Biofilm 5 (2023) 100127 Contents lists available at ScienceDirect Biofilm journal homepage: www.sciencedirect.com/journal/biofilm Development of Martian saline seep models and their implications for planetary protection Madelyn K. Mettler a,b,1, Hannah M. Goemann a,c,1, Rebecca C. Mueller a,d, Oscar A. Vanegas e, Gabriela Lopez f, Nitin Singh g, Kasthuri Venkateswaran g, Brent M. Peyton a,b,h,* a Center for Biofilm Engineering, Montana State University, Bozeman, MT, USA b Department of Chemical and Biological Engineering, Montana State University, Bozeman, MT, USA c Department of Microbiology and Cell Biology, Montana State University, Bozeman, MT, USA d USDA Agricultural Research Service, Western Regional Research Center, Albany, CA, USA e University of Georgia, Athens, GA, USA f Georgia State University, Atlanta, GA, USA g NASA Jet Propulsion Laboratory, California Institute of Technology, Pasadena, CA, USA h Thermal Biology Institute, Montana State University, Bozeman, MT, USA A R T I C L E I N F O A B S T R A C T Keywords: While life on Mars has not been found, Earth-based microorganisms may contaminate the Red Planet during Biofilm rover expeditions and human exploration. Due to the survival advantages conferred by the biofilm morphology Mars to microorganisms, such as resistance to UV and osmotic stress, biofilms are particularly concerning from a Halophile planetary protection perspective. Modeling and data from the NASA Phoenix mission indicate that temporary Drip flow reactor Next-generation sequencing liquid water might exist on Mars in the form of high salinity brines. These brines could provide colonization opportunities for terrestrial microorganisms brought by spacecraft or humans. To begin testing for potential establishment of microbes, results are presented from a simplified laboratory model of a Martian saline seep inoculated with sediment from Hailstone Basin, a terrestrial saline seep in Montana (USA). The seep was modeled as a sand-packed drip flow reactor at room temperature fed media with either 1 M MgSO4 or 1 M NaCl. Biofilms were established within the first sampling point of each experiment. Endpoint 16S rRNA gene community analysis showed significant selection of halophilic microorganisms by the media. Additionally, we detected 16S rRNA gene sequences highly similar to microorganisms previously detected in two spacecraft assembly clean- rooms. These experimental models provide an important foundation for identifying microbes that could hitch- hike on spacecraft and may be able to colonize Martian saline seeps. Future model optimization will be vital to informing cleanroom sterilization procedures. 1. Introduction for life to exist [3]. Achieving these levels of bioburden (for example, < 30 bacterial spores for equipment entering Special Regions of Mars) Robotic space exploration is becoming increasingly attainable for depends on contamination prevention in spacecraft assembly clean- government agencies and the private sector with the advancement of rooms on Earth [3]. Despite the rigorous decontamination measures technology, especially regarding further exploration of Mars. Such ad- typically employed by cleanroom facilities [4,5], next-generation vancements carry increased responsibility for planetary protection as sequencing efforts have recently revealed the presence of outlined by the International Committee on Space Research’s (COSPAR) non-culturable organisms frequently contaminating cleanrooms at Panel on Planetary Protection Policies [1,2]. The most current iteration higher levels than was previously thought possible [6–8]. These or- of policies include strict guidelines on acceptable limits of bioburden on ganisms are of particular concern for planetary protection as they Martian rovers and other technical equipment, particularly in Mars potentially pose increased threats for forward contaminantion of Special Regions, regions deemed to have the most favorable conditions Martian soils. * Corresponding author. Center for Biofilm Engineering, Montana State University, Bozeman, MT, USA. E-mail address: bpeyton@montana.edu (B.M. Peyton). 1 These authors have contributed equally to this work. https://doi.org/10.1016/j.bioflm.2023.100127 Received 28 October 2022; Received in revised form 4 April 2023; Accepted 21 April 2023 Available online 13 May 2023 2590-2075/© 2023 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by- nc-nd/4.0/). M.K. Mettler et al. B io fi lm 5 (2023) 100127 The Martian landscape is notoriously hostile to life as we know it 2. Methods with its thin atmosphere, high levels of ultraviolet (UV) radiation, lack of pure liquid water, and low temperatures. Like Earth, Mars is highly 2.1. Inoculation material heterogenous with varying temperatures, topography, and soil compo- sition throughout the planet. The temperatures on the surface of Mars The saline seep reactor was inoculated with sediment from HSB average − 63 ◦C compared to Earth’s average of 13 ◦C [9], although which was sampled in early June 2021 (coordinates: 46.00 N, − 109.18 mathematical temperature modeling predicts surface temperatures in W; Fig. 1). Additional sampling site details can be found in the Sup- the lower latitudes can approach 22 ◦C during a sunny summer sol [10, plemental Information (Table S1) and in a U.S. Geological Survey report 11]. Water in the form of ice exists in many places on Mars, but negli- from 1979 [35]. Sediment was collected using sterile 50 mL conical vials gible concentrations are present as water vapor in the atmosphere [12]. and spatulas rinsed with 95% ethanol. After sampling, vials for DNA As thoroughly reviewed by Wray [13], the past decades have seen much analysis were placed on dry ice for transit back to the laboratory and debate on the presence of liquid water on Mars, with subsurface glacial then transferred into a − 80 ◦C freezer for storage and DNA extractions. lakes [14,15] and recurring slope linae [16,17] receiving attention for Samples for DFR inoculation were kept on wet ice (approx. 4 ◦C) for their potential to host microbial life. It is largely agreed that the sub- transit to the laboratory. surface glacial reservoirs are likely stable high-salinity liquid brines existing at temperatures below the freezing point of pure water. How- 2.2. Reactor setup and operation ever, whether the temporary recurring slope lineae are caused by dust or water is actively disputed [18,19]. Deliquescence with changes in sea- The Martian seep analog used a DFR filled with sand with an average sonal relative humidity may also play an important role in the presence diameter of 2 mm (Fig. 2). The DFR is a well-characterized biofilm of briny liquids on Mars [13,20,21]. It is also possible that microenvi- reactor with an American Society for Testing and Materials (ASTM) ronments habitable to life exist that are not accurately captured by the standard method for growing low-shear biofilms, but was modified for measurement devices used in exploration vehicles such as rovers and this study [33]. A small brass filter was placed interior to the effluent orbiters. Although concrete evidence for the ability for life to exist on tubing to prevent sand from exiting the reactor through the effluent port. Mars has yet to be revealed, these findings have sparked many questions The reactor, including the filter, sand, and connected effluent tubing, regarding the possibility for terrestrial organisms brought by spacecraft was autoclaved for 20 min at 123 ◦C and 21 psi prior to the experiments to colonize Martian soil. As discussed by the COSPAR Panel on Planetary to ensure sterility. The silicone tubing for influent media was autoclaved Protection, polyextremophilic biofilm-forming organisms are of high separately and then attached to the reactor via sterile 23-gauge needles. concern for planetary forward contamination [22]. Media was pumped into the system using peristaltic pumps operating at Biofilms are microbial communities and are often surface-associated an average flow rate of 0.2 mL/min to each channel which was the lower which provides many survival advantages over planktonic cells limit of the pump’s abilities. Two reactors (a total of ten channels) were including increased resistance to antimicrobials, desiccation, and other used in the high-carbon experiments while only one reactor (a total of environmental stresses such as UV radiation and osmotic lysis [23–29]. six channels) was used for the low-carbon experiments (Table 1). Six of Biofilms pose a significant risk for planetary protection as they can the channels in the high-carbon experiments were under CO2 conditions contaminate spacecraft and subsequently the surfaces on which these to closer mimic the Martian atmosphere compared the Earth’s atmo- spacecraft land. Recent experiments have shown that some poly- sphere (four channels), while all six channels for the low-carbon ex- extremophiles can survive for at least several months under Martian periments were under CO2 conditions [36]. A 0.2 μm air filter and the conditions [30,31]. Crisler et al. (2012) cultivated planktonic halophilic Tygon tubing for CO2 were autoclaved prior to reactor assembly and microorganisms under multiple Martian-relevant stressors (up to 2 M inoculation. The DFR model was kept at room temperature to ensure the MgSO4 and low pH) and through up to 15 freeze-thaw cycles until no accumulation of an observable biofilm within the time constraints of the viable organisms were able to be recovered [32]. However, given that experiments. For both experiments, several internal channels were biofilms confer increased tolerance to environmental stressors, there covered with foil and tape (Fig. 2) to block light from entering the remains a wide knowledge gap in the ability of the biofilm morphology reactor, mimicking subsurface light conditions. Approximately 5 g of to enhance bacterial survival in Martian-relevant conditions. HSB sediment was suspended via vortex mixing for at least 1 min in 40 Here we developed proof-of-concept experimental models which aim mL of media in a sterile 50 mL conical vial to remove microorganisms to 1) identify biofilm-forming halophiles on Earth that may survive in from the sediment surface. After settling for 15–30 min when most of the potential Martian briny liquid conditions and 2) determine whether any particulates had fallen to the bottom of the vial, 7 mL of supernatant was of these organisms overlap with those which have been detected in used to inoculate each channel with the microbial community sus- spacecraft assembly cleanrooms. The project utilized a simple Martian pended in the respective media for each channel. The supernatant was saline seep analog created using a drip flow biofilm reactor (DFR; Bio- pipetted over the entire length of each channel. Continuous flow of Surface Technologies, Bozeman, MT) [33]. The DFR was inoculated with media was initiated approximately 24 h after the inoculation of the re- sediment from Hailstone Basin (HSB), a naturally occurring saline seep actors. The DFR was left on a flat surface, allowing for slight pooling of in south-central Montana. The high soil salinity is due to natural evap- media at the bottom of the channels. oration leading to the accumulation of salts (including sodium, mag- nesium, and sulfates) and heavy metals in the soil [34,35]. Additionally, 2.3. Media as it is located in south-central Montana, the soil is exposed to freezing and subfreezing temperatures for much of the year, increasing its rele- Three different media were used throughout the experiments. The vance to Martian conditions. To our knowledge, this research is the first initial high-carbon experiments used two different media: high-carbon microbiological study published about HSB. We used highly saline MgSO4 medium, which included 1 M MgSO4 and 0.1 M NaCl, while media to select for halophiles from the HSB, and used 16S rRNA-based the high-carbon NaCl medium featured 1 M NaCl and 0.1 M MgSO4. In microbial community sequencing from these experiments provide a both media, there was tryptone, KCl, glucose, and yeast extract based on guide for which microorganisms might survive the harsh Martian con- the media used by Caton et al. (2004) to provide excess nutrients and ditions. We then utilized publicly available datasets from cleanroom promote rapid biofilm accumulation [37]. The low-carbon experiments sampling efforts to identify overlap between the model communities and used a third medium, low-carbon MgSO4 that included 1 M MgSO4 and known cleanroom microbes. Comparing these microbes to those found 0.1 M NaCl but only yeast extract as the sole carbon source at 0.1% of the in spacecraft assembly cleanrooms highlights possible taxa that could be concentration used in the high carbon media. Media were prepared in of concern from a planetary protection perspective. 20 L carboys and autoclaved prior to use. Table 2 summarizes the 2 M.K. Mettler et al. B io fi lm 5 (2023) 100127 Fig. 1. A) Researcher examining a potential sampling location at Hailstone Basin near Rapelje, Montana. Samples were collected on the left side of the image where the soil has an upper white layer of salt. B) Close up image of sampled area. The dark spots mark where the upper salt layer was removed during collection. C) Map of the ephemeral Hailstone Lake with black dot marking the approximate sampling location. Fig. 2. A) Drip flow reactor set up at start of low-carbon experiment and B) sand in the reactor prior to start of experiment. Table 1 Conditions for each channel for all experiments. High-carbon experiment Channel 1 2 3 4 5 6 7 8 9 10 Medium MgSO4 MgSO4 NaCl NaCl MgSO4 MgSO4 NaCl NaCl NaCl NaCl Light or dark Light Dark Dark Light Light Light Dark Dark Light Light Under CO2 No No No No Yes Yes Yes Yes Yes Yes Low-carbon experiment Channel 1 2 3 4 5 6a Medium MgSO4 MgSO4 MgSO4 MgSO4 MgSO4 MgSO4 Light or dark Light Dark Dark Light Light Light Under CO2 Yes Yes Yes Yes Yes Yes a Channel 6 in the low-carbon experiment served as an uninoculated control. components of the media used in both experiments. 2.4. Sampling for culture-based measurements For sampling, three grains of sand were removed from the surface layer of sand in each channel: one from the top (near the influent), 3 M.K. Mettler et al. B io fi lm 5 (2023) 100127 Table 2 2.6. Culture-independent DNA sequencing analysis Concentrations of components in the saline media. High-Carbon High-Carbon Low-Carbon To determine the composition and diversity of sediment microbial MgSO4 (g/L) NaCl (g/L) MgSO4 (g/L) communities in the original sediment collected from HSB, and the MgSO ●7H O 246.5 58.4 246.5 cultured biofilms, we performed 16S rRNA gene amplicon sequencing. 4 2 (epsomite) For the biofilm samples, DNA was extracted directly from surface sand NaCl 5.8 24.6 5.8 grains harvested from the reactor using flame-sterilized tweezers. KCl 2.0 2.0 2.0 Genomic DNA was extracted from 0.5 g samples using the Fast DNA Yeast extract 1.0 1.0 0.001 Glucose 1.0 1.0 – SPIN Kit for soil (MP Biomedicals, Santa Ana, CA), according to the Tryptone 5.0 5.0 – manufacturer’s instructions. The analysis targeted the V4 region of the 16S rRNA gene using 515F-A (GTGYCAGCMGCCGCGGTAA) and 806R–B (GGACTACVSGGGTATCTAAT) primers from the Earth Micro- middle, and distal (near effluent) areas of the channels. Surface sand biome Project with adapters for Illumina-based sequencing on the MiSeq grains were not submerged in the pooled media at the bottom of the platform. PCR reaction and sequencing preparation details are provided reactor. Sampled sand grains had fresh media flowing across the surface, in the supplemental information. To compare the DFR biofilm commu- so any microorganisms on the sand grains were likely to be attached nities to organisms detected in spacecraft assembly clean rooms, we biofilm. The sand was removed with flame-sterilized tweezers and cross-referenced the microorganisms detected in the DFR biofilms with placed into sterile 50 mL conical vials containing 10 mL of sterile 10% those detected via metabarcoding of the 16S rRNA gene in two studies of w/v NaCl. The biofilm on the sand grains was then disaggregated via an spacecraft-assembly cleanroom sampling [7,39]. Sequences were alternating series of 1 min vortex mixing and 1 min sonication, for a total downloaded from NCBI Bioprojects PRJEB15908 (8 samples) and of 5 min. Afterward, the disaggregated biofilm was filtered onto poly- PRJEB8763 (13 samples) from a cleanroom facility at Thales Alenia carbonate filters and stained for direct microscopy counts. The high- Space (TAS, European Space Agency) in Turin, Italy, and the Jet Pro- carbon experiments also included spread plating 100 μL of dis- pulsion Laboratory (JPL, NASA) in Pasadena, CA, USA. The JPL samples aggregated biofilm on plates made with the respective media from the were partitioned into two treatment groups, one for total (T) community sampled channels with 7.5 g/L Gelrite. Spread plates were incubated on members detected in the JPL cleanrooms and one for samples treated the benchtop at room temperature for two days prior to counting the with propidium monoazide (PMA) to capture only living cells. colony forming units (CFU). No distinctions between morphologies were Sequenced reads (paired-end 300 bp) were merged and combined made in the colony counts. with the TAS and JPL datasets. The combined sequences were then Sanger sequencing was also conducted during the high-carbon ex- trimmed, quality filtered, and dereplicated with USEARCH. Zero-radius periments on colonies grown on GelRite plates collected from the reactor operational taxonomic units (ZOTUs) were identified with UNOISE3 channels. DNA was extracted from individual colonies using the One- (v.11.0.667, [40]). The 16S ZOTUs were classified using SINTAX against Tube Tissue DNA Extraction Kit (Bio Basic, Ontario, Canada), accord- a modified version of the Genome Taxonomy Database (GTDB v. 202, ing to the manufacturer’s instructions, but at one-tenth the volume [41]) where the number of sequences were reduced to one representa- recommended for colony extraction. The full-length 16S rRNA gene was tive for each species with additional outgroup for mitochondria and amplified using the universal primers 27F (5′-AGAGTTT- chloroplast 16S sequences added to eliminate eukaryotic sequences. GATCCTGGCTCAG-3′) and 1492R (5′-GGTTACCTTGTTACGACTT-3′). Community composition was assessed using R Statistical Software PCR condition details are provided in the supplemental information. (v.4.1.2; [42]). Phylogenetic and taxonomic metrics were computed Sequences with high quality forward and reverse sequences were with the phyloseq and vegan packages [43,44] and relative abundances of merged using the online tool merger in the EMBOSS toolkit and se- ZOTUs were calculated with the microbiome package [45]. PERMA- quences were classified using BLASTn. To determine if there was overlap NOVA was used to determine statistical effects in Bray-Curtis commu- between these culturable organisms and the ZOTUs identified via nity dissimilarities [44]. Due to large differences in sequencing depth amplicon sequencing we used USEARCH to set the Sanger reads as a between the publicly available data and the DFR model libraries, ordi- database against our full set of amplicon ZOTUs including the DFR nation and community overlap was performed on the full dataset rare- models, JPL and TAS sequences (described in section 2.6 below). Any fied to 2079 reads per sample, while the DFR model sequences were reads that matched to the Sanger database at 97% identity and had e- separately rarefied to 24843 reads per sample for comparisons between values less than 1e-06 were counted as positive hits. the high-carbon and low-carbon experiments. A reference phylogeny was constructed using full length and near full length sequences of iso- 2.5. Microscopy lates and metagenome assemblies downloaded from GenBank. Refer- ence sequences were aligned using mafft [46], and a maximum Prior to imaging, cells from the disaggregated biofilms were stained likelihood tree was constructed using RAxML with the GTR + gamma with 25x SybrGold nucleic acid gel stain as initially described by Chen model [47]. Environmental sequences were then mapped onto the et al. (2001) with an increased incubation time and higher final stain reference tree using pplacer [48] with reference-aligned sequences. The concentration [38]. In short, the stain was prepared and mixed with the phylogenetic tree was then visualized and annotated using iTOL [49]. sample at a final concentration of 4x and allowed to incubate in the dark The overlap of taxa between datasets was determined using vennDia- for 15 min. After incubation, stained cells were vacuum filtered onto 0.2 gram in the MicEco package in R-Studio [50]. μm black polycarbonate filters which were then affixed to glass micro- scope slides for counting. A Nikon Eclipse E800 epifluorescent micro- 3. Results scope with a green fluorescein isothiocyanate (FITC) filter was used to view and count the cells. Field emission scanning electron microscopy 3.1. Biofilm accumulation (FE SEM) was performed on several samples using a Zeiss Supra 55VP FE SEM. For FE SEM imaging, sand grains were removed from the reactor Biofilm growth and accumulation was detected in both the high- directly, placed in a sterile petri dish, and allowed to air dry in a carbon and low-carbon experiments. The high-carbon experiments biosafety cabinet. The sand grains were then gently poured on Ted Pella included both direct microscopy counts and heterotrophic plate counts carbon tape and placed in the FE SEM for imaging. (Fig. 3A, S1). The plate counts were about one order of magnitude lower than the direct counts for each treatment. The general trend for each experiment was a gradual increase in biofilm accumulation. According 4 M.K. Mettler et al. B io fi lm 5 (2023) 100127 to plate counts, the high-carbon experiments reached steady state with 3.2. FE SEM imaging biofilm density of 7–9.5 log 2 10 CFU/cm around 17 days after the start of continuous flow. The microscopy counts show steady state (biofilm Fig. 4 A and B show the sand grains before inoculation and after density of 8–9 log10 cells/cm2) was achieved by 10 days after the start of biofilm accumulation. The FE SEM images of the high-carbon experi- continuous flow, which was the first sampling point. On average, the ment (Fig. 4C) revealed several morphologies within the sample. The channels receiving normal atmospheric air (channels 1–4) had more topography of the sand grain can be seen as well as coccoidal and rod- biofilm accumulation by the end of the experiments compared to the shaped cells of varying sizes. Additionally, there is visible extracellular channels receiving pure CO2 (channels 5–10). The low-carbon experi- polymeric substances (EPS) blanketing the cells and the sand grain ments using low-carbon MgSO4 medium showed a one to two orders of surface (Fig. 4B) further verifying successful biofilm formation. magnitude reduction in direct cell counts compared to the high-carbon experiment using high-carbon MgSO4 medium with an endpoint bio- 3.3. Microbial community analysis film density of 5–8.5 log10 cells/cm2 (Fig. 3B). Once again, the general trend was slightly increased biofilm density across all channels over The final dataset including the samples from the high-carbon and time. The uninoculated control channel initially fostered some cells, but low-carbon experiments, the two cleanrooms [7,39], and the original after 34 days of flow, cell density dropped near the detection limit, sediment from HSB contained 1052 ZOTUs. Bray-Curtis ordination of settling at or below 2 log10 cells/cm2. the community dissimilarities indicated strong separation among the The biofilms in the low-carbon experiments took 20 days after the five sub-datasets (Fig. 5, pseudoF = 5.35, R2 = 0.552, p < 0.001). Within start of continuous flow to reach steady state according to direct cell the model experiments, community composition differences were driven counts. Channel 4 fostered the most biofilm accumulation by the end of primarily by the high-carbon vs. low-carbon media (pseudoF = 4.8, R2 the experiment peaking at 8.2 log10 cells/cm2 which is similar in density = 0.423, p = 0.001, Fig. 5, Table S3), while we did not detect significant to the biofilm in the high-carbon experiment using high-carbon MgSO4 community effects of CO2 vs. ambient air or light vs. dark conditions. medium and CO2 conditions (channels 5 and 6, 8.1 and 7.4 log10 cells/ The HSB sediment used to inoculate the DFR consisted of 281 cm2 respectively). As with the high-carbon experiments, in the low- detected ZOTUs representing 25 distinct phyla of which the major phyla carbon experiments there was no observed influence of light exposure included the archaeal Bacteroidota (24%), and bacterial Halobacteriota on biofilm density between channels. (22%), Proteobacteria (18.4%, classes Gammaproteobacteria 96%, Fig. 3. Biofilm density of A) high-carbon experiment (HC) and B) low-carbon experiment (LC) as measured by epifluorescent microscopy direct counts. All channels in low-carbon experiment were fed MgSO4 medium and CO2. Note the different axes for each experiment. 5 M.K. Mettler et al. B io fi lm 5 (2023) 100127 Fig. 4. A) Sand grains prior to sterilization and inoculation. B) Visible accumulation of biofilm on sand grains in channel 8 of high-carbon experiment. C) FE SEM image of biofilm attached to sand grain from panel B. Fig. 5. Non-metric multi-dimensional scaling (NMDS) analysis using Bray-Curtis dissimilarities comparing the two cleanrooms (JPL and TAS), saline seep models, and HSB sediment. The JPL samples are differentiated by total community (JPL T) and PMA-treated (JPL PMA) and the DFR models are separated by their respective media (high-carbon = HC, low-carbon = LC). Alphaproteobacteria 4%), Firmicutes (14.4%), Desulfobacterota (4.8%), in high-carbon MgSO4 medium (Fig. 6B). Chloroflexota (4.2%), Deinococcota (3.2%), Caldatribacteriota (1.8%), Cultured members of the DFR biofilm communities were identified and Actinobacteria (2.3%, Table S2). In the high-carbon experiment, the by submitting full-length 16S Sanger sequencing to NCBI BlastN. Of the biofilm community consisted of 365 ZOTUs representing 9 distinct phyla 24 colonies submitted for sequencing, nine were identified as Pseu- dominated by Gammaproteobacteria, Bacilli, Clostridia, Alphaproteobac- doalteromonas sp., eight were Halomonas sp., four were Yersinia sp., two teria, and Actinomycetia classes (Fig. 6A). In the low-carbon experiments, were Bacillus sp. and one Serratia sp. (Table 3). Thirteen of these cultured we detected 404 ZOTUs representing 22 phyla, similarly dominated by organisms had matches at or above 97% identity amongst the uncul- Gammaproteobacteria, Bacilli, and Actinomycetia classes. Notably, in the tured 16S amplicon sequences. The Sanger read MP3.2 identified as a low-carbon experiments the genus Halomonas (Gammaproteobacteria) Halomonas sp. was most closely related to H. alkaliphila, and ZOTUs 1 accounted for 56 ± 1.8% of the community on average, compared to 26 and 2 (H. titanicae, H. alkaliphila, or H. meridiana) matched with 100% ± 0.66% in the high-carbon experiments (Fig. 6B). In the high-carbon identity although Zotu1 had one single nucleotide polymorphism (SNP) experiments additional comparisons were made between the media between its 253 nt read and the 937 nt Sanger sequence. ZOTU 132 was (dominant MgSO4 or NaCl) with the main differences between the two also identified as a match to MP3.2 at 97% identity and was identified as media being greater presence of genera Pseudidiomarina and Microbulifer H. salifodinae. An additional 16 ZOTUs matched with 97% identity to 6 M.K. Mettler et al. B io fi lm 5 (2023) 100127 Fig. 6. Relative abundance of A) the 14 highest relative abundance classes across the high-carbon experiment (HC), low-carbon experiment (LC) and Hailstone Basin soil (HSB). B) The 10 highest relative abundance genera present in each of the three media tested in the DFR. Sanger sample MP2.1 and were primarily identified as either H. arcis or flexibility with channel-to-channel treatments, and containment of H. gomseomensis. Zotu3 (Bacillus subtilis or Bacillus atrophaeus) matched substrate to model Martian regolith seeps. We used a simple, with 100% identity to MP1.2 with an additional 17 ZOTUs matching to proof-of-concept experiment design to eliminate confounding factors MP1.2 at 97–99.6% identity and most likely belonged to B. subtilis or such as the influence of substratum on biofilm formation that would Metabacillus halosaccharovorans. Three ZOTUs (7, 9 and 17) matched to occur using a more complicated regolith simulant. The model was kept MP3.1 (Pseudoalteromonas translucida). Lastly, seven ZOTUs matched to at room temperature as temperatures up to 22 ◦C are found on the MP5.2_merged (Serratia rubidaea) at 97.2–100% identity and while surface of Mars during a summer sol [10,11]. The high-carbon experi- three were likely Serratia sp., the other four were identified as Escher- ment was designed as a baseline for model development to compare ichia coli, Lonsdalea quercina, Franconibacter daqui, and Pantoea vagans. several aspects that are strong drivers of microbial growth without being Sanger reads 10.3, 10.5, 3.1, 5.1, 8.2, 9.1, and 9.2 also matched to carbon-limited to confirm biofilm formation. Biofilm density of the various ZOTUs belonging to Halomonas sp., Pseudoalteromonas sp. and high-carbon experiments was above 9.5 log cells/cm2 10 for several Yersinia sp. but were not matched across the full length of the V4 region channels at various time points and biofilms were visible on the sand (App. 1). grains to the naked eye (Fig. 4B). As expected, microscopy counts were Visualization of the ZOTUs in a phylogenetic tree confirmed the higher than the plate counts, since not all organisms are culturable on compositional trends of the model communities and indicated the Gelrite plates. Further, plates were incubated at ambient laboratory presence of overlapping ZOTUs between the models and the cleanroom conditions (exposed to O2), so it is likely that the microorganisms datasets (Fig. 7). The 16S Sanger sequences were also mapped to the tree cultured on plates were only a subset of the biofilm present in the and confirmed the BLASTn analysis of their likely taxonomies as Hal- channels under CO2. omonas sp., Pseudoalteromonas sp., Bacillus sp. and several Yersinia sp. Once biofilm formation was confirmed with the high-carbon exper- (Table S4). Venn diagram analysis of the models vs. the cleanroom iment, the low-carbon medium was designed to provide carbon at a level ZOTUs confirmed the overlap of five ZOTUs between the original HSB more comparable to known Martian regolith. In 2022, it was reported sediment, the two DFR models, and the two cleanroom datasets that Curiosity rover detected organic carbon in concentrations as high as (Fig. S8A) which were classified in Fig. S8B as Halomonas titanicae, 730 μg C/g [51]. In both experiments, the reactor media were designed Actinotalea sp., Ralstonia solanacearum, Paracoccus marinus, and Preistia to mimic possible chemical conditions of Martian saline seeps. Sulfates endophyticus. Interestingly, an additional five ZOTUs were found to such as magnesium sulfate, including hydrated forms, specifically overlap between the models and both cleanrooms that were not found in epsomite, have been detected on Mars which could lead to the formation the HSB sediment which included Bacillus subtilis, Staphylococcus aureus, of dense brines [52–54]. Additionally, potassium ions have been Cutibacterium acnes, Micrococcus aloeverae, and Rhodococcus qingshengii. detected via remote sensing on Mars, and potassium chloride has been Additional shared taxa between the datasets are summarized in Ap- used in other laboratory models of Martian salts [55,56]. MgSO4-do- pendix 2. minated and NaCl-dominated media were compared due to the chaot- ropic nature of high levels of Mg2+ ions which we expected to be more 4. Discussion challenging to microbial growth [57]. Regardless of media formulation, each of the DFR biofilms achieved high steady state densities quickly 4.1. Biofilm accumulation indicates a promising reactor design after the start of continuous flow. Drip flow reactors were designed to grow low-shear biofilms and are often used for modeling medical and wound biofilms [78,79]. To our 4.2. Halophiles dominate the DFR model biofilms knowledge this study is the first to utilize a DFR to model Martian terrestrial conditions, an important first step in adapting this technology Culturing conditions strongly influenced biofilm community for astrobiological efforts. The DFR allows for long residence times, composition, with the communities separating by high-carbon or low- carbon and dominant medium salt. All DFR model biofilms were 7 M.K. Mettler et al. B io fi lm 5 (2023) 100127 Fig. 7. Phylogenetic tree of cleanroom and model seep taxa. Colored strips represent ZOTUs present in each dataset. From inner to outer ring: Reference tree taxa, TAS cleanroom, JPL cleanroom, low-carbon model seep, high-carbon model seep. The high-carbon model is further separated by taxa present in only the NaCl- dominated medium (light blue), MgSO4-dominated medium (bright red) or both (purple). Culturable organisms identified by Sanger sequencing from biofilm col- onies in the high-carbon experiment are indicated by stars (red) at branch ends. (For interpretation of the references to colour in this figure legend, the reader is referred to the Web version of this article.) dominated by Halomonas sp. with additional matches to Pseudoalter- P. haloplanktis, P. shioyasakiensis, and P. pyrdzensis. P. haloplanktis, omonas sp.. Halomonas sp. are almost uniformly halophilic and are originally isolated from Antarctic seawater, has been a focal species for known for producing sulfate-rich exopolysaccharides which may also understanding microbial adaptation to cold temperatures and is a explain their robust growth in the MgSO4-dominated media [58]. known biofilm producer [63]. Several likely species of Halomonas were detected including the highly Bacillus subtilis was the most common Bacillus species detected in the halotolerant H. titanicae, H. alkaliphila, and H. meridiana [59,60] DFR biofilms, especially in the high-carbon experiment. Many strains of although identification at the species levels is challenging due to the B. subtilis are halotolerant and capable of forming endospores [64] high level of conservation in the V4 region of the 16S rRNA gene beyond which could facilitate their survival in cleanrooms and onto spacecraft the genus level [61]. Furthermore, the presence of several possible until potentially more favorable conditions are encountered in Martian Halomonas sp. among the culturable DFR biofilm community members saline seeps. Resuscucitation of B. subtilis endospores can be largely identified by Sanger sequencing confirms the ability of members of this prevented by Martian levels of UV radiation [65,66]. However, minimal genus to grow in the variety of conditions related to a Martian saline coverage by Martian regolith is highly protective of UV radiation thus seep environment tested here. The genus Pseudoalteromonas also consists potentially allowing for resuscitation [67]. We detected B. subtilis among of several halotolerant psychrophiles including P. haloplanktis which is both the amplicon (total) and Sanger (culturable) DFR biofilm com- considered a model cold-adapted bacterium [62]. Three ZOTUs from the munities, confirming its robust growth across conditions and cultur- DFR models were identified as matches to the Sanger read 3.1 (Pseu- ability after exposure to our hypothetical Martian conditions. doalteromonas translucida) although they were more closely identified as 8 M.K. Mettler et al. B io fi lm 5 (2023) 100127 4.3. Implications for planetary protection 4.4. Improvements for future models To prevent microbial contamination, spacecraft are assembled in Now that the concept of biofilm formation in the modified DFR with cleanrooms which are facilities with strict air quality standards main- non-standard microbial communities has been demonstrated, the model tained via control of airborne particles, temperature, and humidity [68]. can be improved by incorporation of more relevant Martian conditions. There are varying classes of cleanrooms which are defined by the Though Martian regolith has varying compositions, a laboratory simu- number of airborne particulates present. The International Standards lant should be employed with a more defined chemistry, specifically Organization (ISO) sets cleanroom standards to which both the National including iron. Further, our initial model experiments were conducted at Aeronautics and Space Administration (NASA) and the European Space room temperature, representing the upper end of the surface tempera- Agency (ESA) adhere [4–6]. Particulates are controlled with high effi- tures found on Mars. For greater accuracy, future experiments should ciency particulate air (HEPA) filters, sticky mats, and suits that users are take place in a refrigerator or freezer, with cyclic temperature changes. required to wear. Even with such measures in place, particulates, Such changes would likely select for psychrophilic halophiles which are including microorganisms, persist in cleanrooms [6–8]. Cleaning and hypothesized to be the most capable of survival in Martian conditions. disinfecting procedures differ between organizations resulting in vary- The DFR biofilms were exposed to either ambient atmosphere or fed ing microbial communities between cleanrooms. Further, cleaning pure CO2 while the Martian atmosphere is composed of 96% CO2 with Ar procedures of spaceflight hardware vary between organizations as dis- and N2 being the next most abundant at about 2% each and other trace cussed by Venkateswaran et al. and can include the use of 70% iso- compnents [36]. The inclusion of Ar and N2, as well as the use of tubing propanol wipes (NASA) or multiple-solvent cleaning (JPL) [69]. not permeable to oxygen would increase relevance to the Martian at- Having verified the successful cultivation of biofilm within the DFR mosphere. Additionally, there was no UV stress included in these ex- under various conditions, we then determined whether there was periments. Perhaps one of the greatest barriers for terrestrial life overlap between the DFR biofilm communities and taxa that have been proliferating on Mars is the high flux of UVB and UVC, often considered detected in spacecraft assembly cleanrooms. Halophiles Halomonas entirely sterilizing for Earth microorganisms [76]. Though the pores of titanicae and Planococcus marinus were found to overlap between the the regolith could offer some protection from such radiation [77], DFR models and the cleanroom datasets. Importantly, H. titanicae was including UV dosage in future experiments would create a more identified in both the PMA-treated (viable) and total community of JPL powerful and realistic model. samples, and was detected in both the total and culturable communities Future experiments should also carefully consider the methods of of the DFR biofilms (Fig. 7). This suggests that not only is H. titanicae inoculation and reactor sampling. While we throroghly vortexed the capable of contaminating and remaining viable in cleanrooms, but also inoculum prior to inoculation, is is possible that some organisms has significant potential for survival in Martian brines. B. subtilis was remained attached to sediment particles and were not inoculated into also found to overlap between the models and the cleanrooms although the reactor. We also detected several taxa in the DFR biofilms that did it was not detected in the original HSB sediment. The failure to form not overlap with the original HSB sediment. These community members biofilm or successfully sequence the uninoculated DFR control channel could have been present below the level of detection in HSB sediment or suggests that these organisms were likely truly present within the could have been introduced to the reactor during sampling as the reactor reactor conditions and not introduced via contamination during the was not placed in a biosafety cabinet. We also suggest including an sequencing process. Due to the low biomass of organisms in cleanrooms expanded library of terrestrial inoculum and/or targeted studies of and subsequent low sequencing depth, there are likely additional un- specific taxa. discovered taxa in cleanrooms. Together this evidence suggests a need for preventative measures 5. Conclusions against halophiles and endosporulators in preparation for space flight as (i) they have previously been detected in cleanrooms suggesting they Parallel experiments using a high-carbon Martian saline seep analog may have high potential for spacecraft contamination and, (ii) they have were completed to confirm feasibility of the reactor system and growth potential for proliferation in Martian seep conditions. Many of the of HSB organisms in the DFR. Subsequent experiments with a low- overlapping taxa found in both the cleanrooms and the DFR experiments carbon saline seep analog containing media with low carbon availabil- are unlikely to survive the Martian conditions not tested in the DFR, ity similar to that detected on Mars were completed. Biofilm accumu- namely low temperatures and UV radiation. However, the presence of B. lation occurred in all three media tested and biofilms reached steady subtilis and additional Bacillus spore-formers provide potential forward state densities within several weeks of starting the experiments. The contamination routes. Further, the lower temperature limit for medium composition was the greatest driver of the resulting biofilm H. titanicae is 4 C for consistent growth which is within the known community composition. Light and atmospheric conditions were not Martian temperature range [10,11,70,71]. H. titanicae resistance to UV observed to affect community composition. Additionally, several taxa radiation is unknown, however, the biofilm morphology can exhibit were present in the reactor experiments that have been detected by greater tolerance to UV radiation and other environmental stresses than sequencing in spacecraft assembly clean rooms. Microbes in or on individual cells of the same species [72]. While these organisms are not spacecraft (built in assembly clean rooms) have the potential to guaranteed to survive space travel, let alone the conditions found on contribute to forward contamination. The overlap between cleanroom Mars, these models can help inform cleanroom sterilization processes by and DFR biofilm taxa represents microorganisms that should inform illuminating the microorganisms of most significant concern. Further cleanroom sterilization practice targets to prevent the transmission of improvement of the models will provide even more accurate informa- microbes that may be capable of Martian saline seep colonization. In tion for targeted cleanroom procedures. Additional methods for the addition, future improvements of these models may help inform disinfection and elimination of halophiles in cleanrooms must be COSPAR policies on planetary protection to ensure responsible space developed. For the time being, methods from hide and fish curi- exploration. ng/preservation including exposure to alternating electrical currents, ozone, and other chemistries could be employed [73–75]. These Funding methods likely require alteration to comply with cleanroom practices and spaceflight hardware. This research was supported by the National Science Foundation Research Experience for Undergraduates program (#2050856) and a fellowship from the Montana Space Grant Consortium. 9 M.K. Mettler et al. B io fi lm 5 (2023) 100127 CRediT authorship contribution statement [14] Lauro SE, et al. Multiple subglacial water bodies below the south pole of Mars unveiled by new MARSIS data. Nat Astron 2021;5(1):63–70. [15] Orosei R, et al. Radar evidence of subglacial liquid water on Mars. Science 2018; Madelyn K. Mettler: conducted experimental design, was a primary 361(6401):490–3. manuscript author, carried out the experiments and data analysis.. [16] Chevrier VF, Rivera-Valentin EG. Formation of recurring slope lineae by liquid Hannah M. Goemann: was a primary manuscript author, carried out brines on present-day Mars. Geophys Res Lett 2012;39(21). [17] Stillman DE. Chapter 2 - unraveling the mysteries of recurring slope lineae. In: the bioinformatics data analysis. Rebecca C. Mueller: carried out the Soare RJ, Conway SJ, Clifford SM, editors. Dynamic Mars. Elsevier; 2018. p. 51–85. experiments and data analysis. Oscar A. Vanegas: conducted experi- [18] McEwen AS, et al. Mars: abundant recurring slope lineae (RSL) following the mental design; carried out the experiments and data analysis. Gabriela planet-encircling dust event (PEDE) of 2018. J Geophys Res: Planets 2021;126(4): Lopez: conducted experimental design; carried out the experiments and e2020JE006575. [19] McEwen AS, et al. Seasonal flows on warm martian slopes. Science 2011;333 data analysis. Brent M. Peyton: conducted experimental design, carried (6043):740. out the experiments and data analysis; all authors contributed to editing [20] Möhlmann D, Thomsen K. Properties of cryobrines on mars. Icarus 2011;212(1): and revising the manuscript. Nitin Singh: contributed to conceptual 123–30. [21] Abotalib AZ, Heggy E. A deep groundwater origin for recurring slope lineae on plan for early data analysis and to editing and revising the manuscript. Mars. Nat Geosci 2019;12(4):235–41. Kasturi Venkateswaran: contributed to conceptual plan for early data [22] Rettberg P, et al. Planetary protection and mars special regions—a suggestion for analysis and to editing and revising the manuscript. updating the definition. Astrobiology 2016;16(2):119–25. [23] Flemming H-C, Neu TR, Wozniak DJ. The EPS matrix: the “House of biofilm cells”. J Bacteriol 2007;189(22):7945–7. [24] Donlan RM. Biofilms: microbial life on surfaces. Emerg Infect Dis 2002;8(9): Declaration of competing interest 881–90. [25] Esbelin J, Santos T, Hébraud M. Desiccation: an environmental and food industry stress that bacteria commonly face. Food Microbiol 2018;69:82–8. The authors declare the following financial interests/personal re- [26] Roberson EB, Firestone MK. Relationship between desiccation and lationships which may be considered as potential competing interests: exopolysaccharide production in a soil Pseudomonas sp. Appl Environ Microbiol Madelyn Mettler reports financial support was provided by Montana 1992;58(4):1284–91. [27] Mah T-FC, O’Toole GA. Mechanisms of biofilm resistance to antimicrobial agents. Space Grant Consortium. Gabriela Lopez reports financial support was Trends Microbiol 2001;9(1):34–9. provided by National Science Foundation. Oscar Vanegas reports [28] Yin W, et al. Biofilms: the microbial "protective clothing" in extreme environments. financial support was provided by National Science Foundation. Int J Mol Sci 2019;20(14). [29] Elasri MO, Miller RV. Study of the response of a biofilm bacterial community to UV radiation. Appl Environ Microbiol 1999;65(5):2025–31. Data availability [30] Beblo-Vranesevic K, et al. Surviving Mars: new insights into the persistence of facultative anaerobic microbes from analogue sites. Int J Astrobiol 2022;21(2): 110–27. Data will be made available on request. [31] Volger R, et al. Mining moon & mars with microbes: biological approaches to extract iron from Lunar and Martian regolith. Planet Space Sci 2020;184:104850. Acknowledgments [32] Crisler JD, et al. Bacterial growth at the high concentrations of magnesium sulfate found in Martian soils. Astrobiology 2012;12(2):98–106. [33] International A. Astm e2647-20: standard test method for quantification of We would like to thank the US Fish and Wildlife Service for allowing Pseudomonas aeruginosa biofilm grown using drip flow biofilm reactor with low us to sample at Hailstone National Wildlife Refuge (permit #21-006). shear and continuous flow. 2020. [34] Miller MR, et al. Saline seep development and control in the north American great plains - hydrogeological aspects. Agric Water Manag 1981;4(1):115–41. Appendix A. Supplementary data [35] Lewis BD, Custer SG, Miller MR. Saline-seep development in the Hailstone Basin, northern stillwater county, Montana. In: Water-resources investigations report; 1979 [Reston, VA]. Supplementary data to this article can be found online at https://doi. [36] Mahaffy PR, et al. Abundance and isotopic composition of gases in the martian org/10.1016/j.bioflm.2023.100127. atmosphere from the curiosity rover. Science 2013;341(6143):263–6. [37] Caton TM, et al. Halotolerant aerobic heterotrophic bacteria from the great salt plains of Oklahoma. Microb Ecol 2004;48(4):449–62. References [38] Chen F, et al. Application of digital image analysis and flow cytometry to enumerate marine viruses stained with SYBR gold. Appl Environ Microbiol 2001; [1] COSPAR. COSPAR policy on planetary protection. 2021. 67(2):539–45. [2] Coustenis A, Kminek G, Hedman N. The COSPAR panel on planetary protection. [39] Koskinen K, et al. Microbial biodiversity assessment of the European Space 2021. p. 2232. Agency’s ExoMars 2016 mission. Microbiome 2017;5(1):143. [3] Rummel JD. Special regions in Mars exploration: problems and potential. Acta [40] Edgar RC. UNOISE2: improved error-correction for Illumina 16S and ITS amplicon Astronaut 2009;64(11):1293–7. sequencing. bioRxiv; 2016. [4] International Standards Organization. Cleanrooms and associated controlled [41] Parks DH, et al. GTDB: an ongoing census of bacterial and archaeal diversity environments — Part 1: classification of air cleanliness by particle concentration. through a phylogenetically consistent, rank normalized and complete genome- 2015. based taxonomy. Nucleic Acids Res 2022;50(D1):D785–94. [5] International Standards Organization. Cleanrooms and associated controlled [42] R Core Team R. A language and environment for statistical computing. Vienna, environments — Part 2: monitoring to provide evidence of cleanroom performance Austria: R Foundation for Statistical Computing; 2021. related to air cleanliness by particle concentration. 2015. [43] McMurdie PJ, Holmes S. phyloseq: an R package for reproducible interactive [6] Moissl-Eichinger C. Extremophiles in spacecraft assembly clean rooms. In: Stan- analysis and graphics of microbiome census data. PLoS One 2013;8(4):e61217. Lotter H, Fendrihan S, editors. Adaption of microbial life to environmental [44] Oksanen J, Kindt R, Simpson GL. The vegan Package. 2009. extremes: novel research results and application. Vienna: Springer Vienna; 2012. [45] Lahti L, Shetty S. Tools for microbiome analysis in R. e. al.. 2017. p. 231 61. [46] Katoh K, Standley DM. MAFFT multiple sequence alignment software version 7: – [7] Mahnert A, et al. Cleanroom maintenance significantly reduces abundance but not improvements in performance and usability. Mol Biol Evol 2013;30(4):772–80. diversity of indoor microbiomes. PLoS One 2015;10(8):e0134848. [47] Stamatakis A. RAxML-VI-HPC: maximum likelihood-based phylogenetic analyses [8] Hendrickson R, et al. Clean room microbiome complexity impacts planetary with thousands of taxa and mixed models. Bioinformatics 2006;22(21):2688–90. protection bioburden. Microbiome 2021;9(1):238. [48] Matsen FA, Kodner RB, Armbrust EV, pplacer. Linear time maximum-likelihood [9] Williams DR. NASA Mars fact sheet. 2021 December 23. Available from: https and Bayesian phylogenetic placement of sequences onto a fixed reference tree. ://nssdc.gsfc.nasa.gov/planetary/factsheet/marsfact.html; 2021. BMC Bioinf 2010;11(1):538. [10] Pajola M, et al. Geology, in-situ resource-identification and engineering analysis of [49] Letunic I, Bork P. Interactive Tree of Life (iTOL) v5: an online tool for phylogenetic the vernal crater area (Arabia Terra): a suitable Mars human landing site tree display and annotation. Nucleic Acids Res 2021;49(W1):W293–6. candidate. Planet Space Sci 2022;213:105444. [50] Russel J. MicEco: various functions for microbial community data. 2022. [11] Spanovich N, et al. Surface and near-surface atmospheric temperatures for the [51] Stern JC, et al. Organic carbon concentrations in 3.5-billion-year-old lacustrine Mars Exploration Rover landing sites. Icarus 2006;180(2):314–20. mudstones of Mars. Proc Natl Acad Sci USA 2022;119(27):e2201139119. [12] Smith MD, et al. The climatology of carbon monoxide and water vapor on Mars as [52] Visscher AM, et al. Growth performance and root transcriptome remodeling of observed by CRISM and modeled by the GEM-Mars general circulation model. Arabidopsis in response to Mars-like levels of magnesium sulfate. PLoS One 2010;5 Icarus 2018;301:117–31. (8):e12348. [13] Wray JJ. Contemporary liquid water on mars? Annu Rev Earth Planet Sci 2021;49 (1):141–71. 10 M.K. Mettler et al. B io fi lm 5 (2023) 100127 [53] Sheppard RY, et al. Updated perspectives and hypotheses on the mineralogy of [67] Mancinelli RL, Klovstad M. Martian soil and UV radiation: microbial viability lower Mt. Sharp, mars, as seen from orbit. J Geophys Res: Planets 2021;126(2): assessment on spacecraft surfaces. Planet Space Sci 2000;48(11):1093–7. e2020JE006372. [68] Whyte W. Cleanroom technology: fundamentals of design, testing and operation. [54] Bristow TF, et al. Brine-driven destruction of clay minerals in Gale crater, Mars. John Wiley & Sons; 2010. Science 2021;373(6551):198–204. [69] Venkateswaran K, et al. Evaluation of various cleaning methods to remove Bacillus [55] Nuding DL, et al. The aqueous stability of a Mars salt analog: instant Mars. spores from spacecraft hardware materials. Astrobiology 2004;4(3):377–90. J Geophys Res: Planets 2015;120(3):588–98. [70] Valero A, et al. Modelling the growth boundaries of Staphylococcus aureus: effect [56] Hecht MH, et al. Detection of perchlorate and the soluble chemistry of martian soil of temperature, pH and water activity. Int J Food Microbiol 2009;133(1–2): at the Phoenix lander site. Science 2009;325(5936):64–7. 186–94. [57] Seckbach J, Oren A, Stan-Lotter H. Polyextremophiles: life under multiple forms of [71] Sánchez-Porro C, et al. Halomonas titanicae sp. nov., a halophilic bacterium stress, vol. 27. Springer; 2013. isolated from the RMS Titanic. Int J Syst Evol Microbiol 2010;60(Pt 12):2768–74. [58] Bejar V, et al. Characterization of exopolysaccharides produced by 19 halophilic [72] de Carvalho C. Biofilms: microbial strategies for surviving UV exposure. Adv Exp strains of the species Halomonas eurihalina. J Biotechnol 1998;61:135–41. Med Biol 2017;996:233–9. [59] Romano I, et al. Halomonas alkaliphila sp. nov., a novel halotolerant alkaliphilic [73] Hussain SA, Sarker MI, Yosief HO. Efficacy of alkyltrimethylammonium bromide bacterium isolated from a salt pool in Campania (Italy). J Gen Appl Microbiol for decontaminating salt-cured hides from the red heat causing moderately 2006;52(6):339–48. halophilic bacteria. Lett Appl Microbiol 2020;70(3):159–64. [60] Tourova TP, et al. Genomic and physiological characterization of halophilic [74] Birbir Y, et al. Annihilation of extremely halophilic archaea in hide preservation bacteria of the genera Halomonas and marinobacter from petroleum reservoirs. salt using alternating electric current. Johnson Matthey Technol Rev 2015;59(2): Microbiology 2022;91(3):235–48. 109–19. [61] Rossi-Tamisier M, et al. Cautionary tale of using 16S rRNA gene sequence [75] Stevens DA, et al. Halomonas, a newly recognized human pathogen causing similarity values in identification of human-associated bacterial species. Int J Syst infections and contamination in a dialysis center: three new species. Medicine Evol Microbiol 2015;65(Pt_6):1929–34. 2009;88(4):244–9. [62] Parrilli E, et al. The art of adapting to extreme environments: the model system [76] Cockell CS, et al. The ultraviolet environment of mars: biological implications past, Pseudoalteromonas. Phys Life Rev 2021;36:137–61. present, and future. Icarus 2000;146(2):343–59. [63] Médigue C, et al. Coping with cold: the genome of the versatile marine Antarctica [77] Hallsworth JE. Mars’ surface is not universally biocidal. Environ Microbiol 2021; bacterium Pseudoalteromonas haloplanktis TAC125. Genome Res 2005;15(10): 23(7):3345–50. 1325–35. [78] Manner S, Goeres D, Skogman M, et al. Prevention of Staphylococcus aureus biofilm [64] Errington J, Aart LTV. Microbe Profile: Bacillus subtilis: model organism for formation by antibiotics in 96-Microtiter Well Plates and Drip Flow Reactors: cellular development, and industrial workhorse. Microbiology (Read) 2020;166(5): critical factors influencing outcomes. Sci Rep 2017;7:43854. https://doi.org/ 425–7. 10.1038/srep43854. [65] Schuerger AC, et al. Survival of endospores of Bacillus subtilis on spacecraft [79] Woods J, Boegli L, Kirker KR, Agostinho AM, Durch AM, deLancey Pulcini E, et al. surfaces under simulated martian environments:: implications for the forward Development and application of a polymicrobial, in vitro, wound biofilm model. contamination of Mars. Icarus 2003;165(2):253–76. J Appl Microbiol 2012;112(5):998–1006. https://doi.org/10.1111/j.1365-2672. [66] Osman S, et al. Effect of shadowing on survival of bacteria under conditions 2012.05264.x. simulating the Martian atmosphere and UV radiation. Appl Environ Microbiol 2008;74(4):959–70. 11