Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Comparison of Bacillus subtilis transcriptome profiles from two separate missions to the International Space Station

Comparison of Bacillus subtilis transcriptome profiles from two separate missions to the... www.nature.com/npjmgrav ARTICLE OPEN Comparison of Bacillus subtilis transcriptome profiles from two separate missions to the International Space Station 1 1 1 Michael D. Morrison , Patricia Fajardo-Cavazos and Wayne L. Nicholson The human spaceflight environment is notable for the unique factor of microgravity, which exerts numerous physiologic effects on macroscopic organisms, but how this environment may affect single-celled microbes is less clear. In an effort to understand how the microbial transcriptome responds to the unique environment of spaceflight, the model Gram-positive bacterium Bacillus subtilis was flown on two separate missions to the International Space Station in experiments dubbed BRIC-21 and BRIC-23. Cells were grown to late-exponential/early stationary phase, frozen, then returned to Earth for RNA-seq analysis in parallel with matched ground control samples. A total of 91 genes were significantly differentially expressed in both experiments; 55 exhibiting higher transcript levels in flight samples and 36 showing higher transcript levels in ground control samples. Genes upregulated in flight samples notably included those involved in biofilm formation, biotin and arginine biosynthesis, siderophores, manganese transport, toxin production and resistance, and sporulation inhibition. Genes preferentially upregulated in ground control samples notably included those responding to oxygen limitation, e.g., fermentation, anaerobic respiration, subtilosin biosynthesis, and anaerobic regulatory genes. The results indicated differences in oxygen availability between flight and ground control samples, likely due to differences in cell sedimentation and the toroidal shape assumed by the liquid cultures in microgravity. npj Microgravity (2019) 5:1 ; doi:10.1038/s41526-018-0061-0 INTRODUCTION formation and architecture; and resistance to antibiotics or abiotic 5,6,9 stresses. More recent efforts tended toward gene expression In certain respects, human spaceflight habitats resemble other studies using genome-wide techniques such as microarrays to confined built environments, such as submersible vehicles, understand how the global pattern of RNA synthesis (i.e., the aircraft, hospital isolation wards, or remote research installations. transcriptome) responds to the spaceflight environment. To date, However, the spaceflight environment is unique because it microarray studies have reported a wide range of responses to contains two additional altered physical parameters: reduced spaceflight including increased transcription of genes encoding (micro-)gravity and increased ionizing radiation from solar and 10,11 12 general metabolism, secondary metabolite biosynthesis, galactic sources. Extensive investigations conducted in spaceflight 11,13 13,14 synthesis of ribosomal proteins, and virulence factors. on macroscopic organisms have resulted in a relatively good Regardless of the output measured, it has proven difficult to understanding of the biological effects of microgravity and derive consistent conclusions from these disparate studies due to radiation at levels ranging from the whole body down to the several confounding factors. 2 3 organ, cellular, and molecular level in humans, animals, and First, until recently spaceflight transcriptome studies have been plants. While microorganisms have also been the subject of performed on only a small selection of Gram-negative bacteria focused research in the spaceflight environment, it has proven (Salmonella enterica serovar Typhimurium, Pseudomonas aerugi- more difficult to understand their responses to spaceflight nosa, Rhodospirillum rubrum), limiting the ability to generalize 5–7 stress. From a theoretical perspective, exposure to microgravity conclusions to a broader range of microbes. Second, spaceflight results in a number of alterations in a microbial cell’s immediate experiments have been conducted under widely different: (i) surroundings, such as loss of convective mass and heat transfer, culture conditions (e.g., media formulations, agar vs. liquid, reduction in mechanical shear forces, and alterations in the way aeration, temperature); (ii) growth stage of the cultures at harvest; liquids behave at air and solid interfaces. Changes in such (iii) spaceflight hardware employed; (iv) pre- and post-flight 6,7 fundamental physical forces alter the rates at which gases, treatment of samples; and (v) types of assays conducted. Third, nutrients, signaling molecules, and waste products are exchanged experimental variation derives from the measurements them- between microbes and their surroundings. It has been proposed selves (technical effects) or from the natural variation inherent in that upon perception of these alterations in their environment, biological systems (biological effects). In an effort to control for microbes mount a complex set of stress responses (the so-called variation, prior microbial spaceflight experiments have included “spaceflight syndrome” ). multiple replicates; however, most experiments reported in the Considerable effort has been expended to understand microbial literature have been flown on a single mission only. Intense responses to spaceflight and their underlying causes. In early competition for limited cargo space destined to research plat- studies, various phenotypic outputs from microbes grown in space forms such as the International Space Station (ISS) generally were measured, such as: growth rate and yield; virulence; biofilm results in the choice a new experiment taking precedence over a Department of Microbiology and Cell Science, University of Florida, Merritt Island, FL, USA Correspondence: Wayne L. Nicholson (WLN@ufl.edu) Received: 20 June 2018 Accepted: 6 November 2018 Published in cooperation with the Biodesign Institute at Arizona State University, with the support of NASA M.D. Morrison et al. Fig. 2 Venn diagrams showing number of genes significantly differentially expressed in the BRIC-21, BRIC-23, or both missions. Total genes (top), genes expressed higher in FL than GC samples (FL > GC; middle) or genes expressed higher in GC than FL samples (GC > FL; bottom) are depicted Fig. 1 Principal Component Analysis of the datasets from BRIC-21 FL (red triangles) and GC (green circles) samples, and BRIC-23 FL (blue Overview of the datasets squares) and GC (purple diamonds) samples In order to assess the quality and reproducibility of the datasets obtained, principal component analysis (PCA) was performed. In repetition of a previously flown experiment. Because of this, the the PCA, the first and second principal components explained 52 intrinsic mission-to-mission variability in the response of micro- and 21% of the variance, respectively. Four distinct population biological systems to the spaceflight environment has remained clusters were identified corresponding to the four environmental largely unexplored. conditions tested (Fig. 1). In the BRIC-21 FL and GC samples, the In 2015, we were afforded the opportunity to send an three replicates were grouped rather tightly, indicating relatively experimental package to the ISS to test the responses of the good agreement. In the BRIC-23 FL and GC samples, the nine Gram-positive bacterium Bacillus subtilis to the human spaceflight replicates were somewhat more disperse, but still formed distinct environment. This was the 21st mission to the ISS using Biological groups (Fig. 1). Examination of Principal Component 1 revealed Research in Canister-Petri Dish Fixation Unit (BRIC-PDFU) hard- that the major source of variation in the datasets derived from the ware, and the experiment was dubbed BRIC-21. From the BRIC-21 differences in the two missions themselves, while variation in experiment we have previously reported in detail measurements Principal Component 2 was due to differences between the FL and of the growth, antibiotic resistance, frequency and spectrum of GC datasets in each experiment (Fig. 1). mutagenesis exhibited by B. subtilis flight (FL) samples in The B. subtilis strain 168 transcriptome consists of 4397 total 16,17 comparison to matched ground control (GC) samples. In genes, of which 4280 encode proteins. Analysis of the BRIC-21 addition, we also performed RNA-seq analyses to compare the RNA-seq data resulted in identification of 293 total genes whose transcriptomes of BRIC-21 FL vs. GC samples, as we will report in expression differed significantly in FL vs. GC samples, representing this communication. In 2016 we had the good fortune, in ~6.8% of the protein-coding genome. Of these genes, 177 were significantly higher in FL samples, and 116 were significantly collaboration with the NASA GeneLab group, to fly a second higher in GC samples. These data are summarized in Supple- mission to the ISS (dubbed BRIC-23) using the same B. subtilis mental Table S1. Analysis of the BRIC-23 RNA-seq data resulted in strain, media, and hardware, and again to perform RNA-seq identification of 255 total genes whose expression differed analyses on the samples. Here we provide a comparative analysis significantly in FL vs. GC samples, representing ~6.0% of the of the B. subtilis transcriptome profiles from the BRIC-21 and BRIC- protein-coding transcriptome. Of these genes, 163 were signifi- 23 spaceflight missions. We report on the complete transcriptome cantly higher in FL samples, and 92 were significantly higher in GC profiling of a Gram-positive bacterium grown in the human samples. These data are summarized in Supplemental Table S2. spaceflight environment (a prior study focused on a subset of We reasoned that comparison of the transcriptome datasets primary and secondary metabolite genes in Streptomyces coelico- from the BRIC-21 and BRIC-23 experiments would identify genes lor ). In this study, we show the effect of exposure to the human that were significantly up- or down-regulated in both missions, spaceflight environment on the B. subtilis transcriptome by thus defining genes whose expression was consistently altered in identifying sets of genes expressed in common in both the response to spaceflight. A comparison of the datasets obtained BRIC-21 and BRIC-23 missions. from the BRIC-21 and BRIC-23 experiments is depicted graphically as Venn diagrams (Fig. 2). A total of 91 genes were significantly differentially expressed in both experiments. Fifty-five of the shared genes exhibited higher transcript levels in FL samples and RESULTS 36 genes showed higher transcript levels in GC samples (Fig. 2). RNA-seq was used to characterize the transcriptomic response of The finding that only ~1/3 of the significantly differentially B. subtilis cultures exposed to the human spaceflight environment expressed transcripts were shared in both the BRIC-21 and BRIC-23 of the ISS (FL samples) vs. matched GC samples on two separate missions indicated a substantial amount of between-experiment missions, BRIC-21 (n = 3) and BRIC-23 (n = 9). The data were variation. What could be the source of this variation? While we analyzed using the bioinformatics pipeline described in the attempted to keep discrepancies between the two experiments to section 'Methods' and the results are presented below. a minimum, one notable difference between the BRIC-21 and npj Microgravity (2019) 1 Published in cooperation with the Biodesign Institute at Arizona State University, with the support of NASA 1234567890():,; M.D. Morrison et al. BRIC-23 experiments was the difference in incubation times (25 vs. have documented biofilms containing Bacillus spp. in the 36 h), which may have contributed to the ~2/3 discordance in the Space Shuttle water system. Laboratory strains of B. subtilis two datasets. Unfortunately, BRIC-PDFU hardware does not allow such as strain 168 and its descendants do not form robust direct measurement of growth to be determined in situ during biofilms due to mutations that have accumulated during spaceflight. However, because the transcription of genes encoding their domestication, and biofilm formation was not noted fundamental growth processes (e.g., replication, transcription, in FL or GC samples from BRIC-21 or BRIC-23. Nonetheless, translation) exhibits a positive correlation with growth rate, as a numerous biofilm-related genes were observed to be proxy for growth rate we compared the fold changes for significantly upregulated in both BRIC-21 and BRIC-23 FL transcripts involved in DNA replication (dnaE), RNA transcription samples, including genes of the srfAA-AB-AC-AD-ycxA (rpoABCD), and protein synthesis (rps and rpl genes encoding small operon, which encodes the pathway for production of the 29,30 and large ribosomal subunit proteins). Neither limma or DESeq2 cyclic lipopeptide surfactin and genes of the tapA-sipW- identified a significant difference in the transcript expression tasA operon encoding the major protein matrix component levels for these genes in either the BRIC-21 or BRIC-23 datasets, of biofilms (Table 1). B. subtilis biofilms also contain as a suggesting that the FL and GC samples were exhibiting similar major component the exopolysaccharide poly-N-acetylglu- growth rates at the time of harvest for RNA extraction. cosamine, produced in a biosynthetic pathway encoded In pre-flight ground-based experiments we determined that by the epsABCDEFGHIJKLMNO operon. Our analysis found under ISS ambient temperature (~23 °C), 25 and 36 h of incubation that only the epsB, epsC, and epsI genes were significantly corresponded to late-exponential phase and the “transition state” upregulated in FL samples of both BRIC-21 and BRIC-23 between exponential and stationary-phase growth, respectively. (Table 1). However, closer examination of the datasets Entrance into the transition state in B. subtilis results in the revealed that in the BRIC-23 experiment, 12 of the 15 eps transcriptional activation of nearly 300 genes which comprise a genes (epsABCDEFGHIJKL) were significantly upregulated in regulon under control of the AbrB protein; we reasoned that FL samples (Supplemental Table S2). In the BRIC-21 comparison of AbrB-controlled transcripts between the BRIC-21 experiment, these 12 genes also displayed a significant and BRIC-23 datasets might provide insight into the growth phase upregulation in FL samples (p < 0.01), but only three of these of these two populations. We found that 60 and 57 AbrB- genes (epsB, espC, and epsI) met our >2-fold change cutoff, dependent transcripts were significantly altered in the BRIC-21 while the remaining nine genes exhibited fold changes and BRIC-23 datasets, respectively, and that 27 transcripts were slightly below the cutoff (Supplemental Table S1). The last significantly altered in both experiments (Fig. 3). From this analysis three genes in the eps operon, epsM, epsN, and epsO, were it therefore appeared that cultures in both experiments were at or found not to be significantly differentially expressed in FL near the transition phase of growth. samples from either mission. In addition, examination of the BRIC-21 dataset revealed that the bslA, ycdA, and luxS genes, also involved in swarming motility and biofilm forma- Genes upregulated in FL samples 33,34 tion, were significantly upregulated in FL samples The 55 genes whose expression was significantly upregulated in (Supplemental Table S1). FL samples from both BRIC-21 and BRIC-23 are listed in Table 1. (ii) Biotin biosynthesis. Biotin (vitamin H) is an essential cofactor They are arranged according to their BSU locus tag, i.e., in the for enzymes such as acetyl-CoA carboxylase and pyruvate order that they are located on the B. subtilis 168 chromosome carboxylase, which are important in fatty acid metabolism map. Examination of the data revealed upregulation of blocks of and central metabolism, respectively. Production of biotin genes associated with particular phenotypes in B. subtilis: from pimelic acid is accomplished by the gene products of the bioWAFDBI biosynthetic operon. Examination of the data (i) Biofilm formation. Under particular environmental condi- from Table 1 revealed significant upregulation in FL samples tions, various microorganisms can produce biofilms consist- of the bioW, bioA, bioD, and bioB genes in both the BRIC-21 ing of cells embedded in a matrix of extracellular polymeric and BRIC-23 missions, as well as upregulation of the bioY substances (EPS) consisting of polysaccharides, proteins, gene encoding the energy coupling factor (ECF) transporter nucleic acids, and lipids. It has been reported that biotin-specific S-protein (Table 1). In addition, inspection of spaceflight exposure promoted biofilm formation in Pseu- 23 24 the BRIC-21 dataset showed significant upregulation in FL domonas aeruginosa and Escherichia coli, and promoted samples of the bioF and bioI biotin biosynthetic genes, the invasive growth of the yeasts Saccharomyces cerevisiae and yuiG (bioYB) gene encoding a putative second biotin-specific Candida albicans. Bacterial biofilms have been implicated ECF transporter S-protein, the yhfT gene whose product is in contamination and biofouling of potable water systems in involved in surfactin production, and the yhfS encoding a long-duration space habitats, and data from spaceflight putative acetyl-CoA C-acetyltransferase (Supplemental Table S1); along with the bio genes, these additional genes belong to a regulon under control of a repressor named BirA, thus are likely upregulated as a block. (iii) Siderophores. Iron is an important cofactor for several enzymes, but is only sparingly soluble in most aerobic environments. To acquire iron, most microorganisms produce and excrete siderophores which bind iron with high affinity for subsequent import by specific transport systems. B. subtilis produces the siderophore bacillibactin (2,3-dihydroxybenzoate) encoded by the besA-dhbACEBF operon. FL samples from both BRIC-21 and BRIC-23 were found to significantly upregulate the besA and dhbC transcripts (Table 1), and subsequent examination of the BRIC-21 data revealed that the entire besA-dhbACEBF operon was significantly upregulated in FL samples (Supplementary Fig. 3 Venn diagram showing genes of the transition-state AbrB Table S1). Another siderophore produced by B. subtilis called regulon significantly differentially expressed in the BRIC-21, BRIC-23, or both missions pulcherriminic acid or pulcherrimin is encoded by two small Published in cooperation with the Biodesign Institute at Arizona State University, with the support of NASA npj Microgravity (2019) 1 M.D. Morrison et al. Table 1. List of genes significantly upregulated in FL samples both in BRIC-21 and BRIC-23 experiments Locus Tag Gene BRIC-21 BRIC-23 Annotated function Regulon Name DESeq2 limma DESeq2 limma BSU02050 ybdO 1.94 2.03 1.25 1.24 Unknown AbrB, SigD BSU02140 glpT 1.43 1.48 1.06 1.10 Glycerol-3-phosphate transporter GlpP, CcpA, PhoP, SigA BSU02310 ybfO 1.46 1.51 1.89 1.99 Similar to erythromycin esterase AbrB, SigW BSU02700 lipA 1.44 1.50 1.23 1.27 Extracellular lipase Unknown BSU02710 yczC 1.93 1.99 1.08 1.07 Unknown Unknown BSU03200 putB 3.16 3.67 1.02 1.02 Proline dehydrogenase CodY PutR, SigA, Spo0A BSU03480 srfAA 2.61 2.91 1.84 1.85 Surfactin synthase subunit 1 Abh, CodY, ComA, PerR, PhoP, SigA, Spx BSU03490 srfAB 3.06 3.28 2.19 2.22 Surfactin synthase subunit 2 Abh, CodY, ComA, PerR, PhoP, SigA, Spx BSU03510 srfAC 3.16 3.37 2.47 2.51 Surfactin synthase subunit 3 Abh, CodY, ComA, PerR, PhoP, SigA, Spx BSU03520 srfAD 3.18 3.46 2.63 2.67 Surfactin synthase thioesterase subunit Abh, CodY, ComA, PerR, PhoP, SigA, Spx BSU03530 ycxA 2.37 2.44 2.08 2.08 Unknown ComA BSU_tRNA_36 trnD-Thr 1.02 1.13 1.03 1.06 Threonyl transfer RNA Unknown BSU09710 bmrC 1.01 1.05 1.33 1.33 Multidrug resistance ABC transporter ATP-binding protein AbrB, BmrB BSU09720 bmrD 1.05 1.09 1.21 1.17 Multidrug resistance ABC transporter ATP-binding protein AbrB, BmrB BSU10370 bioY 3.81 4.10 1.40 1.43 S-protein of biotin ECF transporter BirA BSU11381 appA/1 2.36 2.45 1.31 1.30 Oligopeptide ABC transporter, inactive pseudogene in strain CodY, ScoC, TnrA BSU11382 appA/2 2.52 2.61 1.17 1.10 Oligopeptide ABC transporter, inactive pseudogene in strain CodY, ScoC, TnrA BSU12010 manP 2.38 2.67 1.38 1.47 PTS system-mannose-specific transporter subunit EIIBCA ManR, SigA BSU15960 sivC 2.89 3.06 2.67 2.71 Inhibitor of entry into sporulation via KinB or KinC AbrB, SigD BSU18000 citB 1.83 1.99 1.17 1.18 aconitase CcpA, CcpC, CitB, CodY, FsrA, SigA BSU21330 yomK 1.47 1.58 1.90 1.93 Unknown SPβ prophage BSU21420 bhlA 1.09 1.34 −1.05 −1.28 Holin-like protein SPβ prophage BSU24620 tasA 2.22 2.33 3.11 3.38 major component of biofilm matrix, forms amyloid fibers AbrB, LutR, RemA, SigA, SinR BSU24630 sipW 2.37 2.51 2.81 3.52 Bifunctional signal peptidase I that controls surface-adhered AbrB, LutR, RemA, SigA, biofilm formation and processes TasA and TapA SinR BSU24640 tapA 2.77 2.93 2.78 2.96 TasA anchoring/assembly protein AbrB, LutR, RemA, SigA, SinR BSU26490 yrkJ 1.73 1.85 1.96 1.92 Unknown Unknown BSU26500 yrkI 2.32 2.50 2.22 2.48 Unknown Unknown BSU26510 yrkH 2.63 2.80 2.50 2.89 Unknown Unknown BSU26530 yrkF 3.22 3.62 2.77 3.84 Unknown Unknown BSU26540 yrkE 3.16 3.47 2.76 3.65 Unknown Unknown BSU29440 argH 2.57 2.67 1.25 1.22 Argininosuccinate lyase AhrC BSU29450 argG 2.50 2.61 1.29 1.30 Argininosuccinate synthase AhrC BSU30200 bioB 4.35 4.64 1.20 1.16 Biotin synthase BirA BSU30210 bioD 4.77 5.05 1.67 1.73 Dethiobiotin synthase BirA BSU30230 bioA 4.57 4.75 1.36 1.31 Lysine-8-amino-7-oxononanoate aminotransferase BirA BSU30240 bioW 5.63 5.94 2.31 2.41 6-carboxyhexanoate–CoA ligase BirA BSU30740 mntD 3.95 4.11 3.13 3.32 Manganese ABC transporter (permease) MntR BSU30750 mntC 3.97 4.12 2.98 3.31 Manganese ABC transporter (membrane protein) MntR BSU30760 mntB 3.83 3.98 2.82 3.15 Manganese ABC transporter (ATP-binding protein) MntR BSU30770 mntA 3.12 3.32 2.70 2.76 Manganese ABC transporter (Mn-binding lipoprotein) MntR BSU31250 tlpA 1.42 1.47 1.32 1.33 Methyl-accepting chemotaxis protein AbrB, SigD BSU31990 dhbC 2.48 2.57 1.05 1.08 Isochorismate synthase; siderophore bacillibactin synthesis AbrB, Fur, Kre, SigA, SigI BSU32010 besA 2.31 2.51 1.39 1.40 Trilactone hydrolase, catalyzes ferri-bacillibactin hydrolysis AbrB, Fur leading to cytosolic iron release npj Microgravity (2019) 1 Published in cooperation with the Biodesign Institute at Arizona State University, with the support of NASA M.D. Morrison et al. Table 1 continued Locus Tag Gene BRIC-21 BRIC-23 Annotated function Regulon Name DESeq2 limma DESeq2 limma BSU32450 pucL 1.07 1.32 −1.14 −1.32 Urate oxidase PucR, SigA, TnrA BSU33140 yvqJ 1.20 1.50 1.29 1.34 Similar to to macrolide-efflux protein AbrB BSU33490 cadA 2.37 3.38 1.59 2.23 Cadmium transporting ATPase, resistance to cadmium CzrA, SigA BSU33770 sdpC 4.41 4.61 1.56 1.61 Toxin, collapses the proton motive force and induces AbrB, Rok, Spo0A autolysis, kills non-sporulating cells, induces activity of SigW BSU34290 epsI 1.01 1.10 1.66 1.68 Glycosyltransferase, synthesis of extracellular poly-N- AbrB, EAR riboswitch, acetylglucosamine RemA, SigA, SinR BSU34350 epsC 1.12 1.17 1.48 1.48 UDP-sugar epimerase, required for extracellular AbrB, EAR riboswitch, polysaccharide synthesis RemA, SigA, SinR BSU34360 epsB 1.01 1.04 2.29 2.41 Extracellular polysaccharide synthesis, protein tyrosine kinase AbrB, EAR riboswitch, RemA, SigA, SinR BSU35070 yvmC 1.93 2.11 1.15 1.13 Cyclodipeptide synthase; biosynthesis of the extracellular AbrB, PchR iron chelate pulcherrimin BSU35080 pchR 3.70 3.91 1.55 1.53 Transcriptional repressor (MarR family), controls the CcpA, PchR expression of genes involved in pulcherriminic acid biosynthesis BSU37780 rocA 1.10 1.31 −1.14 −1.35 3-hydroxy-1-pyrroline-5-carboxylate dehydrogenase; AbrB, AhrC, CodY, RocR, arginine, ornithine and citrulline utilization SigL BSU37800 sivA 1.62 1.79 2.14 2.20 Inhibitor of KinA autophosphorylation, and subsequently of AbrB entry into sporulation BSU40180 yydF 4.55 4.77 2.51 2.82 Secreted peptide, controls LiaR-LiaS activity AbrB, Rok, SigA Values are log -fold FL:GC expression ratios. Gene names, annotated functions, and regulons are from Subtiwiki (http://subtiwiki.uni-goettingen.de/v3/index. php) , accessed on September 24, 2018 operons, yvmC-cypX and pchR-yvmA. The yvmC and pchR could be exhibiting a higher degree of resistance to genes were found to be significantly upregulated in FL antibiotics or toxic compounds. However, in a separate samples of both BRIC-21 and BRIC-23 (Table 1), and further publication we reported that replicate samples from the inspection revealed that both the yvmC-cypX and pchR-yvmA BRIC-21 experiment were exposed post-flight to a battery of operons were significantly upregulated in FL samples from 72 antibiotics and growth inhibitors, and no significant BRIC-21 (Supplemental Table S1). difference in resistance levels was found in FL vs. GC (iv) Arginine biosynthesis. The amino acid arginine is produced in samples. a pathway encoded by the argCJBD-carAB-argF operon (vii) Inhibitors of sporulation initiation. Sporulation in B. subtilis is which converts glutamate to citrulline, and the argGH triggered when cells sense the depletion of nutrients in their operon, which converts citrulline to arginine. The argGH environment. Three genes [sivA, sivB (bslA), and sivC; siv for operon was found to be significantly upregulated in both sporulation-inhibitory vegetative genes] have recently been BRIC-21 and BRIC-23 flight samples. Closer examination of described which actively inhibit the initiation of sporulation the datasets revealed significant upregulation of argCJBD- when B. subtilis is growing in the presence of sufficient carAB-argF operon transcripts in BRIC-21 FL samples nutrients. Both the sivA and sivC genes were found to be (Supplemental Table S1) and significant upregulation of significantly upregulated in FL samples of both BRIC-21 and the arginyl-tRNA genes trnJ-Arg and trnE-Arg in BRIC-23 FL BRIC-23 (Table 1), and as mentioned above, the sivB (bslA) samples (Supplemental Table S2). gene associated with biofilm formation was upregulated in (v) Manganese transport. The major manganese ABC-type BRIC-21 FL samples (Supplemental Table S1). transporter in B. subtilis, encoded by the mntABCD operon, was found to be strongly upregulated in both BRIC-21 and Additional genes upregulated in FL samples BRIC-23 FL samples (Table 1). In addition, closer examination of the datasets revealed that the manganese-proton Examination of Table 1 revealed a number of additional genes symporter mntH was significantly upregulated in BRIC-23 upregulated in both BRIC-21 and BRIC-23 FL samples for which no FL samples (Supplementary Table S2). Paradoxically, in BRIC- clear phenotypic consequence could be discerned. First, a number 21 FL samples we also observed significant upregulation of of genes encoding products of unknown function were upregu- the ydfM gene, encoding a putative Mn(II) efflux pump lated, including ybdO, yczC, ycxA, the yrkEFHIJ operon (Table 1). (Supplementary Table S1). Second, genes encoding transporters for glycerol-3-phosphate (vi) Resistance and toxin genes. A number of genes encoding (glpT), mannose (manP), and oligopeptides (appA/1 and appA/2) resistance and toxic functions were observed to be were upregulated, as were genes encoding an extracellular lipase upregulated in FL samples of both BRIC-21 and BRIC-23, (lipA), proline dehydrogenase (putB), aconitase (citB), a single including: ybfO, encoding a putative erythromycin esterase; methyl-accepting chemotaxis protein (tlpA), urate oxidase (pucL), the bmrCD operon, encoding a multidrug efflux transporter; and a secreted peptide controlling LiaR-LiaS activity (yydF) (Table yvqJ, encoding a putative macrolide-efflux protein; cadA, 1). Each of these belong to its own cohort of genes devoted to encoding a cadmium efflux pump; and sdpC, which encodes different functions in B. subtilis, but expression of the other a lytic toxin (Table 1). At first glance, increase in the members of the cohort were not significantly altered. As an aforementioned transcripts might suggest that FL samples example, the threonyl transfer RNA gene trnD-Thr is transcribed as Published in cooperation with the Biodesign Institute at Arizona State University, with the support of NASA npj Microgravity (2019) 1 M.D. Morrison et al. Table 2. List of genes significantly upregulated in GC samples both in BRIC-21 and BRIC-23 experiments Locus Tag Gene BRIC-21 BRIC-23 Function Regulon Name DESeq2 limma DESeq2 limma BSU03050 ldh 3.26 3.61 1.52 2.31 L-lactate dehydrogenase Rex, SigA BSU03060 lctP 3.18 3.35 1.18 1.78 L-lactate permease Rex, SigA BSU03290 nasE 1.63 1.66 1.12 1.20 Assimilatory nitrite reductase (subunit) Fur, NsrR, ResD, SigA, TnrA BSU03300 nasD 2.45 2.66 1.42 2.02 Assimilatory nitrite reductase (subunit) Fur, NsrR, ResD, SigA, TnrA BSU05720 ydhE 1.32 1.33 1.23 1.27 Similar to macrolide glycosyltransferase LiaR BSU06240 bdhA 1.45 1.46 1.19 1.32 Acetoin reductase/butanediol dehydrogenase AbrB BSU10230 yhfH 1.19 1.29 1.59 1.74 Unknown Unknown BSU17710 tatAC 1.30 1.33 1.28 1.33 Component of the twin-arginine translocation pathway Unknown BSU19180 des 1.07 1.07 1.23 1.35 Phosphlipid desaturase DesR, SigA BSU19190 desK 1.08 1.07 1.72 1.85 Two-component sensor kinase, regulation of cold shock DesR, SigA expression of des BSU20580 yoqM 1.90 1.94 1.80 2.18 Unknown SPβ prophage BSU20760 yopU 1.10 1.15 1.29 1.73 Unknown SPβ prophage BSU20770 yopT 1.10 1.14 1.00 1.11 Unknown SPβ prophage BSU21050 yonN 1.07 1.10 1.27 1.42 DNA-binding protein HU 2 SPβ prophage BSU21320 yomL 1.59 1.64 1.24 1.54 Unknown SPβ prophage BSU21329 youB 1.52 1.57 1.16 1.30 Unknown SPβ prophage BSU29310 cmoJ 1.02 1.12 1.14 1.43 Alkyl monooxygenase, required for the conversion of S-methyl- AscR, CymR, SigA cysteine to cysteine BSU29340 tcyN 1.42 1.55 1.17 1.37 Cystine ABC transporter (ATP-binding protein) AscR, CymR, SigA BSU29360 tcyL 1.33 1.50 1.06 1.21 Cystine ABC transporter (permease) AscR, CymR, SigA BSU30660 ytkA 1.58 1.60 1.35 1.43 Unknown unknown BSU37250 narI 2.58 2.67 1.42 2.03 Nitrate reductase (gamma subunit) Fnr, SigA BSU37260 narJ 2.86 3.01 1.36 2.15 Chaperone for the nitrate reductase (protein J) Fnr, SigA BSU37270 narH 3.15 3.32 1.28 2.08 Nitrate reductase (beta subunit) Fnr, SigA BSU37280 narG 3.24 3.41 1.17 1.95 Nitrate reductase (alpha subunit) Fnr, SigA BSU37310 fnr 2.14 2.20 1.56 1.92 Transcriptional regulator of anaerobice genes Fnr, NsrR, ResD, SigA BSU37320 narK 3.03 3.16 1.32 1.90 Nitrite extrusion protein Fnr, NsrR, SigA BSU37350 sboA 2.91 3.17 1.48 1.88 Subtilosin-A AbrB, ResD, Rok, SigA BSU37360 sboX 2.70 3.31 1.38 1.63 Bacteriocin-like product AbrB, ResD, Rok, SigA BSU37410 albE 1.87 2.29 1.17 1.22 Antilisterial bacteriocin (subtilosin) production AbrB, ResD, Rok, SigA BSU37420 albF 2.07 2.61 1.42 1.52 Antilisterial bacteriocin (subtilosin) production AbrB, ResD, Rok, SigA BSU37430 albG 2.44 2.57 1.32 1.40 Antilisterial bacteriocin (subtilosin) production AbrB, ResD, Rok, SigA BSU37440 ywhL 2.13 2.77 1.35 1.39 Unknown unknown BSU38060 ywcJ 3.09 3.28 1.87 2.06 Similar to nitrite transporter Rex BSU38070 sacT 1.60 1.66 1.12 1.16 Transcriptional antiterminator for the sacP-sacA-ywdA operon DnaA, SacT BSU38730 cydD 3.03 3.27 1.52 2.08 ABC transporter required for expression of cytochrome bd (ATP- CcpA, Rex, ResD, SigF binding protein) BSU38740 cydC 3.27 3.51 1.26 1.89 ABC transporter required for expression of cytochrome bd (ATP- CcpA, Rex, ResD, SigF binding protein) Values are log -fold GC:FL expression ratios. Gene names, annotated functions, and regulons are from Subtiwiki (http://subtiwiki.uni-goettingen.de/v3/index. php), accessed on September 24, 2018 part of a 16-tRNA gene operon located just downstream from the Genes upregulated in GC samples ribosomal RNA rrnD gene cluster in B. subtilis, but it is unclear The 36 genes whose expression was significantly upregulated in why only this transcript, and not the entire operon, was GC samples from both BRIC-21 and BRIC-23 are listed in Table 2. significantly upregulated. Third, two genes encoded by the They are arranged according to their BSU locus tag, i.e., in the prophage SPβ (yomK and bhlA) were upregulated in FL samples order that they are located on the B. subtilis 168 chromosome. (Table 1), but numerous SPβ-related genes were also upregulated Examination of the data revealed upregulation of several blocks of in GC samples (Table 2), so no coherent pattern of gene genes associated with the response of B. subtilis to oxygen expression could be ascertained. npj Microgravity (2019) 1 Published in cooperation with the Biodesign Institute at Arizona State University, with the support of NASA M.D. Morrison et al. limitation. A previous study reported that the global response of and transmitted to the ResD response regulator, encoded by the B. subtilis transcriptome to strict anaerobiosis resulted in the the last two genes of the resABCDE operon. ResD activates induction or repression of hundreds of genes involved in a variety expression of a large set of genes and operons including its of cell functions including carbon metabolism, electron transport, own (resABCDE), cydABCD, nasBCDEF, and sboAXalbABCDEFG. iron uptake, antibiotic production, and stress responses. Our data One of the genes activated by ResD is fnr, itself a regulator revealed that only a subset of the entire anaerobic regulon was of anaerobic gene expression. Fnr activates its own activated, indicating that cells in GC samples underwent only expression (narK-fnr), as well as that of narGHJI and another partial oxygen deprivation, and these genes are described below. regulator encoded by the arfM gene. In addition, changes in cell physiology associated with the switch to oxygen (i) Fermentation. TSYG medium contains the fermentable sugar limitation activate a transcriptional regulator called AlsR, glucose, and B. subtilis is capable of mixed-acid fermentation 47,52 which activates expression of alsSD directly and bdhA using separate pathways for production of the end products 53 indirectly. Meanwhile, the regulatory protein Rex senses acetate, lactate, acetoin, 2,3-butanediol, and ethanol. In GC changes in the NAD /NADH ratio brought on by oxygen samples from both the BRIC-21 and BRIC-23 missions we limitation and responds by activating a number of genes observed significant upregulation of the ldh-lctP operon for including the cydABCD and ldh-lctP operons. We were 45,46 fermentation of lactate from pyruvate, as well as the prompted by these observations to search for regulators of alsSD operon for fermentation of acetoin from pyruvate anaerobic gene expression in the datasets, and found and the bdhA gene for fermentation of 2,3-butanediol from significant upregulation of the arfM gene in BRIC-21 acetoin (Table 2). Interestingly, we did not note any (Supplemental Table S1), the resD and resE genes in BRIC- significant changes in the expression of genes involved in 23 (Supplemental Table S2), and the narK-fnr operon in both acetate or ethanol fermentation in our samples (Supple- BRIC-21 and BRIC-23 GC samples (Table 2). mental Tables S1 and S2). (vi) Phospholipid desaturase. Most prokaryotes regulate mem- (ii) Anaerobic respiration. Growth under oxygen-limiting condi- brane fluidity in part by controlling the degree of saturation tions results in two major modifications of the B. subtilis of membrane phospholipids. B. subtilis accomplishes this respiratory electron transport chain. First, oxygen depletion using a fatty acid desaturase encoded by des, the first gene activates synthesis of the high-affinity cytochrome bd 54,55 in the des-desKR operon. Expression of des is activated ubiquinol oxidase encoded by the cydABCD operon. We by exposure to low temperature, mediated through a two- observed that the cydC and cydD genes were significantly component system composed of a membrane-bound upregulated in GC samples of BRIC-21 and BRIC-23 (Table 2), sensor kinase and a response regulator encoded by the and further examination of the datasets revealed that the 55 desK and desR genes, respectively. Transcription of des and entire cydABCD operon was strongly upregulated in GC desK were seen to be significantly upregulated in GC samples from BRIC-21 (Supplemental Table S1). Second, samples of both BRIC-21 and BRIC-23 (Table 2), and closer anaerobiosis results in activation of genes responsible for examination of the datasets revealed that the entire des- utilizing nitrate and nitrite as alternative terminal electron desKR operon was upregulated in BRIC-23 GC samples acceptors. In B. subtilis GC samples we noted significant (Supplemental Table S2). While the biological significance of upregulation of the narGHJI operon encoding nitrate this observation is at present unclear, it is interesting to note reductase, the nasDE operon encoding nitrite reductase, that in previous work we reported that des-desKR transcrip- and the narK-fnr operon encoding the NarK nitrite extrusion tion was also activated by exposure of B. subtilis to low protein and the Fnr regulator of anaerobic gene expression atmospheric pressure. (Table 2). It is interesting to note that the ywcJ transcript, (vii) Metabolism of a cysteine analog. A pathway in B. subtilis was encoding a putative nitrite transporter previously identified recently described which can convert the cysteine analog S- as a member of the B. subtilis anaerobic regulon, was also methyl-cysteine directly to cysteine as a sulfur source, upregulated in GC samples. Because TSYG medium does not encoded by the snaA-tcyJKLMN-cmoOIJ-ribR-sndA-ytnM provide a significant source of nitrate or nitrite, it is likely operon. We observed that three genes from this operon that although expression of the nar and nas genes was (cmoJ, tcyL, and tcyN) were upregulated in GC samples from induced by oxygen limitation, they did not serve a useful both BRIC-21 and BRIC-23 (Table 2). Further examination of physiological function for cells in the GC samples. the datasets revealed additional genes of the operon (iii) Subtilosin production. Further indication for oxygen limita- significantly upregulated in BRIC-21 (snaA, tcyJ, tcyK, tcyM, tion in BRIC-21 and BRIC-23 GC samples was evidenced by cmoI, and ribR) and BRIC-23 (cmoO, sndA, and ytnM)GC induction of the genes encoding the antilisterial antibiotic samples (Supplemental Tables S1 and S2). subtilosin A, which was previously shown to be induced by anaerobiosis (Table 2). The pathway for subtilosin A biosynthesis is encoded by the sboAXalbABCDEFG operon, and in both BRIC-21 and BRIC-23 GC samples the sboA, sboX, Additional genes upregulated in GC samples albE, albF, and albG transcripts were significantly upregu- Examination of Table 2 revealed a number of genes significantly lated (Table 2). Further inspection of the datasets revealed up-regulated in GC samples from both BRIC-21 and BRIC-23 for that the entire sboAXalbABCDEFG operon was significantly which no clear phenotypic consequence could be discerned. First, upregulated in BRIC-21 GC samples (Supplemental Table S1). three genes encoding products of unknown function (yhfH, ytkA, (iv) Other genes belonging to the anaerobic regulon. Two genes and ywhL) were upregulated in GC samples (Table 2). Second, encoding products of unknown or only putatively annotated upregulation of genes encoding a putative macrolide glycosyl- function (ydhE and ytkA) were seen to be upregulated in GC transferase (ydhE), a component of the twin-arginine translocase samples of both BRIC-21 and BRIC-23 (Table 2). Activation of (tatAC), and a transcriptional antiterminator (sacT) (Table 2), but transcripts for these genes under anaerobic conditions was these genes form parts of larger cohorts of genes which were not noted previously. (v) Genes regulating anaerobiosis. The response to oxygen themselves significantly upregulated. Third, transcripts for a limitation in B. subtilis has been well studied and is number of genes encoded by the SPβ prophage (yoqM, yopU, controlled by a complex regulatory hierarchy. Oxygen yopT, yonN, yomL, and youB) were induced, the function of most of limitation is sensed by the membrane sensor kinase ResE which are unknown. Published in cooperation with the Biodesign Institute at Arizona State University, with the support of NASA npj Microgravity (2019) 1 M.D. Morrison et al. DISCUSSION interface, thus limiting their access to oxygen. Third, under the influence of gravity, dissolved gases, nutrients, and waste products In the present study, we analyzed the effects of spaceflight on the are transported in liquids by convection and diffusion; in Bacillus subtilis transcriptome in two separate spaceflight experi- microgravity, convection is negated and transport becomes ments designated BRIC-21 and BRIC-23. This is the first spaceflight dominated by diffusive processes. Oxygen transfer by convection bacterial transcriptome reported from a Gram-positive bacterium, would be expected to be greater in GC than FL cultures, but this and the first transcriptome study of two separate spaceflight effect appears to be outweighed by the first two factors. missions using the same bacterial strain, growth media, and Unfortunately, because B. subtilis prefers an oxygen-rich environ- hardware. The results uncovered several differences between FL ment for optimum growth, inclusion of an air-containing head- and GC samples that were exhibited in both BRIC-21 and BRIC-23 space in the PDFU cultures is unavoidable. In order to address this missions. hardware limitation, we are currently working on the design and construction of PDFU inserts that will maintain a constant Biofilm-associated transcripts geometry of liquid and air space both in microgravity and in 1- The observation that transcripts associated with biofilm formation xg controls. were significantly upregulated both in BRIC-21 and BRIC-23 FL samples is in agreement with the results from prior spaceflight Nutrient utilization-associated transcripts experiments using several different microbes indicating enhanced 23–27 As stated above, convection ceases in microgravity and the biofilm production in spaceflight. Although the domesticated transport of nutrients or waste products through liquid media laboratory strain B. subtilis 168 does not form robust biofilms, we becomes diffusion-limited. We observed significantly upregu- propose that biofilm formation in spaceflight-grown B. subtilis lated transcripts for biotin and arginine biosynthetic genes in FL could readily be studied in detail by using undomesticated samples, indicating that availability of these two nutrients differed biofilm-producing strains such as NCIB 3610 and its 58,59 in FL vs. GC cultures. However, we did not note significant derivatives. differences between FL and GC samples in the expression of any other biosynthetic pathways, nor was there a significant difference Oxygen limitation-associated transcripts in the final growth yield of cells between FL and GC cultures. We noted significant upregulation in GC samples of transcripts Therefore, it does not appear that exposure to microgravity led to associated with fermentation, anaerobic respiration, and subtilosin a generalized nutrient deficiency in FL samples. production, and upregulation in FL samples of transcripts In our analysis we treated the BRIC-21 and BRIC-23 missions as associated with siderophore production. These observations replicate experiments, but it should be noted that the two indicate that FL and GC samples were experiencing different experiments differed in incubation times, thus at harvest the BRIC- degrees of oxygen availability in the two experiments. How could 21 and BRIC-23 samples were at slightly different growth phases exposure to microgravity vs. 1 xg result in different levels of (late-exponential and transition-phase, respectively). It is therefore oxygen available to the liquid cultures? First, we noted that during to be expected that the resulting transcriptome profiles would deintegration of frozen BRIC-21 samples from PDFUs, FL samples differ somewhat, due to the substantial reorganization of global incubated in microgravity had assumed a toroidal shape (Fig. 4), gene expression which occurs during the transition from while GC samples formed a disk-shaped layer in the bottom of the exponential to stationary-phase growth in B. subtilis. In addition, Petri dish as expected. By assuming a toroidal configuration in it should be kept in mind that measuring the transcriptome microgravity, the liquid FL cultures may present a greater surface captures only one aspect of physiology; it does not take into area to the air phase than in GC samples. Second, under the account regulatory controls exerted at the level of numerous influence of gravity, cells in GC samples would tend to sediment posttranscriptional processes (translation, protein processing and toward the bottom of the Petri dish, away from the liquid/air modification, metabolic regulation of enzyme activity, assembly of subcellular structures, etc.) which must take place in order for a microbe to manifest its final phenotype. With these caveats in mind, combining the datasets from the two experiments allowed us to perform a more robust analysis and led to the identification of a common set of genes that were consistently differentially expressed between FL and GC samples in both experiments. These genes, particularly those involved in biofilm formation, will be interesting candidates for future study. METHODS Bacterial strain, media, and growth conditions The strain used in this study was Bacillus subtilis subsp. subtilis strain 168 (trpC2) from our laboratory stock collection. Medium used throughout was Trypticase Soy Yeast Extract (TSY) medium consisting of (g/L): tryptone, 15; soytone, 5; NaCl, 5; yeast extract, 3; K HPO , 2.5; glucose, 2.5; final pH 7. For 2 4 semisolid plates, agar was added to TSY to a final concentration of 15.0 g/L. Glycerol was added to TSY liquid medium to 10% (v/v) final concentration, resulting in TSYG medium. B. subtilis spores were routinely prepared by cultivation in liquid Schaeffer sporulation medium at 37 °C with vigorous aeration. The culture was harvested when phase-contrast microscopic examination revealed that it consisted of >90% free spores, usually after 3–4 days of incubation. Spores were purified by lysozyme treatment, buffer washing, and heat shock (80 °C, 10 min) as described previously, determined by phase-contrast microscopy to be >99% free of cell debris and unsporulated cells, and stored at 4 °C in deionized water. The spore Fig. 4 Removal of a typical frozen BRIC-21 FL sample from its PDFU suspension was heat-activated (65 °C, 20 min) before use. From a working npj Microgravity (2019) 1 Published in cooperation with the Biodesign Institute at Arizona State University, with the support of NASA M.D. Morrison et al. 8 7 suspension (10 /mL) of spores in water, aliquots of 0.1 mL (~10 CFU) were Read processing, alignment, and quantification applied to the bottoms of sterile 60-mm diameter Petri dishes (Falcon Cat. Quality control, mapping, and gene level quantification of Illumina No. 1007, Fisher Scientific) and air-dried for 48–72 h at room temperature sequences were performed using the Galaxy suite available through the protected from light. Samples were integrated into Biological Research in University of Florida’s High-Performance Research Computing Center. The Canister Dual-Chamber Petri Dish Fixation Units (BRIC-PDFU) spaceflight first 12 bases were trimmed off all reads using FASTQ Trimmer v0.014 to hardware using aseptic technique as described in detail previously. remove random hexamer primers, and read quality of the resulting sequences were checked using the FastQC program. Corresponding paired-end read files were mapped to the Bacillus subtilis strain 168 BRIC-PDFU hardware genome [National Center for Biotechnology Information (NCBI) RefSeq Biological Research in Canister (BRIC)-Petri Dish Fixation Unit (PDFU) 67 accession number NC_000964.3] using Bowtie2 v2.3.2. Mapping quality 63,64 hardware has been described in detail previously. Each BRIC canister 68 was evaluated using SAMStat followed by gene level quantification using enclosed 5 PDFUs, and each PDFU contained a space to accommodate the 69 htseq-count v0.6.1. Gene counts for BRIC-21 and BRIC-23 were separated bottom half of a single 60-mm diameter Petri dish to which air-dried into separate count matrices for differential expression analysis. spores were deposited. A separate syringe compartment contained TSYG liquid medium. Growth is initiated by injection of 8.5 mL of TSYG medium Differential expression and functional analyses into the Petri dish compartment, leaving a headspace of ~18 mL of ambient air which exchanges with the static culture via passive diffusion. To reduce false positive results, two Bioconductor packages, limma 70,71 72 The headspace is the only source of additional oxygen, and because the v3.32.10 and DESeq2 v1.16.1, were used to determine differential system is hermetically sealed, O is consumed and CO accumulates in the 2 2 expression. These two packages have been shown to exhibit consistent culture. Once initiated, there is no mechanism to access cultures for detection precision in comparisons using a large or very small number of removal of samples during flight. replicates making these packages suitable for our BRIC-21 (n = 3) and BRIC-23 (n = 9) comparisons. Per limma recommendation, genes with a sum of less than nine counts across all samples were removed before Experimental timeline analysis. For limma analyses, filtered genes were normalized between The schedules of pre-flight, flight, and post-flight activities are described in samples using TMM normalization and differential expression analysis was 16,17 20 detail elsewhere for the BRIC-21 and BRIC-23 experiments. Experi- conducted using the built-in voom transformation. Because DESeq2 mental details for the two FL experiments were essentially identical in requires non-normalized count data, analysis was performed on the raw terms of hardware (BRIC-PDFU), inoculum size (10 spores), medium used filtered genes. P values from both analyses were corrected for multiple (TSYG), and media volumes injected (8.5 mL). Slight differences in ambient testing using the Benjamini–Hochberg method. To be considered ISS temperature were recorded inside the BRIC-21 (average ~22.8 °C) and differentially expressed, genes had to have a >2-fold difference with a P BRIC-23 (average ~22.3 °C) canisters. The only notable difference in the two value < 0.01 between FL and GC expression using both limma and DESeq2 experiments was the time of incubation for BRIC-21 and BRIC-23 samples, methods. Genes meeting these criteria were annotated using the NCBI 25 h and 36 h, respectively. In both missions, growth was terminated by database and the Subtiwiki web server. Functional analysis was transferring BRIC canisters to the −80 °C freezer onboard the ISS. In both conducted using the Search Tool for the Retrieval of Interacting Genes/ cases, samples remained solidly frozen until returned to the lab. Proteins (STRING) protein network database. STRING protein–protein Asynchronous GC samples were treated identically to FL samples in terms interaction networks for up- and downregulated genes were generated of hardware, configuration, schedule, temperature profiles, pre- and post- separately, and Kyoto Encyclopedia of Genes and Genomes (KEGG) 16,17,20 flight handling. pathways were found using the built-in enrichment tools. To check the variability between all the BRIC-21 and BRIC-23 datasets, a PCA was performed in the statistical environment R v3.4.4. Gene loading scores and BRIC-21 RNA isolation, processing, and sequencing sample eigenvalues for each principal component were calculated using Frozen FL and GC samples (n = 3) were partially thawed at room the princomp function in the R package stats v3.4.4. Sample eigenvalues for temperature. Partially thawed samples were transferred into 15-mL conical the first two principal components were plotted in R using the default tubes, and placed on ice until completely thawed. Cells were recovered by plotting function. centrifugation (7000 × g, 20 min, 0 °C) in a benchtop centrifuge. Super- natants were transferred into sterile 15-mL conical tubes and stored at −70 °C. Cell pellets were immediately processed for total RNA extraction DATA AVAILABILITY TM and treatment with RNase-free DNase using the RiboPure RNA The RNA-seq datasets have been deposited in the NASA GeneLab Data System under Purification Kit (Thermo Fisher Scientific Inc, Waltham, MA) following the the accession numbers GLDS-185 (BRIC-21) and GLDS-138 (BRIC-23). manufacture’s protocols. RNA samples were quantified using a Qubit fluorometer (Invitrogen, Thermo Fisher Scientific Inc, Waltham, MA) and quality evaluated using the RNA 6000 Nano Chip on an Agilent 2100 ACKNOWLEDGEMENTS Bioanalyzer (Agilent Technologies, Santa Clara, CA). RNA Integrity Number (RIN) values ranged from 9.8 to 10.0, indicating high-quality total RNA The authors wish to thank the BRIC-21 and BRIC-23 payload development teams at preparations suitable for further processing. Samples were sent to Hudson NASA Kennedy Space Center (D. Dimapilis, A.D. Flowers, C. Grosse, J. Harp, D. Houze, Alpha Institute for Biotechnology (Huntsville, AL, USA) for ribosomal RNA H. Levine, G. Newsham, S. Manning-Roach, J. Richards, S. Richards, J. Smodell, and G. depletion, library preparation, and 100-nt paired-end sequencing on an Washington) and the GeneLab Process Verification Team (V. Boyko, K. Chakravarty, S. Illumina 2500 instrument. Costes, H. Fogle, M. Dinh, J. Galazka, S. Gebre, S. Lai Polo, and S. Reinsch) at NASA Ames Research Center for their excellent technical assistance. This work was supported by the NASA Research Opportunities in Space Biology grant NNX14AT38G BRIC-23 RNA isolation, processing, and sequencing to W.L.N. and P.F.-C. BRIC-23 sample processing was performed by the GeneLab Sample Processing Lab (NASA Ames Research Center, Mountain View, CA); detailed protocols are described in the BRIC-23 GeneLab dataset GLDS-138 (https:// AUTHOR CONTRIBUTIONS genelab-data.ndc.nasa.gov/genelab/accession/GLDS-138/). Briefly, thawed W.L.N. and P.F.-C. conceived and designed the study. All authors performed the FL and GC samples (n = 9) were centrifuged at 16,000 × g for 5 min at 0 °C. spaceflight experiments and RNA processing. M.D.M. and W.L.N. performed the data Total RNA extraction and RNase-free DNase treatment was also performed TM analysis. All authors contributed to writing the paper. using the RiboPure RNA Purification Kit (Thermo Fisher Scientific Inc, Waltham, MA) following the manufacture’s protocol. RNA concentrations were measured using a Qubit 3.0 Fluorometer (Thermo Fisher Scientific, ADDITIONAL INFORMATION Waltham, MA), and RNA quality was assessed on an Agilent 2100 Bioanalyzer (Agilent, Santa Clara, CA). Ribosomal RNA depletion was Supplementary information accompanies the paper on the npj Microgravity website (https://doi.org/10.1038/s41526-018-0061-0). performed using the Ribo-Zero rRNA Removal Kit for Gram-positive bacteria (Illumina) and sample cDNA libraries were synthesized using KAPA Competing interests: The authors declare no competing interests. RNA HyperPrep reagents (Roche) and amplified with 10 PCR cycles. Library sequencing was performed on an Illumina HiSeq 4000 platform. Published in cooperation with the Biodesign Institute at Arizona State University, with the support of NASA npj Microgravity (2019) 1 M.D. Morrison et al. Publisher’s note: Springer Nature remains neutral with regard to jurisdictional claims 29. Vlamakis, H., Chai, Y. R., Beauregard, P., Losick, R. & Kolter, R. Sticking together: in published maps and institutional affiliations. building a biofilm the Bacillus subtilis way. Nat. Rev. Microbiol. 11, 157–168 (2013). 30. Lopez, D., Fischbach, M. A., Chu, F., Losick, R. & Kolter, R. Structurally diverse natural products that cause potassium leakage trigger multicellularity in Bacillus subtilis. Proc. Natl Acad. Sci. USA 106, 280–285 (2009). REFERENCES 31. Driks, A. Tapping into the biofilm: insights into assembly and disassembly of a 1. Mora, M. et al. Microorganisms in confined habitats: microbial monitoring and novel amyloid fibre in Bacillus subtilis. Mol. Microbiol 80, 1133–1136 (2011). control of intensive care units, operating rooms, cleanrooms and the Interna- 32. Roux, D. et al. Identification of poly-N-acetylglucosamine as a major poly- tional Space Station. Front. Microbiol. 7, 20 (2016). saccharide component of the Bacillus subtilis biofilm matrix. J. Biol. Chem. 290, 2. Pietsch, J. et al. The effects of weightlessness on the human organism and 19261–19272 (2015). mammalian cells. Curr. Mol. Med. 11, 350–364 (2011). 33. Kobayashi, K. & Iwano, M. BslA(YuaB) forms a hydrophobic layer on the surface of 3. Lychakov, D. V. Behavioural and functional vestibular disorders after space flight: Bacillus subtilis biofilms. Mol. Microbiol 85,51–66 (2012). 2. Fish, amphibians and birds. J. Evol. Biochem. Physiol. 52,1–16 (2016). 34. Lombardia, E., Rovetto, A. J., Arabolaza, A. L. & Grau, R. R. A LuxS-dependent cell- 4. Vandenbrink, J. P. & Kiss, J. Z. Space, the final frontier: a critical review of recent to-cell language regulates social behavior and development in Bacillus subtilis. J. experiments performed in microgravity. Plant Sci. 243, 115–119 (2016). Bacteriol. 188, 4442–4452 (2006). 5. Rosenzweig, J. A. et al. Spaceflight and modeled microgravity effects on microbial 35. Perkins, J. B. & Pero, J. in Bacillus subtilis and its Closest Relatives: from Genes to growth and virulence. Appl. Microbiol Biotechnol. 85, 885–891 (2010). Cells (eds Sonenshein, A. L., Hoch, J. A., & Losick, R.), 271–286 (ASM Press, 2002). 6. Rosenzweig, J. A., Ahmed, S., Eunson, J. & Chopra, A. K. Low-shear force asso- 36. Rodionov, D. A., Mironov, A. A. & Gelfand, M. S. Conservation of the biotin regulon ciated with modeled microgravity and spaceflight does not similarly impact the and the BirA regulatory signal in Eubacteria and Archaea. Genome Res 12, virulence of notable bacterial pathogens. Appl. Microbiol Biotechnol. https://doi. 1507–1516 (2002). org/10.1007/s00253-014-6025-8 (2014). 37. Miethke, M. Molecular strategies of microbial iron assimilation: from high-affinity 7. Taylor, P. W. Impact of space flight on bacterial virulence and antibiotic sus- complexes to cofactor assembly systems. Metallomics 5,15–28 (2013). ceptibility. Infect. Drug Resist 8, 249–262 (2015). 38. May, J. J., Wendrich, T. M. & Marahiel, M. A. The dhb operon of Bacillus subtilis 8. National Research Council. Recapturing a Future for Space Exploration: Life and encodes the biosynthetic template for the catecholic siderophore 2,3-dihydrox- Physical Sciences for a New Era, 442 (National Academies Press: Washington, DC, ybenzoate-glycine-threonine trimeric ester bacillibactin. J. Biol. Chem. 276, 2011). 7209–7217 (2001). 9. Horneck, G., Klaus, D. M. & Mancinelli, R. L. Space microbiology. Microbiol Mol. Biol. 39. Randazzo, P., Aubert-Frambourg, A., Guillot, A. & Auger, S. The MarR-like protein Rev. 74, 121–156 (2010). PchR (YvmB) regulates expression of genes involved in pulcherriminic acid bio- 10. Zea, L. et al. A molecular genetic basis explaining altered bacterial behavior in synthesis and in the initiation of sporulation in Bacillus subtilis. BMC Microbiol. 16, space. PLoS One 11, e0164359 (2016). 16 (2016). 11. Mastroleo, F. et al. Experimental design and environmental parameters affect 40. Belitsky, B. R. in Bacillus subtilis and Its Closest Relatives: from Genes to Cells (eds Rhodospirillum rubrum S1H response to space flight. ISME J. 3, 1402–1419 (2009). Sonenshein, A. L., Hoch, J. A. & Losick, R.), 203–223 (ASM Press, 2002). 12. Huang, B., Liu, N., Rong, X., Ruan, J. & Huang, Y. Effects of simulated microgravity 41. Garti-Levi, S., Eswara, A., Smith, Y., Fujita, M. & Ben-Yehuda, S. Novel modulators and spaceflight on morphological differentiation and secondary metabolism of controlling entry into sporulation in Bacillus subtilis. J. Bacteriol. 195, 1475–1483 Streptomyces coelicolor A3(2). Appl. Microbiol. Biotechnol. 99, 4409–4422 (2015). (2013). 13. Crabbé, A. et al. Transcriptional and proteomic responses of Pseudomonas aer- 42. Rudner, R. et al. Two transfer RNA gene clusters associated with ribosomal RNA uginosa PAO1 to spaceflight conditions involve Hfq regulation and reveal a role operons rrnD and rrnE. Bacillus subtilis. J. Bacteriol. 175, 503–509 (1993). for oxygen. Appl. Environ. Microbiol 77, 1221–1230 (2011). 43. Ye, R. et al. Global gene expression profiles of Bacillus subtilis grown under 14. Wilson, J. W. et al. Space flight alters bacterial gene expression and virulence and anaerobic conditions. J. Bacteriol. 182, 4458–4465 (2000). reveals a role for global regulator Hfq. Proc. Natl Acad. Sci. USA 104, 16299–16304 44. Nakano, M. & Zuber, P. in Bacillus subtilis and Its Closest Relatives: from Genes to (2007). Cells (eds Sonenshein, A. L., Hoch, J. A., & Losick, R.), 393–404 (ASM Press, 2002). 15. Altman, N. & Krzywinski, M. Sources of variation. Nat. Methods 12,5–6 (2015). 45. Cruz Ramos, H. et al. Fermentative metabolism of Bacillus subtilis: physiology and 16. Morrison, M. D., Fajardo-Cavazos, P. & Nicholson, W. L. Cultivation in space flight regulation of gene expression. J. Bacteriol. 182, 3072–3080 (2000). produces minimal alterations in the susceptibility of Bacillus subtilis cells to 72 46. Marino, M., Hoffmann, T., Schmid, R., Mobitz, H. & Jahn, D. Changes in protein different antibiotics and growth-inhibiting compounds. Appl. Environ. Microbiol. synthesis during the adaptation of Bacillus subtilis to anaerobic growth condi- 83, e01584–01517 (2017). tions. Microbiol. SGM 146,97–105 (2000). 17. Fajardo-Cavazos, P., Leehan, J. D. & Nicholson, W. L. Alterations in the spectrum of 47. Renna, M. C., Najimudin, N., Winik, L. R. & Zahler, S. A. Regulation of the Bacillus spontaneous rifampicin-resistance mutations in the Bacillus subtilis rpoB gene subtilis alsS, alsD, and alsR genes involved in post-exponential-phase production after cultivation in the human spaceflight environment. Front. Microbiol. 9, 192 of acetoin. J. Bacteriol. 175, 3863–3875 (1993). (2018). 48. Nicholson, W. L. The Bacillus subtilis ydjL (bdhA) gene encodes acetoin reductase/ 18. Markowitz, V. M. et al. IMG: the integrated microbial genomes database and 2,3-butanediol dehydrogenase. Appl. Environ. Microbiol 74, 6832–6838 (2008). comparative analysis system. Nucleic Acids Res. 40, D115–D122 (2012). 49. von Wachenfeldt, C. & Hederstedt, L. in Bacillus subtilis and Its Closest Relatives: 19. Matsumoto, Y., Murakami, Y., Tsuru, S., Ying, B. W. & Yomo, T. Growth rate- from Genes to Cells (eds Sonenshein, A. L., Hoch, J. A., & Losick, R.), 163–179 (ASM coordinated transcriptome reorganization in bacteria. BMC Genom. 14, 10 (2013). Press, 2002). 20. Fajardo-Cavazos, P. & Nicholson, W. L. Establishing standard protocols for bac- 50. Zheng, G. L., Yan, L. Z., Vederas, J. C. & Zuber, P. Genes of the sbo-alb locus of terial culture in Biological Research in Canister (BRIC) hardware. Gravit. Space Res. Bacillus subtilis are required for production of the antilisterial bacteriocin sub- 4,58–69 (2016). tilosin. J. Bacteriol. 181, 7346–7355 (1999). 21. Banse, A. V., Chastanet, A., Rahn-Lee, L., Hobbs, E. C. & Losick, R. Parallel pathways 51. Härtig, E. & Jahn, D. Regulation of the anaerobic metabolism in Bacillus subtilis. of repression and antirepression governing the transition to stationary phase in Adv. Microb. Physiol. 61, 195–216 (2012). Bacillus subtilis. Proc. Natl Acad. Sci. USA 105, 15547–15552 (2008). 52. Frädrich, C. et al. The transcription factor AlsR binds and regulates the promoter 22. Flemming, H. C. & Wingender, J. The biofilm matrix. Nat. Rev. Microbiol. 8, of the alsSD operon responsible for acetoin formation in Bacillus subtilis. J. Bac- 623–633 (2010). teriol. 194, 1100–1112 (2012). 23. Kim, W. et al. Spaceflight promotes biofilm formation by Pseudomonas aerugi- 53. Oliveira, R. R. & Nicholson, W. L. The LysR-type transcriptional regulator (LTTR) nosa. PLoS One 8, e62437 (2013). AlsR indirectly regulates expression of the Bacillus subtilis bdhA gene encoding 24. Zea, L. et al. Phenotypic changes exhibited by E. coli cultured in space. Front. 2,3-butanediol dehydrogenase. Appl. Microbiol Biotechnol. 97, 7307–7316 (2013). Microbiol. 8, 12 (2017). 54. Mansilla, M. C. & de Mendoza, D. The Bacillus subtilis desaturase: a model to 25. Van Mulders, S. E. et al. The influence of microgravity on invasive growth in understand phospholipid modification and temperature sensing. Arch. Microbiol. Saccharomyces cerevisiae. Astrobiology 11,45–55 (2011). 183, 229–235 (2005). 26. Crabbé, A. et al. Spaceflight enhances cell aggregation and random budding in 55. Mendoza, D. Temperature sensing by membranes. Annu Rev. Microbiol 68, Candida albicans. Plos One 8, 24 (2013). 101–116 (2014). 27. Koenig, D. W. & Pierson, D. L. Microbiology of the space shuttle water system. 56. Fajardo-Cavazos, P. et al. Evolution of Bacillus subtilis to enhanced growth at low Water Sci. Technol. 35,59–64 (1997). pressure: up-regulated transcription of des-desKR encoding the fatty acid desa- 28. McLoon, A. L., Guttenplan, S. B., Kearns, D. B., Kolter, R. & Losick, R. Tracing the turase system. Astrobiology 12, 258–270 (2012). domestication of a biofilm-forming bacterium. J. Bacteriol. 193, 2027–2034 (2011). npj Microgravity (2019) 1 Published in cooperation with the Biodesign Institute at Arizona State University, with the support of NASA M.D. Morrison et al. 57. Chan, C.-M., Danchin, A., Marliere, P. & Sekowska, A. Paralogous metabolism: S- 71. Ritchie, M. E. et al. limma powers differential expression analyses for RNA- alkyl-cysteine degradation in Bacillus subtilis. Environ. Microbiol 16, 101–117 sequencing and microarray studies. Nucleic Acids Res. 43, e47 (2015). (2014). 72. Love, M. I., Huber, W. & Anders, S. Moderated estimation of fold change and 58. Branda, S. S., Gonzalez-Pastor, J. E., Ben-Yehuda, S., Losick, R. & Kolter, R. Fruiting dispersion for RNA-seq data with DESeq2. Genome Biol. 15, 550 (2014). body formation by Bacillus subtilis. Proc. Natl Acad. Sci. USA 98, 11621–11626 73. Seyednasrollah, F., Laiho, A. & Elo, L. L. Comparison of software packages for (2001). detecting differential expression in RNA-seq studies. Brief. Bioinform. 16,59–70 59. Konkol, M. A., Blair, K. M. & Kearns, D. B. Plasmid-encoded ComI inhibits com- (2015). petence in the ancestral 3610 strain of Bacillus subtilis. J. Bacteriol. 195, 74. Law, C. W., Chen, Y., Shi, W. & Smyth, G. K. Voom: precision weights unlock linear 4085–4093 (2013). model analysis tools for RNA-seq read counts. Genome Biol. 15, R29 (2014). 60. Phillips, Z. E. V. & Strauch, M. A. Bacillus subtilis sporulation and stationary phase 75. Benjamini, Y. & Hochberg, Y. Controlling the false discovery rate: a practical and gene expression. Cell Mol. Life Sci. 59, 392–402 (2002). powerful approach to multiple testing. J. R. Stat. Soc. B 57, 289–300 (1995). 61. Schaeffer, P., Millet, J. & Aubert, J. P. Catabolic repression of bacterial sporulation. 76. Michna, R. H., Zhu, B., Mäder, U. & Stülke, J. SubtiWiki 2.0–an integrated database Proc. Natl Acad. Sci. USA 54, 704–711 (1965). for the model organism Bacillus subtilis. Nucleic Acids Res. 44, D654–662 (2016). 62. Nicholson, W. L. & Setlow, P. in Molecular Biological Methods for Bacillus (eds 77. Szklarczyk, D. et al. STRING v10: protein-protein interaction networks, integrated Harwood, C. R. & Cutting, S. M.), 391–450 (J. Wiley & Sons, 1990). over the tree of life. Nucleic Acids Res. 43, D447–452 (2015). 63. Paul, A.-L. et al. Spaceflight transcriptomes: unique responses to a novel envir- 78. Zhu, B. Y. & Stülke, J. SubtiWiki in 2018: from genes and proteins to functional onment. Astrobiology 12,40–56 (2012). network annotation of the model organism Bacillus subtilis. Nucleic Acids Res. 46, 64. Wells, B., McCray, R. H., Best, M. D. & Levine, H. G. A flight-rated Petri dish D743–D748 (2018). apparatus providing two stage fluid injection for aseptic biological investigations in space. in 31st International Conference on Environmental Systems 2001-01-2286 (SAE International, 2001). Open Access This article is licensed under a Creative Commons 65. Blankenberg, D. et al. Manipulation of FASTQ data with Galaxy. Bioinformatics 26, Attribution 4.0 International License, which permits use, sharing, 1783–1785 (2010). adaptation, distribution and reproduction in any medium or format, as long as you give 66. FastQC: a quality control tool for high throughput sequence data. v. 0.11.5. http:// appropriate credit to the original author(s) and the source, provide a link to the Creative www.bioinformatics.babraham.ac.uk/projects/fastqc/ (Babraham Bioinformatics, Commons license, and indicate if changes were made. The images or other third party 2010). material in this article are included in the article’s Creative Commons license, unless 67. Langmead, B., Trapnell, C., Pop, M. & Salzberg, S. L. Ultrafast and memory-efficient indicated otherwise in a credit line to the material. If material is not included in the alignment of short DNA sequences to the human genome. Genome Biol. 10, R25 article’s Creative Commons license and your intended use is not permitted by statutory (2009). regulation or exceeds the permitted use, you will need to obtain permission directly 68. Lassmann, T., Hayashizaki, Y. & Daub, C. O. SAMStat: monitoring biases in next from the copyright holder. To view a copy of this license, visit http://creativecommons. generation sequencing data. Bioinformatics 27, 130–131 (2011). org/licenses/by/4.0/. 69. Anders, S., Pyl, P. T. & Huber, W. HTSeq–a Python framework to work with high- throughput sequencing data. Bioinformatics 31, 166–169 (2015). © The Author(s) 2019 70. Liu, R. et al. Why weight? Modelling sample and observational level variability improves power in RNA-seq analyses. Nucleic Acids Res. 43, e97 (2015). Published in cooperation with the Biodesign Institute at Arizona State University, with the support of NASA npj Microgravity (2019) 1 http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png npj Microgravity Springer Journals

Comparison of Bacillus subtilis transcriptome profiles from two separate missions to the International Space Station

Loading next page...
 
/lp/springer-journals/comparison-of-bacillus-subtilis-transcriptome-profiles-from-two-cHchkvNeqJ

References (88)

Publisher
Springer Journals
Copyright
Copyright © 2019 by The Author(s)
Subject
Life Sciences; Life Sciences, general; Classical and Continuum Physics; Biotechnology; Immunology; Space Sciences (including Extraterrestrial Physics, Space Exploration and Astronautics) ; Applied Microbiology
eISSN
2373-8065
DOI
10.1038/s41526-018-0061-0
Publisher site
See Article on Publisher Site

Abstract

www.nature.com/npjmgrav ARTICLE OPEN Comparison of Bacillus subtilis transcriptome profiles from two separate missions to the International Space Station 1 1 1 Michael D. Morrison , Patricia Fajardo-Cavazos and Wayne L. Nicholson The human spaceflight environment is notable for the unique factor of microgravity, which exerts numerous physiologic effects on macroscopic organisms, but how this environment may affect single-celled microbes is less clear. In an effort to understand how the microbial transcriptome responds to the unique environment of spaceflight, the model Gram-positive bacterium Bacillus subtilis was flown on two separate missions to the International Space Station in experiments dubbed BRIC-21 and BRIC-23. Cells were grown to late-exponential/early stationary phase, frozen, then returned to Earth for RNA-seq analysis in parallel with matched ground control samples. A total of 91 genes were significantly differentially expressed in both experiments; 55 exhibiting higher transcript levels in flight samples and 36 showing higher transcript levels in ground control samples. Genes upregulated in flight samples notably included those involved in biofilm formation, biotin and arginine biosynthesis, siderophores, manganese transport, toxin production and resistance, and sporulation inhibition. Genes preferentially upregulated in ground control samples notably included those responding to oxygen limitation, e.g., fermentation, anaerobic respiration, subtilosin biosynthesis, and anaerobic regulatory genes. The results indicated differences in oxygen availability between flight and ground control samples, likely due to differences in cell sedimentation and the toroidal shape assumed by the liquid cultures in microgravity. npj Microgravity (2019) 5:1 ; doi:10.1038/s41526-018-0061-0 INTRODUCTION formation and architecture; and resistance to antibiotics or abiotic 5,6,9 stresses. More recent efforts tended toward gene expression In certain respects, human spaceflight habitats resemble other studies using genome-wide techniques such as microarrays to confined built environments, such as submersible vehicles, understand how the global pattern of RNA synthesis (i.e., the aircraft, hospital isolation wards, or remote research installations. transcriptome) responds to the spaceflight environment. To date, However, the spaceflight environment is unique because it microarray studies have reported a wide range of responses to contains two additional altered physical parameters: reduced spaceflight including increased transcription of genes encoding (micro-)gravity and increased ionizing radiation from solar and 10,11 12 general metabolism, secondary metabolite biosynthesis, galactic sources. Extensive investigations conducted in spaceflight 11,13 13,14 synthesis of ribosomal proteins, and virulence factors. on macroscopic organisms have resulted in a relatively good Regardless of the output measured, it has proven difficult to understanding of the biological effects of microgravity and derive consistent conclusions from these disparate studies due to radiation at levels ranging from the whole body down to the several confounding factors. 2 3 organ, cellular, and molecular level in humans, animals, and First, until recently spaceflight transcriptome studies have been plants. While microorganisms have also been the subject of performed on only a small selection of Gram-negative bacteria focused research in the spaceflight environment, it has proven (Salmonella enterica serovar Typhimurium, Pseudomonas aerugi- more difficult to understand their responses to spaceflight nosa, Rhodospirillum rubrum), limiting the ability to generalize 5–7 stress. From a theoretical perspective, exposure to microgravity conclusions to a broader range of microbes. Second, spaceflight results in a number of alterations in a microbial cell’s immediate experiments have been conducted under widely different: (i) surroundings, such as loss of convective mass and heat transfer, culture conditions (e.g., media formulations, agar vs. liquid, reduction in mechanical shear forces, and alterations in the way aeration, temperature); (ii) growth stage of the cultures at harvest; liquids behave at air and solid interfaces. Changes in such (iii) spaceflight hardware employed; (iv) pre- and post-flight 6,7 fundamental physical forces alter the rates at which gases, treatment of samples; and (v) types of assays conducted. Third, nutrients, signaling molecules, and waste products are exchanged experimental variation derives from the measurements them- between microbes and their surroundings. It has been proposed selves (technical effects) or from the natural variation inherent in that upon perception of these alterations in their environment, biological systems (biological effects). In an effort to control for microbes mount a complex set of stress responses (the so-called variation, prior microbial spaceflight experiments have included “spaceflight syndrome” ). multiple replicates; however, most experiments reported in the Considerable effort has been expended to understand microbial literature have been flown on a single mission only. Intense responses to spaceflight and their underlying causes. In early competition for limited cargo space destined to research plat- studies, various phenotypic outputs from microbes grown in space forms such as the International Space Station (ISS) generally were measured, such as: growth rate and yield; virulence; biofilm results in the choice a new experiment taking precedence over a Department of Microbiology and Cell Science, University of Florida, Merritt Island, FL, USA Correspondence: Wayne L. Nicholson (WLN@ufl.edu) Received: 20 June 2018 Accepted: 6 November 2018 Published in cooperation with the Biodesign Institute at Arizona State University, with the support of NASA M.D. Morrison et al. Fig. 2 Venn diagrams showing number of genes significantly differentially expressed in the BRIC-21, BRIC-23, or both missions. Total genes (top), genes expressed higher in FL than GC samples (FL > GC; middle) or genes expressed higher in GC than FL samples (GC > FL; bottom) are depicted Fig. 1 Principal Component Analysis of the datasets from BRIC-21 FL (red triangles) and GC (green circles) samples, and BRIC-23 FL (blue Overview of the datasets squares) and GC (purple diamonds) samples In order to assess the quality and reproducibility of the datasets obtained, principal component analysis (PCA) was performed. In repetition of a previously flown experiment. Because of this, the the PCA, the first and second principal components explained 52 intrinsic mission-to-mission variability in the response of micro- and 21% of the variance, respectively. Four distinct population biological systems to the spaceflight environment has remained clusters were identified corresponding to the four environmental largely unexplored. conditions tested (Fig. 1). In the BRIC-21 FL and GC samples, the In 2015, we were afforded the opportunity to send an three replicates were grouped rather tightly, indicating relatively experimental package to the ISS to test the responses of the good agreement. In the BRIC-23 FL and GC samples, the nine Gram-positive bacterium Bacillus subtilis to the human spaceflight replicates were somewhat more disperse, but still formed distinct environment. This was the 21st mission to the ISS using Biological groups (Fig. 1). Examination of Principal Component 1 revealed Research in Canister-Petri Dish Fixation Unit (BRIC-PDFU) hard- that the major source of variation in the datasets derived from the ware, and the experiment was dubbed BRIC-21. From the BRIC-21 differences in the two missions themselves, while variation in experiment we have previously reported in detail measurements Principal Component 2 was due to differences between the FL and of the growth, antibiotic resistance, frequency and spectrum of GC datasets in each experiment (Fig. 1). mutagenesis exhibited by B. subtilis flight (FL) samples in The B. subtilis strain 168 transcriptome consists of 4397 total 16,17 comparison to matched ground control (GC) samples. In genes, of which 4280 encode proteins. Analysis of the BRIC-21 addition, we also performed RNA-seq analyses to compare the RNA-seq data resulted in identification of 293 total genes whose transcriptomes of BRIC-21 FL vs. GC samples, as we will report in expression differed significantly in FL vs. GC samples, representing this communication. In 2016 we had the good fortune, in ~6.8% of the protein-coding genome. Of these genes, 177 were significantly higher in FL samples, and 116 were significantly collaboration with the NASA GeneLab group, to fly a second higher in GC samples. These data are summarized in Supple- mission to the ISS (dubbed BRIC-23) using the same B. subtilis mental Table S1. Analysis of the BRIC-23 RNA-seq data resulted in strain, media, and hardware, and again to perform RNA-seq identification of 255 total genes whose expression differed analyses on the samples. Here we provide a comparative analysis significantly in FL vs. GC samples, representing ~6.0% of the of the B. subtilis transcriptome profiles from the BRIC-21 and BRIC- protein-coding transcriptome. Of these genes, 163 were signifi- 23 spaceflight missions. We report on the complete transcriptome cantly higher in FL samples, and 92 were significantly higher in GC profiling of a Gram-positive bacterium grown in the human samples. These data are summarized in Supplemental Table S2. spaceflight environment (a prior study focused on a subset of We reasoned that comparison of the transcriptome datasets primary and secondary metabolite genes in Streptomyces coelico- from the BRIC-21 and BRIC-23 experiments would identify genes lor ). In this study, we show the effect of exposure to the human that were significantly up- or down-regulated in both missions, spaceflight environment on the B. subtilis transcriptome by thus defining genes whose expression was consistently altered in identifying sets of genes expressed in common in both the response to spaceflight. A comparison of the datasets obtained BRIC-21 and BRIC-23 missions. from the BRIC-21 and BRIC-23 experiments is depicted graphically as Venn diagrams (Fig. 2). A total of 91 genes were significantly differentially expressed in both experiments. Fifty-five of the shared genes exhibited higher transcript levels in FL samples and RESULTS 36 genes showed higher transcript levels in GC samples (Fig. 2). RNA-seq was used to characterize the transcriptomic response of The finding that only ~1/3 of the significantly differentially B. subtilis cultures exposed to the human spaceflight environment expressed transcripts were shared in both the BRIC-21 and BRIC-23 of the ISS (FL samples) vs. matched GC samples on two separate missions indicated a substantial amount of between-experiment missions, BRIC-21 (n = 3) and BRIC-23 (n = 9). The data were variation. What could be the source of this variation? While we analyzed using the bioinformatics pipeline described in the attempted to keep discrepancies between the two experiments to section 'Methods' and the results are presented below. a minimum, one notable difference between the BRIC-21 and npj Microgravity (2019) 1 Published in cooperation with the Biodesign Institute at Arizona State University, with the support of NASA 1234567890():,; M.D. Morrison et al. BRIC-23 experiments was the difference in incubation times (25 vs. have documented biofilms containing Bacillus spp. in the 36 h), which may have contributed to the ~2/3 discordance in the Space Shuttle water system. Laboratory strains of B. subtilis two datasets. Unfortunately, BRIC-PDFU hardware does not allow such as strain 168 and its descendants do not form robust direct measurement of growth to be determined in situ during biofilms due to mutations that have accumulated during spaceflight. However, because the transcription of genes encoding their domestication, and biofilm formation was not noted fundamental growth processes (e.g., replication, transcription, in FL or GC samples from BRIC-21 or BRIC-23. Nonetheless, translation) exhibits a positive correlation with growth rate, as a numerous biofilm-related genes were observed to be proxy for growth rate we compared the fold changes for significantly upregulated in both BRIC-21 and BRIC-23 FL transcripts involved in DNA replication (dnaE), RNA transcription samples, including genes of the srfAA-AB-AC-AD-ycxA (rpoABCD), and protein synthesis (rps and rpl genes encoding small operon, which encodes the pathway for production of the 29,30 and large ribosomal subunit proteins). Neither limma or DESeq2 cyclic lipopeptide surfactin and genes of the tapA-sipW- identified a significant difference in the transcript expression tasA operon encoding the major protein matrix component levels for these genes in either the BRIC-21 or BRIC-23 datasets, of biofilms (Table 1). B. subtilis biofilms also contain as a suggesting that the FL and GC samples were exhibiting similar major component the exopolysaccharide poly-N-acetylglu- growth rates at the time of harvest for RNA extraction. cosamine, produced in a biosynthetic pathway encoded In pre-flight ground-based experiments we determined that by the epsABCDEFGHIJKLMNO operon. Our analysis found under ISS ambient temperature (~23 °C), 25 and 36 h of incubation that only the epsB, epsC, and epsI genes were significantly corresponded to late-exponential phase and the “transition state” upregulated in FL samples of both BRIC-21 and BRIC-23 between exponential and stationary-phase growth, respectively. (Table 1). However, closer examination of the datasets Entrance into the transition state in B. subtilis results in the revealed that in the BRIC-23 experiment, 12 of the 15 eps transcriptional activation of nearly 300 genes which comprise a genes (epsABCDEFGHIJKL) were significantly upregulated in regulon under control of the AbrB protein; we reasoned that FL samples (Supplemental Table S2). In the BRIC-21 comparison of AbrB-controlled transcripts between the BRIC-21 experiment, these 12 genes also displayed a significant and BRIC-23 datasets might provide insight into the growth phase upregulation in FL samples (p < 0.01), but only three of these of these two populations. We found that 60 and 57 AbrB- genes (epsB, espC, and epsI) met our >2-fold change cutoff, dependent transcripts were significantly altered in the BRIC-21 while the remaining nine genes exhibited fold changes and BRIC-23 datasets, respectively, and that 27 transcripts were slightly below the cutoff (Supplemental Table S1). The last significantly altered in both experiments (Fig. 3). From this analysis three genes in the eps operon, epsM, epsN, and epsO, were it therefore appeared that cultures in both experiments were at or found not to be significantly differentially expressed in FL near the transition phase of growth. samples from either mission. In addition, examination of the BRIC-21 dataset revealed that the bslA, ycdA, and luxS genes, also involved in swarming motility and biofilm forma- Genes upregulated in FL samples 33,34 tion, were significantly upregulated in FL samples The 55 genes whose expression was significantly upregulated in (Supplemental Table S1). FL samples from both BRIC-21 and BRIC-23 are listed in Table 1. (ii) Biotin biosynthesis. Biotin (vitamin H) is an essential cofactor They are arranged according to their BSU locus tag, i.e., in the for enzymes such as acetyl-CoA carboxylase and pyruvate order that they are located on the B. subtilis 168 chromosome carboxylase, which are important in fatty acid metabolism map. Examination of the data revealed upregulation of blocks of and central metabolism, respectively. Production of biotin genes associated with particular phenotypes in B. subtilis: from pimelic acid is accomplished by the gene products of the bioWAFDBI biosynthetic operon. Examination of the data (i) Biofilm formation. Under particular environmental condi- from Table 1 revealed significant upregulation in FL samples tions, various microorganisms can produce biofilms consist- of the bioW, bioA, bioD, and bioB genes in both the BRIC-21 ing of cells embedded in a matrix of extracellular polymeric and BRIC-23 missions, as well as upregulation of the bioY substances (EPS) consisting of polysaccharides, proteins, gene encoding the energy coupling factor (ECF) transporter nucleic acids, and lipids. It has been reported that biotin-specific S-protein (Table 1). In addition, inspection of spaceflight exposure promoted biofilm formation in Pseu- 23 24 the BRIC-21 dataset showed significant upregulation in FL domonas aeruginosa and Escherichia coli, and promoted samples of the bioF and bioI biotin biosynthetic genes, the invasive growth of the yeasts Saccharomyces cerevisiae and yuiG (bioYB) gene encoding a putative second biotin-specific Candida albicans. Bacterial biofilms have been implicated ECF transporter S-protein, the yhfT gene whose product is in contamination and biofouling of potable water systems in involved in surfactin production, and the yhfS encoding a long-duration space habitats, and data from spaceflight putative acetyl-CoA C-acetyltransferase (Supplemental Table S1); along with the bio genes, these additional genes belong to a regulon under control of a repressor named BirA, thus are likely upregulated as a block. (iii) Siderophores. Iron is an important cofactor for several enzymes, but is only sparingly soluble in most aerobic environments. To acquire iron, most microorganisms produce and excrete siderophores which bind iron with high affinity for subsequent import by specific transport systems. B. subtilis produces the siderophore bacillibactin (2,3-dihydroxybenzoate) encoded by the besA-dhbACEBF operon. FL samples from both BRIC-21 and BRIC-23 were found to significantly upregulate the besA and dhbC transcripts (Table 1), and subsequent examination of the BRIC-21 data revealed that the entire besA-dhbACEBF operon was significantly upregulated in FL samples (Supplementary Fig. 3 Venn diagram showing genes of the transition-state AbrB Table S1). Another siderophore produced by B. subtilis called regulon significantly differentially expressed in the BRIC-21, BRIC-23, or both missions pulcherriminic acid or pulcherrimin is encoded by two small Published in cooperation with the Biodesign Institute at Arizona State University, with the support of NASA npj Microgravity (2019) 1 M.D. Morrison et al. Table 1. List of genes significantly upregulated in FL samples both in BRIC-21 and BRIC-23 experiments Locus Tag Gene BRIC-21 BRIC-23 Annotated function Regulon Name DESeq2 limma DESeq2 limma BSU02050 ybdO 1.94 2.03 1.25 1.24 Unknown AbrB, SigD BSU02140 glpT 1.43 1.48 1.06 1.10 Glycerol-3-phosphate transporter GlpP, CcpA, PhoP, SigA BSU02310 ybfO 1.46 1.51 1.89 1.99 Similar to erythromycin esterase AbrB, SigW BSU02700 lipA 1.44 1.50 1.23 1.27 Extracellular lipase Unknown BSU02710 yczC 1.93 1.99 1.08 1.07 Unknown Unknown BSU03200 putB 3.16 3.67 1.02 1.02 Proline dehydrogenase CodY PutR, SigA, Spo0A BSU03480 srfAA 2.61 2.91 1.84 1.85 Surfactin synthase subunit 1 Abh, CodY, ComA, PerR, PhoP, SigA, Spx BSU03490 srfAB 3.06 3.28 2.19 2.22 Surfactin synthase subunit 2 Abh, CodY, ComA, PerR, PhoP, SigA, Spx BSU03510 srfAC 3.16 3.37 2.47 2.51 Surfactin synthase subunit 3 Abh, CodY, ComA, PerR, PhoP, SigA, Spx BSU03520 srfAD 3.18 3.46 2.63 2.67 Surfactin synthase thioesterase subunit Abh, CodY, ComA, PerR, PhoP, SigA, Spx BSU03530 ycxA 2.37 2.44 2.08 2.08 Unknown ComA BSU_tRNA_36 trnD-Thr 1.02 1.13 1.03 1.06 Threonyl transfer RNA Unknown BSU09710 bmrC 1.01 1.05 1.33 1.33 Multidrug resistance ABC transporter ATP-binding protein AbrB, BmrB BSU09720 bmrD 1.05 1.09 1.21 1.17 Multidrug resistance ABC transporter ATP-binding protein AbrB, BmrB BSU10370 bioY 3.81 4.10 1.40 1.43 S-protein of biotin ECF transporter BirA BSU11381 appA/1 2.36 2.45 1.31 1.30 Oligopeptide ABC transporter, inactive pseudogene in strain CodY, ScoC, TnrA BSU11382 appA/2 2.52 2.61 1.17 1.10 Oligopeptide ABC transporter, inactive pseudogene in strain CodY, ScoC, TnrA BSU12010 manP 2.38 2.67 1.38 1.47 PTS system-mannose-specific transporter subunit EIIBCA ManR, SigA BSU15960 sivC 2.89 3.06 2.67 2.71 Inhibitor of entry into sporulation via KinB or KinC AbrB, SigD BSU18000 citB 1.83 1.99 1.17 1.18 aconitase CcpA, CcpC, CitB, CodY, FsrA, SigA BSU21330 yomK 1.47 1.58 1.90 1.93 Unknown SPβ prophage BSU21420 bhlA 1.09 1.34 −1.05 −1.28 Holin-like protein SPβ prophage BSU24620 tasA 2.22 2.33 3.11 3.38 major component of biofilm matrix, forms amyloid fibers AbrB, LutR, RemA, SigA, SinR BSU24630 sipW 2.37 2.51 2.81 3.52 Bifunctional signal peptidase I that controls surface-adhered AbrB, LutR, RemA, SigA, biofilm formation and processes TasA and TapA SinR BSU24640 tapA 2.77 2.93 2.78 2.96 TasA anchoring/assembly protein AbrB, LutR, RemA, SigA, SinR BSU26490 yrkJ 1.73 1.85 1.96 1.92 Unknown Unknown BSU26500 yrkI 2.32 2.50 2.22 2.48 Unknown Unknown BSU26510 yrkH 2.63 2.80 2.50 2.89 Unknown Unknown BSU26530 yrkF 3.22 3.62 2.77 3.84 Unknown Unknown BSU26540 yrkE 3.16 3.47 2.76 3.65 Unknown Unknown BSU29440 argH 2.57 2.67 1.25 1.22 Argininosuccinate lyase AhrC BSU29450 argG 2.50 2.61 1.29 1.30 Argininosuccinate synthase AhrC BSU30200 bioB 4.35 4.64 1.20 1.16 Biotin synthase BirA BSU30210 bioD 4.77 5.05 1.67 1.73 Dethiobiotin synthase BirA BSU30230 bioA 4.57 4.75 1.36 1.31 Lysine-8-amino-7-oxononanoate aminotransferase BirA BSU30240 bioW 5.63 5.94 2.31 2.41 6-carboxyhexanoate–CoA ligase BirA BSU30740 mntD 3.95 4.11 3.13 3.32 Manganese ABC transporter (permease) MntR BSU30750 mntC 3.97 4.12 2.98 3.31 Manganese ABC transporter (membrane protein) MntR BSU30760 mntB 3.83 3.98 2.82 3.15 Manganese ABC transporter (ATP-binding protein) MntR BSU30770 mntA 3.12 3.32 2.70 2.76 Manganese ABC transporter (Mn-binding lipoprotein) MntR BSU31250 tlpA 1.42 1.47 1.32 1.33 Methyl-accepting chemotaxis protein AbrB, SigD BSU31990 dhbC 2.48 2.57 1.05 1.08 Isochorismate synthase; siderophore bacillibactin synthesis AbrB, Fur, Kre, SigA, SigI BSU32010 besA 2.31 2.51 1.39 1.40 Trilactone hydrolase, catalyzes ferri-bacillibactin hydrolysis AbrB, Fur leading to cytosolic iron release npj Microgravity (2019) 1 Published in cooperation with the Biodesign Institute at Arizona State University, with the support of NASA M.D. Morrison et al. Table 1 continued Locus Tag Gene BRIC-21 BRIC-23 Annotated function Regulon Name DESeq2 limma DESeq2 limma BSU32450 pucL 1.07 1.32 −1.14 −1.32 Urate oxidase PucR, SigA, TnrA BSU33140 yvqJ 1.20 1.50 1.29 1.34 Similar to to macrolide-efflux protein AbrB BSU33490 cadA 2.37 3.38 1.59 2.23 Cadmium transporting ATPase, resistance to cadmium CzrA, SigA BSU33770 sdpC 4.41 4.61 1.56 1.61 Toxin, collapses the proton motive force and induces AbrB, Rok, Spo0A autolysis, kills non-sporulating cells, induces activity of SigW BSU34290 epsI 1.01 1.10 1.66 1.68 Glycosyltransferase, synthesis of extracellular poly-N- AbrB, EAR riboswitch, acetylglucosamine RemA, SigA, SinR BSU34350 epsC 1.12 1.17 1.48 1.48 UDP-sugar epimerase, required for extracellular AbrB, EAR riboswitch, polysaccharide synthesis RemA, SigA, SinR BSU34360 epsB 1.01 1.04 2.29 2.41 Extracellular polysaccharide synthesis, protein tyrosine kinase AbrB, EAR riboswitch, RemA, SigA, SinR BSU35070 yvmC 1.93 2.11 1.15 1.13 Cyclodipeptide synthase; biosynthesis of the extracellular AbrB, PchR iron chelate pulcherrimin BSU35080 pchR 3.70 3.91 1.55 1.53 Transcriptional repressor (MarR family), controls the CcpA, PchR expression of genes involved in pulcherriminic acid biosynthesis BSU37780 rocA 1.10 1.31 −1.14 −1.35 3-hydroxy-1-pyrroline-5-carboxylate dehydrogenase; AbrB, AhrC, CodY, RocR, arginine, ornithine and citrulline utilization SigL BSU37800 sivA 1.62 1.79 2.14 2.20 Inhibitor of KinA autophosphorylation, and subsequently of AbrB entry into sporulation BSU40180 yydF 4.55 4.77 2.51 2.82 Secreted peptide, controls LiaR-LiaS activity AbrB, Rok, SigA Values are log -fold FL:GC expression ratios. Gene names, annotated functions, and regulons are from Subtiwiki (http://subtiwiki.uni-goettingen.de/v3/index. php) , accessed on September 24, 2018 operons, yvmC-cypX and pchR-yvmA. The yvmC and pchR could be exhibiting a higher degree of resistance to genes were found to be significantly upregulated in FL antibiotics or toxic compounds. However, in a separate samples of both BRIC-21 and BRIC-23 (Table 1), and further publication we reported that replicate samples from the inspection revealed that both the yvmC-cypX and pchR-yvmA BRIC-21 experiment were exposed post-flight to a battery of operons were significantly upregulated in FL samples from 72 antibiotics and growth inhibitors, and no significant BRIC-21 (Supplemental Table S1). difference in resistance levels was found in FL vs. GC (iv) Arginine biosynthesis. The amino acid arginine is produced in samples. a pathway encoded by the argCJBD-carAB-argF operon (vii) Inhibitors of sporulation initiation. Sporulation in B. subtilis is which converts glutamate to citrulline, and the argGH triggered when cells sense the depletion of nutrients in their operon, which converts citrulline to arginine. The argGH environment. Three genes [sivA, sivB (bslA), and sivC; siv for operon was found to be significantly upregulated in both sporulation-inhibitory vegetative genes] have recently been BRIC-21 and BRIC-23 flight samples. Closer examination of described which actively inhibit the initiation of sporulation the datasets revealed significant upregulation of argCJBD- when B. subtilis is growing in the presence of sufficient carAB-argF operon transcripts in BRIC-21 FL samples nutrients. Both the sivA and sivC genes were found to be (Supplemental Table S1) and significant upregulation of significantly upregulated in FL samples of both BRIC-21 and the arginyl-tRNA genes trnJ-Arg and trnE-Arg in BRIC-23 FL BRIC-23 (Table 1), and as mentioned above, the sivB (bslA) samples (Supplemental Table S2). gene associated with biofilm formation was upregulated in (v) Manganese transport. The major manganese ABC-type BRIC-21 FL samples (Supplemental Table S1). transporter in B. subtilis, encoded by the mntABCD operon, was found to be strongly upregulated in both BRIC-21 and Additional genes upregulated in FL samples BRIC-23 FL samples (Table 1). In addition, closer examination of the datasets revealed that the manganese-proton Examination of Table 1 revealed a number of additional genes symporter mntH was significantly upregulated in BRIC-23 upregulated in both BRIC-21 and BRIC-23 FL samples for which no FL samples (Supplementary Table S2). Paradoxically, in BRIC- clear phenotypic consequence could be discerned. First, a number 21 FL samples we also observed significant upregulation of of genes encoding products of unknown function were upregu- the ydfM gene, encoding a putative Mn(II) efflux pump lated, including ybdO, yczC, ycxA, the yrkEFHIJ operon (Table 1). (Supplementary Table S1). Second, genes encoding transporters for glycerol-3-phosphate (vi) Resistance and toxin genes. A number of genes encoding (glpT), mannose (manP), and oligopeptides (appA/1 and appA/2) resistance and toxic functions were observed to be were upregulated, as were genes encoding an extracellular lipase upregulated in FL samples of both BRIC-21 and BRIC-23, (lipA), proline dehydrogenase (putB), aconitase (citB), a single including: ybfO, encoding a putative erythromycin esterase; methyl-accepting chemotaxis protein (tlpA), urate oxidase (pucL), the bmrCD operon, encoding a multidrug efflux transporter; and a secreted peptide controlling LiaR-LiaS activity (yydF) (Table yvqJ, encoding a putative macrolide-efflux protein; cadA, 1). Each of these belong to its own cohort of genes devoted to encoding a cadmium efflux pump; and sdpC, which encodes different functions in B. subtilis, but expression of the other a lytic toxin (Table 1). At first glance, increase in the members of the cohort were not significantly altered. As an aforementioned transcripts might suggest that FL samples example, the threonyl transfer RNA gene trnD-Thr is transcribed as Published in cooperation with the Biodesign Institute at Arizona State University, with the support of NASA npj Microgravity (2019) 1 M.D. Morrison et al. Table 2. List of genes significantly upregulated in GC samples both in BRIC-21 and BRIC-23 experiments Locus Tag Gene BRIC-21 BRIC-23 Function Regulon Name DESeq2 limma DESeq2 limma BSU03050 ldh 3.26 3.61 1.52 2.31 L-lactate dehydrogenase Rex, SigA BSU03060 lctP 3.18 3.35 1.18 1.78 L-lactate permease Rex, SigA BSU03290 nasE 1.63 1.66 1.12 1.20 Assimilatory nitrite reductase (subunit) Fur, NsrR, ResD, SigA, TnrA BSU03300 nasD 2.45 2.66 1.42 2.02 Assimilatory nitrite reductase (subunit) Fur, NsrR, ResD, SigA, TnrA BSU05720 ydhE 1.32 1.33 1.23 1.27 Similar to macrolide glycosyltransferase LiaR BSU06240 bdhA 1.45 1.46 1.19 1.32 Acetoin reductase/butanediol dehydrogenase AbrB BSU10230 yhfH 1.19 1.29 1.59 1.74 Unknown Unknown BSU17710 tatAC 1.30 1.33 1.28 1.33 Component of the twin-arginine translocation pathway Unknown BSU19180 des 1.07 1.07 1.23 1.35 Phosphlipid desaturase DesR, SigA BSU19190 desK 1.08 1.07 1.72 1.85 Two-component sensor kinase, regulation of cold shock DesR, SigA expression of des BSU20580 yoqM 1.90 1.94 1.80 2.18 Unknown SPβ prophage BSU20760 yopU 1.10 1.15 1.29 1.73 Unknown SPβ prophage BSU20770 yopT 1.10 1.14 1.00 1.11 Unknown SPβ prophage BSU21050 yonN 1.07 1.10 1.27 1.42 DNA-binding protein HU 2 SPβ prophage BSU21320 yomL 1.59 1.64 1.24 1.54 Unknown SPβ prophage BSU21329 youB 1.52 1.57 1.16 1.30 Unknown SPβ prophage BSU29310 cmoJ 1.02 1.12 1.14 1.43 Alkyl monooxygenase, required for the conversion of S-methyl- AscR, CymR, SigA cysteine to cysteine BSU29340 tcyN 1.42 1.55 1.17 1.37 Cystine ABC transporter (ATP-binding protein) AscR, CymR, SigA BSU29360 tcyL 1.33 1.50 1.06 1.21 Cystine ABC transporter (permease) AscR, CymR, SigA BSU30660 ytkA 1.58 1.60 1.35 1.43 Unknown unknown BSU37250 narI 2.58 2.67 1.42 2.03 Nitrate reductase (gamma subunit) Fnr, SigA BSU37260 narJ 2.86 3.01 1.36 2.15 Chaperone for the nitrate reductase (protein J) Fnr, SigA BSU37270 narH 3.15 3.32 1.28 2.08 Nitrate reductase (beta subunit) Fnr, SigA BSU37280 narG 3.24 3.41 1.17 1.95 Nitrate reductase (alpha subunit) Fnr, SigA BSU37310 fnr 2.14 2.20 1.56 1.92 Transcriptional regulator of anaerobice genes Fnr, NsrR, ResD, SigA BSU37320 narK 3.03 3.16 1.32 1.90 Nitrite extrusion protein Fnr, NsrR, SigA BSU37350 sboA 2.91 3.17 1.48 1.88 Subtilosin-A AbrB, ResD, Rok, SigA BSU37360 sboX 2.70 3.31 1.38 1.63 Bacteriocin-like product AbrB, ResD, Rok, SigA BSU37410 albE 1.87 2.29 1.17 1.22 Antilisterial bacteriocin (subtilosin) production AbrB, ResD, Rok, SigA BSU37420 albF 2.07 2.61 1.42 1.52 Antilisterial bacteriocin (subtilosin) production AbrB, ResD, Rok, SigA BSU37430 albG 2.44 2.57 1.32 1.40 Antilisterial bacteriocin (subtilosin) production AbrB, ResD, Rok, SigA BSU37440 ywhL 2.13 2.77 1.35 1.39 Unknown unknown BSU38060 ywcJ 3.09 3.28 1.87 2.06 Similar to nitrite transporter Rex BSU38070 sacT 1.60 1.66 1.12 1.16 Transcriptional antiterminator for the sacP-sacA-ywdA operon DnaA, SacT BSU38730 cydD 3.03 3.27 1.52 2.08 ABC transporter required for expression of cytochrome bd (ATP- CcpA, Rex, ResD, SigF binding protein) BSU38740 cydC 3.27 3.51 1.26 1.89 ABC transporter required for expression of cytochrome bd (ATP- CcpA, Rex, ResD, SigF binding protein) Values are log -fold GC:FL expression ratios. Gene names, annotated functions, and regulons are from Subtiwiki (http://subtiwiki.uni-goettingen.de/v3/index. php), accessed on September 24, 2018 part of a 16-tRNA gene operon located just downstream from the Genes upregulated in GC samples ribosomal RNA rrnD gene cluster in B. subtilis, but it is unclear The 36 genes whose expression was significantly upregulated in why only this transcript, and not the entire operon, was GC samples from both BRIC-21 and BRIC-23 are listed in Table 2. significantly upregulated. Third, two genes encoded by the They are arranged according to their BSU locus tag, i.e., in the prophage SPβ (yomK and bhlA) were upregulated in FL samples order that they are located on the B. subtilis 168 chromosome. (Table 1), but numerous SPβ-related genes were also upregulated Examination of the data revealed upregulation of several blocks of in GC samples (Table 2), so no coherent pattern of gene genes associated with the response of B. subtilis to oxygen expression could be ascertained. npj Microgravity (2019) 1 Published in cooperation with the Biodesign Institute at Arizona State University, with the support of NASA M.D. Morrison et al. limitation. A previous study reported that the global response of and transmitted to the ResD response regulator, encoded by the B. subtilis transcriptome to strict anaerobiosis resulted in the the last two genes of the resABCDE operon. ResD activates induction or repression of hundreds of genes involved in a variety expression of a large set of genes and operons including its of cell functions including carbon metabolism, electron transport, own (resABCDE), cydABCD, nasBCDEF, and sboAXalbABCDEFG. iron uptake, antibiotic production, and stress responses. Our data One of the genes activated by ResD is fnr, itself a regulator revealed that only a subset of the entire anaerobic regulon was of anaerobic gene expression. Fnr activates its own activated, indicating that cells in GC samples underwent only expression (narK-fnr), as well as that of narGHJI and another partial oxygen deprivation, and these genes are described below. regulator encoded by the arfM gene. In addition, changes in cell physiology associated with the switch to oxygen (i) Fermentation. TSYG medium contains the fermentable sugar limitation activate a transcriptional regulator called AlsR, glucose, and B. subtilis is capable of mixed-acid fermentation 47,52 which activates expression of alsSD directly and bdhA using separate pathways for production of the end products 53 indirectly. Meanwhile, the regulatory protein Rex senses acetate, lactate, acetoin, 2,3-butanediol, and ethanol. In GC changes in the NAD /NADH ratio brought on by oxygen samples from both the BRIC-21 and BRIC-23 missions we limitation and responds by activating a number of genes observed significant upregulation of the ldh-lctP operon for including the cydABCD and ldh-lctP operons. We were 45,46 fermentation of lactate from pyruvate, as well as the prompted by these observations to search for regulators of alsSD operon for fermentation of acetoin from pyruvate anaerobic gene expression in the datasets, and found and the bdhA gene for fermentation of 2,3-butanediol from significant upregulation of the arfM gene in BRIC-21 acetoin (Table 2). Interestingly, we did not note any (Supplemental Table S1), the resD and resE genes in BRIC- significant changes in the expression of genes involved in 23 (Supplemental Table S2), and the narK-fnr operon in both acetate or ethanol fermentation in our samples (Supple- BRIC-21 and BRIC-23 GC samples (Table 2). mental Tables S1 and S2). (vi) Phospholipid desaturase. Most prokaryotes regulate mem- (ii) Anaerobic respiration. Growth under oxygen-limiting condi- brane fluidity in part by controlling the degree of saturation tions results in two major modifications of the B. subtilis of membrane phospholipids. B. subtilis accomplishes this respiratory electron transport chain. First, oxygen depletion using a fatty acid desaturase encoded by des, the first gene activates synthesis of the high-affinity cytochrome bd 54,55 in the des-desKR operon. Expression of des is activated ubiquinol oxidase encoded by the cydABCD operon. We by exposure to low temperature, mediated through a two- observed that the cydC and cydD genes were significantly component system composed of a membrane-bound upregulated in GC samples of BRIC-21 and BRIC-23 (Table 2), sensor kinase and a response regulator encoded by the and further examination of the datasets revealed that the 55 desK and desR genes, respectively. Transcription of des and entire cydABCD operon was strongly upregulated in GC desK were seen to be significantly upregulated in GC samples from BRIC-21 (Supplemental Table S1). Second, samples of both BRIC-21 and BRIC-23 (Table 2), and closer anaerobiosis results in activation of genes responsible for examination of the datasets revealed that the entire des- utilizing nitrate and nitrite as alternative terminal electron desKR operon was upregulated in BRIC-23 GC samples acceptors. In B. subtilis GC samples we noted significant (Supplemental Table S2). While the biological significance of upregulation of the narGHJI operon encoding nitrate this observation is at present unclear, it is interesting to note reductase, the nasDE operon encoding nitrite reductase, that in previous work we reported that des-desKR transcrip- and the narK-fnr operon encoding the NarK nitrite extrusion tion was also activated by exposure of B. subtilis to low protein and the Fnr regulator of anaerobic gene expression atmospheric pressure. (Table 2). It is interesting to note that the ywcJ transcript, (vii) Metabolism of a cysteine analog. A pathway in B. subtilis was encoding a putative nitrite transporter previously identified recently described which can convert the cysteine analog S- as a member of the B. subtilis anaerobic regulon, was also methyl-cysteine directly to cysteine as a sulfur source, upregulated in GC samples. Because TSYG medium does not encoded by the snaA-tcyJKLMN-cmoOIJ-ribR-sndA-ytnM provide a significant source of nitrate or nitrite, it is likely operon. We observed that three genes from this operon that although expression of the nar and nas genes was (cmoJ, tcyL, and tcyN) were upregulated in GC samples from induced by oxygen limitation, they did not serve a useful both BRIC-21 and BRIC-23 (Table 2). Further examination of physiological function for cells in the GC samples. the datasets revealed additional genes of the operon (iii) Subtilosin production. Further indication for oxygen limita- significantly upregulated in BRIC-21 (snaA, tcyJ, tcyK, tcyM, tion in BRIC-21 and BRIC-23 GC samples was evidenced by cmoI, and ribR) and BRIC-23 (cmoO, sndA, and ytnM)GC induction of the genes encoding the antilisterial antibiotic samples (Supplemental Tables S1 and S2). subtilosin A, which was previously shown to be induced by anaerobiosis (Table 2). The pathway for subtilosin A biosynthesis is encoded by the sboAXalbABCDEFG operon, and in both BRIC-21 and BRIC-23 GC samples the sboA, sboX, Additional genes upregulated in GC samples albE, albF, and albG transcripts were significantly upregu- Examination of Table 2 revealed a number of genes significantly lated (Table 2). Further inspection of the datasets revealed up-regulated in GC samples from both BRIC-21 and BRIC-23 for that the entire sboAXalbABCDEFG operon was significantly which no clear phenotypic consequence could be discerned. First, upregulated in BRIC-21 GC samples (Supplemental Table S1). three genes encoding products of unknown function (yhfH, ytkA, (iv) Other genes belonging to the anaerobic regulon. Two genes and ywhL) were upregulated in GC samples (Table 2). Second, encoding products of unknown or only putatively annotated upregulation of genes encoding a putative macrolide glycosyl- function (ydhE and ytkA) were seen to be upregulated in GC transferase (ydhE), a component of the twin-arginine translocase samples of both BRIC-21 and BRIC-23 (Table 2). Activation of (tatAC), and a transcriptional antiterminator (sacT) (Table 2), but transcripts for these genes under anaerobic conditions was these genes form parts of larger cohorts of genes which were not noted previously. (v) Genes regulating anaerobiosis. The response to oxygen themselves significantly upregulated. Third, transcripts for a limitation in B. subtilis has been well studied and is number of genes encoded by the SPβ prophage (yoqM, yopU, controlled by a complex regulatory hierarchy. Oxygen yopT, yonN, yomL, and youB) were induced, the function of most of limitation is sensed by the membrane sensor kinase ResE which are unknown. Published in cooperation with the Biodesign Institute at Arizona State University, with the support of NASA npj Microgravity (2019) 1 M.D. Morrison et al. DISCUSSION interface, thus limiting their access to oxygen. Third, under the influence of gravity, dissolved gases, nutrients, and waste products In the present study, we analyzed the effects of spaceflight on the are transported in liquids by convection and diffusion; in Bacillus subtilis transcriptome in two separate spaceflight experi- microgravity, convection is negated and transport becomes ments designated BRIC-21 and BRIC-23. This is the first spaceflight dominated by diffusive processes. Oxygen transfer by convection bacterial transcriptome reported from a Gram-positive bacterium, would be expected to be greater in GC than FL cultures, but this and the first transcriptome study of two separate spaceflight effect appears to be outweighed by the first two factors. missions using the same bacterial strain, growth media, and Unfortunately, because B. subtilis prefers an oxygen-rich environ- hardware. The results uncovered several differences between FL ment for optimum growth, inclusion of an air-containing head- and GC samples that were exhibited in both BRIC-21 and BRIC-23 space in the PDFU cultures is unavoidable. In order to address this missions. hardware limitation, we are currently working on the design and construction of PDFU inserts that will maintain a constant Biofilm-associated transcripts geometry of liquid and air space both in microgravity and in 1- The observation that transcripts associated with biofilm formation xg controls. were significantly upregulated both in BRIC-21 and BRIC-23 FL samples is in agreement with the results from prior spaceflight Nutrient utilization-associated transcripts experiments using several different microbes indicating enhanced 23–27 As stated above, convection ceases in microgravity and the biofilm production in spaceflight. Although the domesticated transport of nutrients or waste products through liquid media laboratory strain B. subtilis 168 does not form robust biofilms, we becomes diffusion-limited. We observed significantly upregu- propose that biofilm formation in spaceflight-grown B. subtilis lated transcripts for biotin and arginine biosynthetic genes in FL could readily be studied in detail by using undomesticated samples, indicating that availability of these two nutrients differed biofilm-producing strains such as NCIB 3610 and its 58,59 in FL vs. GC cultures. However, we did not note significant derivatives. differences between FL and GC samples in the expression of any other biosynthetic pathways, nor was there a significant difference Oxygen limitation-associated transcripts in the final growth yield of cells between FL and GC cultures. We noted significant upregulation in GC samples of transcripts Therefore, it does not appear that exposure to microgravity led to associated with fermentation, anaerobic respiration, and subtilosin a generalized nutrient deficiency in FL samples. production, and upregulation in FL samples of transcripts In our analysis we treated the BRIC-21 and BRIC-23 missions as associated with siderophore production. These observations replicate experiments, but it should be noted that the two indicate that FL and GC samples were experiencing different experiments differed in incubation times, thus at harvest the BRIC- degrees of oxygen availability in the two experiments. How could 21 and BRIC-23 samples were at slightly different growth phases exposure to microgravity vs. 1 xg result in different levels of (late-exponential and transition-phase, respectively). It is therefore oxygen available to the liquid cultures? First, we noted that during to be expected that the resulting transcriptome profiles would deintegration of frozen BRIC-21 samples from PDFUs, FL samples differ somewhat, due to the substantial reorganization of global incubated in microgravity had assumed a toroidal shape (Fig. 4), gene expression which occurs during the transition from while GC samples formed a disk-shaped layer in the bottom of the exponential to stationary-phase growth in B. subtilis. In addition, Petri dish as expected. By assuming a toroidal configuration in it should be kept in mind that measuring the transcriptome microgravity, the liquid FL cultures may present a greater surface captures only one aspect of physiology; it does not take into area to the air phase than in GC samples. Second, under the account regulatory controls exerted at the level of numerous influence of gravity, cells in GC samples would tend to sediment posttranscriptional processes (translation, protein processing and toward the bottom of the Petri dish, away from the liquid/air modification, metabolic regulation of enzyme activity, assembly of subcellular structures, etc.) which must take place in order for a microbe to manifest its final phenotype. With these caveats in mind, combining the datasets from the two experiments allowed us to perform a more robust analysis and led to the identification of a common set of genes that were consistently differentially expressed between FL and GC samples in both experiments. These genes, particularly those involved in biofilm formation, will be interesting candidates for future study. METHODS Bacterial strain, media, and growth conditions The strain used in this study was Bacillus subtilis subsp. subtilis strain 168 (trpC2) from our laboratory stock collection. Medium used throughout was Trypticase Soy Yeast Extract (TSY) medium consisting of (g/L): tryptone, 15; soytone, 5; NaCl, 5; yeast extract, 3; K HPO , 2.5; glucose, 2.5; final pH 7. For 2 4 semisolid plates, agar was added to TSY to a final concentration of 15.0 g/L. Glycerol was added to TSY liquid medium to 10% (v/v) final concentration, resulting in TSYG medium. B. subtilis spores were routinely prepared by cultivation in liquid Schaeffer sporulation medium at 37 °C with vigorous aeration. The culture was harvested when phase-contrast microscopic examination revealed that it consisted of >90% free spores, usually after 3–4 days of incubation. Spores were purified by lysozyme treatment, buffer washing, and heat shock (80 °C, 10 min) as described previously, determined by phase-contrast microscopy to be >99% free of cell debris and unsporulated cells, and stored at 4 °C in deionized water. The spore Fig. 4 Removal of a typical frozen BRIC-21 FL sample from its PDFU suspension was heat-activated (65 °C, 20 min) before use. From a working npj Microgravity (2019) 1 Published in cooperation with the Biodesign Institute at Arizona State University, with the support of NASA M.D. Morrison et al. 8 7 suspension (10 /mL) of spores in water, aliquots of 0.1 mL (~10 CFU) were Read processing, alignment, and quantification applied to the bottoms of sterile 60-mm diameter Petri dishes (Falcon Cat. Quality control, mapping, and gene level quantification of Illumina No. 1007, Fisher Scientific) and air-dried for 48–72 h at room temperature sequences were performed using the Galaxy suite available through the protected from light. Samples were integrated into Biological Research in University of Florida’s High-Performance Research Computing Center. The Canister Dual-Chamber Petri Dish Fixation Units (BRIC-PDFU) spaceflight first 12 bases were trimmed off all reads using FASTQ Trimmer v0.014 to hardware using aseptic technique as described in detail previously. remove random hexamer primers, and read quality of the resulting sequences were checked using the FastQC program. Corresponding paired-end read files were mapped to the Bacillus subtilis strain 168 BRIC-PDFU hardware genome [National Center for Biotechnology Information (NCBI) RefSeq Biological Research in Canister (BRIC)-Petri Dish Fixation Unit (PDFU) 67 accession number NC_000964.3] using Bowtie2 v2.3.2. Mapping quality 63,64 hardware has been described in detail previously. Each BRIC canister 68 was evaluated using SAMStat followed by gene level quantification using enclosed 5 PDFUs, and each PDFU contained a space to accommodate the 69 htseq-count v0.6.1. Gene counts for BRIC-21 and BRIC-23 were separated bottom half of a single 60-mm diameter Petri dish to which air-dried into separate count matrices for differential expression analysis. spores were deposited. A separate syringe compartment contained TSYG liquid medium. Growth is initiated by injection of 8.5 mL of TSYG medium Differential expression and functional analyses into the Petri dish compartment, leaving a headspace of ~18 mL of ambient air which exchanges with the static culture via passive diffusion. To reduce false positive results, two Bioconductor packages, limma 70,71 72 The headspace is the only source of additional oxygen, and because the v3.32.10 and DESeq2 v1.16.1, were used to determine differential system is hermetically sealed, O is consumed and CO accumulates in the 2 2 expression. These two packages have been shown to exhibit consistent culture. Once initiated, there is no mechanism to access cultures for detection precision in comparisons using a large or very small number of removal of samples during flight. replicates making these packages suitable for our BRIC-21 (n = 3) and BRIC-23 (n = 9) comparisons. Per limma recommendation, genes with a sum of less than nine counts across all samples were removed before Experimental timeline analysis. For limma analyses, filtered genes were normalized between The schedules of pre-flight, flight, and post-flight activities are described in samples using TMM normalization and differential expression analysis was 16,17 20 detail elsewhere for the BRIC-21 and BRIC-23 experiments. Experi- conducted using the built-in voom transformation. Because DESeq2 mental details for the two FL experiments were essentially identical in requires non-normalized count data, analysis was performed on the raw terms of hardware (BRIC-PDFU), inoculum size (10 spores), medium used filtered genes. P values from both analyses were corrected for multiple (TSYG), and media volumes injected (8.5 mL). Slight differences in ambient testing using the Benjamini–Hochberg method. To be considered ISS temperature were recorded inside the BRIC-21 (average ~22.8 °C) and differentially expressed, genes had to have a >2-fold difference with a P BRIC-23 (average ~22.3 °C) canisters. The only notable difference in the two value < 0.01 between FL and GC expression using both limma and DESeq2 experiments was the time of incubation for BRIC-21 and BRIC-23 samples, methods. Genes meeting these criteria were annotated using the NCBI 25 h and 36 h, respectively. In both missions, growth was terminated by database and the Subtiwiki web server. Functional analysis was transferring BRIC canisters to the −80 °C freezer onboard the ISS. In both conducted using the Search Tool for the Retrieval of Interacting Genes/ cases, samples remained solidly frozen until returned to the lab. Proteins (STRING) protein network database. STRING protein–protein Asynchronous GC samples were treated identically to FL samples in terms interaction networks for up- and downregulated genes were generated of hardware, configuration, schedule, temperature profiles, pre- and post- separately, and Kyoto Encyclopedia of Genes and Genomes (KEGG) 16,17,20 flight handling. pathways were found using the built-in enrichment tools. To check the variability between all the BRIC-21 and BRIC-23 datasets, a PCA was performed in the statistical environment R v3.4.4. Gene loading scores and BRIC-21 RNA isolation, processing, and sequencing sample eigenvalues for each principal component were calculated using Frozen FL and GC samples (n = 3) were partially thawed at room the princomp function in the R package stats v3.4.4. Sample eigenvalues for temperature. Partially thawed samples were transferred into 15-mL conical the first two principal components were plotted in R using the default tubes, and placed on ice until completely thawed. Cells were recovered by plotting function. centrifugation (7000 × g, 20 min, 0 °C) in a benchtop centrifuge. Super- natants were transferred into sterile 15-mL conical tubes and stored at −70 °C. Cell pellets were immediately processed for total RNA extraction DATA AVAILABILITY TM and treatment with RNase-free DNase using the RiboPure RNA The RNA-seq datasets have been deposited in the NASA GeneLab Data System under Purification Kit (Thermo Fisher Scientific Inc, Waltham, MA) following the the accession numbers GLDS-185 (BRIC-21) and GLDS-138 (BRIC-23). manufacture’s protocols. RNA samples were quantified using a Qubit fluorometer (Invitrogen, Thermo Fisher Scientific Inc, Waltham, MA) and quality evaluated using the RNA 6000 Nano Chip on an Agilent 2100 ACKNOWLEDGEMENTS Bioanalyzer (Agilent Technologies, Santa Clara, CA). RNA Integrity Number (RIN) values ranged from 9.8 to 10.0, indicating high-quality total RNA The authors wish to thank the BRIC-21 and BRIC-23 payload development teams at preparations suitable for further processing. Samples were sent to Hudson NASA Kennedy Space Center (D. Dimapilis, A.D. Flowers, C. Grosse, J. Harp, D. Houze, Alpha Institute for Biotechnology (Huntsville, AL, USA) for ribosomal RNA H. Levine, G. Newsham, S. Manning-Roach, J. Richards, S. Richards, J. Smodell, and G. depletion, library preparation, and 100-nt paired-end sequencing on an Washington) and the GeneLab Process Verification Team (V. Boyko, K. Chakravarty, S. Illumina 2500 instrument. Costes, H. Fogle, M. Dinh, J. Galazka, S. Gebre, S. Lai Polo, and S. Reinsch) at NASA Ames Research Center for their excellent technical assistance. This work was supported by the NASA Research Opportunities in Space Biology grant NNX14AT38G BRIC-23 RNA isolation, processing, and sequencing to W.L.N. and P.F.-C. BRIC-23 sample processing was performed by the GeneLab Sample Processing Lab (NASA Ames Research Center, Mountain View, CA); detailed protocols are described in the BRIC-23 GeneLab dataset GLDS-138 (https:// AUTHOR CONTRIBUTIONS genelab-data.ndc.nasa.gov/genelab/accession/GLDS-138/). Briefly, thawed W.L.N. and P.F.-C. conceived and designed the study. All authors performed the FL and GC samples (n = 9) were centrifuged at 16,000 × g for 5 min at 0 °C. spaceflight experiments and RNA processing. M.D.M. and W.L.N. performed the data Total RNA extraction and RNase-free DNase treatment was also performed TM analysis. All authors contributed to writing the paper. using the RiboPure RNA Purification Kit (Thermo Fisher Scientific Inc, Waltham, MA) following the manufacture’s protocol. RNA concentrations were measured using a Qubit 3.0 Fluorometer (Thermo Fisher Scientific, ADDITIONAL INFORMATION Waltham, MA), and RNA quality was assessed on an Agilent 2100 Bioanalyzer (Agilent, Santa Clara, CA). Ribosomal RNA depletion was Supplementary information accompanies the paper on the npj Microgravity website (https://doi.org/10.1038/s41526-018-0061-0). performed using the Ribo-Zero rRNA Removal Kit for Gram-positive bacteria (Illumina) and sample cDNA libraries were synthesized using KAPA Competing interests: The authors declare no competing interests. RNA HyperPrep reagents (Roche) and amplified with 10 PCR cycles. Library sequencing was performed on an Illumina HiSeq 4000 platform. Published in cooperation with the Biodesign Institute at Arizona State University, with the support of NASA npj Microgravity (2019) 1 M.D. Morrison et al. Publisher’s note: Springer Nature remains neutral with regard to jurisdictional claims 29. Vlamakis, H., Chai, Y. R., Beauregard, P., Losick, R. & Kolter, R. Sticking together: in published maps and institutional affiliations. building a biofilm the Bacillus subtilis way. Nat. Rev. Microbiol. 11, 157–168 (2013). 30. Lopez, D., Fischbach, M. A., Chu, F., Losick, R. & Kolter, R. Structurally diverse natural products that cause potassium leakage trigger multicellularity in Bacillus subtilis. Proc. Natl Acad. Sci. USA 106, 280–285 (2009). REFERENCES 31. Driks, A. Tapping into the biofilm: insights into assembly and disassembly of a 1. Mora, M. et al. Microorganisms in confined habitats: microbial monitoring and novel amyloid fibre in Bacillus subtilis. Mol. Microbiol 80, 1133–1136 (2011). control of intensive care units, operating rooms, cleanrooms and the Interna- 32. Roux, D. et al. Identification of poly-N-acetylglucosamine as a major poly- tional Space Station. Front. Microbiol. 7, 20 (2016). saccharide component of the Bacillus subtilis biofilm matrix. J. Biol. Chem. 290, 2. Pietsch, J. et al. The effects of weightlessness on the human organism and 19261–19272 (2015). mammalian cells. Curr. Mol. Med. 11, 350–364 (2011). 33. Kobayashi, K. & Iwano, M. BslA(YuaB) forms a hydrophobic layer on the surface of 3. Lychakov, D. V. Behavioural and functional vestibular disorders after space flight: Bacillus subtilis biofilms. Mol. Microbiol 85,51–66 (2012). 2. Fish, amphibians and birds. J. Evol. Biochem. Physiol. 52,1–16 (2016). 34. Lombardia, E., Rovetto, A. J., Arabolaza, A. L. & Grau, R. R. A LuxS-dependent cell- 4. Vandenbrink, J. P. & Kiss, J. Z. Space, the final frontier: a critical review of recent to-cell language regulates social behavior and development in Bacillus subtilis. J. experiments performed in microgravity. Plant Sci. 243, 115–119 (2016). Bacteriol. 188, 4442–4452 (2006). 5. Rosenzweig, J. A. et al. Spaceflight and modeled microgravity effects on microbial 35. Perkins, J. B. & Pero, J. in Bacillus subtilis and its Closest Relatives: from Genes to growth and virulence. Appl. Microbiol Biotechnol. 85, 885–891 (2010). Cells (eds Sonenshein, A. L., Hoch, J. A., & Losick, R.), 271–286 (ASM Press, 2002). 6. Rosenzweig, J. A., Ahmed, S., Eunson, J. & Chopra, A. K. Low-shear force asso- 36. Rodionov, D. A., Mironov, A. A. & Gelfand, M. S. Conservation of the biotin regulon ciated with modeled microgravity and spaceflight does not similarly impact the and the BirA regulatory signal in Eubacteria and Archaea. Genome Res 12, virulence of notable bacterial pathogens. Appl. Microbiol Biotechnol. https://doi. 1507–1516 (2002). org/10.1007/s00253-014-6025-8 (2014). 37. Miethke, M. Molecular strategies of microbial iron assimilation: from high-affinity 7. Taylor, P. W. Impact of space flight on bacterial virulence and antibiotic sus- complexes to cofactor assembly systems. Metallomics 5,15–28 (2013). ceptibility. Infect. Drug Resist 8, 249–262 (2015). 38. May, J. J., Wendrich, T. M. & Marahiel, M. A. The dhb operon of Bacillus subtilis 8. National Research Council. Recapturing a Future for Space Exploration: Life and encodes the biosynthetic template for the catecholic siderophore 2,3-dihydrox- Physical Sciences for a New Era, 442 (National Academies Press: Washington, DC, ybenzoate-glycine-threonine trimeric ester bacillibactin. J. Biol. Chem. 276, 2011). 7209–7217 (2001). 9. Horneck, G., Klaus, D. M. & Mancinelli, R. L. Space microbiology. Microbiol Mol. Biol. 39. Randazzo, P., Aubert-Frambourg, A., Guillot, A. & Auger, S. The MarR-like protein Rev. 74, 121–156 (2010). PchR (YvmB) regulates expression of genes involved in pulcherriminic acid bio- 10. Zea, L. et al. A molecular genetic basis explaining altered bacterial behavior in synthesis and in the initiation of sporulation in Bacillus subtilis. BMC Microbiol. 16, space. PLoS One 11, e0164359 (2016). 16 (2016). 11. Mastroleo, F. et al. Experimental design and environmental parameters affect 40. Belitsky, B. R. in Bacillus subtilis and Its Closest Relatives: from Genes to Cells (eds Rhodospirillum rubrum S1H response to space flight. ISME J. 3, 1402–1419 (2009). Sonenshein, A. L., Hoch, J. A. & Losick, R.), 203–223 (ASM Press, 2002). 12. Huang, B., Liu, N., Rong, X., Ruan, J. & Huang, Y. Effects of simulated microgravity 41. Garti-Levi, S., Eswara, A., Smith, Y., Fujita, M. & Ben-Yehuda, S. Novel modulators and spaceflight on morphological differentiation and secondary metabolism of controlling entry into sporulation in Bacillus subtilis. J. Bacteriol. 195, 1475–1483 Streptomyces coelicolor A3(2). Appl. Microbiol. Biotechnol. 99, 4409–4422 (2015). (2013). 13. Crabbé, A. et al. Transcriptional and proteomic responses of Pseudomonas aer- 42. Rudner, R. et al. Two transfer RNA gene clusters associated with ribosomal RNA uginosa PAO1 to spaceflight conditions involve Hfq regulation and reveal a role operons rrnD and rrnE. Bacillus subtilis. J. Bacteriol. 175, 503–509 (1993). for oxygen. Appl. Environ. Microbiol 77, 1221–1230 (2011). 43. Ye, R. et al. Global gene expression profiles of Bacillus subtilis grown under 14. Wilson, J. W. et al. Space flight alters bacterial gene expression and virulence and anaerobic conditions. J. Bacteriol. 182, 4458–4465 (2000). reveals a role for global regulator Hfq. Proc. Natl Acad. Sci. USA 104, 16299–16304 44. Nakano, M. & Zuber, P. in Bacillus subtilis and Its Closest Relatives: from Genes to (2007). Cells (eds Sonenshein, A. L., Hoch, J. A., & Losick, R.), 393–404 (ASM Press, 2002). 15. Altman, N. & Krzywinski, M. Sources of variation. Nat. Methods 12,5–6 (2015). 45. Cruz Ramos, H. et al. Fermentative metabolism of Bacillus subtilis: physiology and 16. Morrison, M. D., Fajardo-Cavazos, P. & Nicholson, W. L. Cultivation in space flight regulation of gene expression. J. Bacteriol. 182, 3072–3080 (2000). produces minimal alterations in the susceptibility of Bacillus subtilis cells to 72 46. Marino, M., Hoffmann, T., Schmid, R., Mobitz, H. & Jahn, D. Changes in protein different antibiotics and growth-inhibiting compounds. Appl. Environ. Microbiol. synthesis during the adaptation of Bacillus subtilis to anaerobic growth condi- 83, e01584–01517 (2017). tions. Microbiol. SGM 146,97–105 (2000). 17. Fajardo-Cavazos, P., Leehan, J. D. & Nicholson, W. L. Alterations in the spectrum of 47. Renna, M. C., Najimudin, N., Winik, L. R. & Zahler, S. A. Regulation of the Bacillus spontaneous rifampicin-resistance mutations in the Bacillus subtilis rpoB gene subtilis alsS, alsD, and alsR genes involved in post-exponential-phase production after cultivation in the human spaceflight environment. Front. Microbiol. 9, 192 of acetoin. J. Bacteriol. 175, 3863–3875 (1993). (2018). 48. Nicholson, W. L. The Bacillus subtilis ydjL (bdhA) gene encodes acetoin reductase/ 18. Markowitz, V. M. et al. IMG: the integrated microbial genomes database and 2,3-butanediol dehydrogenase. Appl. Environ. Microbiol 74, 6832–6838 (2008). comparative analysis system. Nucleic Acids Res. 40, D115–D122 (2012). 49. von Wachenfeldt, C. & Hederstedt, L. in Bacillus subtilis and Its Closest Relatives: 19. Matsumoto, Y., Murakami, Y., Tsuru, S., Ying, B. W. & Yomo, T. Growth rate- from Genes to Cells (eds Sonenshein, A. L., Hoch, J. A., & Losick, R.), 163–179 (ASM coordinated transcriptome reorganization in bacteria. BMC Genom. 14, 10 (2013). Press, 2002). 20. Fajardo-Cavazos, P. & Nicholson, W. L. Establishing standard protocols for bac- 50. Zheng, G. L., Yan, L. Z., Vederas, J. C. & Zuber, P. Genes of the sbo-alb locus of terial culture in Biological Research in Canister (BRIC) hardware. Gravit. Space Res. Bacillus subtilis are required for production of the antilisterial bacteriocin sub- 4,58–69 (2016). tilosin. J. Bacteriol. 181, 7346–7355 (1999). 21. Banse, A. V., Chastanet, A., Rahn-Lee, L., Hobbs, E. C. & Losick, R. Parallel pathways 51. Härtig, E. & Jahn, D. Regulation of the anaerobic metabolism in Bacillus subtilis. of repression and antirepression governing the transition to stationary phase in Adv. Microb. Physiol. 61, 195–216 (2012). Bacillus subtilis. Proc. Natl Acad. Sci. USA 105, 15547–15552 (2008). 52. Frädrich, C. et al. The transcription factor AlsR binds and regulates the promoter 22. Flemming, H. C. & Wingender, J. The biofilm matrix. Nat. Rev. Microbiol. 8, of the alsSD operon responsible for acetoin formation in Bacillus subtilis. J. Bac- 623–633 (2010). teriol. 194, 1100–1112 (2012). 23. Kim, W. et al. Spaceflight promotes biofilm formation by Pseudomonas aerugi- 53. Oliveira, R. R. & Nicholson, W. L. The LysR-type transcriptional regulator (LTTR) nosa. PLoS One 8, e62437 (2013). AlsR indirectly regulates expression of the Bacillus subtilis bdhA gene encoding 24. Zea, L. et al. Phenotypic changes exhibited by E. coli cultured in space. Front. 2,3-butanediol dehydrogenase. Appl. Microbiol Biotechnol. 97, 7307–7316 (2013). Microbiol. 8, 12 (2017). 54. Mansilla, M. C. & de Mendoza, D. The Bacillus subtilis desaturase: a model to 25. Van Mulders, S. E. et al. The influence of microgravity on invasive growth in understand phospholipid modification and temperature sensing. Arch. Microbiol. Saccharomyces cerevisiae. Astrobiology 11,45–55 (2011). 183, 229–235 (2005). 26. Crabbé, A. et al. Spaceflight enhances cell aggregation and random budding in 55. Mendoza, D. Temperature sensing by membranes. Annu Rev. Microbiol 68, Candida albicans. Plos One 8, 24 (2013). 101–116 (2014). 27. Koenig, D. W. & Pierson, D. L. Microbiology of the space shuttle water system. 56. Fajardo-Cavazos, P. et al. Evolution of Bacillus subtilis to enhanced growth at low Water Sci. Technol. 35,59–64 (1997). pressure: up-regulated transcription of des-desKR encoding the fatty acid desa- 28. McLoon, A. L., Guttenplan, S. B., Kearns, D. B., Kolter, R. & Losick, R. Tracing the turase system. Astrobiology 12, 258–270 (2012). domestication of a biofilm-forming bacterium. J. Bacteriol. 193, 2027–2034 (2011). npj Microgravity (2019) 1 Published in cooperation with the Biodesign Institute at Arizona State University, with the support of NASA M.D. Morrison et al. 57. Chan, C.-M., Danchin, A., Marliere, P. & Sekowska, A. Paralogous metabolism: S- 71. Ritchie, M. E. et al. limma powers differential expression analyses for RNA- alkyl-cysteine degradation in Bacillus subtilis. Environ. Microbiol 16, 101–117 sequencing and microarray studies. Nucleic Acids Res. 43, e47 (2015). (2014). 72. Love, M. I., Huber, W. & Anders, S. Moderated estimation of fold change and 58. Branda, S. S., Gonzalez-Pastor, J. E., Ben-Yehuda, S., Losick, R. & Kolter, R. Fruiting dispersion for RNA-seq data with DESeq2. Genome Biol. 15, 550 (2014). body formation by Bacillus subtilis. Proc. Natl Acad. Sci. USA 98, 11621–11626 73. Seyednasrollah, F., Laiho, A. & Elo, L. L. Comparison of software packages for (2001). detecting differential expression in RNA-seq studies. Brief. Bioinform. 16,59–70 59. Konkol, M. A., Blair, K. M. & Kearns, D. B. Plasmid-encoded ComI inhibits com- (2015). petence in the ancestral 3610 strain of Bacillus subtilis. J. Bacteriol. 195, 74. Law, C. W., Chen, Y., Shi, W. & Smyth, G. K. Voom: precision weights unlock linear 4085–4093 (2013). model analysis tools for RNA-seq read counts. Genome Biol. 15, R29 (2014). 60. Phillips, Z. E. V. & Strauch, M. A. Bacillus subtilis sporulation and stationary phase 75. Benjamini, Y. & Hochberg, Y. Controlling the false discovery rate: a practical and gene expression. Cell Mol. Life Sci. 59, 392–402 (2002). powerful approach to multiple testing. J. R. Stat. Soc. B 57, 289–300 (1995). 61. Schaeffer, P., Millet, J. & Aubert, J. P. Catabolic repression of bacterial sporulation. 76. Michna, R. H., Zhu, B., Mäder, U. & Stülke, J. SubtiWiki 2.0–an integrated database Proc. Natl Acad. Sci. USA 54, 704–711 (1965). for the model organism Bacillus subtilis. Nucleic Acids Res. 44, D654–662 (2016). 62. Nicholson, W. L. & Setlow, P. in Molecular Biological Methods for Bacillus (eds 77. Szklarczyk, D. et al. STRING v10: protein-protein interaction networks, integrated Harwood, C. R. & Cutting, S. M.), 391–450 (J. Wiley & Sons, 1990). over the tree of life. Nucleic Acids Res. 43, D447–452 (2015). 63. Paul, A.-L. et al. Spaceflight transcriptomes: unique responses to a novel envir- 78. Zhu, B. Y. & Stülke, J. SubtiWiki in 2018: from genes and proteins to functional onment. Astrobiology 12,40–56 (2012). network annotation of the model organism Bacillus subtilis. Nucleic Acids Res. 46, 64. Wells, B., McCray, R. H., Best, M. D. & Levine, H. G. A flight-rated Petri dish D743–D748 (2018). apparatus providing two stage fluid injection for aseptic biological investigations in space. in 31st International Conference on Environmental Systems 2001-01-2286 (SAE International, 2001). Open Access This article is licensed under a Creative Commons 65. Blankenberg, D. et al. Manipulation of FASTQ data with Galaxy. Bioinformatics 26, Attribution 4.0 International License, which permits use, sharing, 1783–1785 (2010). adaptation, distribution and reproduction in any medium or format, as long as you give 66. FastQC: a quality control tool for high throughput sequence data. v. 0.11.5. http:// appropriate credit to the original author(s) and the source, provide a link to the Creative www.bioinformatics.babraham.ac.uk/projects/fastqc/ (Babraham Bioinformatics, Commons license, and indicate if changes were made. The images or other third party 2010). material in this article are included in the article’s Creative Commons license, unless 67. Langmead, B., Trapnell, C., Pop, M. & Salzberg, S. L. Ultrafast and memory-efficient indicated otherwise in a credit line to the material. If material is not included in the alignment of short DNA sequences to the human genome. Genome Biol. 10, R25 article’s Creative Commons license and your intended use is not permitted by statutory (2009). regulation or exceeds the permitted use, you will need to obtain permission directly 68. Lassmann, T., Hayashizaki, Y. & Daub, C. O. SAMStat: monitoring biases in next from the copyright holder. To view a copy of this license, visit http://creativecommons. generation sequencing data. Bioinformatics 27, 130–131 (2011). org/licenses/by/4.0/. 69. Anders, S., Pyl, P. T. & Huber, W. HTSeq–a Python framework to work with high- throughput sequencing data. Bioinformatics 31, 166–169 (2015). © The Author(s) 2019 70. Liu, R. et al. Why weight? Modelling sample and observational level variability improves power in RNA-seq analyses. Nucleic Acids Res. 43, e97 (2015). Published in cooperation with the Biodesign Institute at Arizona State University, with the support of NASA npj Microgravity (2019) 1

Journal

npj MicrogravitySpringer Journals

Published: Jan 7, 2019

There are no references for this article.