Bacillus subtilis Spore Resistance to Simulated Mars Surface Conditions

In a Mars exploration scenario, knowing if and how highly resistant Bacillus subtilis spores would survive on the Martian surface is crucial to design planetary protection measures and avoid false positives in life-detection experiments. Therefore, in this study a systematic screening was performed to determine whether B. subtilis spores could survive an average day on Mars. For that, spores from two comprehensive sets of isogenic B. subtilis mutant strains, defective in DNA protection or repair genes, were exposed to 24 h of simulated Martian atmospheric environment with or without 8 h of Martian UV radiation [M(+)UV and M(-)UV, respectively]. When exposed to M(+)UV, spore survival was dependent on: (1) core dehydration maintenance, (2) protection of DNA by α/β-type small acid soluble proteins (SASP), and (3) removal and repair of the major UV photoproduct (SP) in spore DNA. In turn, when exposed to M(-)UV, spore survival was mainly dependent on protection by the multilayered spore coat, and DNA double-strand breaks represent the main lesion accumulated. Bacillus subtilis spores were able to survive for at least a limited time in a simulated Martian environment, both with or without solar UV radiation. Moreover, M(-)UV-treated spores exhibited survival rates significantly higher than the M(+)UV-treated spores. This suggests that on a real Martian surface, radiation shielding of spores (e.g., by dust, rocks, or spacecraft surface irregularities) might significantly extend survival rates. Mutagenesis were strongly dependent on the functionality of all structural components with small acid-soluble spore proteins, coat layers and dipicolinic acid as key protectants and efficiency DNA damage removal by AP endonucleases (ExoA and Nfo), non-homologous end joining (NHEJ), mismatch repair (MMR) and error-prone translesion synthesis (TLS). Thus, future efforts should focus on: (1) determining the DNA damage in wild-type spores exposed to M(+/-)UV and (2) assessing spore survival and viability with shielding of spores via Mars regolith and other relevant materials.


INTRODUCTION
Mars is a cold and dry planet, with intense UV (190-400 nm) and ionizing radiation in the form of galactic cosmic radiation (GCR) and solar particle events (SPE) (Guo et al., 2018). The Martian atmosphere is also highly oxidizing due to the OH radicals and oxygen atoms produced by photolysis which result in surface oxidation and the formation of O 2 , O 3 and H 2 O 2 (Gargaud et al., 2011). In addition, the Mars surface exhibits: temperature shifts from −125 • C to +20 • C; extremely low water vapor pressure (Davila et al., 2010;Fox-Powell et al., 2016); and very low atmospheric pressure. These extreme conditions are stressful to all known life forms, causing physiological, biochemical and structural damage, which can be lethal for most terrestrial organisms (Jakosky et al., 2003). At the molecular level, this damage can affect membrane lipids, proteins, RNA and, most importantly, DNA. Specific DNA damage includes single strand breaks (SSB), double strand breaks (DSB), and photolesions such as cyclobutane-type pyrimidine dimers (CPDs), 6-4 photoproducts (6-4 PPs) and the thymine dimer 5-thyminyl-5,6-dihydrothymine, commonly known as the spore photoproduct (SP) (Setlow, 2014).
Nonetheless, despite complex stress-induced damage, spores of the Gram-positive bacterium Bacillus subtilis have repeatedly demonstrated their resistance to many space-related extremes, becoming one of the model organisms in the field of Space Microbiology. Studies have shown Bacillus spores survive in extreme dryness, high levels of UV and ionizing radiation, and outer space conditions in Low Earth Orbit (LEO), where they were exposed to solar UV, high vacuum, GCR, and temperature fluctuations (Dose et al., 1995;Horneck et al., 2001Horneck et al., , 2010Nicholson and Schuerger, 2005;Fajardo-Cavazos et al., 2010;Moeller et al., 2012b).
Because of their extreme resistance, spores of B. subtilis, and other spore-forming bacteria, present a challenge for bio-sterilization in spacecraft facilities, calling for the development of new and more efficient sterilization regimens (Stapelmann et al., 2013;Khodadad et al., 2017). Bacillus subtilis spores were also shown to survive in Mars analog soils, confirming a potential forward contamination risk to Mars sites with liquid brines (Schuerger et al., 2017).
Resistance of spores to extreme conditions does not rely on one single mechanism, but rather on a combination of several strategies (Setlow, 2014). The first line of action is "damage prevention." The overall spore structure is composed of the core, inner membrane, cortex, coat, and crust layers (Figure 1), and has a wide number of properties and components that protect spores from many stress factors. Specifically, the spore core has low water content (25-55% of wet weight), due in some fashion to the spore's peptidoglycan cortex, that provides resistance to wet heat. Within the core, high levels (∼25% of core dry weight) of pyridine-2,6-dicarboxylic acid -dipicolinic acid (DPA), in a 1:1 chelate with Ca 2+ (Ca-DPA) help to protect spores from desiccation and DNA-damaging agents and maintain spore dormancy (Magge et al., 2008). The core's high levels of α/β-type small, acid-soluble spore proteins (SASP) (Magge et al., 2008) that saturate spore DNA are one of the main factors protecting spores from genotoxic chemicals, desiccation, dry and wet heat, as well as UV and γ-radiation (Mason and Setlow, 1986;Moeller et al., 2008). Moreover, the thick proteinaceous coat and crust layers, as well as the inner membrane, function as barriers to many toxic chemicals minimizing their ability to access the spore core where DNA and most spore enzymes are located. The spore coats also contain melanin-like pigments that absorb UV radiation, and there is evidence that such pigments can play a significant role in spore resistance to UV-B and UV-A radiation (Hullo et al., 2001;Moeller et al., 2008Moeller et al., , 2014Setlow, 2014).
The continuous and ongoing efforts to characterize the geochemistry, mineralogy and consequent habitability of the Martian surface (Skelley et al., 2005;Davila et al., 2010;Fox-Powell et al., 2016) have led to recent findings of the presence of water on Mars. This finding suggested that ancient Martian environments could have supported microbial life, and therefore Mars has become the focus of space exploration and life-detection studies (Grotzinger et al., 2014(Grotzinger et al., , 2015Fox-Powell et al., 2016).
To help ensure the legitimacy of life-detection studies and to prevent forward contamination, there are international planetary protection policies restricting the number of microorganisms on spacecraft surfaces, and Special Regions of Mars have been identified where proliferation of known microbes could take place (Schuerger et al., 2013;Rummel et al., 2014;Rettberg et al., 2016). Hence, it is of concern that extremely resistant microorganisms, including B. subtilis, have been detected in spacecraft-associated facilities (Venkateswaran et al., 2014;Checinska et al., 2015;Moissl-Eichinger et al., 2016), and that these organisms (and most importantly, their spores), might pose a threat to the forward contamination of surface terrains, or the search for past or present life on Mars Horneck et al., 2010;Goetz et al., 2016). In spite of its importance, there is a paucity of experimental data on the molecular mechanisms of spore survival of Earth microorganisms in the Martian environment. Consequently, if we are to design adequate planetary protection measures and prevent forward contamination, it is of utmost importance to expand our knowledge on how microorganisms are able to resist Mars' environmental conditions, and thus, potentially survive on this planet.
In the current study, a systematic screening was performed to determine if and how B. subtilis spores could survive an average day on Mars. A number of spores of B. subtilis strains FIGURE 1 | Bacillus subtilis spore structure depicting the main resistance mechanisms analyzed in the current study. Each protection (in yellow) or DNA repair (in green) mechanism is represented by a symbol. Each symbol is coupled with a small description of the gene that is mutated, the protein it codes for, followed by the main cellular event it is involved in. The location of the symbol corresponds to the main place of action within the spore. More information on the mechanisms of DNA protection, repair, dehydration, and coat assembly is provided in the section "Introduction." lacking protective elements and/or DNA repair proteins were exposed to 24 h of simulated Martian surface conditions with or without 8 h of UV radiation, and spore survival and mutagenesis were measured. The results of this study reveal the molecular mechanisms behind B. subtilis spore resistance in a Martian environment and assess the possibility of microbial contamination due to spores on the Martian surface.

Bacterial Strains, Growth, Sporulation, and Spore Purification
The two sets of B. subtilis strains used in this work are listed in Tables 1, 2, and all are isogenic with their respective wild-type strains, either PS832, PY79 or 168. One set of spores was chosen to determine the role of various spore protection mechanisms, including SASP, Ca-DPA, the spore core hydration level and the spore coat and crust, in spore survival (Table 1); the other set was used to study the importance of different DNA repair mechanisms ( Table 2).

Sample Preparation
Spore suspensions were prepared in sterile distilled water such that a 50 µL aliquot contained 5 × 10 8 spores. Each sample for Antibiotic resistance: Cat R , resistance to chloramphenicol (5 µg mL −1 ); Erm R , resistance to erythromycin (2 µg mL −1 ); Neo R resistance to neomycin (10 µg mL −1 ); Spc R , resistant to spectinomycin (100 µg mL −1 ); Tet R , resistance to tetracycline (10 µg mL −1 ). exposure was prepared by applying 50 µl of spores onto a 10 mm × 20 mm aluminum coupon (Model M4985, Seton, Inc., Branford, CT, United States) to ensure that the spores spread homogenously on the coupons by complete covering of the surface, yielding spore multilayer samples with a thickness of ∼25 spore layers (Tauscher et al., 2006). In our study, coupons were chosen to simulate surface materials of a spore-contaminated spacecraft. Each set of spore samples was tested in three replicates of each genotype with the same spore concentration. Spore samples were air-dried under ambient laboratory conditions (20 • C, 33 ± 5% relative humidity) for 1 day prior to exposure to simulated Mars surface conditions.

Spore Exposure in the Mars Simulation Chamber
Spore-inoculated coupons were exposed for 24 h to simulated Martian conditions in a cylindrical Mars Simulation Chamber  Table 3. Regarding irradiation conditions, the 8 h of radiation exposure represents a worst-case scenario for high UV flux (note that no ionizing radiation was simulated), and thus likely to give the maximum UV effects on B. subtilis spores under Martian conditions. In parallel, two additional sample sets were prepared; one was stored for the same time under ambient laboratory conditions (Earth atmosphere, pressure, room temperature, and protected from light) and the remaining sample set was stored at 4 • C in a refrigerator. The MSC was developed as part of an ongoing series of Mars astrobiology and planetary protection projects, and has been described previously (Schuerger et al., , 2011.
of the spores (data not shown). The PVA procedure has no geno-or cytotoxic effect on the spore viability (Horneck et al., 2001). Spore survival was determined from serial dilutions in distilled water as colony-forming units after incubation overnight at 37 • C on nutrient broth (NB) agar plates (Difco, Detroit, MI, United States) (Moeller et al., 2007b(Moeller et al., , 2010. Spore survival was determined by observing standard colony formation of macroscopic visible colonies on NB agar containing the appropriate selective antibiotic, as described above (Horneck et al., 2001). The relative sensitivity of spores of each mutant strain was determined with respect to that of the corresponding wild-type spores, and in some cases with splB spores, results were compared statistically using the Student's t-test and differences with P-values of ≤0.05 were considered statistically significant.

Detection of Sporulation Deficiency
To verify mutation induction caused by exposure to Martian conditions, 250 B. subtilis colonies arising from survivors of each Martian exposure tested were picked and streak-purified on SSM-agar plates solidified with 1.5% agar, containing the appropriate antibiotic(s), and incubated at 37 • C for 7 days. Sporulation deficiencies were determined visually by changes in colony morphology and pigmentation. Sporulated B. subtilis colonies show brownish pigmentation after extended incubation on sporulation plates, whereas a decrease in pigmentation and a translucent appearance are characteristic of asporogenous or Spo B. subtilis mutants (Piggot and Coote, 1976;Hullo et al., 2001;Fajardo-Cavazos et al., 2005). The frequency of Spo − mutants was expressed as the ratio of the Spo − colonies to the total 250 colonies picked after 7 days of incubation on SMM plates. To verify the Spo − mutation rates, plate from spores that had been exposed in colonies were individually transferred into 5 mL of SSM media and incubated for 24 h at 37 • C. Sporulation was then induced by diluting the overnight culture 1:100 into 5 mL of SSM medium. To determine the number of spores formed, after 24 h of cultivation, appropriate dilutions of cultures were plated on NB agar before and after a heat-shock (80 • C; 10 min) to kill growing or sporulating cells but not spores, as described (Maughan et al., 2007). Each analysis of the selected Spo − mutants was repeated at least three times.

Numerical and Statistical Analysis
The surviving fraction of B. subtilis spores was determined from the quotient N/N 0 , with N = the number of colony-forming  . c Gas composition in the MSC system was ordered from Boggs Gases, Inc. (Titusville, FL, United States) as a commercial mixture of the top five gases in the Martian atmosphere (see Schuerger et al., 2008).
units (CFU) of the Mars-exposed sample and N 0 that of the untreated controls. The Spo − mutant frequencies from the control and M(+/−)UV exposed spores were determined from three replicate samples. The frequency of Spo − mutations in samples induced by exposure to the M(+/−)UV conditions was determined as [M/N -m S ], with M = the total number of mutants from the exposed samples; N = 250; and m S = frequency of spontaneous Spo − mutations in unexposed samples. The sporulation frequency of the induced asporogenous mutants was determined by dividing the CFU after heat shock (spores) by the CFU before heat shock (growing/sporulating cells and spores). The data shown are expressed as averages ± standard deviations, and results were compared statistically using the Student's t-test. Values were analyzed in multigroup pairwise combinations, and differences with P-values of ≤0.05 were considered statistically significant (Moeller et al., 2005(Moeller et al., , 2006(Moeller et al., , 2007a(Moeller et al., ,b, 2008Horneck et al., 2008).

RESULTS
To know which spore components and molecular mechanisms are involved in B. subtilis spore resistance to simulated Mars surface conditions, two sets of B. subtilis spores were exposed to a simulated Martian atmospheric environment with or without 8 h of UV radiation (M(+/−)UV). The first set comprised spores deficient in spore protective components (Table 1), and the second set comprised spores deficient in various DNA repair mechanisms ( Table 2). A summary registering which mutant genotypes, and respective missing mechanisms of protection or repair, revealed the highest and/or lowest sensitivity to M(+/−)UV tested conditions is presented in Figure 2.

Spore Protection
When exposed to both M(+/−)UV conditions, B. subtilis spores lacking proteins responsible for spore coat assembly were significantly more sensitive than wild-type spores (Table 4 and Supplementary Figures S1-S3). The outer and inner spore coats provided significant protection against the Martian environment, with cotE PY79 spores, lacking the outer coat, being less sensitive [15-fold M(+)UV and 18-fold M(−)UV] than safA spores, lacking the inner spore coat [∼240-fold M(+)UV and 63-fold M(-)UV], when compared with the wild-type spores. Spores lacking both outer and inner spore coat layers (cotE safA spores) exhibited astonishing increases in sensitivity of ∼1000-fold in M(+)UV, and ∼200-fold in M(−)UV, compared to wild-type spores ( Table 4 and Supplementary Figures S1-S3).
Despite the striking effects of inner and outer coat defects on spore resistance to M(+/−)UV, the loss of the spore crust layer (cotVW, and cotX cotYZ spores) had no significant effects on spore survival under the tested conditions (Table 4 and Supplementary Figures S1-S3).
A second group of crucial protective components in spores is the α/β-type SASP that saturate spore DNA and protect it from damage. Spores lacking SASP-α and -β (sspA sspB spores) are thus lacking ∼80% of the α/β-type SASP pool (Hathout et al., 2003). When exposed to M(+/−)UV sspA sspB spores had increased sensitivity when compared with the wild-type, being significantly more sensitive to M (+)UV (273-fold, with a P-value of 0.0015) than to M(−)UV (17-fold, with a P-value of 0.0021) ( Table 4 and Supplementary Figures S4-S7). Interestingly, sspE spores, which lack the most prominent SASP, SspE), had no significant effect on spore survival in both M(+/−)UV (with a P-value of 0.4936, same as wild-type), but had increased sensitivity when additionally lacking SASP-α and -β (sspE sspA sspB spores). Results show sspE sspA sspB spores with 435-fold and 39-fold sensitivity in M(+)UV (with a P-value of 0.0012) and M(−)UV (with a P-value of 0.0013), respectively, when compared with wild-type spores ( Table 4 and Supplementary Figures S4-S7).
A third spore protective factor is the low water content in the spore core. Spores with higher core water content (dacB, and sleB spoVF spores) exhibited lower resistance to conditions M(+/−)UV, when compared to wild-type spores ( Table 4 and Supplementary Figures S4-S7). Notably, spores lacking α/β-type SASP and either DacB (dacB sspA sspB spores, with a P-value of 0.0086) or CaDPA (sleB spoV sspA sspB spores, with a P-value of 0.0053) were more sensitive to the Martian environment than either dacB or sleB spoVF spores. Results also show that addition of DPA to the sporulation medium suppressed sleB spoVF spores' decreased resistance while sporulating, reaching near wild-type survivability levels ( Table 4 and Supplementary Figures S4-S7).
Spores lacking an outer coat with an additional SASP deficiency (cotE sspA sspB spores, with a P-value of 0.0052), were more sensitive to Mars conditions than spores lacking either protective component alone (cotE, and sspA sspB spores) ( Table 4 and Supplementary Figures S4-S7). An additional deficiency in Ca-DPA (cotE sleB spoVF sspA sspB spores), and consequent higher core water content, resulted in rapid FIGURE 2 | Sporulation deficiency (in %) of B. subtilis spores deficient in protection mechanisms exposed to simulated Martian conditions, measured as Spocolonies per 250 colonies of survivors of DNA repair deficient spores exposed to M(+)UV (white bars) or M(-)UV (gray bars). ( * ) depicts significance after paired t-test P < 0.05, when compared with the respective wild-type. Data are expressed as averages and standard deviations. killing with a 10 5 -fold, in M(+)UV and 10 4 -fold in M(−)UV, greater sensitivity compared with the wild-type (with P-values of 0.0001 or 0.0001, respectively). However, the effects of the sleB spoVF mutations were again suppressed when these spores were prepared with DPA added to the sporulation medium with 10 3 -fold greater sensitivity compared with the wild-type in M(+)UV and 646-fold in M(−)UV (Table 4 and Supplementary Figures S4-S7).

Spore DNA Repair
Bacillus subtilis spores rely on a complex network of mechanisms to repair DNA damage accumulated during periods of dormancy, and ensure genomic integrity. When spores were exposed to M(+)UV, SP lyase deficient spores (splB spores, with a P-values of 0.0004) were ∼300-fold more sensitive than wild-type spores, whereas spores lacking NHEJ (ligD ku, with a P-values of 0.0049) or HR (recA, with a P-values of 0.0037) were only ∼35 and ∼80-fold more sensitive than wild-type spores ( Table 5 and Supplementary Figures S8-S12). A number of single or double mutations in other DNA repair genes resulted in smaller amounts of sensitization of spores to M(+/−)UV, including exoA nfo, uvrAB, mfd, sbcDC, polY1 polY2, and mutSL mutations. Mutation of the disA gene (lacking DNA integrity scanning protein) had only minimal (but not significant) effects on spore survival in M(+/−)UV reaching near wild-type levels of survivability (with a P-value of 0.0943). Sensitivity of recA and ligD ku mutant spores was revealed to be in the same order of magnitude in both tested environments M(+/−)UV, being of ∼80-to 90-fold for recA spores in M(+)UV, and ∼30-fold for ligD ku in M(+)UV and M(−)UV (Table 5 and Supplementary Figures S8-S12).
Analysis of the M(+/−)UV survival rates of splB spores additionally lacking other DNA repair genes, revealed that almost all DNA repair mutations caused significant increases in spore sensitivity, when compared to that of the splB single mutant spores. Spores lacking SP lyase (SP repair) and strand specific DNA repair (mfd splB spores, with a P-value of 0.0001) as well as spores lacking SP repair and NER (uvrAB splB TABLE 4 | Spore surviving fraction and increased sensitivity of mutant spores lacking protection mechanisms exposed to M(+)UV or M(−)UV.

Surviving fraction
Increased sensitivity compared to wild-type spores (fold)

Protective component M(+)UV M(−)UV M(+)UV M(−)UV
Wild-type (wt, PS832) (6.6 ± 0.8) × 10 −2 (7.3 ± 0.1) × 10 −1 1.0 ± 0.1 1.0 ± 0. ( * ) DPA supplementation during sporulation. + Statistically significant different from values for wild-type spores (P ≤ 0.05); individual P-values are given in brackets below the initial values. # Statistically significant difference between values for these mutant spores compared to values for sspA sspB spores (P ≤ 0.05); individual P-values are given in brackets below the initial values. The surviving fraction was determined after a 24 h exposure to M(+/−)UV relative to that of control spores of each genotype, which were stored in air at room temperature (20 ± 2 • C), at relative humidity of 40 ± 5% and protected from UV radiation. Increased sensitivity was determined relative to the respective wild-type spores as the ratio of the surviving fraction of wild-type over the surviving fraction of the various mutant spores. Three biological replicates were analyzed for each condition.
spores, with a P-value of 0.0001) exhibited dramatic increases in M(+)UV sensitivity of 60-and 250-fold, respectively (Table 5 and Supplementary Figure S13).

Sporulation Deficiency
When assessing sporulation deficiency through mutagenesis in survivors of spores of various strains after M(+/−)UV exposure, 5 | Spore surviving fraction and increased sensitivity of mutant spores lacking DNA-repair proteins exposed to M(+UV) and M(-UV).

Survival fraction Increased sensitivity compared to wild-type spores
Increased sensitivity compared to splB spores

DNA repair M(+)UV M(−)UV M(+)UV M(−)UV M(+)UV M(−)UV
Wild-type (wt, 168) (3.6 ± 0.7) × 10 −2 (7.1 ± 0.9) × 10 −1 1.0 ± 0.2 1.0 ± 0.1 n.a. n.a. n.a., not applicable. + Statistically significant difference from values for wild-type spores (P ≤ 0.05); individual P-values are given in brackets below the initial values. # Statistically significant difference between values for these mutant spores compared to values for splB spores (P ≤ 0.05); individual P-values are given in brackets below the initial values. The surviving fraction was determined after 24 h exposure to M(+/−)UV relative to that of control spores of each genotype, which were stored in air at room temperature (20 ± 2 • C), at relative humidity of 40 ± 5% and protected from UV radiation. Increased sensitivity was determined relative to the respective wild-type or splB spores as the ratio of the surviving fraction of wild-type or splB spores over the surviving fraction of the various mutant spore. Three biological replicates were analyzed for each condition.
∼3% of survivors of wild-type spores exposed to M(−)UV had accumulated Spo − mutants. Under Mars(+)UV conditions, spore components exhibited importance in mutagenesis, in order from highest to lowest frequency: major SASPs > intact spore coats > DPA > reduced spore core water. Spo − ratio (calculated as in sections "Detection of Sporulation Deficiency" and "Numerical and Statistical Analysis") was similar to that for disA, ywjD, and splB spores exposed to M(−)UV. Survivors of all other DNA repair mutant strains exposed to M(−)UV exhibited increased mutagenesis, from ∼20% in ligD ku spores to ∼30% in exoA nfo and mutSL spores. As expected, levels of Spo − mutants were increased with M(+)UV exposure of spores FIGURE 3 | Sporulation deficiency (in %) of B. subtilis spores deficient in DNA repair mechanisms exposed to simulated Martian conditions, measured as Spocolonies per 250 colonies of survivors of DNA repair deficient spores exposed to M(+)UV (white bars) or M(-)UV (gray bars). ( * ) depicts significance after paired t-test P < 0.05, when compared with the respective wild-type. Data are expressed as averages and standard deviations.
of all DNA repair mutants, reaching ∼60% in mutSL spores that lack the ability to repair DNA via MMR (Figures 2, 3).
Additional inactivation of splB with other repair mechanisms did not result in higher mutagenicity neither under UV exposed or UV-shielded conditions.

DISCUSSION
Because of our extensive understanding of the genetics and molecular biology of B. subtilis spore protection and repair mechanisms, these spores are of great value in investigating spore resistance to extreme environments, methods for sterilization and disinfection, and in verifying planetary protection protocols. B. subtilis spore survival in a simulated Martian surface environment is dependent on complex systems that rely on two different strategies: "damage prevention" and "damage repair." The current study demonstrates that, when exposed to M(+)UV, B. subtilis spore survival was dependent on the ability to maintain spore core dehydration; to effectively protect spore DNA through binding of major α/β-type SASP, and that spore damage by Martian UV generates primarily SP. However, when FIGURE 4 | Major factors involved in B. subtilis spore resistance to simulated Mars surface conditions. The main mutant genotypes (left) and missing mechanisms of protection or repair (right) are presented, providing a comparison between B. subtilis spore sensitivity after exposure to the M(-)UV and M(+)UV Martian environments (see Tables 4, 5 for information on all tested genotypes). Fold sensitivity was calculated as "mutant versus wild-type" measuring spore survival by colony formation. Different fold-sensitivity values are represented in a color code from <10 to <10 6 -fold, all comparing sensitivities of wild-type and mutant spores as determined in Tables 4, 5.
exposed to M(−)UV, the most important factor in spore survival was the multilayered spore coat, as seen with cotE safA spores (Figure 4); and lethal damage by M(−)UV was largely due to DNA as seen by increased M(−)UV sensitivity of recA spores (Figure 4).

Spore Outer Coat as the Most Important Protection Mechanism
Altogether, the ability of B. subtilis spores to maximally survive M(+/−)UV was due to the ability to prevent DNA damage through several protective components, including the coat, low core water content, Ca-DPA and α/β-type SASP. Yet, it was the spore outer coat (cotE) that was the most important protection mechanism, as spores lacking DPA, α/β-type SASP, and outer coat (cotE sleB spoVF sspA sspB spores) were 100fold more sensitive than spores lacking DPA and α/β-type SASP, but with an intact outer coat (sspA sspB sleB spoVF spores). The latter observation is consistent with previous work showing that the spore coat is important for spore resistance to solar radiation, particularly UV-B and UV-A (Riesenman and Nicholson, 2000;Moeller et al., 2014). It is notable that the B. subtilis spore crust, the spore's outermost layer played no significant role in spore resistance to M(+/−)UV. The precise role of the spore crust in spore properties and the detailed role of individual spore crust proteins is not yet well understood. This is particularly important as individual crust mutants were recently suggested to yield different phenotypes with respect to the double crust mutant spores (tested in this study) (McKenney and Eichenberger, 2012;Krajcikova et al., 2017).

Spore Decreased Core Water Content Is Key for Survival
During spore formation, spore core dehydration and mineralization is established, in part, as DPA is taken up into the forespore and chelates Ca 2+ ions (Ca-DPA), displacing core water (reviewed in Setlow, 2014). The compression of the forespore that takes place later in sporulation also displaces significant amounts of core water, further reducing core water content (Magge et al., 2008). How the latter takes place is not known, but likely involves the spore cortex peptidoglycan in some fashion (Zhang et al., 2012). In the current study, two different core water deficiencies were tested: (1) dacB spores, which have an altered cortex and thus present elevated core water levels and (2) sleB spoVF spores, which lack Ca-DPA due to the spoVF mutation, are stabilized against spontaneous spore germination by the sleB mutation, and have elevated core water because Ca-DPA has been replaced by water (Paidhungat et al., 2001). Results demonstrated dacB spores to be significantly more sensitive in both M(+/−)UV, when compared with the wild-type spores. This increased sensitivity was perhaps due to greater molecular damage (at least some to DNA) induced by oxidative stress when in the high vacuum/desiccation of the Martian environment. Calcium-DPA-deficient sleB spoVF spores were shown to be more sensitive to M(+/−)UV than Ca-DPA-replete spores, in particular to exposed to M(+)UV, as shown previously (Setlow et al., 2006;Magge et al., 2008). This happens because sleB spoVF sporulating cells are unable to synthesize DPA, but exogenously added DPA can enter the spore, reaching near wild-type levels. Although, whether the latter effects are due only to the spore elevated core water content, or to some direct protective effect of Ca-DPA is not clear.

SspE May Provide Some Protection When SspA and SspB Are Missing
Protection of the DNA in the spore core is also dependent on the high levels of α/β-type small, acid-soluble spore proteins (SASP) (Magge et al., 2008). These act by saturating spore DNA, and are extremely important when exposed to desiccation and UV radiation (Mason and Setlow, 1986;Moeller et al., 2008). As expected, spores lacking SASP-α and -β (sspA sspB spores), and thus lacking ∼80% of the α/β-type SASP pool (Hathout et al., 2003), had increased sensitivity to both M(+/−)UV (when compared with the wild-type), being significantly more sensitive to UV-irradiated, rather than to non-irradiated Martian environments. In contrast, sspE mutants lacking the most prominent SASP, SspE, which bounds poorly to DNA in wild-type spores, had no significant effect on spore survival in both M(+/−)UV. However, SspE may provide some protection when SspA and SspB are missing, as suggested by the increased sensitivity in sspE sspA sspB spores, when compared with sspA sspB mutants. Removing α/β-type SASP in spore coat-or cortex-defective spores (cotE sspA sspB; dacB sspA sspB and sleB spoVF sspA sspB spores) increased spore sensitivity to M(+/−)UV, confirming DNA-binding α/β-type SASP as a key factor in B. subtilis spore resistance to M(+/−)UV, presumably by the α/β-type SASP binding to spore DNA and converting the spore chromosome into a monogenomic toroidal shaped A-DNA structure (Setlow and Li, 2015).

Double Strand Breaks and Base Modifications in M(−)UV
The UV-exposed Martian surface conditions have direct and indirect effects on cells, either through the direct transfer of radiation energy, and consequent damage of biomolecules or through generation of reactive nitrogen species (RNS), or reactive oxygen species (ROS) that then induce biomolecular damage (Lenhart et al., 2012). Ultraviolet-induced damage is typically seen as DNA SSB or DSB, as well as photolesions such as CPDs, 6-4 PPs or SP (Setlow and Li, 2015). Spores lacking HR (recA), NHEJ (lig ku), or BER (exoA nfo) were significantly more sensitive to M(−)UV than wild-type spores ( Table 5), indicating that DSB and base modifications comprise a substantial fraction of the DNA damage suffered, likely due to the extreme desiccation in M(−)UV (Rebeil et al., 1998;Setlow and Li, 2015;Nicholson et al., 2018).

Spore Photoproduct as Major Damage in M(+)UV
The formation of SP as a major product of UV-damage with M(+)UV exposure was expected, and has been shown previously (Xue and Nicholson, 1996). Accumulated SPs have been shown to be repaired by SP lyase (SPL), and also by the NER pathway -mechanisms that are crucial in spore UV resistance (Setlow and Li, 2015). The current study is the first to analyze the relative sensitivities of various SP repair mutant strains of B. subtilis spores to the Martian environment, including results with spores lacking other DNA repair mechanisms. Notably, in M(−)UV accumulated SP in spores exposed to M(+)UV were shown to be repaired by both SplB and the NER pathway, mechanisms that are crucial in spore resistance to natural UV environments (Xue and Nicholson, 1996;Setlow and Li, 2015).

YwdJ and Mfd Might Participate in SP Repair
In the current study, ywjD spores lacking the UV-damage endonuclease YwjD, showed no increased sensitivity to M(+/−)UV. Yet, ywjD splB spores were more sensitive to M(+)UV than splB single mutant spores. This suggests that YwjD might participate in SP repair, functioning as an alternative DNA repair enzyme, and is in line with previous studies (Ramirez-Guadiana et al., 2012). While ywjD spores showed no increased sensitivity to M(+/−)UV, spores were more sensitive to M(+)UV than splB spores, suggesting that YwdJ can also participate in SP repair. Moreover, mfd splB spores, lacking both SP lyase and the spores also much to M(+)UV splB spores indicating that transcription, had also increased sensitivity to M(+)UV, when compared with splB single mutant spores. Thus, transcription-coupled repair might be involved in SP repair. This is likely due to the role Mfd plays in NER (Gomez-Marroquin et al., 2016). The lack of the DNA exonuclease SbcDC involved in inter-strand cross-link repair (ISCLR) (Mascarenhas et al., 2006;Lenhart et al., 2012) also demonstrated increased sensitivity in M(+)UV (sbcDC splB spores), when compared with splB single mutant spores. This was not observed, however, in polY1 polY2 spores, lacking both DNA polymerases PolY1 and PolY2, which mediate DNA repair by translesion synthesis.

Sporulation Deficiency
Strains lacking mutSL or exoA nfo shown an increased Spo − rate after exposure to M(+/−)UV, suggesting their critical involvement of MMR and BER in DNA repair in order to ensure sporulation. An increased loss of viability during sporulation of strains lacking the ability to repair DNA damage by MMR had already been suggested (Modrich, 1996;Salas-Pacheco et al., 2005;Ibarra et al., 2008;Fukui, 2010), indicating mutSL contribution to genome stability and overall spore resistance. In turn, exoA nfo genes are known to encode for apurinic/apyrimidinic endonucleases involved in the repair of oxidative DNA damage through BER (Ibarra et al., 2008;Moeller et al., 2011;Campos et al., 2014). This means that spores exposed to M(+/−)UV, ensure sporulation through efficient MMR by mutSL, and repair oxidative damage by BER (exoA nfo). Especially, the absence of the proteins LigD, Ku, ExoA, Nfo, SbcDC, and MutSL showed significant increased mutation frequencies of Spo − , indicating their crucial role in DNA repair, genome stability and restoration. In the current study, however, the interaction between Nfo and ExoA and the DNA integrity scanning protein DisA (Campos et al., 2014) was not assessed, and would be advised for future studies on the process of oxidative DNA damage repair after exposure to simulated Martian conditions (Campos et al., 2014). This sporulation deficiency analysis is informative on the types of error-free or error-prone mechanisms leading to spore survival. For instance, Figure 2 shows that RecA-mediated homologous recombination (HR) and wild type have similar proportions of Spo − mutants, indicating that spore survival in a recA-mutant is error-free. Considering that other repair mechanisms such as SP, NER, NHEJ and MMR are still at least partially functional in a recA-deficient background, this is the best argument presented in the paper to say that UV-induced photolesions such as DNA strand breaks, dimers or AP sites are the major lesion caused by Martian exposure.

CONCLUSION
When considering a Mars exploration scenario one can expect spore killing by the Martian environment to be mostly UV-driven, as the other environmental conditions (atmospheric composition, low pressure and low temperature) were shown to have only minimal effects on wild-type spore viability. Most importantly, the current study demonstrates that wild-type B. subtilis spores could survive in a Mars surface environment, if somehow shielded from UV (e.g., by dust, rocks, or spacecraft surface irregularities) It should be noted, however, that this study determined survivability by the ability to form colony forming units, and any defects in growth after exposure were not analyzed. Besides, increased spore sensitivity has been reported when in contact with Mars analog soils (Schuerger et al., 2003;Moeller et al., 2010); and vegetative cells of B. subtilis were found to be more sensitive the presence of perchlorates (found in Mars subglacial brines) irradiated with a Martian UV-flux (Wadsworth and Cockell, 2017). Thus, future efforts should focus on assessing spore survival and viability in real longduration Mars mission scenarios. This can be done by: (1) directly determining DNA damage in wild-type spores exposed to M(+/−)UV, (2) address whether exposed mutants have growth defects, after germination, (3) taking into consideration the shielding of spores via Mars regolith and other relevant materials, and (4) assess the effect of Mars surface photochemistry on spore sensitivity.

AUTHOR CONTRIBUTIONS
The study was conceived by RM. Experiments were conducted by MC, FF, FC, RM, and PE. The simulation experiments in the described Mars chamber were conducted by AS. The manuscript was written by MC with input from RM, FF, FC, WN, AS, PE, and PS.

ACKNOWLEDGMENTS
We acknowledge Andrea Schröder, Marina Raguse, and Katja Nagler for their assistance. We thank Mario Pedraza-Reyes and Jörg Stülke for cordially providing several B. subtilis strains.