, 109:6998–7000.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.05 MB

Interactions Between Iron Sulfide Minerals and Organic Carbon: Implications for Biosignature Preservation and Detection

Abstract

1. Introduction

2. Materials and Methods

2.1. Glassware, medium and solutions

2.2. Iron sulfide mineralization experiments

2.3. X-ray diffraction

2.4. Scanning electron microscopy

2.5. STXM and NEXAFS spectroscopy

2.6. STXM and NEXAFS data processing

2.7. Image analysis

3. Results

3.1. Characterization of the mineral phase of iron sulfide mineral aggregates precipitated in prebiotic and biotic conditions

3.2. Characterization of the organic phase of iron sulfide mineral aggregates precipitated in prebiotic and biotic conditions

3.3. Organic carbon distribution in prebiotic iron sulfide mineral aggregates

3.4. Filamentous structures in prebiotic iron sulfide minerals

3.5. Organic carbon distribution in biotic iron sulfide minerals

3.6. Affinity of iron sulfide minerals for organic molecules

4. Discussion

5. Conclusions

Footnotes

Acknowledgments

Author Disclosure Statement

Funding Information

Supplementary Material

Abbreviations Used

References

Supplementary Material