Biomarker Preservation and Survivability Under Extreme Dryness and Mars-Like UV Flux of a Desert Cyanobacterium Capable of Trehalose and Sucrose Accumulation

Unraveling how long life can persist under extreme dryness and what kind of environmental extremes can be faced by dried microorganisms is relevant to understand Mars habitability and to search for life on planets with transient liquid water availability. Because trehalose and sucrose stabilize dried anhydrobiotes, an in silico survey of the genome of the desert cyanobacterium Chroococcidiopsis sp. CCMEE 029 was performed to identify pathways for trehalose and sucrose biosynthesis. The expression of the identified genes was induced in response to desiccation, and trehalose and sucrose accumulation was detected in dried cells. This adaptation strategy enabled viability and biomarker permanence under extreme dryness and Mars-like UV flux. Chroococcidiopsis survivors were scored in 7-year dried biofilms mixed with phyllosilicatic Mars regolith simulant and exposed to 5.5 × 103 kJ/m2 of a Mars-like UV flux. No survivors occurred after exposure to 5.5 × 105 kJ/m2 although, in dead cells, photosynthetic pigments, and nucleic acids, both DNA and RNA, were still detectable. This suggests that dried biofilms mixed with phyllosilicatic Martian regolith simulant are suitable candidates to identify biosignatures embedded in planetary analog minerals as planned in the future BioSignatures and habitable Niches (BioSigN) space mission to be performed outside the International Space Station.


INTRODUCTION
Unraveling how long life can persist under extreme dryness and which environmental extremes can be faced in the dried state is relevant for long-term models of Mars habitability (Davila and Schulze-Makuch, 2016) and for searching for life on planets with transient availability of liquid water (Schulze-Makuch et al., 2017;Wilhelm et al., 2018). Moreover, according to the so called geogenetic latency hypothesis, subsurface microbes could survive planetary surface extinction and be re-exposed to the surface via geological processes, when conditions allow water to flow (Boston et al., 2019).
Indeed dryness is one of the main life-threatening factors; nevertheless, upon desiccation, a small group of taxonomically diverse organisms enter a reversible metabolic dormancy, a phenomenon known as anhydrobiosis (Crowe et al., 1992). Anhydrobiotes survive water removal as specialized structures, such as bacterial spores, cyanobacterial akinetes, and certain crustacean cysts and insect larvae, and a few among cyanobacteria, lichens, fungi, rotifers, nematodes, and tardigrades enter anhydrobiosis in the vegetative state (Alpert, 2006).
Hot and cold deserts are relevant Martian field analogs due to their aridity, temperatures, and/or geological features (Martins et al., 2017) and are considered of great interest to appreciate biosignature preservation under conditions similar to those of Mars (Aerts et al., 2020). So, if life ever existed on Mars' surface, it must have faced the presumed three main climatic stages of Mars: this beginning with a water-rich period, followed by a cold and semiarid one, and ending with the presentday arid and cold environment (Fairén et al., 2010). Because everything we know about biology we have learned from life on Earth (McKay, 2010), desiccation-tolerant microorganisms might be the best-case biologic scenario for understanding the habitability of Mars (Wilhelm et al., 2018) and identifying protective biomolecules to be used as a biomarker database (Jorge-Villar and Edwards, 2013).
Desert cyanobacteria of the genus Chroococcidiopsis, being metabolically active for a few hours per year, provide some evidence of their anhydrobiotic potential (Friedmann et al., 1993;Warren-Rhodes et al., 2006). Under laboratory conditions, they recovered after 4 and 7 years of air-dried storage (Billi, 2009;Fagliarone et al., 2017;Mosca et al., 2019) although new evidence will be provided by the 500-Year Microbiology Experiment aimed to investigate the desiccation longevity of dried Chroococcidiopsis and Bacillus subtilis spores, sealed in glass vials with silica gel beads (Cockell, 2015;Ulrich et al., 2018).
Insights into the environmental extremes that desert strains of Chroococcidiopsis can face have been revealed by challenging dried cells with the exposure to Mars simulations, either in the laboratory or in space. Under Mars laboratory simulations, dried biofilms survived 1.5 × 10 3 kJ/m 2 of a Mars-like UV flux followed by 7 years of air-dried storage (Mosca et al., 2019). During the EXPOSE-R2 space mission, dried samples survived 18-month exposure to Mars simulations in low Earth orbit outside the International Space Station (Rabbow et al., 2017). Under these conditions, samples face extreme dryness induced by a space vacuum and Mars simulated pressure combined with solar UV radiation and cosmic ionizing radiation . In the Biofilm Organisms Surfing Space (BOSS) space experiment, within dried biofilms, top cell layers shielded bottom cell layers against UV radiation (Billi et al., 2019a). Whereas in the BIOlogy and Mars EXperiment (BIOMEX) space experiment, dried cells were shielded by mixing with Martian regolith simulants (Billi et al., 2019b).
In the present work, by taking advantage of the BIOMEX and BOSS results, biofilms of Chroococcidiopsis sp. CCMEE 029 were obtained by growing cells mixed with phyllosilicatic Martian regolith simulant (P-MRS) and then exposed to Mars laboratory simulations. The hypothesis is that, within the biofilm structure, the mixing of cells with minerals guarantees better UV radiation shielding. Indeed, the identification of suitable biosignatures embedded in planetary analog minerals is one the aims of the future BioSignatures and habitable Niches (BioSigN) space experiment to be performed outside the International Space Station (de Vera and The Life Detection Group of BIOMEX/BIOSIGN, 2019).
No doubt, unraveling the biochemical mechanisms that stabilize dried Chroococcidiopsis cells will help identify key molecules for dryness adaptation that might enable survival and/or biomarker permanence under Mars-like conditions. In this scenario, a crucial role should be played by trehalose and sucrose, two non-reducing sugars accumulated by anhydrobiotes, which replace water molecules, prevent membrane phase transition, and allow cytoplasmic vitrification (Sun and Leopold, 1997;Crowe et al., 1998). Indeed vitrification is supposed to underlie Chroococcidiopsis survival after exposure to subfreezing temperatures in salt solutions as inferred for Europa's icy surface (Cosciotti et al., 2019). Trehalose is also proposed to act as free-radical scavenger, thus avoiding oxidative damage (Benaroudj et al., 2001). Indeed, in a radioresistant fungus, the overexpression of a trehalose-synthase gene yields an increased trehalose accumulation and enhanced resistance to gamma irradiation, UV light, and heavy metal ions (Liu et al., 2017). Notably, glucose addition to Bacillus subtilis spores increased survival after 6-year exposure to a space vacuum inside the NASA Long Duration Exposure Facility (Horneck et al., 1994).
Here an in silico analysis of the genome of Chroococcidiopsis sp. CCMEE 029 was performed in order to identify genes involved in the following trehalose and sucrose biosynthesis: (i) the TreY/TreZ pathway catalyzing the transglycosylation of maltodextrins into maltooligosyl trehalose and trehalose hydrolytic release; (ii) the TreS pathway transforming maltose into trehalose; (iii) the TPS/TPP pathway catalyzing the formation of trehalose-6-phosphate and its dephosphorylation into trehalose; (iv) the SPS/SPP pathway with a sucrosephosphate synthase and sucrose-phosphate phosphatase; and (v) the SUS pathway catalyzing the reversible transfer of a glucosyl moiety between fructose and a nucleoside diphosphate (for a review, see Avonce et al., 2006;Kolman et al., 2015).
The role of the identified genes in the desiccation tolerance of Chroococcidiopsis was evaluated by monitoring their expression by real-time quantitative PCR (RT-qPCR) in 10-and 60min dried cells. Trehalose and sucrose content was also determined in dried cells. After a prolonged air-dried storage (7 years), dried biofilms mixed with P-MRS and exposed to Mars laboratory simulations were investigated for viability and biomarker permanence by mean of SYTOX-Green staining and RT-qPCR, respectively.

Organism, Culture Conditions, and Desiccation
The cyanobacterium Chroococcidiopsis sp. CCMEE 029 from the Negev Desert (Israel) is maintained at the Department of Biology as part of the Culture Collection of Microorganisms from Extreme Environments (CCMEE) established by E. Imre Friedmann and Roseli Ocampo-Friedmann. The strain was grown under routine conditions at 25 • C in liquid BG-11 medium under a photon flux density of 40 µmol/m 2 /s provided by fluorescent cool-white bulbs.
Biofilms were obtained by plating about 2 × 10 8 cells mixed with 0.2 g of a Phyllosilicatic Mars Regolith Simulant (P-MRS; see Baqué et al., 2016, for regolith composition) on top of BG-11 agarized medium in Petri dishes sealed with Parafilm and allowed to grow for 2 months.
Desiccation was performed as follows: (i) over silica gel: liquid-culture aliquots were immobilized on 0.2-µm polycarbonate filters (Millipore) and stored for 10 and 60 min in a glass vacuum desiccator with anhydrous silica gel at room temperature (RT) in the dark; (ii) through air-drying: liquidculture aliquots were dried under a laminar flow hood for 24 h and stored at RT in the dark. Biofilms were allowed to air-dry for 15 days under routine conditions by removing the Parafilm from the Petri dishes (Billi et al., 2019a).

Mars Laboratory Simulations
Dried biofilms mixed with minerals were accommodated in the DLR 16-well aluminum sample carriers at the Planetary and Space Simulation facilities (Microgravity User Support Center, DLR Cologne, Germany) and exposed to Mars simulations between September 2013 and December 2013 as planned in the context of the EXPOSE-R2 space mission . The top layer of the carrier samples were exposed in triplicate to 5.5 × 10 3 kJ/m 2 , 1.4 × 10 5 kJ/m 2 , 2.7 × 10 5 kJ/m 2 , and 5.5 × 10 5 kJ/m 2 with a SOL200 lamp (1271.2 W/m 2 200-400 nm) for 1 h 12 min, 30 h, 60 h, and 120 h under Earth's atmosphere, and in the bottom layer, samples were kept in the dark (see Table 3, de Vera et al., 2019). In another simulation, samples were exposed in the top layer of the carrier in triplicate to 5.7 × 10 5 kJ/m 2 with a SOL200 lamp (1,271 W/m 2 200-400 nm) for 98 h 73 min under a CO 2 Mars-like atmosphere (at 1 kPa), and in the bottom-layer carrier, samples were kept in the dark under a Mars-like atmosphere (see Table 4, de Vera et al., 2019). After the Mars simulations, dried samples were kept in the dark at RT for about 7 years.

Identification of Genes Encoding Trehalose and Sucrose Biosynthetic Enzymes
The genome of Chroococcidiopsis sp. CCMEE 029 was sequenced by using Illumina/Solexa technology (CD Genomics, NY, USA), gene annotation was performed by using the PROKKA v.1.11 software (Seemann, 2014), and the interface was provided by Galaxy-based framework Orione (Cuccuru et al., 2014). Genes encoding trehalose and sucrose biosynthetic enzymes were identified by a BlastN (Nucleotide Query Searching a Nucleotide Database) search for nucleotide sequences of sequenced cyanobacterial orthologs as previously reported for DNA repair genes (Mosca et al., 2019).

RNA Extraction and RT-qPCR
Total RNA was extracted from each sample by using 1 ml of TRI Reagent R (Merck) and treatment with RQ1 RNase-Free DNase µl, including 1 µg of cDNA template, 6 µl of iTaq TM universal SYBR R Green supermix (BioRad Laboratories, Hercules, CA, USA), and 400 nM of the appropriate primer (Table 1). Primer specificity was confirmed by melting curve analysis. 16S rRNA (GenBank accession number AF279107) was used as a reference gene (Pinto et al., 2012). PCR cycling conditions were performed in a StepOnePlus TM Real-Time PCR System (Thermo Fisher Scientific, Waltham, MA, USA) as follows: a cycle of 95 • C for 10 min, then 40 cycles of 95 • C for 15 s, and 60 • C for 1 min, followed by a ramp from 60 to 95 • C for the melting curve stage.
(i) Gene expression in response to desiccation. Real-time reactions were performed by using total RNA extracted from 10-and 60-min dried cells and liquid cultures. Relative mRNA levels were calculated by the comparative cycle threshold (Ct) method. Primer specificity was confirmed by melting curve analysis. 16S rRNA (GenBank accession number AF279107) was used as a reference gene (Pinto et al., 2012). Values obtained for liquid control were set as one, and values of dried cells were considered to be upregulated (>1) or downregulated (<1). For each gene target, n ≥ 3 qPCR reactions were conducted, each one in duplicate. (ii) Permanence of 16S rRNA and mRNA in dried biofilms exposed to Mars laboratory simulations. RT-qPCR was performed by using total RNA extracted from 7-yeardried biofilms mixed with P-MRS and exposed to Mars

HPLC Analysis
The quantitative analysis of trehalose and sucrose content was performed by reverse phase ion pair (RPIP)-HPLC48 on a Luna 5lC18 reversed phase column (150 × 4.6 mm) (Phenomenex, Torrance, CA, USA) in acetonitrile (8.5%) and tetra-n-butylammonium hydrogen sulfate (1.2 mM; Fluka, St. Louis, MO, USA) and by applying a stepwise gradient of 0.2 M NaCl from 1 to 62%. The HPLC system (Dionex, Sunnyvale, CA, USA) was run at a flow rate of 1.1 ml/min and the fluorescent labeling reaction was performed by post-column addition of 2-cyanoacetamide (0.25%; Sigma) and NaOH (0.25%) at a flow rate of 0.35 ml/min by a PC10 post-column pneumatic delivery package (Dionex, Sunnyvale, CA, USA). Signals were quantified by comparison with known amounts of standard disaccharides analyzed in parallel runs. Data were processed by using ThermoScientific Chromeleon Chromatography Data System (CDS) software.

Cell Viability
Cell viability was assessed by staining with the cell-impermeant nucleic acid dye SYTOX Green (Molecular Probes, Inc., Eugene, OR, US) at a final concentration of 50 µM for 5 min in the dark as previously reported (Billi, 2009). Images were taken with an Olympus FV1000 confocal laser-scanning microscope (CLSM) by exciting the cells with a 488-nm laser and collecting the emissions between 510 and 530 nm. Photosynthetic pigment autofluorescence was excited with a 635-nm laser and emissions collected from the 645-800 nm emission range.

Statistical Analysis
Experiments were carried out with at least three independent replicates, data are shown with standard deviation and significance assessed by using Student's t-test.

Genes Encoding Trehalose Biosynthetic Enzymes
The in silico analysis identified in Chroococcidiopsis sp. CCMEE 029's genome genes for trehalose biosynthesis according to TreY/TreZ and TreS pathways ( Two treS genes encoding a maltose alpha-Dglucosyltransferase (TreS) were identified in Chroococcidiopsis sp. CCMEE 029's genome ( Table 2). The treS1 had a 3393-bp open reading frame predicted to encode a long TreS and shared the highest similarity (BlastN output: query cover 99%, e-value 0.0, and total score 2,859) with the ortholog of Microcoleus sp. PCC 7113 (GenBank: CP003630). Although the treS2 contained a 1,641-bp open reading frame predicted to encode a short TreS, it showed the highest similarity (BlastN output: query cover 100%, e-value 0.0, and total score 1,657) with the ortholog of Scytonema sp. NIES-4073 (GenBank: AP018268.1). The in silico analysis did not identify any gene involved in the TPS/TPP pathway for trehalose biosynthesis.

Genes Encoding Sucrose Biosynthetic Enzymes
The in silico analysis of the genome of Chroococcidiopsis sp. CCMEE 029 identified genes for the sucrose synthesis according to SPS/SPP and SUS pathways ( Table 3) putative sucrose-phosphate phosphatase (SPP) and exhibited the highest similarity (BlastN output: query cover 99%, e-value 2e-127, and total score 469) with the gene of Gloeocapsopsis sp. AAB1 encoding a sucrose-6F-phosphate phosphohydrolase (GenBank: KJ183087.1).

Expression of Trehalose and Sucrose Biosynthesis Genes Upon Desiccation
The expression of trehalose biosynthetic genes was evaluated by RT-qPCR in 10-and 60-min dried Chroococcidiopsis cells over silica gel and compared to their expression in liquid controls ( Figure 1A). Regarding the TreY/TreZ pathway, after 10 min of desiccation, the expression of the treY and treY genes was induced 5-and 6-fold, respectively, and after 60 min, of desiccation, the two genes resulted induced 8-and 9fold, respectively ( Figure 1A). For the TreS pathway, the treS long gene showed high mRNA levels after 10 and 60 min of desiccation, whereas the expression of the treS short gene was significantly induced at both desiccation points although at a lower level compared to the treS long gene (Figure 1A).
Genes involved in sucrose biosynthesis were induced after 10 and 60 min of desiccation compared to liquid control cultures ( Figure 1B). Regarding the SPS/SPP pathway, the expression of the spp gene was induced at both desiccation points, and the mRNA levels of the sps gene increased after 10 and 60 min of desiccation. For the SUS pathway, the susB gene was more highly induced after 10 min than after 60 min of desiccation, and the mRNA levels of the susA gene increased at both desiccation points although at a lower level compared to the susB gene ( Figure 1B).

Trehalose and Sucrose Content in Dried Cells
In dried Chroococcidiopsis cells, the amount of trehalose and sucrose was quantified by HLPC analysis, and liquid cultures were used as a control (Table 4). Cells desiccated over silica gel for 3 months showed a trehalose and sucrose amount of about 1.934 and 5.387 mg/g dry weight, respectively. This value was 33and 5-fold increased from that detected in hydrated cells, which for trehalose and sucrose was about 0.059 and 1.127 mg/g dry weight, respectively. Cells air-dried for 24 h showed a trehalose and sucrose content of about 1.745 and 4.850 mg/g dry weight, respectively ( Table 4).

RNA Permanence Under Desiccation and Mars-Like UV Flux
The effects of short-term desiccation on RNA integrity was determined in 10-and 60-min dried cells Chroococcidiopsis by evaluating the cycle threshold (Ct) values in RT-qPCR assay (Figure 2), considering that a reduced RNA copy number corresponded to a high Ct value. A progressive reduction of 16S rRNA copy numbers occurred in 10-and 60-min dried cells that showed average Ct values of 23 and 25, respectively, hydrated cells had an average Ct value of 22 (Figure 2). The effects of long-term desiccation following exposure to Mars-like UV flux was determined in dried Chroococcidiopsis biofilms mixed with P-MRS and exposed to UV doses ranging from 5.5 × 10 3 kJ/m 2 to 5.5 × 10 5 kJ/m 2 (Figure 2). The absence of a reduction in the average Ct values with the increased UV doses suggests the permanence of a reduced 16S rRNA copy number (Figure 2). FIGURE 2 | Permanence of 16S rRNA in 10-and 60-min dried Chroococcidiopsis cells and in 7-year dried Chroococcidiopsis biofilms mixed with P-MRS, according to Ct values in the RT-qPCR assay. Samples: hydrated control cells; 10-and 60-min dried cells; 7-year dried biofilms mixed with P-MRS and exposed to increasing Mars-like UV radiation doses. Ct values are shown as medians (lines), 25th percentile to 75th percentile (boxes), and ranges (whiskers).

RNA Permanence Under Desiccation, Mars-Like UV Flux and Mars-Like Atmosphere
The Ct value method was used to evaluate the mRNA permanence in dried Chroococcidiopsis biofilms mixed with P-MRS and exposed to a Mars-like UV flux (5.7 × 10 5 kJ/m 2 ) combined with a Mars-like atmosphere and stored 7 years in the air-dried state. As shown in Figure 3, the four genes encoding sucrose biosynthetic enzyme were used as gene targets in determining Ct values in dried biofilms because they were previously reported to be induced upon desiccation (Figure 2B). Compared to hydrated cells, each gene showed reduced Ct value in 7-year dried biofilms exposed in the bottom-layer carrier, e.g., kept in the dark and exposed to a Mars-like atmosphere (Figure 3). When compared to dried biofilms exposed in the bottom-layer carrier, each gene showed slightly reduced Ct values in dried biofilms exposed in the top-layer carrier, i.e., exposed to 5.7 × 10 5 kJ/m 2 of a Mars-like UV flux combined with a Mars-like atmosphere (Figure 3). Similar results were obtained for the four genes encoding trehalose biosynthetic enzymes (not shown).

Cell Viability of Dried Biofilms Under Mars-Like UV Flux
In hydrated Chroococcidiopsis cells, the staining with the cellimpermeant SYTOX-Green stain revealed a few dead cells with damaged cell membranes showing SYTOX-Green stained nucleoids and red photosynthetic pigment autofluorescence ( Figure 4A). In 7-year dried biofilms mixed with P-MRS and exposed to 5.5 × 10 3 kJ/m 2 of a Mars-like UV, about 20% of the alive cells were SYTOX-Green negative ( Figure 4B). Only dead cells showing SYTOX-Green stained nucleoids and red photosynthetic pigment autofluorescence occurred in 7-year dried biofilms mixed with P-MRS and exposed to 1.4 × 10 5 kJ/m 2 and 2.7 × 10 5 kJ/m 2 (not shown) as well as to 5.5 × 10 5 kJ/m² ( Figure 4C).
FIGURE 3 | Permanence of mRNA in 7-year dried Chroococcidiopsis biofilms mixed with P-MRS, according to Ct values in the RT-qPCR assay. Control: hydrated cells under laboratory routine growth conditions; bottom: dried biofilms exposed to Mars-like atmosphere; top: dried biofilms exposed to Mars-like UV flux (5.7 × 10 5 kJ/m 2 ) and Mars-like atmosphere.

DISCUSSION
In order to investigate the role of trehalose and sucrose in survivability and biomarker preservation under Mars laboratory simulations of the anhydrobiotic cyanobacterium Chroococcidiopsis, an in silico analysis of the genome of the CCMEE 029 strain was performed. The genome survey identified genes for trehalose synthesis according to the TreY/TreZ and TreS pathways as well as genes for sucrose synthesis according to the SPS/SPP and SUS pathways. The presence of the treY and treZ genes provided Chroococcidiopsis with the capability of catalyzing a two-step reaction to convert maltodextrins (maltooligosaccharides, glycogen, and starch) into trehalose. But the absence of a trehalase gene (treH) might contribute to trehalose accumulation. Indeed, the treH gene absence was reported for two desiccation-tolerant Leptolyngbya strains (Shimura et al., 2015;Murik et al., 2017) although treH mutants of Anabaena sp. PCC 7120 showed an increased trehalose amount and enhanced desiccation tolerance (Higo et al., 2006). The presence of two treS genes, encoding a long and short trehalose synthase, provided Chroococcidiopsis with the capability of transforming maltose into trehalose, according to the TreS pathway (Klähn and Hagemann, 2011). The genome survey of Gloeocapsopsis sp. UTEX B3054 and Leptolyngbya ohadii (Murik et al., 2017;Urrejola et al., 2019) revealed the presence of two treS gene, a feature suggested to be unique of desiccationtolerant cyanobacteria (Murik et al., 2017). The long treS gene of Chroococcidiopsis sp. CCMEE 029 showed motifs encoding maltogenic amylase (IPR032091), maltokinase (IPR012811), and protein kinase (IPR011009) domains, responsible for trehalose production through glycogen degradation, that occurred in the long TreS homolog of Leptolyngbya ohadii (Murik et al., 2017).
The presence the sps and sps genes encoding a sucrosephosphate synthase and sucrose-phosphate phosphatase provided Chroococcidiopsis with the capability of using the SPS/SPP sucrose biosynthetic pathway (Kolman et al., 2015). But the presence of two sus genes encoding a sucrose synthase also suggests the presence of the SUS pathway. Remarkably, Sus-encoding genes were identified during the genome survey of a selection of heterocyst-forming cyanobacteria, and they were present in the genome of a few unicellular cyanobacteria (Kolman et al., 2012). Indeed, representatives of the Chroococcidiopsis genus and heterocyst-differentiating cyanobacteria have been reported to be each other's closest living relatives (Fewer et al., 2002).
The relevance of trehalose and sucrose accumulation in the Chroococcidiopsis adaptation to dryness was supported by the expression of the identified genes during 10 and 60 min of desiccation over silica gel. Compared to the treY gene, the expression of the treZ gene was slightly higher after 60 min of dehydration, in agreement with the trehalose hydrolytic release (Avonce et al., 2006). The expression of the TreS-short encoding gene was 2-fold higher compared to TreS-long encoding gene as previously noticed in Leptolyngbya ohadii (Murik et al., 2017). Among the identified sucrose biosynthetic, the expression of sps, sps, and susB gens was induced after 10 and 60 min of desiccation over silica, whereas the susA gene was slightly induced only in 10 min dried cells. The low expression of this sucrose degradation-encoding gene might have contributed to increased sucrose content during desiccation.
Overall, the presence of multiple biosynthetic pathways might confer the advantage of accumulating sugars relevant for dryness adaptation under changeable environmental conditions leading to limited availability of a given substrate. As a result of this genomic feature, Chroococcidiopsis cells dried over silica gel for 3 months showed a trehalose and sucrose content about 33-fold (38 nmol/mg dry weight) and 5-fold (106 nmol/mg dry weight) increased compared to hydrated cells. Cells airdried for 24 h had a trehalose and sucrose content 30-and 4-fold increased.
Indeed, a different level of desiccation tolerance might be reflected in terms of compatible solute content. For example, Desmonostoc salinum CCM-UFV059 dried on silica gel had a trehalose and sucrose content of 40 and 15 nmol/mg dry weight, respectively (Viggiano de Alvarenga et al., 2020). Anabaena sp. PCC7120 did not recover desiccation on silica gel and showed no changes in sucrose content (Viggiano de Alvarenga et al., 2020), did not recover after 8 months of air-drying (Yoshimura et al., 2006), and showed a low trehalose amount (0.05-0.1% of dry weight) and a higher sucrose accumulation (1-2% of dry weight) (Higo et al., 2006).
In Chroococcidiopsis, the accumulation of trehalose and sucrose upon desiccation might have played a key role in cell survival and biomarker permanence in 7-year dried biofilms mixed with P-MRS and exposed to Mars laboratory simulations. Survivors were identified by SYTOX-Green staining after exposure to 5.5 × 10 3 kJ/m 2 of a Mars-like UV flux (Figure 4). It was previously reported that dried biofilms survived 1.5 × 10 3 kJ/m 2 of a Mars-like UV flux (Baqué et al., 2013;Mosca et al., 2019), and dried cells mixed with Mars regolith simulants died under 5.7 × 10 3 kJ/m 2 of a Mars-like UV flux (Baqué et al., 2016). Hence, in the present work, a tight contact between cells and minerals within the biofilm structure should have guaranteed the shielding against 5.5 × 10 3 kJ/m 2 of a Mars-like UV flux. Moreover, the occurrence of live cells in 7-year dried biofilms mixed with P-MRS exposed to 5.5 × 10 3 kJ/m 2 of a Marslike UV (Figure 4), corresponding to 4 days on the Martian surface (Cockell et al., 2000), further supports the possibility that, during Mars's climatic history, a biofilm-like life form mixed with minerals could have survived a few hours' exposure to Mars UV flux while being transported from one protected niche to another (Westall et al., 2013).
On the other hand, although only dead cells occurred after 5.5 × 10 5 kJ/m 2 , they showed the permanence of biomarkers: 16S RNA as revealed by RT-qPCR (Figure 2) and chlorophyll and genomic DNA as shown by CLSM analysis (Figure 4).
The drying process caused a decrease in the 16S rRNA copy number compared to hydrated cells although no further decrease occurred in 7-year dried biofilms mixed with P-MRS and exposed to increasing Mars-like UV (Figure 2). This might be due, in addition to the biofilm structure (Baqué et al., 2013), to the UV shielding provided by P-MRS and the trehalose presence acting as a free-radical scavenger (Benaroudj et al., 2001).
Moreover, the RNA stability was further reduced by 7 years of air-dried storage and, to a larger extent, by the exposure to Marslike UV flux. In fact, 7-year dried biofilms mixed with P-MRS and exposed to a Mars-like UV flux (5.5 × 10 5 kJ/m 2 ) combined with a Mars-like atmosphere, showed an overall reduced copy number of a selection of mRNAs (Figure 3), corresponding to the induced expression of the sucrose and trehalose biosynthetic genes upon desiccation (Figure 1). For example, compared to dried samples kept in the dark, i.e., in the bottom-layer carrier of the exposure facility, a reduction of the copy number of the sps mRNA occurred in 7-year dried biofilms mixed with P-MRS and exposed to 5.5 × 10 5 kJ/m 2 , i.e., in the top-layer carrier of the exposure facility (Figure 3).
These results are relevant because 5.5 × 10 5 kJ/m 2 (200-400 nm) corresponds to the UV dose expected during 1 year of exposure in low Earth orbit, a period of time generally planned for astrobiological experiments outside the International Space Station (de Vera and The Life Detection Group of BIOMEX/BIOSIGN, 2019; de Vera et al., 2019). Moreover, because this dose corresponds to 383 sols (half-year) on the Martian surface (Cockell et al., 2000), the detectability of biomarkers, such as RNA, genomic DNA, and chlorophyll, in dead biofilms suggests that biological signals might still be preserved over a long period of time if sufficiently stabilized under dryness and protected from UV radiation.
In conclusion, this work contributes to expand our knowledge on the adaptation strategies to extremely dry conditions and suggests that sucrose and trehalose accumulation might reduce macromolecular susceptibility to chemical and physical degradation taking place after cell death (Eigenbrode, 2008). This has implications for future life detection missions on Mars; moreover, the biomarker detectability in biofilms mixed with P-MRS exposed to a Mars-like UV flux followed by 7 years of air-dried storage is of interest in the context of the future BioSigN space mission, which will be performed outside the International Space Station in order to identify suitable biosignatures embedded in planetary analog minerals (de Vera and The Life Detection Group of BIOMEX/BIOSIGN, 2019).

AUTHOR CONTRIBUTIONS
DB and J-PV supervised the study. CF performed the experiments. MB prepared biofilm samples. SC performed the HPLC analysis. AN carried out the bioinformatic analyses. DB wrote the manuscript. All authors read and approved the final manuscript.

ACKNOWLEDGMENTS
The authors thank CINECA for granting computer time (Application Code HP10CKZEGT) and Elena Romano, Centre of Advanced Microscopy P. B. Albertano, University of Rome Tor Vergata, for her skillful assistance in using the confocal laser scanning microscope. The Mars-laboratory simulations were supported by the European Space Agency for the EXPOSE-R2 space mission and performed by Elke Rabbow at DLR (Cologne, Germany).