RAPID DE NOVO EVOLUTION OF TOLERANCE TO DESICCATION AND ULTRAVIOLET-C RADIATION IN A NON-EXTREMOPHILE: IMPLICATIONS FOR POSSIBLE LIFE ON MARS AND VENUS By Brian D. Wade A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of Crop and Soil Sciences — Master of Science 2023 ABSTRACT Mars was once Earth-like, with habitable environments ranging from benign to extreme. Venus might also have been Earth-like, perhaps having an ocean for two billion years. But today, the surface of Mars is a dry-ice-cold, radiation-soaked, hyper-arid desert with an oxidizing regolith under a hypobaric atmosphere, while that of Venus is a lead-meltingly hot, crushingly hyperbaric desert. Understandably, most astrobiologists consider the surfaces of both Mars and Venus to be uninhabitable. However, if life ever existed on the surface of those worlds, could it have adapted to their drastic environmental change and evolved mechanisms to persist today in Mars’s shallow subsurface and Venus’s middle to lower clouds? If so, desiccation and, to a lesser extent, ultraviolet-C (UV-C) radiation would exert tremendous selective pressures on any such Martian and Venusian life. While the mechanisms of tolerance to those stressors in bacteria are fairly well understood, the evolutionary dynamics that can produce those tolerances have been largely unexplored. Therefore, I performed an evolution experiment in which replicated populations of a desiccation- and UV-C radiation-sensitive strain of Escherichia coli were exposed to daily pulses of either desiccation only, UV-C radiation only, or both stressors combined. Tolerance to those stressors, both separately and combined, evolved within a mere 500 generations. I also hypothesized that cross-tolerance would evolve, i.e., treatment with one stressor would result in a correlated gain of tolerance to the non-treatment stressor, but this hypothesis was rejected. Thus, the evolution of co-tolerance required selection with both stressors combined. My results show that a non-extremophile can readily and rapidly adapt to two Mars- and Venus-relevant stressors. The implications of my findings for our neighboring planets are that life might persist today in Mars’s shallow subsurface and in Venus’s clouds, provided that the drastic environmental changes that occurred on those worlds allowed adaptation to one or two stressors at a time. Copyright by BRIAN D. WADE 2023 This thesis is dedicated to my family and friends, both living and passed. Thank you for being there when I needed you and cheering me on the whole time. I love and will forever remember you all. And finally, a special dedication to my young kids, Thomas and Natalie. May you never stop wondering at our pale blue dot and the worlds beyond. I love you with everything I am. iv ACKNOWLEDGEMENTS First, and foremost, I thank my wife, Tracy, for helping me get back up after the many times I have fallen during this very long journey, and for encouraging me to keep going despite some awfully deep canyons that stood in my way. And thanks to my kids for making sure I took breaks to play while writing. I’ll never forget those moments. To my mom and dad, thank you for letting me explore my own way in life, even though it has taken me far away from you at times. I’m sure I’ve worried you many times, and in many ways, during this journey. I’m sorry, but thank you for being strong. I’m a scientist and, I hope, a decent human being because of your upbringing. My committee has been incredibly patient and supportive. To my advisor, Prof. Richard Lenski, and Profs. Gemma Reguera, Kazem Kashefi, and Terence Marsh (de facto member!), you all have made me a better scientist. Thank you! To Prof. Michael Velbel, thank you for keeping me grounded on Mars! My level of knowledge of the Red Planet has improved considerably while getting to know you. And I’ll always cherish the many wonderful conversations we’ve had about Mars, life (both Martian and Earthly!), and the inner workings of science. And finally, to all the Lenskians whom I’ve had the pleasure of knowing throughout my time here: Neerja Hajela, Dr. Zachary Blount, Dr. Jeff Barrick, Dr. Justin Meyer, Dr. Caroline Turner, Dr. Nkrumah Grant, Dr. Kyle Card, Dr. Minako Izutsu, Devin Lake, Dr. Mike Wiser, Dr. Rohan Maddamsetti, and Dr. Sean Sleight. You have all helped shape me as a scientist and as a person. v TABLE OF CONTENTS LIST OF FIGURES.......................................................................................................................vii INTRODUCTION...........................................................................................................................1 MATERIALS AND METHODS....................................................................................................4 RESULTS......................................................................................................................................14 DISCUSSION................................................................................................................................19 CONCLUSIONS...........................................................................................................................29 BIBLIOGRAPHY.........................................................................................................................30 APPENDIX...................................................................................................................................46 vi LIST OF FIGURES Figure 1: Stress–growth procedure................................................................................................46 Figure 2: Fitness of the evolved groups when stressed by desiccation.........................................47 Figure 3: Fitness of the evolved groups when stressed by UV-C radiation..................................48 Figure 4: Fitness of the evolved groups when not stressed (control conditions)..........................49 Figure 5: Total fitness of the evolved groups’ replicate populations............................................50 Figure 6: Differences in cell pellets..............................................................................................52 vii INTRODUCTION Investigations of Mars’s surface via landed spacecraft have shown that it once supported multiple habitable environments, ranging from benign to extreme (Squyres et al., 2004, 2008; Grotzinger et al., 2014, 2015; Hurowitz et al., 2017; Arvidson and Catalano, 2018; Losa-Adams et al., 2021; Vasavada, 2022). But eventually, Mars lost most of its atmosphere (Lammer et al., 2018; Jakosky, 2021; Vasavada, 2022), resulting in a surface that today is a frigid, radiation-soaked, hyper-arid desert with an oxidizing regolith under a hypobaric atmosphere (Gómez‐Elvira et al., 2014; Hassler et al., 2014; Lasne et al., 2016; Martínez et al., 2017; Fischer et al., 2019; Lange et al., 2022; Harri et al., 2023; Munguira et al., 2023; Polkko et al., 2023). When exposed to high-fidelity simulations of Mars’s present-day surface conditions, some terrestrial organisms, including Serratia liquefaciens and Bacillus subtilis, can survive for some period; however, they are unable to grow and reproduce (Schuerger and Nicholson, 2006; Hansen et al., 2009; Smith et al., 2009; Berry et al., 2010; Kerney and Schuerger, 2011; Schuerger et al., 2020; Schwendner et al., 2020). Thus, many astrobiologists argue that the surface of Mars is uninhabitable (e.g., Tarnas et al., 2021). However, if life ever existed on Mars and was present near the surface, then its evolution would have been shaped by a very different environmental history than that of life on Earth (McKay and Stoker, 1989; Clark, 1998; Fairén et al., 2005, 2017; Schulze-Makuch et al., 2005, 2013; Dohm et al., 2011; Davila and Schulze-Makuch, 2016; Cabrol, 2018, 2021; Salvatore and Levy, 2021). This historically contingent evolution would perhaps have facilitated the capacity for Martian life to not only survive, but also grow and reproduce in a specific environment — namely, the shallow subsurface — where sporadic liquid water and permissive temperatures might allow a patchy and 1 transitory existence (Warren-Rhodes et al., 2006, 2007a, 2007b, 2019, 2022; Schulze-Makuch et al., 2018; Davila et al., 2020). The environmental history of Venus is opposite that of Mars. Whereas Mars’s surface went from being Earth-like to a dry-ice-cold, hypobaric desert after losing most of its atmosphere, Venus’s surface went from being possibly Earth-like to a lead-meltingly hot, crushingly hyperbaric desert after its atmosphere suffered a runaway greenhouse effect (Petropoulos and Telonis, 1988; Lammer et al., 2018; Gillmann et al., 2022; Westall et al., 2023). While the surface of Venus is now uninhabitable, its lower to middle cloud layers might be conducive to an aerial biosphere (Cockell, 1999; Schulze-Makuch et al., 2004, 2013; Grinspoon and Bullock, 2007; Limaye et al., 2018, 2021a, 2021b; Bains et al., 2021, 2023; Schulze-Makuch, 2021; Seager et al., 2021; Westall et al., 2023). Therefore, as with the plausible, historically contingent evolution of hypothetical Martian life that I put forth above, Venusian life, if it ever existed, might have adapted from living on a once-habitable surface to persisting solely in the clouds. Desiccation and, to a lesser extent, ultraviolet-C (UV-C) radiation would exert intense selective pressures on any life that might persist in Mars’s shallow subsurface (Hansen et al., 2009; Smith et al., 2009; Berry et al., 2010; Schulze-Makuch et al., 2018; Warren-Rhodes et al., 2019; Davila et al., 2020) or Venus’s middle clouds (Schulze-Makuch et al., 2004, 2013; Bains et al., 2021, 2023; Limaye et al., 2021b; Seager et al., 2021). The mechanisms of tolerance to desiccation and UV-C radiation in bacteria are fairly well understood (Krisko and Radman, 2010, 2013a, 2013b; Slade and Radman, 2011; Lebre et al., 2017; Esbelin et al., 2018; Greffe and Michiels, 2020; Laskowska and Kuczyńska-Wiśnik, 2020; Daly, 2023). However, the evolutionary dynamics that can produce those tolerances have been largely unexplored. I therefore carried out an evolution 2 experiment in which I propagated replicate populations of Escherichia coli under substantial stress from desiccation and UV-C radiation, both separately and in combination. 3 MATERIALS AND METHODS Ancestral strain and growth conditions I performed the evolution experiment starting from two sub-strains, REL606 and REL607, of Escherichia coli strain B (Lenski et al., 1991; Daegelen et al., 2009). These bacteria are strictly asexual, lacking plasmids or functional bacteriophages (Lenski et al., 1991). Lenski et al. (1991) derived REL607 from REL606, which are nearly identical to each other, the key difference being that REL607 has a mutation conferring the ability to catabolize the sugar L-(+)-arabinose (Ara+), whereas REL606 does not (Ara‒). These sub-strains are easily distinguished from one another by the color of their colonies when plated on nutrient-rich agar supplemented with arabinose and the redox indicator tetrazolium chloride (TA agar [Levin et al., 1977]): colonies of REL607 (Ara+) appear white on TA agar, while those of REL606 (Ara−) appear red. This phenotypic marker permits direct measurement of fitness through evolved-vs-ancestor competition assays (see below, competitive fitness assays and cell density measurements). Moreover, the marker has repeatedly been shown to be selectively neutral in medium limited by glucose and other sugars, as well as in various other environmental conditions (Lenski, 1988; Bennett et al., 1992; Travisano et al., 1995; Travisano and Lenski, 1996; Turner et al., 1998; Rozen and Lenski, 2000; Ostrowski et al., 2005; Sleight and Lenski, 2007; Izutsu and Lenski, 2022). Cells were grown at pH 7 in Davis minimal (DM) broth, which is formulated as follows: 0.7% potassium phosphate dibasic trihydrate; 0.2% potassium phosphate monobasic (anhydrous); 0.1% ammonium sulfate; 0.05% sodium citrate tribasic dihydrate; 0.01% magnesium sulfate (anhydrous); 0.0002% thiamine hydrochloride; and 0.04% D-(+)-glucose (Lenski, 1988). The magnesium sulfate, thiamine, and glucose were added to the autoclaved base medium from 0.2- µm-filtered solutions. Cells grew in 800-µL cultures in 96-deep-well plates at 37°C while orbitally 4 shaken at 475 rpm. The 0.04% glucose in the DM broth limited the ancestral strains’ population density at stationary phase to 3.1 × 108 ± 1.7 × 107 cells mL‒1 (95% confidence interval), and thus 2.5 × 108 ± 1.4 × 107 total cells per culture. Treatment and control populations Four groups of six replicate populations evolved for 500 generations (75 daily cycles) under one of four conditions: desiccation only, UV-C radiation only, both stressors combined, or no stress (control). Half of the populations comprising each group were founded by clones of REL606 and the other half by clones of REL607; each population started from a separate colony, with each colony the outgrowth of a single haploid cell. Therefore, the replicate populations were identical at the beginning of the experiment, ensuring that selection would act on de novo mutations and not standing genetic variation (Lenski et al., 1991; Izutsu and Lenski, 2022; Lenski, 2023). The half-and-half founding of populations allowed me to account for any effects of the arabinose-utilization marker on fitness; no such effects occurred. I archived all the evolving populations in 15% glycerol at −80°C after 50 generations, 100 generations, and every 100 generations thereafter. The REL606 and REL607 ancestral strains remained in cryogenic storage during the experiment. Aliquots of both the ancestral and evolved samples were later revived and used to assess changes in the fitness of the evolved populations relative to their ancestors (see below, competitive fitness assays and cell density measurements). Procedures for desiccation and UV-C irradiation I developed the following procedure for the desiccation-treated populations: I took 8-µL aliquots of cultures grown for 22 – 24 h in DM broth, spotted them onto the bottom of 5-mL borosilicate glass beakers, and dried them at 37°C for 1 h in an air-tight plastic box containing anhydrous calcium sulfate (8-mesh granules of Drierite; WA Hammond Drierite, Xenia, Ohio, USA). The 5 aliquots were visibly dry after 45 min; hence, the cells were in a desiccated state for at least 15 min. After desiccation, the cells were re-hydrated by adding 800 µL of fresh DM broth to each beaker and then orbitally shaken at 120 rpm for 15 – 30 min at room temperature (20 – 24°C) to re-suspend them. After re-hydration and re-suspension, I transferred the 800 µL of broth in each beaker to a well in a 96-deep-well plate, and the cultures were grown until the next day under the conditions described above. The survival rate of the ancestors under these conditions was 1.2% ± 0.3% (95% confidence interval). Figure 1(a) illustrates my desiccation procedure. My procedure for the UV-C radiation-treated populations was similar to that for the desiccation-treated populations: 8-µL aliquots of cultures grown for 22 – 24 h in DM broth were spotted onto the bottom of 5-mL borosilicate glass beakers and then irradiated at room temperature (20 – 24°C) with 254-nm-wavelength light using a Spectroline model EF‒140C lamp (Spectronics, Melville, New York, USA). After irradiation, the cells were re-suspended, without shaking, in 792 µL of fresh DM broth. I then transferred the resulting 800 µL of broth in each beaker to a well in a 96-deep-well plate, and the cultures grew alongside the desiccation-treated populations until the next day. The survival rate of the ancestors under these conditions was 2.5% ± 1.1% (95% confidence interval). Figure 1(b) shows my irradiation procedure. Each UV-C radiation-treated population was spotted, irradiated, and then re-suspended before proceeding to the next population to prevent any evaporation of the 8-µL aliquot. Populations treated with UV-C radiation only were irradiated for 16 sec, whereas those subjected to the combined desiccation and UV-C radiation treatment were irradiated for 8 sec. The 4-Watt lamp was positioned 4 cm directly above each aliquot, giving radiation doses of ~4 and 2 J cm‒2, respectively. This halving of the dose was necessary to ensure adequate numbers of surviving cells in the dual-treated populations after sequential exposure to both stressors. Dual-treated populations 6 were first irradiated as described and then desiccated under the same conditions and for the same amount of time as those treated with desiccation only. These populations were grown in a 96-deep- well plate alongside the other stressed populations until the next day. Compensation for a stress-induced bottleneck During the evolution experiment, the stress-treated populations experienced an additional bottleneck when the aliquots were exposed to the stressor(s) that the non-stressed control populations did not experience. To account for those conditions, after 22 – 24 h of growth, I diluted the control populations in fresh, glucose-free DM broth such that 1% of each population would be transferred to the next round of growth. Aliquots from the dilutions were then transferred to separate wells of fresh DM broth in a 96-deep-well plate and grown alongside the stress-treated populations until the next day. This dilution ensured that the control populations’ effective survival rate — and hence the impact of random drift caused by the population bottleneck — was close to that of the stress-treated populations. Procedural controls My experiment also included procedural controls to detect any contamination that might have occurred during the stress procedures and daily transfers to the 96-deep-well plates. I placed uninoculated DM broth in the outermost wells of the 96-deep-well plates, immediately adjacent to the wells containing the stress-treated and control populations. The procedural control wells went through the same stress procedures and daily incubations as the experimental populations. For example, aliquots from the procedural control wells next to the desiccation-treated populations underwent the same desiccation procedure and were then incubated alongside those populations. None of my procedural controls showed any contamination during the evolution experiment, as determined by the absence of colonies when I spread samples on TA agar plates. In addition, I 7 alternated Ara+ and Ara‒ populations in the wells of the 96-deep-well plates to monitor for possible cross-contamination events. Cross-contamination did not occur during the experiment, as verified by the lack of red colonies on TA agar plates of Ara+ populations and the lack of white colonies on TA agar plates of Ara‒ populations. Competitive fitness assays and cell density measurements I performed competitive fitness assays using mixtures of the evolved and ancestral cells having the opposite arabinose-marker states. Thus, evolved populations derived from REL607 (Ara+) were mixed with the ancestral REL606 strain (Ara−), and vice versa. Each fitness assay began by inoculating from their frozen stocks 16 µL of the evolved population in 800 µL of DM broth and 16 µL of the appropriate ancestral competitor in a separate 800 µL of broth. These cultures grew alongside each other for 24 h as described above, and they were then transferred to another pair of separate 800-µL volumes of broth for a second 24 h of growth. These two days of separate growth allowed the bacteria, after being revived from the frozen samples, to acclimate to the environment in which they would compete. After the second day’s growth, aliquots of the two competitors were mixed in equal culture volumes (providing approximately equal numbers of evolved and ancestral cells) for the competition. The two days of growth were necessary not only to allow the cells to recover from being frozen, but also to allow them to consume all the glycerol used as a cryoprotectant. The presence of glycerol could adversely impact the fitness assays and their interpretation in two ways. First, residual glycerol has been shown to alter the growth dynamics of E. coli (Atolia et al., 2020). Second, a variety of organisms, albeit mostly eukaryotes, use glycerol and its derivatives as an osmoprotectant (Borowitzka and Brown, 1974; Adler et al., 1985; Somero, 1992; Roberts, 2005; Yobi et al., 2013; Raymond et al., 2020; Khan et al., 2023), and supplemental glycerol can be used 8 to alleviate stress from low water activity (e.g., Sriram et al., 2011). Therefore, residual glycerol might provide some protection to E. coli during the desiccation process. I performed the competitions by subjecting aliquots of the evolved-ancestor mixture to the relevant stress and then re-grew the mixed culture as described above and illustrated in Figure 1. Cells from the competition mixture were plated on TA agar at three time points to enumerate colonies for the fitness calculations described below. First, a dilution of the competition mixture was plated to determine the initial densities of the evolved and ancestral cells (diluted from NR in Figure 1). Second, the stress was applied to an aliquot of the mixture, and the cells in that aliquot were re-suspended as described above. A dilution of the re-suspended cells was plated, again on TA agar, before placing the culture of re-suspended cells in the incubator for re-growth, to determine the densities of evolved and ancestral cells after the stress (diluted from NS in Figure 1). Third, and lastly, a dilution of the re-grown culture was plated to determine the densities of evolved and ancestral cells after their recovery from stress (diluted from NGʹ in Figure 1). I used the numbers of evolved and ancestral colonies from these three samples to calculate fitness values for the entire stress‒growth cycle and for the two separate fitness components, as described next. Fitness components and their calculations I calculated fitness components from two non-overlapping periods of the competitions that I call “survival” and “growth”. The survival component spanned the period between the first and second platings described above. The growth component spanned the period between the second and third platings. I then summed the survival and growth components to give “total” fitness. Each fitness component was calculated as a selection-rate differential (Lenski et al., 1991; Travisano et al., 1995) per daily stress–growth cycle. Whereas the more commonly encountered relative fitness (W) expresses the ratio of the competitors’ realized growth rates, the selection-rate 9 differential (r) is the difference in their realized growth rates. The selection-rate differential is more appropriate under two conditions, both of which are relevant to my study: (i) when the net growth rate of one or both competitors is negative, as it generally was during the survival phase; and (ii) when the difference in fitness between the competitors is large (Travisano et al., 1995), as was the case when the ancestor competed against the stress-evolved populations in a stressful environment. I used the following three equations to calculate the selection-rate differentials for the survival (rS) and recovery (rG) fitness components, and then for the total fitness (rT): rS = [ln(Es / Ei) ‒ ln(As / Ai)] / day; rG = [ln(Eg / Es) ‒ ln(Ag / As)] / day; rT = (rS + rG) / day; where Ei, Es, Eg and Ai, As, Ag are the densities of the evolved (E) and ancestral (A) cells estimated from the three successive platings, respectively. Statistical analyses I performed 4 replicate competition assays for each of the 6 endpoint evolved populations in each treatment and control group, and I did so under each of three assay conditions: desiccation stress, UV-C radiation stress, and the unstressed control environment. Thus, I performed 288 competition assays (4 treatments × 6 populations × 4 replicate assays × 3 conditions), and each assay provided 3 estimates (864 total estimates): survival, growth, and total fitness. I analyzed the results as 9 sets of data, partitioned by the 3 estimates and the 3 assay conditions. For example, survival of the 24 10 evolved populations (4 treatments × 6 populations) when exposed to desiccation forms one set of data, their growth when exposed to that same stressor is a second dataset, and their total fitness under that same condition constitutes the third set. I performed the statistical analyses using the SAS software, version 9.4 (SAS Institute, Cary, North Carolina, USA). This analysis involved three steps. First, I tested for outliers in each group using Cook’s distance (Cook, 1977) and by visually inspecting plots of residuals. I identified a value as an outlier (i) if its Cook’s distance was > 0.38 and it was evident in the plots of residuals, or (ii) if it was clearly present in the plots of residuals (e.g., a discontinuous bin in the histogram), even if its Cook’s distance was < 0.38. I based the Cook’s distance threshold on equation MC from Kim and Storer (1996). Only 6 of the 864 (0.7%) fitness estimates were judged to be outliers, and none of the sets of 4 replicate assays had multiple outliers. In order to maintain the balanced design, I replaced the value of each outlier with the median of the three other replicate measurements of the population to which it belonged. Second, I checked if each group’s residuals were normally distributed using Shapiro‒Wilk’s test (Shapiro and Wilk, 1965) and by visually inspecting plots of residuals. If a test was significant at the 95% confidence level and that result was evident in the plots of residuals, then I performed a Box‒Cox power transformation (Box and Cox, 1964, 1982) of the entire fitness data set. When I performed this transformation, the group means from the transformed data were used in place of those from the untransformed data for an analysis of variance (ANOVA), which I describe next. Third, and finally, I performed a one-way Welch’s ANOVA (Welch, 1951) of group means for each of the 9 datasets using the MIXED procedure (SAS, 2015). I used Satterthwaite’s approximation (Satterthwaite, 1946) to determine the denominator degrees of freedom. If the Welch’s ANOVA was significant at the 95% confidence level, then I made post hoc comparisons 11 of each treatment group’s mean to that of the control group using Welch’s t test (Welch, 1938). I adjusted the P values of the three, two-tailed Welch’s t tests for multiple comparisons using Dunnett’s T3 correction for heterogeneous variances (Dunnett, 1980). To determine whether the mean total fitness of the treatments varied, I performed a one-way Welch’s ANOVA of the population means for each of the 12 combinations of treatment and assay condition. For example, I ran one ANOVA for the total fitness of the desiccation-treated populations when not stressed (control condition), a second ANOVA for the total fitness of the same populations when exposed to desiccation, and a third ANOVA for their total fitness when exposed to UV-C radiation. These ANOVAs were carried out in the same way as described for the ANOVAs of group means, but without post hoc comparisons. Because Welch’s ANOVA and Welch’s t test are intended for unequal variances, I did not perform a preliminary test for homogeneous variances. In fact, I expected the variances between treatment groups to be heterogeneous due to the severity of the stresses from UV-C radiation and desiccation (relative to the benign control treatment) and the potential for small deviations in my procedures (e.g., slight variations in drying time) to have large effects. When there is a reasonable expectation of heterogeneous variances, an unconditional analysis is more robust against Type 1 errors, with little loss of power, than an analysis that is conditioned on whether the variances are homogeneous (Moser et al., 1989; Moser and Stevens, 1992; Ruxton, 2006; Hayes and Cai, 2007; Derrick et al., 2016; Delacre et al., 2019). Comparisons of treatment and control groups, but not among treatments As I mentioned above (in the sub-section on procedures for desiccation and UV-C irradiation), the survival rates of the ancestors when stressed by desiccation and UV-C radiation were 1.2% ± 0.3% and 2.5% ± 1.1%, respectively. The lower bound of the radiation survival rate’s 95% confidence 12 interval overlaps the upper bound for the desiccation survival rate, but it does not overlap the mean estimate of the desiccation survival rate. Thus, the difference in survival rate between the two stressors is significant. Because of that significant difference in the ancestors’ survival rate, it would not be appropriate to compare one treatment group’s level of fitness to that of another treatment group. Consequently, I only compare the various stress treatments with the control group, and all my conclusions about changes in survival, growth, and total fitness are relative to the evolved control group when exposed to the same conditions as the treatment groups. Although I intended to equalize the ancestors’ survival rates when stressed by UV-C radiation and desiccation, that effort was stymied by the idiosyncrasies of my irradiation apparatus and procedures, most notably the manual on/off switching of the UV lamp, which led to substantial variation in survival (as indicated by the wide 95% confidence interval). 13 Fitnesses of the evolved groups when stressed by desiccation RESULTS Figure 2(a) shows that the control populations, as a group, had no significant change in their survival when stressed by desiccation (i.e., the corresponding 95% confidence interval includes 0). Both the desiccation group and the dual-stressed group (i.e., evolved with desiccation and radiation stresses combined) achieved significant gains relative to the ancestor in their survival when exposed to desiccation. By contrast, the radiation group experienced a significant loss in survival under desiccation. In all three treatment groups, those changes are significant not only relative to the common ancestor (as shown by the confidence intervals), but also in comparison to the control group (as indicated by the asterisks). Figure 2(b) shows that all four groups, including the control, grew faster than the common ancestor. This result is not an indication of adaptation to desiccation per se; rather, as I will show later, it simply reflects faster growth in the culture medium used in all the treatments. After accounting for multiple statistical comparisons, neither the desiccation group nor the dual-stressed group grew faster than the control group, although both treatment groups trended in that direction. However, the radiation group’s significant loss in survival was offset by significantly faster growth than that of the control group. Total fitness represents the sum of the survival and growth fitness components. Figure 2(c) shows that both the desiccation and dual-stressed groups had gains in total fitness that were significantly greater than the control group. The radiation group showed no significant difference in total fitness relative to the control group, as the differences in survival and growth components offset one another. 14 Fitnesses of the evolved groups when stressed by UV-C radiation Figure 3(a) shows that neither the control group nor the desiccation group showed any significant change relative to the common ancestor in survival when stressed by UV-C radiation. However, the radiation and dual-stressed groups had highly significant gains in survival relative to the common ancestor as well as when compared to the control group. Moreover, as shown in Figure 3(b), only the radiation group had significantly faster growth than the control group. Figure 3(c) shows that the radiation and dual-stressed groups had gains in total fitness that were significantly greater than those of the control group. The desiccation group had a total fitness gain relative to the ancestor that was slightly, but significantly, less than that of the evolved control group. This deficit in the desiccation group’s total fitness resulted from the combination of small differences in survival and growth. Fitnesses of the evolved groups when not stressed Figure 4(a) shows that the evolved control group’s survival did not differ from the ancestor under the control conditions. Furthermore, none of the evolved treatment groups differed from the control group in this respect. Figure 4(b) reveals that the control group grew significantly faster than the common ancestor, which demonstrates adaptation to the culture medium, as expected. Unexpectedly, both the radiation and dual-stressed groups not only grow faster in the control environment than the ancestor, but they also show significantly greater gains in this fitness component than the control group. By contrast, the desiccation group exhibited smaller gains than the control group. In terms of total fitness, the residuals for the desiccation group were not normally distributed, perhaps in part due to heterogeneity among the replicate populations (as examined in the next sub- section), and therefore the data were transformed (as explained in the sub-section on statistical 15 analyses). After transformation, all four evolved groups had higher total fitness in the non-stressed environment than the common ancestor, as can be seen in Figure 4(c). The radiation group had slightly, but significantly, higher total fitness than the control group, whereas the desiccation group had significantly smaller gains than the control group. The dual-stressed group’s gains were indistinguishable from the control group. Variation among the replicate populations in each evolved group All the analyses above examined average fitness values for a group of evolved populations; any heterogeneity among the six replicate populations within a treatment group was therefore simply a source of statistical noise. Figures 5(a) ‒ 5(c) show the total fitness relative to the common ancestor for the 24 evolved populations (4 treatments × 6 populations) individually in each of the three assay environments. Thus, I can visualize the variation among the replicate populations in each environment and thereby assess the consistency of their evolutionary responses. Importantly, I replicated the fitness assays for each population, and therefore I could perform ANOVAs to test whether there was significant variation among the populations within a treatment group. Overall, six of the 12 ANOVAs show significant variation among the replicate populations, as indicated by the asterisks above the set of populations in each treatment group. Thus, genetic diversity in the evolutionary responses was an important outcome of the experiment. At the same time, this visualization of the data provides compelling evidence of the specificity of adaptation to the stressful treatments. As Figure 5(a) shows, all 12 populations that experienced daily desiccation events during the evolution experiment (i.e., 6 populations in the desiccation group and 6 in the dual-stressed group) had average total fitness in the desiccation environment (direct response) that was greater than the grand mean of the control populations (shown by the dashed line). In the case 16 of the radiation-only-treated populations, half of them had average total fitness in the desiccation environment (correlated response) greater than the grand mean of the control group. Like the desiccation-treated populations’ direct response to desiccation, Figure 5(b) shows that all 12 populations that experienced daily exposures to UV-C radiation (i.e., 6 populations in the radiation group and 6 in the dual-stressed group) had average total fitness in the radiation environment (direct response) that was greater than the grand mean of the control group. On the other hand, all but one of the desiccation-only-treated populations had lower average total fitness in the radiation environment (correlated response) than the grand mean of the control populations. In contrast to the populations in the stressed environments, the pattern in the unstressed control environment is less clear, as shown in Figure 5(c). The three treatment groups that evolved in stressful conditions include individual populations with average fitness values both above and below the grand mean of the evolved control populations. This ambiguity presumably reflects the fact that the culture medium was the same in all four treatments, so that adaptation to that medium was important in all cases. In summary, the population-level analyses generally support the group-level results reported in the previous sections, which demonstrated adaptation to the specific stresses of desiccation and UV-C radiation. At the same time, the population-level analyses reveal significant variation among the independently evolved populations in most groups in their direct response to the environment where they evolved, in their correlated responses to other environments, or both. A serendipitous finding of differences in cell pellets Toward the end of the evolution experiment, I noticed that the cells of all 12 populations in the desiccation and dual-stressed groups formed diffuse pellets after settling overnight in their spent growth medium, as seen in the upper two rows of Figure 6. By contrast, as seen in the lower two 17 rows, the populations in the radiation and control groups formed condensed pellets, with the possible exception of one population in the radiation-treated group (second from left) that seems to have an intermediate phenotype. Thus, this diffuse-pellet phenotype appears to have evolved as a specific response to desiccation. I have not yet further explored this phenotype. 18 DISCUSSION Hypotheses on the evolutionary dynamics of tolerance to desiccation and radiation My experiment was designed to test four broad hypotheses concerning the evolutionary dynamics of tolerance to desiccation and radiation. (1) The treatment groups that had daily exposures to desiccation and UV-C radiation separately would evolve increased tolerance to the corresponding stressor, while the dual-stressed group would evolve increased tolerance to both stressors. The increased tolerance would be manifest by each treatment group having greater gains in total fitness than the evolved control group, when fitness was measured by competitions against the common ancestor in the corresponding stress environment. (2) The groups treated with only one stressor would have correlated gains in total fitness when exposed to the non-treatment stressor, and thus cross-tolerance would evolve, e.g., the desiccation group would also gain tolerance to radiation. I based this hypothesis on two observations: (i) desiccation and UV-C radiation cause similar damage to cells, primarily through oxidation of proteins (França et al., 2007; Fredrickson et al., 2008; Krisko and Radman, 2010, 2013a, 2013b; Slade and Radman, 2011; Santos et al., 2013; Kragh and Truelstrup Hansen, 2020; Daly, 2023; Łupkowska et al., 2023), and (ii) some organisms exhibit unusually high tolerance to both of these stressors (e.g., members of the bacterial genus Deinococcus [de Groot et al., 2005; Yang et al., 2009, 2010; Slade and Radman, 2011; Dong et al., 2015; Liu et al., 2017]). (3) The greater gains in total fitness in the treatment groups relative to the control group would be driven by improved survival during exposure to the stressors, rather than by improved growth. All the treatment and control populations were propagated in the same culture medium, and 19 therefore all of them had similar opportunities to adapt to that medium. Consequently, adaptation to the medium would be evident by improvement, relative to the ancestor, in the growth component of fitness, and improvement beyond that of the evolved control group would suggest specific adaptation to the stressor. In the case of survival, any improvements in that fitness component would indicate specific adaptation to the stressor. I based this hypothesis of survival being key to evolving tolerance on previous research (cited immediately above) that suggests protective mechanisms, often mediated by the proteome during acute exposure to stressors, are central to tolerating desiccation and radiation, including UV-C, in microbes. (4) Tolerance to desiccation and UV-C radiation would incur trade-offs, i.e., correlated losses in total fitness, when exposed to benign (control) conditions. The trade-off would be manifest in one of two ways: (i) each treatment group’s total fitness grand mean would be negative relative to the ancestor, or (ii) most of each treatment group’s total fitness population means would be less, but not necessarily negative, than the evolved control group’s total fitness grand mean, when fitness was measured by competitions against the common ancestor in the benign environment. I derived these two trade-off criteria from concepts laid out in the works of Lenski and Travisano (1994), Novak et al. (2006), and Bennett and Lenski (2007). I based this hypothesis of correlated losses when not stressed on the fact that similar kinds of trade-offs have been seen in nature (Luo et al., 2008; Chang and Leu, 2011; Friman et al., 2013; Porter and Rice, 2013; Comont et al., 2020; Yang et al., 2023), and they also often arise in evolution experiments (Mongold et al., 1996; Sleight et al., 2006; Bennett and Lenski, 2007; Caspeta and Nielsen, 2015; Boyd et al., 2022; Ballu et al., 2023; Wang et al., 2023). 20 All treatment groups evolved increased tolerance Figures 2(c) and 3(c) show that all the treatment groups increased their total fitness more than the evolved control group when stressed, and thus they evolved tolerance to the corresponding stressor(s). This result agrees with my first hypothesis. Furthermore, this de novo tolerance to desiccation and UV-C radiation evolved within 500 generations, a mere blink of the eye in evolutionary time. Further still, it is important to point out that the dual-stressed group evolved tolerance to the combined stressors within the same number of generations as those groups treated with only one of the stressors. And this rapid evolution of tolerance occurred in a non-extremophile that was very sensitive to those stressors. Taken together, these observations suggest that tolerance to desiccation and UV-C radiation can readily evolve, and it is possible for a non-extremophile to evolve tolerance to more than one harsh stressor at the same time. Figure 5(a) shows that there was significant variation among total fitness of the desiccation group’s populations when stressed by desiccation, suggesting that there were multiple evolutionary paths to desiccation tolerance. Therefore, it is likely that multiple mechanisms exist between those populations for tolerance to desiccation, be it fitness components involving both direct and indirect responses or changes in the genome, physiology, biochemistry, etc. Figures 5(a) ‒ 5(c) reveal that the heterogeneous variation among total fitness of the desiccation group’s populations existed under all conditions. Future studies are needed to explore how the different types of mechanisms might be coupled. For example, one could address whether more similar paths based on fitness components are also more similar in terms of the underlying genetic and/or physiological changes. Conversely, Figure 5(b) shows the lack of significant variation among total fitness of the radiation group’s populations when stressed by UV-C radiation, suggesting that there was only one, or very limited, evolutionary path(s) to UV-C radiation tolerance. Consequently, it is likely 21 that a single mechanism, or very similar ones, exists between those populations for tolerance to UV-C radiation. It can be seen in Figure 5(a) that the homogeneous variation among total fitness of the radiation group’s populations also existed under desiccation, but Figure 5(c) reveals it was heterogeneous under benign (control) conditions. Regarding the dual-stressed group, Figures 5(a) ‒ 5(c) show that its populations’ total fitness did not vary significantly under any condition. When that group was stressed separately by desiccation and UV-C radiation, the lack of variation among its populations’ total fitness indicates that, like evolution under radiation only, there were very limited, or perhaps a single, evolutionary path(s) to dual-tolerance. Thus, as I suggested for tolerance to UV-C radiation, dual-tolerance probably involves very similar mechanisms, or maybe just one. The dynamics of de novo tolerance to desiccation and UV-C radiation described immediately above leads me to suspect that the evolution of dual-tolerance was confined by the apparently limited path(s) to UV-C radiation tolerance. In other words, the multiple paths to achieving desiccation tolerance might have been more difficult hikes than the main trail leading to radiation tolerance. To my knowledge, this is the first published study to experimentally evolve tolerance to desiccation in a prokaryote. However, desiccation tolerance has been experimentally evolved repeatedly in more than one species of fruit fly (Hoffmann and Parsons, 1989; Gibbs et al., 1997; Tejeda et al., 2016). Tolerance to UV-C radiation has been experimentally evolved multiple times in different bacteria (Alcántara-Díaz et al., 2004; Wassmann et al., 2010; Selveshwari et al., 2021; Ellington et al., 2023). And in one experiment, Begyn et al. (2020) evolved endospores under stress from UV-C radiation and found that they became even more recalcitrant, but the vegetative cells remained sensitive. 22 Cross-tolerance did not evolve, but an unexpected trade-off did evolve Figures 2(c) and 3(c) show that neither the desiccation group nor the radiation group had statistically significant correlated gains in total fitness relative to the evolved control group when exposed to the non-treatment stressor, and therefore cross-tolerance did not evolve. Nonetheless, both of those groups had gains relative to the ancestor. Tolerance to both stressors required the combined treatment of the dual-stressed group. This result refutes my second hypothesis. And it is also surprising, not only because there are numerous examples of organisms that are highly tolerant of both desiccation and UV-C radiation (e.g., members of the genus Deinococcus, as cited above), but as I discuss next, an unexpected trade-off occurred between those two stressors. Figure 3(c) reveals that the desiccation group incurred a trade-off in total fitness when stressed by UV-C radiation. I determined that this trade-off occurred based on the second criterion given above in my fourth hypothesis, whereby most of the desiccation group’s population means fall below the evolved control group’s grand mean when exposed to radiation, as can be seen in Figure 5(b). Five of the 6 population means are below the evolved control group’s grand mean, and 4 of those 5 have 95% confidence intervals that do not include the control’s mean. In the case of the radiation group under desiccation, Figure 2(c) shows that it did not incur a trade-off. This dynamic of the desiccation group incurring a trade-off while the radiation group did not, suggests that, under the conditions of my experiment, the de novo evolved mechanisms for tolerance to desiccation and UV-C radiation differed considerably. Survival mechanisms were key to desiccation tolerance, but not radiation tolerance Figure 2(a) shows that the desiccation group had a significant gain in the survival component of fitness when stressed by desiccation, both relative to the ancestor and to the evolved control group. However, Figure 2(b) shows that its gain in the growth component was significant relative to the 23 ancestor, but not to the evolved control group. Therefore, mechanisms for better surviving the stressor were key to evolving desiccation tolerance. Figures 2(a) and 2(b) and Figures 3(a) and 3(b) reveal that survival mechanisms were also key for the dual-stressed group’s evolution of tolerance to desiccation and UV-C radiation. When the radiation group was stressed by UV-C radiation, Figures 3(a) and 3(b) show that it had significant gains, relative to the evolved control group and the ancestor, in both the survival and growth components of fitness. Thus, mechanisms for better surviving the stressor and recovering from it contributed to the evolution of UV-C radiation tolerance. This result refutes my third hypothesis, as I thought that survival mechanisms would be key for evolving tolerance to both stressors, separately and combined. When the radiation group was stressed by desiccation, Figure 2(a) reveals that it incurred a trade-off in survival, according to the first criterion stated above in my fourth hypothesis, whereby the group has a negative mean value relative to the ancestor. However, Figure 2(b) shows that the trade-off was offset by a trade-up in growth. Nevertheless, it can be seen in Figure 2(c) that the offsetting gain in growth was not enough to provide cross-tolerance to desiccation, though there was a gain in total fitness relative to the ancestor. Figures 3(a) and 3(b) show that this dynamic did not play out with the desiccation group when stressed by UV-C radiation, although Figure 3(a) reveals that there was a statistically non-significant loss in survival relative to the ancestor. Desiccation, but not radiation, tolerance incurred a trade-off under benign conditions Figure 4(c) reveals that the evolution of desiccation tolerance came with a trade-off under benign conditions, as per my second criterion. As Figure 5(c) shows, all but one of the desiccation group’s total fitness population means fall below the grand mean of the evolved control group, and of those 5 populations, 4 of them have 95% confidence intervals that do not include the control’s mean. 24 Moreover, half the populations are less fit than the ancestor, and another one’s fitness is the same as that of the ancestor. Figure 4(b) shows that the desiccation group’s trade-off in the benign environment was due to its growth component of fitness being significantly less than that of the evolved control group, but nonetheless slightly better than the ancestor. Conversely, Figure 4(c) reveals that the evolution of UV-C radiation tolerance came with a trade-up under benign conditions, as its total fitness is significantly greater than that of the evolved control group. Figure 4(b) shows that this trade-up in the benign environment was due to a significant gain in the radiation group’s growth component of fitness relative to the evolved control group. It can be seen in Figure 4(b) that, like the radiation group, the dual-stressed group had a significant gain in growth under benign conditions. Nevertheless, Figure 4(c) shows that this gain was insufficient to give a trade-up in that environment, perhaps due to a slight loss in survival (Figure 4(a)). And concerning survival, Figure 4(a) shows that all the treatment groups had slight losses in that fitness component relative to the ancestor, while the evolved control group’s survival remained the same as the ancestor. These results partially refute my fourth, and final, hypothesis, as I thought that all stress-treated groups would incur trade-offs when not stressed. Implications of my study for possible extant life in Mars's shallow subsurface Researchers have experimentally evolved bacteria to tolerate 6 of the harshest stressors existing in the shallow subsurface of Mars: desiccation (this study); UV-C radiation (Alcántara-Díaz et al., 2004; Wassmann et al., 2010; Begyn et al., 2020; Selveshwari et al., 2021; Ellington et al., 2023; this study); ionizing radiation (Wright and Hill, 1968; Licciardello et al., 1969; Davies and Sinskey, 1973; Parisi and Antoine, 1974; Ewing, 1995; Harris et al., 2009; Lim et al., 2009; Byrne et al., 2014; Bruckbauer et al., 2019, 2020); hypobaria (Nicholson et al., 2010); lowered temperature (Mongold et al., 1996); and freeze‒thaw‒growth cycles (Sleight and Lenski, 2007; 25 Kwon et al., 2018). Those experiments have established that non-extremophiles can easily evolve tolerance to Mars-relevant stressors, at least individually. Importantly, my evolution experiment establishes, I believe for the first time, that not only can desiccation tolerance readily evolve, but also that a non-extremophile can quickly adapt to two harsh stressors at the same time. The surface of Mars is a hyper-arid desert (Martínez et al., 2017; Fischer et al., 2019; Polkko et al., 2023), where the relative humidity can be < 5% during the day and near 100% at night, depending on the season and latitude. Nighttime highs of 10% to 30% are typical throughout most of the Martian year, except for the north polar region, where typical highs are 30% to 60%. Those levels of relative humidity and their daily fluctuation would impose severe desiccation stress on any life that might exist near Mars’s surface. Nevertheless, two environments exist on Earth where life persists under extremely arid (and other) conditions that approach those of Mars, the Atacama Desert’s hyper-arid core (Davila et al., 2021; Azua-Bustos et al., 2022; Warren-Rhodes et al., 2022) and Antarctica’s McMurdo Dry Valleys (McKay et al., 2017; Salvatore and Levy, 2021; Warren-Rhodes et al., 2022). A number of studies of those two Mars analogs have demonstrated that desiccation would be a very intense selective pressure on possible Martian life near the surface (e.g., Sun, 2013; Schulze-Makuch et al., 2018). Mars’s surface is soaked in radiation (Gómez‐Elvira et al., 2014; Hassler et al., 2014), with ultraviolet radiation ~4 times that of Earth. Yet there exist terrestrial organisms that could tolerate that level of radiation, e.g., the aforementioned Deinococcus species. My evolution experiment and those by Wassmann et al. (2010) and a few other researchers (cited above) showed that tolerance to UV-C radiation can easily evolve in radiation-sensitive microbes. Furthermore, Godin et al. (2023) found that Bacillus subtilis had a 10% survival rate under broad-spectrum UV radiation when covered by just a dusting (0.3 mm thick) of Mars’s regolith, and Mancinelli and 26 Klovstad (2000) and Schuerger et al. (2003) had similar results with B. subtilis endospores covered by a mere 0.5 mm-thick dust layer. Other researchers have reported that a few millimeters, or even a fraction of a millimeter, of regolith or rock could be enough to attenuate UV-C and broad- spectrum UV radiation to survivable levels (Cockell et al., 2005; Gómez et al., 2010; Schuerger et al., 2012; Mickol et al., 2017; Carrier et al., 2019). Further still, individual regolith grains (Osman et al., 2008) and small-scale geomorphology (Moores et al., 2007) could partially shield microbes from UV radiation. Thus, UV-C radiation would likely be sterilizing only in the uppermost millimeter of Mars’s surface. The evolvability of co-tolerance to desiccation and UV-C radiation, or to any two stressors, would have been highly advantageous as Mars’s climate changed drastically. But today, all 6 Mars-relevant stressors to which tolerance has been experimentally evolved are concurrent in Mars’s shallow subsurface (as opposed to the deep subsurface where conditions are thought to be Earth-like [Tarnas et al., 2021]), and their combined effects would exert tremendous selective pressures on extant life that might exist there. However, while at first this environment, with its multitude of severe stressors, might seem inhospitable, the results of my evolution experiment and those just cited lead me to hypothesize that, if Mars’s drastic climate change allowed adaptation to one or two stressors at a time, then Martian life, if it ever existed, might persist today in the shallow subsurface. Implications of my study for possible extant life in Venus's clouds The relative humidity of Venus’s lower to middle cloud layers is < 0.4% (Hallsworth et al., 2021). Therefore, like Mars, extreme desiccation would exert immense selective pressure on life that might exist in Venus’s clouds. Ionizing and UV-C radiation do not penetrate Venus’s middle cloud layer (Dartnell et al., 2015; Patel et al., 2022). Thus, unlike Mars, possible Venusian cloud life 27 would likely face only minor, intermittent selection from UV-C and ionizing radiation, as these stressors would act only when, or if, cells are lifted to the higher cloud layers. The highly concentrated sulfuric acid droplets within Venus’s potentially habitable cloud layers have a pH less than zero (Grinspoon and Bullock, 2007). Hence, severe acidity would be another intense selective pressure on possible Venusian cloud life, as it is those droplets that life would inhabit. Nevertheless, at least two chemical mechanisms that could reduce the sulfuric acid droplets’ severe acidity have been proposed. Rimmer et al. (2021) suggested that a pH of 1 to 2 could result from the reaction of sulfur dioxide gas dissolving in acidic droplets containing hydroxide salts. The other mechanism, put forth by Bains et al. (2021), builds on that of Rimmer et al. (2021) by suggesting that dissolved sulfur dioxide gas reacts with, in particular, ammonium salts in the acidic droplets, resulting in a pH of ‒1 to 1. If those chemical mechanisms indeed take place in Venus’s clouds, then notwithstanding the tremendous desiccation, there exist terrestrial extreme acidophiles that could persist in the acidic droplets. For example, the archaeon, Picrophilus torridus grows optimally at pH 0.7 and as low as pH ‒0.06 (Schleper et al., 1995)! Moreover, Seager et al. (2023) showed that nucleic acid bases were stable at sulfuric acid concentrations and temperatures that would be encountered in droplets existing in Venus’s middle to lower cloud layers. Taken together, these studies indicate that the severe acidity of Venus’s cloud droplets might have been an evolutionarily surmountable peak. However, the extreme desiccation of Venus’s clouds would have been an even higher and more rugged peak to climb. 28 CONCLUSIONS I have demonstrated that a non-extremophile can readily and rapidly adapt to two harsh, Mars- and Venus-relevant stressors simultaneously. However, co-tolerance to desiccation and UV-C radiation combined required treatment with both stressors, as populations that were treated with just one of the stressors did not evolve cross-tolerance to the other one. Furthermore, the desiccation-treated group incurred a trade-off when exposed to UV-C radiation, and when placed in the benign environment of the control group. Improvements in survival were key to evolving desiccation tolerance, while improvements in both survival and growth were significant for UV-C radiation tolerance. I also found that multiple evolutionary paths led to desiccation tolerance, whereas few paths led to UV-C radiation tolerance. Lastly, I serendipitously discovered that desiccation-treated populations form different cell pellets than those of UV-C radiation-treated and evolved control populations, whereby the former form diffuse cell pellets while the latter form typical condensed pellets. The implications of my findings for our neighboring planets are that life might persist today in Mars’s shallow subsurface and in Venus’s clouds, provided that the drastic environmental changes that occurred on those worlds allowed adaptation to one or two stressors at a time. This is not to say in any way that E. coli, or its experimentally evolved descendants, reflect what microbes might be like on Mars or Venus. I used E. coli simply as a powerful and tractable model organism to explore the evolutionary dynamics of de novo tolerance to two Mars- and Venus-relevant stressors. To that end, future evolution experiments with non-extremophiles like E. coli that simultaneously combine three, or perhaps even four, stressors that are relevant to Mars and/or Venus would significantly expand our understanding of what might have been evolutionarily possible on those planets as their environments became extreme. 29 BIBLIOGRAPHY Adler L, Blomberg A, and Nilsson A (1985) Glycerol metabolism and osmoregulation in the salt- tolerant yeast Debaryomyces hansenii. Journal of Bacteriology 162, 300–306. doi:10.1128/jb.162.1.300-306.1985 Alcántara-Díaz D, Breña-Valle M, and Serment-Guerrero J (2004) Divergent adaptation of light exposures. Mutagenesis 19, 349–354. to cyclic ultraviolet Escherichia coli doi:10.1093/mutage/geh039 Arvidson RE and Catalano JG (2018) Martian habitability as inferred from landed mission observations. In From Habitability to Life on Mars, edited by NA Cabrol and EA Grin, Amsterdam, Netherlands: Elsevier, Inc., pp 77–126. doi:10.1016/B978-0-12-809935-3.00004- 9 Atolia E, Cesar S, Arjes HA, Rajendram M, Shi H, Knapp BD, Khare S, Aranda-Díaz A, Lenski RE, and Huang KC (2020) Environmental and physiological factors affecting high-throughput measurements of bacterial growth. mBio 11, #e01378-20. doi:10.1128/mbio.01378-20 Azua-Bustos A, González-Silva C, and Fairén AG (2022) The Atacama Desert in northern Chile as an analog model of Mars. Frontiers in Astronomy and Space Sciences 8, #810426. doi:10.3389/fspas.2021.810426 Bains W, Petkowski JJ, Rimmer PB, and Seager S (2021) Production of ammonia makes Venusian clouds habitable and explains observed cloud-level chemical anomalies. Proceedings of the National Academy of Sciences of the USA 118, #e2110889118. doi:10.1073/pnas.2110889118 Bains W, Petkowski JJ, and Seager S (2023) Venus’ atmospheric chemistry and cloud life. Astrobiology, ahead of print. characteristics are compatible with Venusian doi:10.1089/ast.2022.0113 Ballu A, Despréaux P, Duplaix C, Dérédec A, Carpentier F, and Walker A-S (2023) Antifungal alternation can be beneficial for durability but at the cost of generalist resistance. Communications Biology 6, #180. doi:10.1038/s42003-023-04550-6 Begyn K, Kim TD, Heyndrickx M, Michiels C, Aertsen A, Rajkovic A, and Devlieghere F (2020) Directed evolution by UV-C treatment of Bacillus cereus spores. International Journal of Food Microbiology 317, #108424. doi:10.1016/j.ijfoodmicro.2019.108424 Bennett AF and Lenski RE (2007) An experimental test of evolutionary trade-offs during temperature adaptation. Proceedings of the National Academy of Sciences of the USA 104, 8649–8654. doi:10.1073/pnas.0702117104 Bennett AF, Lenski RE, and Mittler JE (1992) Evolutionary adaptation to temperature. I. Fitness responses of Escherichia coli to changes in its thermal environment. Evolution 46, 16–30. doi:10.1111/j.1558-5646.1992.tb01981.x 30 Berry BJ, Jenkins DG, and Schuerger AC (2010) Effects of simulated Mars conditions on the survival and growth of Escherichia coli and Serratia liquefaciens. Applied and Environmental Microbiology 76, 2377–2386. doi:10.1128/AEM.02147-09 Borowitzka LJ and Brown AD (1974) The salt relations of marine and halophilic species of the unicellular green alga, Dunaliella. The role of glycerol as a compatible solute. Archives of Microbiology 96, 37–52. doi:10.1007/BF00590161 Box GEP and Cox DR (1964) An analysis of transformations. Journal of the Royal Statistical 26, 211–243. doi:10.1111/j.2517- Society, Series B: Statistical Methodology 6161.1964.tb00553.x Box GEP and Cox DR (1982) An analysis of transformations revisited, rebutted. Journal of the American Statistical Association 77, 209–210. doi:10.1080/01621459.1982.10477788 Boyd SM, Rhinehardt KL, Ewunkem AJ, Harrison SH, Thomas MD, and Graves, Jr JL (2022) Experimental evolution of copper resistance in Escherichia coli produces evolutionary trade- offs in the antibiotics chloramphenicol, bacitracin, and sulfonamide. Antibiotics 11, #711. doi:10.3390/antibiotics11060711 Bruckbauer ST, Trimarco JD, Martin J, Bushnell B, Senn KA, Schackwitz W, Lipzen A, Blow M, Wood EA, Culberson WS, Pennacchio C, and Cox MM (2019) Experimental evolution of extreme resistance to ionizing radiation in Escherichia coli after 50 cycles of selection. Journal of Bacteriology 201, #e00784-18. doi:10.1128/jb.00784-18 Bruckbauer ST, Martin J, Minkoff BB, Veling MT, Lancaster I, Liu J, Trimarco JD, Bushnell B, Lipzen A, Wood EA, Sussman MR, Pennacchio C, and Cox MM (2020) Physiology of highly radioresistant Escherichia coli after experimental evolution for 100 cycles of selection. Frontiers in Microbiology 11, #582590. doi:10.3389/fmicb.2020.582590 Byrne RT, Klingele AJ, Cabot EL, Schackwitz WS, Martin JA, Martin J, Wang Z, Wood EA, Pennacchio C, Pennacchio LA, Perna NT, Battista JR, and Cox MM (2014) Evolution of extreme resistance to ionizing radiation via genetic adaptation of DNA repair. eLife 3, #e01322. doi:10.7554/eLife.01322 Cabrol NA (2018) The coevolution of life and environment on Mars: an ecosystem perspective on the robotic exploration of biosignatures. Astrobiology 18, 1–27. doi:10.1089/ast.2017.1756 Cabrol NA (2021) Tracing a modern biosphere on Mars. Nature Astronomy 5, 210–212. doi:10.1038/s41550-021-01327-x Carrier BL, Abbey WJ, Beegle LW, Bhartia R, and Liu Y (2019) Attenuation of ultraviolet radiation in rocks and minerals: Implications for Mars Science. Journal of Geophysical Research: Planets 124, 2599–2612. doi:10.1029/2018JE005758 Caspeta L and Nielsen J (2015) Thermotolerant yeast strains adapted by laboratory evolution show trade-off at ancestral temperatures and preadaptation to other stresses. mBio 6, #e00431-15. doi:10.1128/mbio.00431-15 31 Chang S-L and Leu J-Y (2011) A tradeoff drives the evolution of reduced metal resistance in natural populations of yeast. PLoS Genetics 7, #e1002034. doi:10.1371/journal.pgen.1002034 Clark BC (1998) Surviving the limits to life at the surface of Mars. Journal of Geophysical Research: Planets 103, 28545–28555. doi:10.1029/98JE02081 Cockell CS (1999) Life on Venus. Planetary and Space Science 47, 1487–1501. doi:10.1016/S0032-0633(99)00036-7 Cockell CS, Schuerger AC, Billi D, Friedmann EI, and Panitz C (2005) Effects of a simulated martian UV flux on the cyanobacterium, Chroococcidiopsis sp. 029. Astrobiology 5, 127–140. doi:10.1089/ast.2005.5.127 Comont D, Lowe C, Hull R, Crook L, Hicks HL, Onkokesung N, Beffa R, Childs DZ, Edwards R, Freckleton RP, and Neve P (2020) Evolution of generalist resistance to herbicide mixtures reveals a trade-off in resistance management. Nature Communications 11, #3086. doi:10.1038/s41467-020-16896-0 Cook RD (1977) Detection of influential observation in linear regression. Technometrics 19, 15– 18. doi:10.1080/00401706.1977.10489493 Daegelen P, Studier FW, Lenski RE, Cure S, and Kim JF (2009) Tracing ancestors and relatives of Escherichia coli B, and the derivation of B strains REL606 and BL21(DE3). Journal of Molecular Biology 394, 634–643. doi:10.1016/j.jmb.2009.09.022 Daly MJ (2023) The scientific revolution that unraveled the astonishing DNA repair capacity of the Deinococcaceae: 40 years on. Canadian Journal of Microbiology, e-First (pre-print). doi:10.1139/cjm-2023-0059 Dartnell LR, Nordheim TA, Patel MR, Mason JP, Coates AJ, and Jones GH (2015) Constraints on Icarus 257, 396–405. I. Cosmic rays. a potential aerial biosphere on Venus: doi:10.1016/j.icarus.2015.05.006 Davies R and Sinskey AJ (1973) Radiation-resistant mutants of Salmonella typhimurium LT2: 133–144. of Bacteriology characterization. Journal 113, and development doi:10.1128/jb.113.1.133-144.1973 Davila AF and Schulze-Makuch D (2016) The last possible outposts for life on Mars. Astrobiology 16, 159–168. doi:10.1089/ast.2015.1380 Davila A, Kahre MA, Quinn R, and Des Marais DJ (2020) The biological potential of present-day Mars. In Planetary Astrobiology, The University of Arizona Space Science Series, edited by V Meadows, G Arney, B Schmidt, and DJ Des Marais, Tucson, Arizona: The University of Arizona Press, pp 169–184. doi:10.2458/azu_uapress_9780816540068-ch007 Davila AF, Warren-Rhodes K, and DiRuggiero J (2021) The Atacama Desert: a window into late Mars surface habitability? In Mars Geological Enigmas: From the Late Noachian Epoch to 32 the Present Day, edited by RJ Soare, SJ Conway, J-P Williams, and DZ Oehler, Amsterdam, Netherlands: Elsevier, Inc., pp 333–355. doi:10.1016/B978-0-12-820245-6.00012-4 Delacre M, Leys C, Mora YL, and Lakens D (2019) Taking parametric assumptions seriously: arguments for the use of Welch’s F-test instead of the classical F-test in one-way ANOVA. International Review of Social Psychology 32, #13. doi:10.5334/irsp.198 Derrick B, Toher D, and White P (2016) Why Welch’s test is Type I error robust. The Quantitative Methods for Psychology 12, 30–38. doi:10.20982/tqmp.12.1.p030 Dohm JM, Miyamoto H, Ori GG, Fairén AG, Davila AF, Komatsu G, Mahaney WC, Williams J- P, Joye SB, Di Achille G, Oehler DZ, Marzo GA, Schulze-Makuch D, Acocella V, Glamoclija M, Pondrelli M, Boston P, Hart KM, Anderson RC, Baker VR, Fink W, Kelleher BP, Furfaro R, Gross C, Hare TM, Frazer AR, Ip F, Allen CCR, Kim KJ, Maruyama S, McGuire PC, Netoff D, Parnell J, Wendt L, Wheelock SJ, Steele A, Hancock RGV, Havics RA, Costa P, and Krinsley D (2011) An inventory of potentially habitable environments on Mars: Geological and biological perspectives. In Analogs for Planetary Exploration, Geological Society of America Special Paper 483, edited by WB Garry and JE Bleacher, Boulder, Colorado: Geological Society of America, pp 317–347. doi:10.1130/2011.2483(21) Dong N, Li H-R, Yuan M, Zhang X-H, and Yu Y (2015) Deinococcus antarcticus sp. nov., isolated from soil. International Journal of Systematic and Evolutionary Microbiology 65, 331–335. doi:10.1099/ijs.0.066324-0 Dunnett CW (1980) Pairwise multiple comparisons in the unequal variance case. Journal of the American Statistical Association 75, 796–800. doi:10.1080/01621459.1980.10477552 Ellington AJ, Schult TJ, Reisch CR, and Christner BC (2023) Comparative genomics and directed evolution reveal genetic determinants of extreme UVC radiation tolerance in bacteria recovered from the stratosphere. bioRxiv, 2023.03.27.534493. doi:10.1101/2023.03.27.534493 Esbelin J, Santos T, and Hébraud M (2018) Desiccation: An environmental and food industry 82–88. Food Microbiology commonly face. 69, stress doi:10.1016/j.fm.2017.07.017 bacteria that Ewing D (1995) The directed evolution of radiation resistance in E. coli. Biochemical and Biophysical Research Communications 216, 549–553. doi:10.1006/bbrc.1995.2657 Fairén AG, Dohm JM, Uceda ER, Rodríguez AP, Baker VR, Fernández-Remolar D, Schulze- Makuch D, and Amils R (2005) Prime candidate sites for astrobiological exploration through the hydrogeological history of Mars. Planetary and Space Science 53, 1355–1375. doi:10.1016/j.pss.2005.06.007 Fairén AG, Parro V, Schulze-Makuch D, and Whyte L (2017) Searching for life on Mars before it is too late. Astrobiology 17, 962–970. doi:10.1089/ast.2017.1703 33 Fischer E, Martínez GM, Rennó NO, Tamppari LK, and Zent AP (2019) Relative humidity on Mars: New results from the Phoenix TECP sensor. Journal of Geophysical Research: Planets 124, 2780–2792. doi:10.1029/2019JE006080 França MB, Panek AD, and Eleutherio ECA (2007) Oxidative stress and its effects during dehydration. Comparative Biochemistry and Physiology Part A: Molecular & Integrative Physiology 146, 621–631. doi:10.1016/j.cbpa.2006.02.030 Fredrickson JK, Li SW, Gaidamakova EK, Matrosova VY, Zhai M, Sulloway HM, Scholten JC, Brown MG, Balkwill DL, and Daly MJ (2008) Protein oxidation: key to bacterial desiccation resistance? The ISME Journal 2, 393–403. doi:10.1038/ismej.2007.116 Friman V-P, Ghoul M, Molin S, Johansen HK, and Buckling A (2013) Pseudomonas aeruginosa adaptation to lungs of cystic fibrosis patients leads to lowered resistance to phage and protist enemies. PLoS ONE 8, #e75380. doi:10.1371/journal.pone.0075380 Gibbs AG, Chippindale AK, and Rose MR (1997) Physiological mechanisms of evolved desiccation resistance in Drosophila melanogaster. Journal of Experimental Biology 200, 1821–1832. doi:10.1242/jeb.200.12.1821 Gillmann C, Way MJ, Avice G, Breuer D, Golabek GJ, Höning D, Krissansen-Totton J, Lammer H, O’Rourke JG, Persson M, Plesa A-C, Salvador A, Scherf M, and Zolotov MY (2022) The long-term evolution of the atmosphere of Venus: processes and feedback mechanisms. Space Science Reviews 218, #56. doi:10.1007/s11214-022-00924-0 Godin PJ, Moore CA, Smith C, and Moores JE (2023) Absorption and scattering of UV and visible light through simulated martian regoliths and rock samples. Astrobiology 23, 280–290. doi:10.1089/ast.2021.0184 Gómez F, Mateo-Martí E, Prieto-Ballesteros O, Martín-Gago J, and Amils R (2010) Protection of chemolithoautotrophic bacteria exposed to simulated Mars environmental conditions. Icarus 209, 482–487. doi:10.1016/j.icarus.2010.05.027 Gómez‐Elvira J, Armiens C, Carrasco I, Genzer M, Gómez F, Haberle R, Hamilton VE, Harri A- M, Kahanpää H, Kemppinen O, Lepinette A, Martín Soler J, Martín‐Torres J, Martínez‐Frías J, Mischna M, Mora L, Navarro S, Newman C, de Pablo MA, Peinado V, Polkko J, Rafkin SCR, Ramos M, Rennó NO, Richardson M, Rodríguez‐Manfredi JA, Romeral Planelló JJ, Sebastián E, de la Torre Juárez M, Torres J, Urquí R, Vasavada AR, Verdasca J, and Zorzano M-P (2014) Curiosity’s rover environmental monitoring station: Overview of the first 100 sols. Journal of Geophysical Research: Planets 119, 1680–1688. doi:10.1002/2013JE004576 Greffe VRG and Michiels J (2020) Desiccation-induced cell damage in bacteria and the relevance for inoculant production. Applied Microbiology and Biotechnology 104, 3757–3770. doi:10.1007/s00253-020-10501-6 Grinspoon DH and Bullock MA (2007) Astrobiology and Venus exploration. In Exploring Venus as a Terrestrial Planet, Geophysical Monograph Series 176, edited by LW Esposito, ER 34 Stofan, and TE Cravens, Washington, DC: American Geophysical Union, pp 191–206. doi:10.1029/176GM12 de Groot A, Chapon V, Servant P, Christen R, Saux MF-L, Sommer S, and Heulin T (2005) Deinococcus deserti sp. nov., a gamma-radiation-tolerant bacterium isolated from the Sahara Desert. International Journal of Systematic and Evolutionary Microbiology 55, 2441–2446. doi:10.1099/ijs.0.63717-0 Grotzinger JP, Sumner DY, Kah LC, Stack K, Gupta S, Edgar L, Rubin D, Lewis K, Schieber J, Mangold N, Milliken R, Conrad PG, DesMarais D, Farmer J, Siebach K, Calef III F, Hurowitz J, McLennan SM, Ming D, Vaniman D, Crisp J, Vasavada A, Edgett KS, Malin M, Blake D, Gellert R, Mahaffy P, Wiens RC, Maurice S, Grant JA, Wilson S, Anderson RC, Beegle L, Arvidson R, Hallet B, Sletten RS, Rice M, Bell III J, Griffes J, Ehlmann B, Anderson RB, Bristow TF, Dietrich WE, Dromart G, Eigenbrode J, Fraeman A, Hardgrove C, Herkenhoff K, Jandura L, Kocurek G, Lee S, Leshin LA, Leveille R, Limonadi D, Maki J, McCloskey S, Meyer M, Minitti M, Newsom H, Oehler D, Okon A, Palucis M, Parker T, Rowland S, Schmidt M, Squyres S, Steele A, Stolper E, Summons R, Treiman A, Williams R, Yingst A, and the MSL Science Team (2014) A habitable fluvio-lacustrine environment at Yellowknife Bay, Gale crater, Mars. Science 343, #1242777. doi:10.1126/science.1242777 Grotzinger JP, Gupta S, Malin MC, Rubin DM, Schieber J, Siebach K, Sumner DY, Stack KM, Vasavada AR, Arvidson RE, Calef III F, Edgar L, Fischer WF, Grant JA, Griffes J, Kah LC, Lamb MP, Lewis KW, Mangold N, Minitti ME, Palucis M, Rice M, Williams RME, Yingst RA, Blake D, Blaney D, Conrad P, Crisp J, Dietrich WE, Dromart G, Edgett KS, Ewing RC, Gellert R, Hurowitz JA, Kocurek G, Mahaffy P, McBride MJ, McLennan SM, Mischna M, Ming D, Milliken R, Newsom H, Oehler D, Parker TJ, Vaniman D, Wiens RC, and Wilson SA (2015) Deposition, exhumation, and paleoclimate of an ancient lake deposit, Gale crater, Mars. Science 350, #aac7575. doi:10.1126/science.aac7575 Hallsworth JE, Koop T, Dallas TD, Zorzano M-P, Burkhardt J, Golyshina OV, Martín-Torres J, Dymond MK, Ball P, and McKay CP (2021) Water activity in Venus’s uninhabitable clouds and other planetary atmospheres. Nature Astronomy 5, 665–675. doi:10.1038/s41550-021- 01391-3 Hansen AA, Jensen LL, Kristoffersen T, Mikkelsen K, Merrison J, Finster KW, and Lomstein BAa (2009) Effects of long-term simulated martian conditions on a freeze-dried and homogenized bacterial permafrost community. Astrobiology 9, 229–240. doi:10.1089/ast.2008.0244 Harri A-M, Paton M, Hieta M, Polkko J, Newman C, Pla-García J, Leino J, Mäkinen T, Kauhanen J, Jaakonaho I, Sánchez-Lavega A, Hueso R, Genzer M, Lorenz RD, Lemmon MT, Vicente- Retortillo A, Tamppari L, Viúdez-Moreiras D, de la Torre Juárez M, Savijärvi H, Rodríguez- Manfredi JA, and Martinez G (2023) Perseverance MEDA atmospheric pressure observations — 168500355.59581586/v1. results. doi:10.22541/essoar.168500355.59581586/v1 Archive, initial Open ESS Harris DR, Pollock SV, Wood EA, Goiffon RJ, Klingele AJ, Cabot EL, Schackwitz W, Martin J, Eggington J, Durfee TJ, Middle CM, Norton JE, Popelars MC, Li H, Klugman SA, Hamilton 35 LL, Bane LB, Pennacchio LA, Albert TJ, Perna NT, Cox MM, and Battista JR (2009) Directed evolution of ionizing radiation resistance in Escherichia coli. Journal of Bacteriology 191, 5240–5252. doi:10.1128/jb.00502-09 Hassler DM, Zeitlin C, Wimmer-Schweingruber RF, Ehresmann B, Rafkin S, Eigenbrode JL, Brinza DE, Weigle G, Böttcher S, Böhm E, Burmeister S, Guo J, Köhler J, Martin C, Reitz G, Cucinotta FA, Kim M-H, Grinspoon D, Bullock MA, Posner A, Gómez-Elvira J, Vasavada A, Grotzinger JP, and the MSL Science Team (2014) Mars’ surface radiation environment measured with the Mars Science Laboratory’s Curiosity rover. Science 343, #1244797. doi:10.1126/science.1244797 Hayes AF and Cai L (2007) Further evaluating the conditional decision rule for comparing two independent means. British Journal of Mathematical and Statistical Psychology 60, 217–244. doi:10.1348/000711005X62576 Hoffmann AA and Parsons PA (1989) An integrated approach to environmental stress tolerance and life-history variation: desiccation tolerance in Drosophila. Biological Journal of the Linnean Society 37, 117–136. doi:10.1111/j.1095-8312.1989.tb02098.x Hurowitz JA, Grotzinger JP, Fischer WW, McLennan SM, Milliken RE, Stein N, Vasavada AR, Blake DF, Dehouck E, Eigenbrode JL, Fairén AG, Frydenvang J, Gellert R, Grant JA, Gupta S, Herkenhoff KE, Ming DW, Rampe EB, Schmidt ME, Siebach KL, Stack-Morgan K, Sumner DY, and Wiens RC (2017) Redox stratification of an ancient lake in Gale crater, Mars. Science 356, #eaah6849. doi:10.1126/science.aah6849 Izutsu M and Lenski RE (2022) Experimental test of the contributions of initial variation and new mutations to adaptive evolution in a novel environment. Frontiers in Ecology and Evolution 10, #958406. doi:10.3389/fevo.2022.958406 Jakosky BM (2021) Atmospheric loss to space and the history of water on Mars. Annual Review of Earth and Planetary Sciences 49, 71–93. doi:10.1146/annurev-earth-062420-052845 Kerney KR and Schuerger AC (2011) Survival of Bacillus subtilis endospores on ultraviolet- irradiated rover wheels and Mars regolith under simulated martian conditions. Astrobiology 11, 477–485. doi:10.1089/ast.2011.0615 Khan S, Siraj S, Shahid M, Haque MM, and Islam A (2023) Osmolytes: Wonder molecules to combat protein misfolding against stress conditions. International Journal of Biological Macromolecules 234, #123662. doi:10.1016/j.ijbiomac.2023.123662 Kim C and Storer BE (1996) Reference values for Cook’s distance. Communications in Statistics — Simulation and Computation 25, 691–708. doi:10.1080/03610919608813337 Kragh ML and Truelstrup Hansen L (2020) Initial transcriptomic response and adaption of Listeria monocytogenes to desiccation on food grade stainless steel. Frontiers in Microbiology 10, #3132. doi:10.3389/fmicb.2019.03132 36 Krisko A and Radman M (2010) Protein damage and death by radiation in Escherichia coli and Deinococcus radiodurans. Proceedings of the National Academy of Sciences of the USA 107, 14373–14377. doi:10.1073/pnas.1009312107 Krisko A and Radman M (2013a) Biology of extreme radiation resistance: the way of Deinococcus #a012765. Spring Harbor Perspectives in Biology 5, radiodurans. Cold doi:10.1101/cshperspect.a012765 Krisko A and Radman M (2013b) Phenotypic and genetic consequences of protein damage. PLoS Genetics 9, #e1003810. doi:10.1371/journal.pgen.1003810 Kwon YW, Bae J-H, Kim S-A, and Han NS (2018) Development of freeze-thaw tolerant Lactobacillus rhamnosus GG by adaptive laboratory evolution. Frontiers in Microbiology 9, #2781. doi:10.3389/fmicb.2018.02781 Lammer H, Zerkle AL, Gebauer S, Tosi N, Noack L, Scherf M, Pilat-Lohinger E, Güdel M, Grenfell JL, Godolt M, and Nikolaou A (2018) Origin and evolution of the atmospheres of early Venus, Earth and Mars. The Astronomy and Astrophysics Review 26, #2. doi:10.1007/s00159-018-0108-y Lange L, Forget F, Banfield D, Wolff M, Spiga A, Millour E, Viúdez-Moreiras D, Bierjon A, Piqueux S, Newman C, Pla-García J, and Banerdt WB (2022) InSight pressure data recalibration, and its application to the study of long-term pressure changes on Mars. Journal of Geophysical Research: Planets 127, #e2022JE007190. doi:10.1029/2022JE007190 Laskowska E and Kuczyńska-Wiśnik D (2020) New insight into the mechanisms protecting bacteria during desiccation. Current Genetics 66, 313–318. doi:10.1007/s00294-019-01036-z Lasne J, Noblet A, Szopa C, Navarro-González R, Cabane M, Poch O, Stalport F, François P, Atreya SK, and Coll P (2016) Oxidants at the surface of Mars: a review in light of recent exploration results. Astrobiology 16, 977–996. doi:10.1089/ast.2016.1502 Lebre PH, De Maayer P, and Cowan DA (2017) Xerotolerant bacteria: surviving through a dry spell. Nature Reviews Microbiology 15, 285–296. doi:10.1038/nrmicro.2017.16 Lenski RE (1988) Experimental studies of pleiotropy and epistasis in Escherichia coli. I. Variation in competitive fitness among mutants resistant to virus T4. Evolution 42, 425–432. doi:10.1111/j.1558-5646.1988.tb04149.x Lenski RE (2023) Revisiting the design of the long-term evolution experiment with Escherichia coli. Journal of Molecular Evolution 91, 241–253. doi:10.1007/s00239-023-10095-3 Lenski RE and Travisano M (1994) Dynamics of adaptation and diversification: a 10,000- generation experiment with bacterial populations. Proceedings of the National Academy of Sciences of the USA 91, 6808–6814. doi:10.1073/pnas.91.15.6808 37 Lenski RE, Rose MR, Simpson SC, and Tadler SC (1991) Long-term experimental evolution in Escherichia coli. I. Adaptation and divergence during 2,000 generations. The American Naturalist 138, 1315–1341. doi:10.1086/285289 Levin BR, Stewart FM, and Chao L (1977) Resource-limited growth, competition, and predation: A model and experimental studies with bacteria and bacteriophage. The American Naturalist 111, 3–24. doi:10.1086/283134 Licciardello JJ, Nickerson JTR, Goldblith SA, Shannon CA, and Bishop WW (1969) Development in Salmonella cultures. Applied Microbiology 18, 24–30. of radiation resistance doi:10.1128/am.18.1.24-30.1969 Lim S, Yun H, Joe M, and Kim D (2009) Development of the radiation-resistant strain of Moraxella osloensis and effect of penicillin G on its growth. Radiation Physics and Chemistry 78, 437–440. doi:10.1016/j.radphyschem.2009.03.023 Limaye SS, Mogul R, Smith DJ, Ansari AH, Słowik GP, and Vaishampayan P (2018) Venus’ spectral signatures and the potential for life in the clouds. Astrobiology 18, 1181–1198. doi:10.1089/ast.2017.1783 Limaye SS, Zelenyi L, and Zasova L (2021a) Introducing the Venus collection—papers from the layer. Astrobiology 21, 1157–1162. the cloud first workshop on habitability of doi:10.1089/ast.2021.0142 Limaye SS, Mogul R, Baines KH, Bullock MA, Cockell C, Cutts JA, Gentry DM, Grinspoon DH, Head JW, Jessup K-L, Kompanichenko V, Lee YJ, Mathies R, Milojevic T, Pertzborn RA, Rothschild L, Sasaki S, Schulze-Makuch D, Smith DJ, and Way MJ (2021b) Venus, an astrobiology target. Astrobiology 21, 1163–1185. doi:10.1089/ast.2020.2268 Liu Z, Kim MC, Wang L, Zhu G, Zhang Y, Huang Y, Wei Z, Danzeng W, and Peng F (2017) Deinococcus taklimakanensis sp. nov., isolated from desert soil. International Journal of Systematic and Evolutionary Microbiology 67, 4311–4316. doi:10.1099/ijsem.0.002168 Losa-Adams E, Gil-Lozano C, Fairén AG, Bishop JL, Rampe EB, and Gago-Duport L (2021) Long-lasting habitable periods in Gale crater constrained by glauconitic clays. Nature Astronomy 5, 936–942. doi:10.1038/s41550-021-01397-x Luo W, Song F, and Xie Y (2008) Trade-off between tolerance to drought and tolerance to flooding in three wetland plants. Wetlands 28, 866–873. doi:10.1672/07-225.1 Łupkowska A, Monem S, Dębski J, Stojowska-Swędrzyńska K, Kuczyńska-Wiśnik D, and Laskowska E (2023) Protein aggregation and glycation in Escherichia coli exposed to desiccation-rehydration #127335. Microbiological stress. doi:10.1016/j.micres.2023.127335 Research 270, Mancinelli RL and Klovstad M (2000) Martian soil and UV radiation: microbial viability assessment on spacecraft surfaces. Planetary and Space Science 48, 1093–1097. doi:10.1016/S0032-0633(00)00083-0 38 Martínez GM, Newman CN, De Vicente-Retortillo A, Fischer E, Renno NO, Richardson MI, Fairén AG, Genzer M, Guzewich SD, Haberle RM, Harri A-M, Kemppinen O, Lemmon MT, Smith MD, de la Torre-Juárez M, and Vasavada AR (2017) The modern near-surface Martian climate: A review of in-situ meteorological data from Viking to Curiosity. Space Science Reviews 212, 295–338. doi:10.1007/s11214-017-0360-x McKay CP and Stoker CR (1989) The early environment and its evolution on Mars: Implications for life. Reviews of Geophysics 27, 189–214. doi:10.1029/RG027i002p00189 McKay CP, Andersen D, and Davila A (2017) Antarctic environments as models of planetary habitats: University Valley as a model for modern Mars and Lake Untersee as a model for 303–318. The Enceladus doi:10.1080/2154896X.2017.1383705 ancient Mars. Journal Polar and 7, Mickol RL, Page JL, and Schuerger AC (2017) Magnesium sulfate salt solutions and ices fail to protect Serratia liquefaciens from the biocidal effects of UV irradiation under martian conditions. Astrobiology 17, 401–412. doi:10.1089/ast.2015.1448 Mongold JA, Bennett AF, and Lenski RE (1996) Evolutionary adaptation to temperature. IV. Adaptation of Escherichia coli at a niche boundary. Evolution 50, 35–43. doi:10.1111/j.1558- 5646.1996.tb04470.x Moores JE, Smith PH, Tanner R, Schuerger AC, and Venkateswaran KJ (2007) The shielding effect of small-scale martian surface geometry on ultraviolet flux. Icarus 192, 417–433. doi:10.1016/j.icarus.2007.07.003 Moser BK and Stevens GR (1992) Homogeneity of variance in the two-sample means test. The American Statistician 46, 19–21. doi:10.1080/00031305.1992.10475839 Moser BK, Stevens GR, and Watts CL (1989) The two-sample t test versus Satterthwaite’s approximate F test. Communications in Statistics — Theory and Methods 18, 3963–3975. doi:10.1080/03610928908830135 Munguira A, Hueso R, Sánchez-Lavega A, de la Torre-Juarez M, Martínez GM, Newman CE, Sebastian E, Lepinette A, Vicente-Retortillo A, Chide B, Lemmon MT, Bertrand T, Lorenz RD, Banfield D, Gómez-Elvira J, Martín-Soler J, Navarro S, Pla-García J, Rodríguez-Manfredi JA, Romeral J, Smith MD, and Torres J (2023) Near surface atmospheric temperatures at Jezero from Mars 2020 MEDA measurements. Journal of Geophysical Research: Planets 128, #e2022JE007559. doi:10.1029/2022JE007559 Nicholson WL, Fajardo-Cavazos P, Fedenko J, Ortíz-Lugo JL, Rivas-Castillo A, Waters SM, and Schuerger AC (2010) Exploring the low-pressure growth limit: Evolution of Bacillus subtilis in the laboratory to enhanced growth at 5 kilopascals. Applied and Environmental Microbiology 76, 7559–7565. doi:10.1128/AEM.01126-10 Novak M, Pfeiffer T, Lenski RE, Sauer U, and Bonhoeffer S (2006) Experimental tests for an evolutionary trade‐off between growth rate and yield in E. coli. The American Naturalist 168, 242–251. doi:10.1086/506527 39 Osman S, Peeters Z, La Duc MT, Mancinelli R, Ehrenfreund P, and Venkateswaran K (2008) Effect of shadowing on survival of bacteria under conditions simulating the Martian atmosphere and UV radiation. Applied and Environmental Microbiology 74, 959–970. doi:10.1128/AEM.01973-07 Ostrowski EA, Rozen DE, and Lenski RE (2005) Pleiotropic effects of beneficial mutations in Escherichia coli. Evolution 59, 2343–2352. doi:10.1111/j.0014-3820.2005.tb00944.x Parisi A and Antoine AD (1974) Increased radiation resistance of vegetative Bacillus pumilus. Applied Microbiology 28, 41–46. doi:10.1128/am.28.1.41-46.1974 Patel MR, Mason JP, Nordheim TA, and Dartnell LR (2022) Constraints on a potential aerial #114796. II. Ultraviolet radiation. Icarus 373, biosphere doi:10.1016/j.icarus.2021.114796 on Venus: Petropoulos B and Telonis P (1988) The lower atmosphere of Venus after Vega 1 and 2 missions. Earth, Moon, and Planets 43, 45–60. doi:10.1007/BF00056477 Polkko J, Hieta M, Harri A-M, Tamppari L, Martínez G, Viúdez-Moreiras D, Savijärvi H, Conrad P, Zorzano Mier MP, De La Torre Juarez M, Hueso R, Munguira A, Leino J, Gómez F, Jaakonaho I, Fischer E, Genzer M, Apestigue V, Arruego I, Banfield D, Lepinette A, Paton M, Rodriguez-Manfredi JA, Sánchez Lavega A, Sebastian E, Toledo D, Vicente-Retortillo A, and the MEDA team (2023) Initial results of the relative humidity observations by MEDA instrument onboard the Mars 2020 Perseverance rover. Journal of Geophysical Research: Planets 128, #e2022JE007447. doi:10.1029/2022JE007447 Porter SS and Rice KJ (2013) Trade-offs, spatial heterogeneity, and the maintenance of microbial diversity. Evolution 67, 599–608. doi:10.1111/j.1558-5646.2012.01788.x Raymond JA, Morgan-Kiss R, and Stahl-Rommel S (2020) Glycerol is an osmoprotectant in two Antarctic Chlamydomonas species from an ice-covered saline lake and is synthesized by an unusual bidomain enzyme. Frontiers in Plant Science 11, #1259. doi:10.3389/fpls.2020.01259 Rimmer PB, Jordan S, Constantinou T, Woitke P, Shorttle O, Hobbs R, and Paschodimas A (2021) Hydroxide salts in the clouds of Venus: their effect on the sulfur cycle and cloud droplet pH. The Planetary Science Journal 2, #133. doi:10.3847/PSJ/ac0156 Roberts MF (2005) Organic compatible solutes of halotolerant and halophilic microorganisms. Saline Systems 1, #5. doi:10.1186/1746-1448-1-5 Rozen DE and Lenski RE (2000) Long‐term experimental evolution in Escherichia coli. VIII. Dynamics of a balanced polymorphism. The American Naturalist 155, 24–35. doi:10.1086/303299 Ruxton GD (2006) The unequal variance t-test is an underused alternative to Student’s t-test and the Mann–Whitney U test. Behavioral Ecology 17, 688–690. doi:10.1093/beheco/ark016 40 Salvatore MR and Levy JS (2021) The McMurdo Dry Valleys of Antarctica: a geological, environmental, and ecological analog to the Martian surface and near surface. In Mars Geological Enigmas: From the Late Noachian Epoch to the Present Day, edited by RJ Soare, SJ Conway, J-P Williams, and DZ Oehler, Amsterdam, Netherlands: Elsevier, Inc., pp 291– 332. doi:10.1016/B978-0-12-820245-6.00011-2 Santos AL, Oliveira V, Baptista I, Henriques I, Gomes NCM, Almeida A, Correia A, and Cunha  (2013) Wavelength dependence of biological damage induced by UV radiation on bacteria. Archives of Microbiology 195, 63–74. doi:10.1007/s00203-012-0847-5 SAS (2015) The MIXED procedure. In SAS/STAT® 14.1 User’s Guide, Cary, North Carolina, USA: SAS Institute Inc., pp 6047–6241. Satterthwaite FE (1946) An approximate distribution of estimates of variance components. Biometrics Bulletin 2, 110–114. doi:10.2307/3002019 Schleper C, Puehler G, Holz I, Gambacorta A, Janekovic D, Santarius U, Klenk H-P, and Zillig W (1995) Picrophilus gen. nov., fam. nov.: a novel aerobic, heterotrophic, thermoacidophilic genus and family comprising archaea capable of growth around pH 0. Journal of Bacteriology 177, 7050–7059. doi:10.1128/jb.177.24.7050-7059.1995 Schuerger AC and Nicholson WL (2006) Interactive effects of hypobaria, low temperature, and CO2 atmospheres inhibit the growth of mesophilic Bacillus spp. under simulated martian conditions. Icarus 185, 143–152. doi:10.1016/j.icarus.2006.06.014 Schuerger AC, Mancinelli RL, Kern RG, Rothschild LJ, and McKay CP (2003) Survival of endospores of Bacillus subtilis on spacecraft surfaces under simulated martian environments: implications for the forward contamination of Mars. Icarus 165, 253–276. doi:10.1016/S0019- 1035(03)00200-8 Schuerger AC, Golden DC, and Ming DW (2012) Biotoxicity of Mars soils: 1. Dry deposition of analog soils on microbial colonies and survival under Martian conditions. Planetary and Space Science 72, 91–101. doi:10.1016/j.pss.2012.07.026 Schuerger AC, Mickol RL, and Schwendner P (2020) The hypopiezotolerant bacterium, Serratia liquefaciens, failed to grow in Mars analog soils under simulated Martian conditions at 7 hPa. Life 10, #77. doi:10.3390/life10060077 Schulze-Makuch D (2021) The case (or not) for life in the Venusian clouds. Life 11, #255. doi:10.3390/life11030255 Schulze-Makuch D, Grinspoon DH, Abbas O, Irwin LN, and Bullock MA (2004) A sulfur-based survival strategy for putative phototrophic life in the venusian atmosphere. Astrobiology 4, 11– 18. doi:10.1089/153110704773600203 Schulze-Makuch D, Irwin LN, Lipps JH, LeMone D, Dohm JM, and Fairén AG (2005) Scenarios for the evolution of life on Mars. Journal of Geophysical Research: Planets 110, #E12S23. doi:10.1029/2005JE002430 41 Schulze-Makuch D, Irwin LN, and Fairén AG (2013) Drastic environmental change and its effects on a planetary biosphere. Icarus 225, 775–780. doi:10.1016/j.icarus.2013.05.001 Schulze-Makuch D, Wagner D, Kounaves SP, Mangelsdorf K, Devine KG, de Vera J-P, Schmitt- Kopplin P, Grossart H-P, Parro V, Kaupenjohann M, Galy A, Schneider B, Airo A, Frösler J, Davila AF, Arens FL, Cáceres L, Cornejo FS, Carrizo D, Dartnell L, DiRuggiero J, Flury M, Ganzert L, Gessner MO, Grathwohl P, Guan L, Heinz J, Hess M, Keppler F, Maus D, McKay CP, Meckenstock RU, Montgomery W, Oberlin EA, Probst AJ, Sáenz JS, Sattler T, Schirmack J, Sephton MA, Schloter M, Uhl J, Valenzuela B, Vestergaard G, Wörmer L, and Zamorano P (2018) Transitory microbial habitat in the hyperarid Atacama Desert. Proceedings of the National Academy of Sciences of the USA 115, 2670–2675. doi:10.1073/pnas.1714341115 Schwendner P, Jobson M-E, and Schuerger AC (2020) Addition of anaerobic electron acceptors to solid media did not enhance growth of 125 spacecraft bacteria under simulated low-pressure Martian conditions. Scientific Reports 10, #18290. doi:10.1038/s41598-020-75222-2 Seager S, Petkowski JJ, Gao P, Bains W, Bryan NC, Ranjan S, and Greaves J (2021) The venusian lower atmosphere haze as a depot for desiccated microbial life: A proposed life cycle for 21, 1206–1223. persistence of doi:10.1089/ast.2020.2244 aerial biosphere. Astrobiology the venusian Seager S, Petkowski JJ, Seager MD, Grimes Jr. JH, Zinsli Z, Vollmer-Snarr HR, Abd El-Rahman MK, Wishart DS, Lee BL, Gautam V, Herrington L, Bains W, and Darrow C (2023) Stability of nucleic acid bases in concentrated sulfuric acid: Implications for the habitability of Venus’ clouds. Proceedings of the National Academy of Sciences of the USA 120, #e2220007120. doi:10.1073/pnas.2220007120 Selveshwari S, Lele K, and Dey S (2021) Genomic signatures of UV resistance evolution in Escherichia coli depend on the growth phase during exposure. Journal of Evolutionary Biology 34, 953–967. doi:10.1111/jeb.13764 Shapiro SS and Wilk MB (1965) An analysis of variance test for normality (complete samples). Biometrika 52, 591–611. doi:10.1093/biomet/52.3-4.591 Slade D and Radman M (2011) Oxidative stress resistance in Deinococcus radiodurans. Microbiology and Molecular Biology Reviews 75, 133–191. doi:10.1128/MMBR.00015-10 Sleight SC and Lenski RE (2007) Evolutionary adaptation to freeze‐thaw‐growth cycles in Escherichia coli. Physiological and Biochemical Zoology 80, 370–385. doi:10.1086/518013 Sleight SC, Wigginton NS, and Lenski RE (2006) Increased susceptibility to repeated freeze-thaw cycles in Escherichia coli following long-term evolution in a benign environment. BMC Evolutionary Biology 6, #104. doi:10.1186/1471-2148-6-104 Smith DJ, Schuerger AC, Davidson MM, Pacala SW, Bakermans C, and Onstott TC (2009) Survivability of Psychrobacter cryohalolentis K5 under simulated martian surface conditions. Astrobiology 9, 221–228. doi:10.1089/ast.2007.0231 42 Somero GN (1992) Adapting to water stress: convergence on common solutions. In Water and Life, edited by GN Somero, CB Osmond, and CL Bolis, Berlin, Heidelberg: Springer, pp 3– 18. doi:10.1007/978-3-642-76682-4_1 Squyres SW, Grotzinger JP, Arvidson RE, Bell III JF, Calvin W, Christensen PR, Clark BC, Crisp JA, Farrand WH, Herkenhoff KE, Johnson JR, Klingelhöfer G, Knoll AH, McLennan SM, McSween Jr. HY, Morris RV, Rice Jr. JW, Rieder R, and Soderblom LA (2004) In situ evidence for an ancient aqueous environment at Meridiani Planum, Mars. Science 306, 1709– 1714. doi:10.1126/science.1104559 Squyres SW, Arvidson RE, Ruff S, Gellert R, Morris RV, Ming DW, Crumpler L, Farmer JD, Des Marais DJ, Yen A, McLennan SM, Calvin W, Bell III JF, Clark BC, Wang A, McCoy TJ, Schmidt ME, and de Souza Jr. PA (2008) Detection of silica-rich deposits on Mars. Science 320, 1063–1067. doi:10.1126/science.1155429 Sriram S, Roopa KP, and Savitha MJ (2011) Extended shelf-life of liquid fermentation derived talc formulations of Trichoderma harzianum with the addition of glycerol in the production medium. Crop Protection 30, 1334–1339. doi:10.1016/j.cropro.2011.06.003 Sun HJ (2013) Endolithic microbial life in extreme cold climate: snow is required, but perhaps less is more. Biology 2, 693–701. doi:10.3390/biology2020693 Tarnas JD, Mustard JF, Sherwood Lollar B, Stamenković V, Cannon KM, Lorand J-P, Onstott TC, Michalski JR, Warr O, Palumbo AM, and Plesa A-C (2021) Earth-like habitable environments in the subsurface of Mars. Astrobiology 21, 741–756. doi:10.1089/ast.2020.2386 Tejeda MT, Arredondo J, Liedo P, Pérez-Staples D, Ramos-Morales P, and Díaz-Fleischer F (2016) Reasons for success: Rapid evolution for desiccation resistance and life-history changes in the polyphagous fly Anastrepha ludens. Evolution 70, 2583–2594. doi:10.1111/evo.13070 Travisano M and Lenski RE (1996) Long-term experimental evolution in Escherichia coli. IV. the specificity of adaptation. Genetics 143, 15–26. Targets of selection and doi:10.1093/genetics/143.1.15 Travisano M, Vasi F, and Lenski RE (1995) Long-term experimental evolution in Escherichia coli. III. Variation among replicate populations in correlated responses to novel environments. Evolution 49, 189–200. doi:10.1111/j.1558-5646.1995.tb05970.x Turner PE, Cooper VS, and Lenski RE (1998) Tradeoff between horizontal and vertical modes of in bacterial plasmids. Evolution 52, 315–329. doi:10.1111/j.1558- transmission 5646.1998.tb01634.x Vasavada AR (2022) Mission overview and scientific contributions from the Mars Science Laboratory Curiosity rover after eight years of surface operations. Space Science Reviews 218, #14. doi:10.1007/s11214-022-00882-7 43 Wang J, Raza W, Jiang G, Yi Z, Fields B, Greenrod S, Friman V-P, Jousset A, Shen Q, and Wei Z (2023) Bacterial volatile organic compounds attenuate pathogen virulence via evolutionary trade-offs. The ISME Journal 17, 443–452. doi:10.1038/s41396-023-01356-6 Warren-Rhodes K, Weinstein S, Piatek JL, Dohm J, Hock A, Minkley E, Pane D, Ernst LA, Fisher G, Emani S, Waggoner AS, Cabrol NA, Wettergreen DS, Grin E, Coppin P, Diaz C, Moersch J, Oril GG, Smith T, Stubbs K, Thomas G, Wagner M, Wyatt M, and Boyle LN (2007a) Robotic ecological mapping: Habitats and the search for life in the Atacama Desert. Journal of Geophysical Research: Biogeosciences 112, #G04S06. doi:10.1029/2006JG000301 Warren-Rhodes K, Phillips M, Davila A, and McKay CP (2022) Insights of extreme desert ecology to the habitats and habitability of Mars. In Microbiology of Hot Deserts, Ecological Studies, Volume 244, edited by J-B Ramond and DA Cowan, Cham, Switzerland: Springer Nature Switzerland AG, pp 235–291. doi:10.1007/978-3-030-98415-1_9 Warren-Rhodes KA, Rhodes KL, Pointing SB, Ewing SA, Lacap DC, Gómez-Silva B, Amundson R, Friedmann EI, and McKay CP (2006) Hypolithic cyanobacteria, dry limit of photosynthesis, and microbial ecology in the hyperarid Atacama Desert. Microbial Ecology 52, 389–398. doi:10.1007/s00248-006-9055-7 Warren-Rhodes KA, Dungan JL, Piatek J, Stubbs K, Gómez-Silva B, Chen Y, and McKay CP (2007b) Ecology and spatial pattern of cyanobacterial community island patches in the Atacama Desert, Chile. Journal of Geophysical Research: Biogeosciences 112, #G04S15. doi:10.1029/2006JG000305 Warren-Rhodes KA, Lee KC, Archer SDJ, Cabrol N, Ng-Boyle L, Wettergreen D, Zacny K, Pointing SB, and the NASA Life in the Atacama Project Team (2019) Subsurface microbial habitats in an extreme desert Mars-analog environment. Frontiers in Microbiology 10, #69. doi:10.3389/fmicb.2019.00069 Wassmann M, Moeller R, Reitz G, and Rettberg P (2010) Adaptation of Bacillus subtilis cells to Archean-like UV climate: relevant hints of microbial evolution to remarkably increased radiation resistance. Astrobiology 10, 605–615. doi:10.1089/ast.2009.0455 Welch BL (1938) The significance of the difference between two means when the population variances are unequal. Biometrika 29, 350–362. doi:10.2307/2332010 Welch BL (1951) On the comparison of several mean values: an alternative approach. Biometrika 38, 330–336. doi:10.2307/2332579 Westall F, Höning D, Avice G, Gentry D, Gerya T, Gillmann C, Izenberg N, Way MJ, and Wilson C (2023) The habitability of Venus. Space Science Reviews 219, #17. doi:10.1007/s11214- 023-00960-4 Wright SJL and Hill EC (1968) The development of radiation-resistant cultures of Escherichia ‘growth-irradiation cycles.’ Microbiology 51, 97–106. coli 1 by a process of doi:10.1099/00221287-51-1-97 44 Yang JW, Chang F-H, Yeh Y-C, Tsai A-Y, Chiang K-P, Shiah F-K, Gong G-C, and Hsieh C (2023) Trade-offs between competitive ability and resistance to top-down control in marine microbes. mSystems 8, #e01017-22. doi:10.1128/msystems.01017-22 Yang Y, Itoh T, Yokobori S, Itahashi S, Shimada H, Satoh K, Ohba H, Narumi I, and Yamagishi A (2009) Deinococcus aerius sp. nov., isolated from the high atmosphere. International Journal 1862–1866. Systematic of doi:10.1099/ijs.0.007963-0 Evolutionary Microbiology and 59, Yang Y, Itoh T, Yokobori S, Shimada H, Itahashi S, Satoh K, Ohba H, Narumi I, and Yamagishi A (2010) Deinococcus aetherius sp. nov., isolated from the stratosphere. International Journal of Systematic and Evolutionary Microbiology 60, 776–779. doi:10.1099/ijs.0.010876-0 Yobi A, Wone BWM, Xu W, Alexander DC, Guo L, Ryals JA, Oliver MJ, and Cushman JC (2013) Metabolomic profiling in Selaginella lepidophylla at various hydration states provides new insights into the mechanistic basis of desiccation tolerance. Molecular Plant 6, 369–385. doi:10.1093/mp/sss155 45 APPENDIX FIGURES Figure 1: Stress–growth procedure. The treatment groups underwent daily pulses of stress from desiccation (a) or UV-C radiation (b), or both stressors during the evolution experiment. From left to right, an 8-µL aliquot of an 800-µL culture grown for 22 ‒ 24 h at 37°C was spotted on the bottom of a small glass beaker, subjected to the stressor(s), re-suspended in fresh medium and then re-grown for 22 ‒ 24 h before repeating this procedure (75 cycles total). For the group treated with both stressors, the desiccation conditions remained the same while the exposure to UV-C radiation was halved (~2 J cm−2). Halving the UV-C radiation exposure ensured sufficient survival of that treatment group after being subjected to both stressors (UV-C radiation followed by desiccation). During competitive fitness assays, the cell densities used to calculate fitness were determined at the points in the procedure labeled with an “N” as explained in the text (see Materials and methods, sub-section “Fitness components and their calculations”). 46 Figure 2: Fitness of the evolved groups when stressed by desiccation. The survival (a) and growth (b) fitness components were summed to give the total fitness (c). Fitness is expressed as a selection-rate differential, where a value of zero corresponds to the null hypothesis of no difference between the evolved group and ancestor. Triangles indicate that the fitness assay stress, in this case, desiccation, was also the treatment for that respective group during the evolution experiment. The error bars represent 95% confidence intervals about the group mean, where the degrees of freedom are the number of replicate populations in each group, minus one (6 – 1 = 5). Multiple two-tailed Welch’s t-tests were performed to compare each treatment group mean to the evolved control group mean, with the number of asterisks signifying the following Dunnett’s T3-adjusted P-values: *, 0.01 < P ≤ 0.05; **, 0.001 < P ≤ 0.01; ***, P ≤ 0.001; “ns”, no significant difference (P > 0.05). 47 Figure 3: Fitness of the evolved groups when stressed by UV-C radiation. The details of the charts are the same as those described in Figure 2’s legend. Concerning the total fitness, the desiccation group’s residuals were not normally distributed. Thus, I performed a Box-Cox power transformation of the entire total fitness data set. The stippled columns and error bars represent the untransformed data, while the asterisks represent the transformed data that were used to make comparisons of each treatment group’s total fitness mean to that of the evolved control group. 48 Figure 4: Fitness of the evolved groups when not stressed (control conditions). The details of the charts are the same as those described in Figure 2’s legend. Concerning the total fitness, the desiccation group’s residuals were not normally distributed. Therefore, I transformed the entire total fitness data set and made comparisons in the same way as described in Figure 3’s legend. 49 Figure 5: Total fitness of the evolved groups’ replicate populations. The populations’ total fitness when stressed by desiccation (a), UV-C radiation (b), and not stressed (c). Triangles indicate that the fitness assay stress was also the treatment for those respective populations during the evolution experiment. The error bars represent 95% confidence intervals about the population mean, where the degrees of freedom are the number of replicate measurements for each population, minus one (4 – 1 = 3). One-way Welch’s ANOVAs were performed to determine if the means of populations within a group varied significantly, with the number of asterisks signifying the following one- tailed P-values: *, 0.01 < P ≤ 0.05; **, 0.001 < P ≤ 0.01; ***, P ≤ 0.001; “ns”, no significant difference (P > 0.05). Stippled columns specify that transformed data was used in the ANOVA, 50 Figure 5 (cont’d) but the columns themselves and error bars represent the untransformed data. The dashed line marks the mean of the evolved control group. 51 Figure 6: Differences in cell pellets. Shown here is a composite image of a single 96-deep-well plate, seen from above, looking down into the wells. A row between the control populations and UV-C radiation-evolved populations was cropped out because that row contained cell pellets of populations not presented in this paper. The outermost wells to the left and right are negative controls (no cells). Populations that evolved under stress from desiccation only and those evolved under stress from desiccation and UV-C radiation combined formed diffuse cell pellets after settling overnight in their spent growth medium, as is evident in the upper two rows. These diffuse cell pellets were not formed by populations that evolved under stress from UV-C radiation only and those evolved under benign conditions (control), instead, those populations formed typical, condensed pellets, as can be seen in the lower two rows. 52