4 Discussion
We first gave a thorough description of our workflow in the field, explaining how we approach subterranean Martian analogue environments in search for biosignatures. We then presented a case study of blue microbial mats and their substrates from three lava tubes in Iceland analyzed with a variety of analytical techniques, namely a portable XRF, 16S rRNA gene sequencing, Raman spectroscopy, and SEM/EDS. Here we discuss the possible origins of the copper-rich secondary mineral precipitates, the significance of the bacteria that inhabit them, and the molecular markers they leave behind therein.
4.1 Cave geology and speleothems
We observed stark differences in elemental concentration (with preliminary in situ XRF readings) between the substrates where the microbial mats were sampled and the background areas, i.e., uncolonized areas nearby the sampling locations (Figure 5a-c). This demonstrates an altered geological substrate that biofilms either contribute to or benefit from in order to thrive in this environment. Our mineralogical and elemental analysis of the lava rock substrate (Figure 7) and blue secondary mineral precipitates (Figures 8-10) show sequestering of elements more generally dispersed and scarce in basaltic lava tubes, particularly copper. The boundaries between the unaltered basalt and the precipitate rim, as seen in the SEM images of the thin sections (Figures 8 & 9), are pristine, dictating that the elements that make up the observed blue mineral precipitates must have filtered in with the water through cracks in the cave walls or ceiling, instead of being leached from the host rock.
4.1.1 Sources of copper in Iceland
The average copper content of recent Icelandic basaltic lava is low, in the range of 10-200 ppm (μg/g) (Eason & Sinton, 2009; Gibson et al., 1982), with occasional higher values found in older Miocene volcaniclastic rocks (Schmincke et al., 1982). Similarly, the copper content of Icelandic groundwater is also low, ranging from 0.00015 to 0.00209 ppm (Gunnarsdottir et al., 2015) where the Cu(II) ion is the more common oxidation state (Schock et al., 1995). However, copper is a semi-volatile element that can partition into a volatile-rich fluid that can physically separate from magma (Candela & Holland, 1984). Copper has been observed to be enriched in volcanic laze plumes from basaltic intraplate volcanoes, which can transfer it directly into the marine biosphere (Mason et al., 2021), and result in highly copper-enriched groundwater in specific areas. In the case of Iceland, copper is enriched in hyaloclastite deposits at the lava-ice interface (Furnes, 1978) and from there could be readily mobilized into groundwater that can infiltrate the lava tubes through cracks. In addition, rain and melting snow could be leaching copper from the volcanic ash that abounds on the lava fields above, bringing it into the tubes through cracks and cave openings. Kiernan et al. (2003) has noted the hydrogeological significance of lava tubes in the Eldhraun lava field, with rainwater and floods transporting glacio-aeolian deposits in these efficient groundwater flow channels.
Volcanic ash, derived from explosive eruptions of evolved intermediate-silicic volcanoes, has an evolved nature relative to basaltic lava flows, characterized by an enrichment in many incompatible elements, including copper. Hoffmann et al. (2012) showed that copper was released from various volcanic ash samples in concentrations up to 0.00065 ppm after just 15 min of contact with seawater. Smith et al. (1982) reported an average concentration of 0.39 ppm of Cu in leachates from volcanic ash in experiments simulating its interaction with rainfall and prolonged exposure with groundwater. Experiments with volcanic ash from the 2000 eruption of the Mt. Hekla volcano in Iceland showed a Cu flux of 0.069 ppm within 8 hours of mixing with deionized water (Jones & Gislason, 2008). Though this eruption would not have reached our areas of interest at the initial deposition stage, the volcanic cloud spewed out 0.1 Tg of ash (Rose et al., 2003), allowing it to be transported north-northeast by wind and depositing up to 5 kg/m2 on the headwaters of the Ytri–Rangá River (Haraldsson, 2001).
There could thus exist many sources of copper in the Ódáðahraun and Eldhraun lava fields, including the hyaloclastite formations in nearby Mt. Laki and Mt. Herðubreið. The precise source of ash in the Ódáðahraun lava field is difficult to determine, as wind plays a large role in transportation of ash and sand, and a large proportion of it (some of which may be copper-rich ash and shattered hyaloclastite material) has been blown long distances over the lavas from sources closer to Vatnajökull glacier or the Jökulsá á Fjöllum glacial river (Arnalds, 2015). Indeed both the Ódáðahraun and Eldhraun lava fields are subject to severe wind erosion and very high aeolian deposition rates (Arnalds, 2010; Arnalds et al., 2001), constantly replenishing the areas with volcanic ash.
4.1.2 Copper abundance on Mars
The surface of Mars is enriched in sulfur (Rudnick & Gao, 2003), and is thus expected to show elevated concentrations of chalcophile elements such as copper in the crust (Payré et al., 2019). Copper abundance values up to 580 ppm were detected by the Curiosity rover in the Liga sedimentary bedrock at Gale Crater (Berger et al., 2017). A chrysocolla bearing unit, along with pseudomalachite and other copper mineral phases, has been identified in the Shalbatana Valley palaeolacustrine system on Mars, and is hypothesized to be a supergene alteration product of copper sulfide minerals interacting with oxygen and water (Popa et al., 2014). This indicates a redox system capable of oxidizing Cu+ to Cu2+ in an oxidizing environment in Mars’ past, which would likely have extended to lava tubes. Copper enrichments seemingly precipitated from groundwater have been found adsorbed to manganese deposits in the Kimberley Formation in Gale Crater on Mars, and are thought to be deposited in oxidizing conditions within fractures in the bedrock (Payré et al., 2019).
4.1.2 Blue speleothems in Iceland
In Figures 8-10 we described copper-rich secondary mineral precipitates as elucidated by SEM/EDS. While Figure 8 and 10 show a single copper phase, Figure 9 boasts a more complex precipitate, with discrete layers of copper silicates, copper phosphates, manganese oxides, and carbonate-bearing species. The copper phases present are expected to be mostly in the Cu(II) oxidation state, as this is the ion more readily available in the groundwater (Schock et al., 1995), and because of the prominent blue color (Cu(I) minerals are generally red/brown). Chrysocolla, an amorphous copper phyllosilicate, was found in samples C2 (Figure 8), B5 (Figure 9), and H7 (Figure 10). A similar finding was previously reported in lava tubes in Kipuka Kanohina Cave Preserve in Hawai’i by Northup et al. (2011), who also found a blue drop filled with a precipitate with an Al:Cu:Si ratio of 0.15:1.8:1 as analyzed by EDS, comparable to a typical chrysocolla ratio of 0.12:1.96:1 (Anthony, 1990). We also identified a layer of pseudomalachite in sample B5, which has been hinted at often being associated with microbial mats (Little & Wagner, 2018). Manganese oxides can be deposited by microbes in waters with manganese concentration as low as 10-20 ppb (Dickinson et al., 1996). Manganese oxidation is associated with the metabolism of several genera of bacteria. Leptothrix (a Proteobacteria associated with strictly low concentrations of organic matter) has been shown to deposit amorphous MnO2 (vernadite) and a black MnO2 precipitate (birnessite) (Gounot, 1994), whileBacillus has been observed to recrystallize birnessite to octahedral Mn3O4 (hausmannite) (Nealson et al., 1988). Boston et al. (2001) found manganese oxides in limestone caves to be linked to biogenic activity, identifying biogenically precipitated manganese ’snow’ in limestone caves using a suite of analyses, including culturing Mn-oxidizing bacteria isolated from the ‘snow’ and observing them produce amorphous manganese oxides in the laboratory. Ultimately, blue, copper-rich speleothems are attractive targets for astrobiological research, as they house bacterial communities resistant to elevated copper content. Should they occur in lava tubes on Mars, they could also be thought of as a biotope within the caves, host to extremophilic organisms. 4.2 Bacterial communities inhabiting blue mineral precipitates The bacteria found in copper-enriched areas in the caves are present there because they can tolerate such an environment, and they actively sequester and adsorb copper ions. The significance of the bacteria identified and their copper-coping mechanisms are discussed below. Proteobacteria was the major phylum identified in the blue samples in the southern cave (C2) and northern cave (H7). Families and genera were generally similar across the samples, with some notable deviations. Differences in environment (average cave temperature, age of lava tube, humidity, etc.) may contribute to the differences in bacteria present inside of the caves from north to south; however, there was a large overlap in the major abundances found. A recent study (Selensky et al., 2021) suggests that surface environment is not a major factor in organic nutrient cycling in lava tubes, which further exemplifies them as isolated environments that may harbor chemolithoautotrophic organisms and biosignatures that are useful in the search for life in extraterrestrial lava tubes. Furthermore, Northup et al. (2016) found that bacterial diversity in Icelandic lava tubes differed substantially from that in surface soil samples, with the most abundant cave phylum being Actinobacteria, followed by Acidobacteria and Proteobacteria.
The majority of genera detected in the blue samples (C2 and H7) represent copper-resistant oligotrophs. Ralstonia, Caulobacter, Cupriavidus, and Corynebacterium accounted for 97% of the relative genus abundance in sample C2 (Figure 6a). In sample H7Ralstonia, Cupriavidus, Caulobacter, andMethylobacterium-Methylorubrum accounted for a total of 99% of the relative abundance (Figure 6b). Ralstonia, Caulobacter, Cupriavidus , and Corynebacterium are reported to have high metal resistance, which explains their presence in high copper concentration regions of the caves (Janssen et al., 2010; Mergeay et al., 2015; Morosov et al., 2018; Yang et al., 2019). Copper is essential for bacteria in trace amounts as it is utilized as a micronutrient as well as an enzymatic cofactor for redox activities (Giachino & Waldron, 2020). Copper resistance is found in different mechanisms within the bacteria, the most ubiquitous being a) oxidation of Cu(I) to Cu(II), a less toxic ion to bacteria; b) copper sequestration by metallothioneins and c) transmembrane copper export (Ladomersky & Petris, 2015). Additionally, the genome of Cupriavidus metallidurans containscop genes, which are highly induced by Cu(II) and encode for copper exporting P1B-type ATPases (Monchy et al., 2020), and cueOgenes encoding for multicopper oxidase (Sanyal et al., 2020).
Much of the literature discusses the ability of both Cupriavidusand Ralstonia to adsorb metals to their cell membrane. Morphological changes can be seen as they adjust to environments with high concentrations of metals (Diels et al., 2009). This is also found with extremely high concentrations of Mn(II), where a strain ofRalstonia picketti was able to survive and also remove the Mn(II) from aqueous solution possibly by biosorption onto hydroxyl and carbonyl groups on the bacterial surface, ultimately producing a precipitate (Huang et al., 2018). These bacteria also induce efficient efflux of copper transport using metallophores that can regulate uptake of metals into the cell instead of relying on passive transport that can be deadly (Kenney & Rosenzweig, 2018; von Rozycki & Nies, 2009).Cupriavidus is found colonizing metal abundant biotopes with their well-adapted metal resistance (Janssen et al., 2010). The biotopic relationship copper deposits have with these bacteria entwines their fates as they grow concurrently alongside each other.
The capability of surviving in oligotrophic environments coupled with high metal-resistance makes our reported genera particularly well-suited for Martian and Martian analogue environments. Two novel species ofCupriavidus were discovered in mudflow deposits from Mt. Pinatubo, where they exhibit chemolithoautotrophic growth using hydrogen, oxygen, and carbon dioxide in an area deprived of organic carbon (Sato et al., 2006). Cupriavidus, Ralstonia, andMethylobacterium were found on the Mars Odyssey Orbiter (prior to flight) and other space industry settings as described by Mijnendonckx et al. (2013). They described the metal resistance of Cupriavidus metallidurans and Ralstonia picketti using select metals as micronutrients. They found Cupriavidus metalliduransisolates to be able to withstand an excess of Cu2+typically toxic to many species of bacteria. These bacteria’s strong resistance to the antimicrobial disinfection and sterilization procedures of the space industry not only demonstrate their candidacy as extremophiles, but further their candidacy as potential contaminants that may be brought to Mars. The features of these bacteria that allow them to persist in difficult environments on Earth and potentially on Mars should also be considered when checking for contaminants in future missions.
4.3 Molecular biosignatures
Although the presence of colorful pigments in an environment deprived of light, such as lava tubes, could appear surprising, carotenoids and other pigments can play different roles depending on the metabolisms of the host cells. Carotenoid molecules have a wide distribution in very diverse organisms (more than 750 chemical structures determined to date), including extremophiles, where they serve several key functions at the cellular level. In addition to serving as photoprotective accessory pigments in phototrophic organisms, they have excellent antioxidant properties acting as reactive oxygen species (ROS) scavengers thus protecting cellular DNA and proteins (Stahl & Sies, 2003). They are also believed to help stabilize membranes at low temperatures (Dieser et al., 2018), particularly relevant for Icelandic microbial communities. It has even been proposed that carotenoids played an important role in the early evolution of life on Earth (Alcaíno et al., 2016; Klassen, 2010) by being involved in membrane stabilization, prior to fatty acids (Ourisson & Nakatani, 1994). These functions and properties, essential for the highly UV-irradiated early Earth, might also be compatible with early Mars organisms (Cockell, 2002; Rothschild, 1990), adding to the relevance of carotenoids as a biosignature target for our search for life on Mars. In addition, although we would not normally expect phototrophic organisms in cave environments, cyanobacteria have recently been reported in deep subsurface rock samples on Earth, where they switch to a light-independent hydrogen-based lithoautotrophic metabolism (Puente-Sánchez et al., 2018).
Carotenoids may prove as a critical biosignature in our search for life in extraterrestrial lava tubes as they are only found in living organisms, are relatively easy to distinguish and detect, and have been shown to be highly resistant to oxidative and radiative stress in simulated Martian conditions (Baqué et al., 2016). Portable Raman instruments could potentially detect carotenoid molecules in cavesin situ by humans or robotic explorers. As lava tubes are shielded from radiation on Mars, carotenoids have the potential to be preserved on even longer timescales. Potential pigmented species found in our DNA analysis of sample C2 occur in the genera Cupriavidusand Methylobacterium. These are thought to be the source of the carotenoids identified in the sample (Ramachandran et al., 2014; Osawa et al., 2015; Heider et al., 2012). The identification of carotenoids in our samples demonstrate a staying power in this copper-rich environment that may lend itself to future identification in caves as evidence of extant or extinct life.