Procedures to isolate and identify A. niger collected from the ISS were described previously (Romsdahl et al., 2018). In brief, sterile swabs soaked in saline solution were used to sample the ISS surface and transported to Earth. Particles retrieved from the swab were spread into potato dextrose agar (PDA) plates and any growing colonies were purified, collected, and further analyzed. One of the collected isolates was identified as A. niger via ITS region sequencing, which was subsequently confirmed via whole-genome sequencing (WGS).

JSC-093350089 was cultivated on glucose minimal medium (GMM) agar plates (6 g/l NaNO₃, 0.52 g/l KCl, 0.52 g/l MgSO₄·7H₂O, 1.52 g/l KH₂PO₄, 10 g/l D-glucose, and 15 g/l agar supplemented with 1 ml/l of Hutner’s trace elements) covered with a cellophane membrane. Each of 10 prepared Petri plates (D = 10 cm) was inoculated with 1 × 10⁷ conidia/plate. Subsequently, plates were sealed with 3 M^™ Micropore^™ Surgical Tape (VWR International, Radnor, PA, United States) and placed in four Biological Research in Canister (BRIC) systems (three and two plates/BRIC). BRICs were divided into two groups, which were exact mimics, and transferred to 4°C. The exact timeline of the experiment is presented in Supplementary Figure 1. The whole experiment lasted 42 days from preparing the payload by a science team prior to launch till the handout of the ISS-grown samples back to the science team after the flight. The first group of BRICs was sent to the ISS and continuously kept at 4°C (1–20 days) prior to being transferred to ambient temperature for the active growth phase ~22°C for 12 days (21–32 days). After that time BRICs were stored at 4°C before returning to Earth (33–42 days). Upon arrival to Earth, BRICs were turned over to the science team for the downstream analyses, which commenced immediately. The second group of BRICs, treated as controls, was kept on Earth at Kennedy Space Center (KSC) and mimicked the ISS experiment timeline with roughly 2 h of delay. BRICs containing control samples were shipped from KSC to research team along with the ISS-grown samples. Lastly, for an additional secondary metabolite analysis, 1 × 10⁷ conidia/plate of the ISS- and ground-grown JSC-093350089 were grown on GMM medium at 28°C for 5 days.

Mycelia and conidia were collected from ground-and ISS-grown JSC-09335008 GMM agar plates. DNA was extracted using the Power Soil DNA Isolation Kit (Mo Bio Laboratories, Carlsbad, California, United States) following the manufacturers protocol. Extracted DNA was checked for quality using Qubit 2.0 Fluorometer and used for paired-end library preparation with TruSeq Nano DNA Library Preparation Kit (Illumina, San Diego, California, United States) followed by WGS at the Duke Center for Genomic and Computational Biology. Samples were sequenced using a HiSeq 4,000 Illumina Sequencer generating 101 base long reads.

Illumina sequence reads were trimmed using Trimmomatic v 0.36 (Bolger et al., 2014) and checked for quality using FastQC v 0.11.5 (Andrews, 2010). The genome and annotation files for A. niger CBS 513.88, (Pel et al., 2007) were downloaded from the FungiDB web portal (Stajich et al., 2012). Reads were mapped to CBS 513.88 the reference genome using the Burrows-Wheeler Aligner (BWA) software package v 0.7.12 (Li and Durbin, 2009) and further processed with SAMtools v 1.6 to generate sorted BAM files (Li et al., 2009). SNPs and INDELs were identified using GATK v 3.7 (DePristo et al., 2011). Duplicates were marked using Picard-tools MarkDuplicates¹ to remove PCR artifacts. Sequence reads containing putative INDELs were realigned using GATK’s IndelRealigner to generate an updated BAM file. Variants within each sample were called using GATK’s Haplotype Caller. GATK’s VariantFiltration was used to filter each VCF file based on stringent cutoffs for quality and coverage {SNPs: QD < 2.0, MQ < 40.0, QUAL < 100, FS > 60.0, MQRankSum < −12.5, SOR > 4.0, ReadPosRankSum < −8.0; Indels: QD < 2.0, FS > 200.0, MQRankSum < −12.5, SOR > 4, InbreedingCoeff < −0.8, ReadPosRankSum < −20.0}, so that only high-quality variants remained.

Mycelia and conidia from GMM agar plates were collected and stored at –80°C prior to protein extraction. Proteins were extracted with the lysis buffer consisting of 100 mm triethylammonium bicarbonate (TEAB) with 1:100 Halt Protease Inhibitor Cocktail (Thermo Scientific, Rockford, IL) and 200 μg/ml phenylmethylsulfonyl fluoride (Sigma-Aldrich, St. Louis, MO, United States). Mycelia and conidia were homogenized by bead beating using Precellys 24 homogenizer (Bertin, Rockville, MD). The lysed fungal material was centrifuged at 17,000 g for 15 min and the protein concentration in the supernatants was measured by the Bradford assay (Bio-Rad Laboratories, Inc. Hercules, CA, United States).

A 100 μg proteins from each sample were precipitated in 20% trichloroacetic acid (TCA) at 4°C. Protein pellets were washed with ice-cold acetone and re-suspended in 25 μl TEAB (100 mM) and 25 μl 2,2,2-trifluoroethanol (TFE). Proteins were reduced with 1 μl of tris(2-carboxyethyl)phosphine (TCEP, 500 mM), alkylated with iodoacetamide (IAA, 30 mM), and digested with 2.5 μg/sample of trypsin/lysC (Promega, Madison, WI, United States) overnight at 37°C. The digested peptides were quantified using the Pierce Quantitative Colorimetric Peptide Assay (Thermo Scientific, Waltham, MA, United States). 40 μg of peptides from each specific sample was labeled with the Thermo Scientific TMTsixplex Isobaric Mass Tagging Kit (JSC-E1 (ground 1) with TMT⁶-128, JSC-E2 (ground 2) with TMT⁶-130, JSC-S1 (ISS 1) with TMT⁶-129, JSC-S2 (ISS 2) with TMT⁶-131) according to the manufacturer’s protocol. All labeled-peptide mixtures were combined into a single tube, mixed, and fractionated using the Thermo Scientific Pierce High pH Reversed-Phase Peptide Fractionation Kit. While this kit usually uses eight fractions with step elution of up to 50% acetonitrile, ninth fraction was added eluting at 100% acetonitrile. Nine fractionated samples were dried using a SpeedVac concentrator and re-suspended in 1% (v/v) formic acid prior to liquid chromatography with tandem mass spectrometry (LC-MS/MS) analysis.

The samples were analyzed on an Orbitrap Fusion Tribrid mass spectrometer with an EASY-nLC 1,000 Liquid Chromatograph, a 75 μm × 2 cm Acclaim PepMap100 C18 trapping column, and a 75 μm × 25 cm PepMap RSLC C18 analytical column, and an Easy-Spray ion source (Thermo Scientific, Waltham, MA, United States). The peptides were eluted at 45°C with a flow rate of 300 nl/min over a 110 min gradient, from 3 to 30% solvent B (100 min), 30–50% solvent B (3 min), 50–90% solvent B (2 min), and 90% solvent B (2 min). The solvent A was 0.1% formic acid in water and the solvent B was 0.1% formic acid in acetonitrile.

The full MS survey scan (m/z 400–1,500) was acquired at a resolution of 120,000 and an automatic gain control (AGC) target of 2 × 10⁵ in the Orbitrap with the 50 ms maximum injection time for MS scans. Monoisotopic precursor ions were selected with charge states 2–7, a ± 10 ppm mass window, and 70 s dynamic exclusion. The MS² scan (m/z 400–2000) was performed using the linear ion trap with the 35% collision-induced dissociation (CID) energy. The ion trap scan rate was set to “rapid,” with an AGC target of 4 × 10³, and a 150 ms maximum injection time. Ten fragment ions from each MS² experiment were subsequently selected for an MS³ experiment. The MS³ scan (m/z 100–500) was performed to generate the TMT reporter ions in the linear ion trap using higher-energy collisional dissociation (HCD) at a 55% collision energy, a rapid scan rate and an AGC target of 5 × 10³, and a maximum injection time of 250 ms.

All MS data (MS¹, MS², and MS³) were searched using the Proteome Discoverer (version 2.2.0.388, Thermo Scientific) with the Sequest-HT searching engines against an Aspergillus niger CBS 513.88 database containing 10,549 sequences (NCBI). The searches were performed with the following parameters: 2 maximum missed cleavage sites, 6 minimum peptide length, 5 ppm tolerance for precursor ion masses, and 0.6 Dalton tolerance for fragment ion masses. The static modification settings included carbamidomethyl of cysteine residues, and dynamic modifications included oxidation of methionine, TMT6plex modification of lysine ε-amino groups and peptide N-termini, and acetyl modification of protein N-terminus. A false discovery rate (FDR) of 1% was set using a target-decoy database search. The reporter ions integration tolerance was 0.5 Da and the co-isolation threshold was 75%. The average signal-to-noise threshold of all reporter peaks was bigger than 10. The total intensity of a reporter ion for a protein was calculated based on the sum of all detected reporter ions of associated peptides from that protein. The ratios between an abundance of a reporter and the average abundance of all reporters were used to estimate the abundance ratio of each protein.

For statistical analysis, the sum of reporter ion intensities for each protein was Log2 transformed and the biological duplicate and technical triplicate measurement for each protein was averaged. Only the proteins that were identified and quantified in both biological and technical replicates were used for the analysis. Student t-test was performed to identify proteins with changed abundance. Proteins with value of p <0.05 were further evaluated for increased and decreased abundance using a cut-off value of ≥ ± 1 fold (Log2) change.

Upon return to Earth single plates from both ground- and ISS-kept BRICs were used to collect agar plugs in triplicate for secondary metabolite (SM) extraction. Plugs were extracted with 3 ml of methanol and 1:1 methanol/dichloromethane each followed by 1 h sonication. Crude extract was evaporated in vacuo to yield a residue that was then suspended in 1 ml of 20% dimethyl sulfoxide/methanol and 10 μl was examined by high-performance liquid chromatography-photodiode array detection-mass spectrometry (HPLC-DAD-MS) analysis. HPLC–MS was carried out using ThermoFinnigan LCQ Advantage ion trap mass spectrometer with an RP C18 column (Alltech Prevail C18 3 mm 2.1 × 100 mm) at a flow rate 125 μl/min. The solvent gradient for LC–MS was 95% acetonitrile/H2O (solvent B) in 5% acetonitrile/H2O (solvent A) both containing 0.05% formic acid, as follows: 0% solvent B from 0 min to 5 min, 0 to 100% solvent B from 5 min to 35 min, 100% solvent B from 35 min to 40 min, 100 to 0% solvent B from 40 min to 45 min, and re-equilibration with 0% solvent B from 45 min to 50 min. The SM profiles of ISS- and ground-grown JSC-093350089, which were regrown at 28°C were obtained following the procedure described above.

To compare the yields of produced SMs in ISS-grown, ground-grown, and regrown samples, the area under the electrospray ionization curve (ESI) was integrated for each compound. SM data collected from three independent biological replicates of ISS- and ground-grown, and regrown JSC-093350089 were used for testing statistical significance of production yields of identified SMs by Welch’s corrected t-test. The data are presented as column charts with corresponding error bars. Data analysis was conducted using GraphPad Prism version 7.

The genomes of ISS-flown and ground-grown JSC-093350089 were compared to identify occurring genetic variations. Obtained reads were aligned to the CBS 513.88 reference genome and single-nucleotide polymorphisms (SNPs) present in the ground control were filtered. This revealed presence of 375 SNPs and 620 INDELs that occurred because of the exposure to conditions aboard the ISS (Table 1). All identified genetic variations are summarized in Supplementary Tables 1 and 2, presenting SNPs and INDELs, respectively. Distribution of non-synonymous point mutations among genes is presented in Table 2. Interestingly, about 80% of these mutations occurred in chromosome VIII and 13% occurred in chromosome XII, while the remaining 7% were distributed among other chromosomes (Supplementary Figure 2A). The majority of missense point mutations (75%) were observed within genes of unknown function. However, several characterized genes containing missense SNPs have DNA-binding activity (An06g01180, An08g11890, and An12g00840), DNA polymerase activity (An08g11520), protein kinase and transferase activity (An08g12110), phospholipase activity (An08g12250), and chromosome anchoring RacA protein binding activity (An12g06420; Table 2). Additional, mutations like one stop lost, one start gained and one 5 prime untranslated region (UTR) mutation were observed. Most of the observed SNPs (~55%) and INDELs (71%) were located in intergenic regions (Table 1). Interestingly, unlike SNPs, INDELs were distributed throughout all chromosomes. However, similarly to SNPs, the highest number of INDELS was found in chromosomes VIII and XII. Among observed INDELS 109 caused frameshift, 14 lead to disruptive inframe deletion, and a few caused start lost and stop gained (Supplementary Table 2; Supplementary Figure 2B).

Differentially expressed proteins in ISS-grown JSC-093350089 strain were investigated following the extraction of total protein from two biological replicates of ISS-grown and ground control counterpart strains. Due to the low yields of extracted proteins, biological replicates were combined and divided into two parts that were then TMT labeled and subtracted to analysis via LC–MS/MS followed by spectrum/sequence matching using A. niger CBS 513.88 protein database (NCBI). Protein abundance ratios in ISS-grown JSC-093350089 were normalized to Earth-grown counterparts, which enabled identification of 70 up- and 142 downregulated proteins (fold change (FC) > ∣2∣, p < 0.05) in response to space conditions (Supplementary Tables 3 and 4, respectively). AspGD Gene Ontology (GO) Slim terms (Cerqueira et al., 2014) were used to study the distribution of differentially expressed proteins in ISS-grown JSC-093350089 when compared to ground controls (Figure 1). Among differentially expressed proteins, 29 were involved in carbohydrate metabolism, 18 in stress responses, 21 in amino acid metabolism, and 15 in protein catabolic processes (Figure 1). Protein GO term enrichment analysis was conducted using FungiDB (Stajich et al., 2012), which revealed that significantly over-represented upregulated biological processes included carbohydrate metabolic processes (28% of all upregulated proteins) and stress response (10%), whereas significantly over-represented downregulated processes included cellular amino acid metabolic processes (13%), proteasomal ubiquitin-independent (10%) and dependent processes (10%), and proteasomal protein catabolic processes (10%; Supplementary Table 5).

The majority of differentially expressed proteins in ISS-grown JSC-093350089 A. niger were involved in carbohydrate metabolism (Table 3). Interestingly, eight of these genes, including cellobiohydrolases A and B (An07g09330 and An01g11660), XlnA 1,4-β-xylanase (An03g00940), and D-xylose reductases XyrA and XdhA (An01g03740 and An12g00030) were regulated by XlnR. XlnR is a transcriptional regulator involved in degradation of polysaccharides, xylan, cellulose, and D-xylose (Hasper et al., 2000). β-glucanases An11g01540 and An02g00850, which are involved in carbon starvation response in A. niger (Nitsche et al., 2012), were at minimum 3-fold upregulated in ISS-grown strain. α-1,2-mannosidases An08g08370, An13g01260 were at least 3.5-fold upregulated, whereas pyruvate decarboxylase PdcA (An02g06820) was nearly 3-fold upregulated. Pyruvate kinase PkiA (An07g08990), pyruvate dehydrogenase Pda1 (An07g09530), and isocitrate lyase AcuD (An01g09270) were at least 2-fold less abundant in ISS-grown samples. Several proteins involved in the stress response were differentially expressed in ISS-grown JSC-093350089 (Table 4). Proteins exhibiting at least 2-fold upregulation included cell wall organization protein EcmA (An04g01230) and An16g07920, whose orthologs play a role in salt stress response. Downregulated stress response proteins included heat shock protein An06g01610, DNA-binding protein HttA (An11g11300), and quinone reductase An12g06300. Lastly, a variety of proteins involved in cellular amino acid processes (Table 5), and protein catabolic processes (Table 6) were downregulated.

Alterations in SM production in response to ISS conditions were assessed by extracting organic compounds from three biological replicates of ISS- and ground-grown JSC-093350089. The extracts were analyzed using HPLC-DAD-MS. Compounds were identified based on mass, UV absorption, and retention time, which were in agreement with literature (Chiang et al., 2011; Figure 2). Assessment of SM production yields revealed slight decrease in production of bicoumanigrin A, fonsecinones B and C, aurasperones C and B, and kotanin in ISS-grown isolate. Production levels of pestalamide B, nigerazine B, and nigragillin exhibited a reduction of approximately 50% when compared to control strains.

Both ISS- and ground-grown JSC-093350089 were regrown at 28°C and SMs production was evaluated. Due to the difference in growth temperature, obtained SM profiles of regrown isolates differed from those grown on the ISS (~22°C; Figure 3), as the trend in the production of SMs was the opposite of the one observed immediately after growth on the ISS. The majority of identified compounds showed slight increase in the production levels. Further, statistically significant increase in the production of the antioxidant pyranonigrin A (Miyake et al., 2007) and kotanin was observed when compared to ground controls.