The evolution of superbugs in space: a genomic perspective on pathogens in the International Space Station environment

Graphical abstract

Highlights

• •Hypervirulent, antibiotic-resistant bacteria in space station pose risks to crew.
• •E. ludwigii and E. cancerogenus from ISS show elevated defense-related gene profiles.
• •Some ISS genomes possess slightly higher ARGs compared to their Earth counterparts.
• •VFs linked to metal uptake and secretion are found exclusively in certain ISS strains.
• •Space borne strains suggest functional adaptations to stressful ISS conditions.

Abstract

Microgravity, pressure, and temperature variations in the International Space Station (ISS) create conditions leading to the emergence of superbugs. Due to technical issues in spacecraft, astronauts are forced to stay in ISS for extended periods; prolonged stay and exposure in stressful ISS environment weakens their immune systems, increasing susceptibility to infections. The presence of hypervirulent and antibiotic-resistant pathogens in space station is a worrisome feature as these might cause serious life-threatening infections in astronauts staying in high stress environments with weakened immune systems. In the present study, we compared antimicrobial resistance genes (ARGs) and virulence factors (VFs) in bacterial genomes from ISS with Earth counterparts. ISS genomes exhibited elevated counts of defense-related genes, particularly in E. ludwigii and E. cancerogenus. Among genes uniquely found in ISS genomes, CRISPR-Cas system components were notably prevalent. Though Earth genomes harbored higher number of ARGs overall, several species from ISS possessed modestly higher ARG counts. VFs profiling showed a slightly lower count in ISS genomes, but P. conspicua, E. ludwigii, and K. pneumoniae from ISS carried exclusive VFs linked to metal ion uptake and secretion systems, suggesting environment-driven functional adaptations. The adaptation of pathogenic bacteria in ISS is alarming and therefore periodic monitoring of bacterial genomic surveillance is important. Our findings shed light on genomic profiles in bacterial strains from both ISS and Earth, enhancing our understanding of the bacterial pathogens’ potential impact on drug resistance and pathogenicity in space-missions and the possible threat of spread from ISS.

1. Introduction

The International Space Station (ISS), orbiting in low Earth orbit, serves as a human-made space habitat that presents a unique environment, distinctly different from Earth, due to its increased radiation exposure and microgravity conditions.¹ Unlike earlier mission that lasted only a few days, current stays on ISS extend over several months. The progress in human space exploration has paved the way for extended missions to the Moon and Mars. However, a significant concern associated with these long-duration missions is the potential rise in virulence and antimicrobial resistance (AMR) among opportunistic pathogens, which could heighten the risk of infections and complicate treatment for astronauts during space expeditions.²

The ISS is often described as an “extreme environment” due to factors such as microgravity, cosmic radiation, and elevated carbon dioxide levels.² These conditions have profound effects on both astronauts and the microbial communities within the ISS, as human immune systems are weakened, and microorganisms adapt to survive in these harsh surroundings. The challenging environment of the ISS significantly affects astronauts by disrupting the bacterial microflora in their gastrointestinal, nasal, and respiratory systems, thereby heightening their vulnerability to infections.1, 3 Research has shown that some microorganisms aboard the ISS can evolve into highly virulent pathogens with resistance to multiple antibiotic classes, which is a significant concern.²

The dominant microorganisms within the ISS predominantly belong to the genera Staphylococcus, Corynebacterium, Bacillus, Propionibacterium, and Micrococcus.4, 5 These microorganisms are largely introduced into the ISS by astronauts and cosmonauts, who serves as hosts to a wide range of microbial species.⁴ A study demonstrated that all tested strains of Enterobacter bugandensis, Acinetobacter pittii, Klebsiella quasipneumoniae, Pantoea conspicua, along with other species within the Enterobacteriaceae family of spacecraft isolates, exhibited resistance to the majority of the antibiotics tested; notably, they showed complete (100%) resistance to oxacillin, penicillin, and rifampin.⁶ A study by Tierney et al. found that the genomes of Acinetobacter pittii contain LexA, a transcriptional regulator that aids in their survival in harsh environments and increases their mutation rates.⁷ Other research has also documented the emergence of AMR in microorganisms isolated form ISS. These findings underscore the importance of ongoing surveillance of the microbiome from the ISS.

In this research, we attempted to compare the functional gene clusters and genes associated with resistance and virulence in the genomes of various bacterial species from the ISS with their Earth counterparts. Our investigation into AMR determinants can not only contribute to improving surveillance but also design mitigation strategies to combat potential MDR pathogen contamination from inbound spaceflights after extended space missions.

2. Materials and methods

2.1. Data retrieval of ISS bacterial whole genome sequences

Whole-genome sequences (WGS) of the selected bacterial organisms were obtained from NASA-Open Science Data Repository (NASA-OSDR) and the NCBI Sequence Read Archive (SRA). Both raw sequence reads and genome assemblies of the ISS origin collected were part of several key studies, including “A Molecular Genetic Basis Explaining Altered Bacterial Behavior in Space”, “Microbial Observatory (ISS-MO): Study of BSL-2 Bacterial Isolates from the International Space Station”, “ISS Enterobacteriales”, and “ISS Enterobacteriales Genomes (NCBI-SRA)”.⁸

2.2. Average Nucleotide Identity screening of Earth counterparts

Clinical and environmental genome sequence counterparts were retrieved from the NCBI repository to provide a comprehensive comparative analysis. Genome dataset for each species were curated by filtering out atypical genomes, metagenome-assembled genomes (MAGs), and large multi-isolate project genomes. Further, the dataset was curated by retaining the genomes of chromosomal and complete assembly level deposited between 2011–2025. The closest Earth counterparts to the ISS-derived genomes were identified by Average Nucleotide Identity (ANI)-based screening using FastANI v1.34.⁹ Each ISS genome was used as a query against the curated set of genomes retrieved from NCBI GenBank database. Earth genome with the highest ANI similarity score was selected as the closest counterpart.

2.3. Genome assembly and quality assessment

The raw reads of isolates from the ISS Enterobacteriales Genomes SRA study required quality control (QC) and preprocessing, while the remaining were genome assemblies. The raw sequence data were initially evaluated for quality using FastQC v0.12.0 to detect potential issues such as adapter contamination, base composition anomalies, and sequence duplication levels.¹⁰ These sequences were pre-processed using Trimmomatic v0.39 (with a quality score cut off of Q30), involving adapter removal, quality filtering, and trimming of low-quality sequences.¹¹ De novo genome assembly of the pre-processed sequences of the 'ISS Enterobacteriales Genomes SRA' raw data was performed using the SPAdes genome assembler v3.15.5.12, 13 The other ISS genomes retrieved as assemblies were proceeded directly to quality assessment. Quality assessment of both ISS and Earth genome assemblies (n = 44) was conducted using QUAST v5.3.0 and BUSCO (Benchmarking Universal Single-Copy Orthologs) v5.8.2 tools to evaluate the contiguity and completeness, respectively.14, 15, 16

2.4. Gene prediction, genome mapping, and functional annotation

Gene prediction and structural annotation of the genomes was performed using Prokka v1.14.6 to identify genomic features.¹⁷ Circular genome maps for the ISS genomes were constructed and visualized using Artemis DNAPlotter v18.2.0.¹⁸ The Cluster of Orthologous Genes (COGs) were identified using COGclassifier v2.0 to further classify the functional categories of annotated genes and to understand their putative functions on various COG categories.

2.5. Phylogenetic tree construction

The 16S rRNA sequences were extracted from FFN files of Prokka annotation. The concatenated 16S rRNA sequences of all ISS and Earth genomes were subjected to multiple sequence alignment (MSA) using MAFFT v7.525 tool with G-INS-i (Needleman-Wunsch algorithm) iterative refinement method.¹⁹ The resulting alignment was used for reconstructing the phylogenetic tree using IQ-TREE v 3.0.1.²⁰ The tree was provided with branch support values using the Ultrafast Bootstrap Approximation (UFBoot) and the SH-like approximate likelihood ratio test (SH-aLRT) of 1000 iterations.²¹ Serratia marcescens was used as the outgroup. ModelFinder, the in-built model-selection method in IQ-TREE, chose TN+F+I+R2 model as the best-fit model based on the Bayesian Information Criterion (BIC) score. Phylogenetic tree was visualized and annotated using Interactive Tree of Life (iTOL) v7.²²

2.6. Identification of resistance and virulence determinants

Antimicrobial resistance genes (ARGs) and virulence factors (VFs) were identified using ABRicate v1.0.1. ARGs were identified by aligning the query sequences against Comprehensive Antibiotic Resistance Database (CARD), using a minimum threshold of 80% identity and 80% coverage. This involved predicting the presence of resistance genes and categorizing them based on the drug classes affected and resistance mechanisms. Similarly, VFs were identified by aligning the genomes against the virulence factor database (VFDB), with the same threshold criteria of 80% identity and 80% coverage.23, 24

3. Results

3.1. Genome dataset and quality assessment

The dataset for this study comprised 44 genomes, including the following species: Acinetobacter pittii, Escherichia coli, Enterobacter bugandensis, Enterobacter cancerogenus, Enterobacter ludwigii, Klebsiella aerogenes, Klebsiella pneumoniae, Klebsiella quasipneumoniae, Kalamiella piersonii, Pantoea conspicua, Staphylococcus aureus, Staphylococcus haemolyticus, and Staphylococcus hominis. The dataset comprised of two groups based on the source of the genomes: ISS (n = 22) and Earth (n = 22). The ISS isolates originated from various locations within the space station, including the Permanent Multipurpose Module (PMM), the Waste and Hygiene Compartment (WHC), the Advanced Resistive Exercise Device (ARED) foot platform, and the port crew quarters' bump-out exterior aft wall. Metadata of the ISS genomes including type of strains, sampling methods, storage conditions, and other relevant details, is provided in Table S1a. ANI scores and the metadata including source (clinical/environmental), host, and geographic location of the Earth counterpart genomes considered for this study through ANI screening were summarized in Table S1b.

Evaluation of assembly quality using QUAST revealed that all ISS genomes were of single contig, reflecting high contiguity. Among the Earth counterparts, total number of contigs ranged from 1 to 5, except for P. conspicua. GC content of A. pittii genomes from both environments was approximately 39%, while Pantoea genera genomes averaged 56.4%. Enterobacter and Klebsiella genera showed a mean GC content of 56.3%, while E. coli had ∼51%. Staphylococcus genomes exhibited an average GC content of 32.7% (Fig. S1). N50 and L50 values were within the optimal ranges for both Earth and ISS genomes, except for Earth P. conspicua genome. Details of all the quality metrics are provided in the Table S2a. Quantitative assessment of genomes using BUSCO revealed that completeness score of Earth assemblies ranged from 92.6% to 99.8%, with an average of 97.26%. ISS genome assemblies showed completeness ranging from 93.6% to 99.9%, with an average of 97.64%. The metrics including single-copy, duplicated, fragmented, and missing BUSCOs were tabulated in Table S2b.

3.2. Gene prediction and genome mapping

The statistics of the number of coding sequences (CDS), genes, CRISPR, miscellaneous RNA, ribosomal RNA, repeat regions, transfer RNA, and transfer-messenger RNA are provided in the Table S3. Total number of genes in genomes of A. pittii, Enterobacter and Pantoea genera from both ISS and Earth ranged from 3,935 to 7,077, with an average of 5,010 genes. The number of CDS in these genomes ranged from 3,824 to 6,786, with an average of 4,777. In the Staphylococcus genomes, the total number genes ranged from 2,357 to 3,158, averaging 2,787. Total CDS varied from 2,197 to 2,957, with a mean of 2,605. The circular genome maps showing the gene features of the ISS genomes, including CDS and RNA genes are depicted in Fig. S2a-f.

3.3. Functional gene categories

Gene counts associated with the functional COG categories of both ISS and Earth genomes are summarized in Table S4. The most prominent functional subsystems, determined by average gene count, were amino acid transport and metabolism (COG E, 360 genes), carbohydrate transport and metabolism (COG G, 335 genes), transcription (COG K, 286 genes), translation, ribosomal structure and biogenesis (COG J, 258 genes), and cell wall/membrane/envelope biogenesis (COG M, 240 genes). Subcategory-specific distribution of COGs across the species were studied and average gene counts were considered for species with multiple genomes.

Several species namely, E. ludwigii (114 genes) and E. cancerogenus (112 genes) exhibited a higher number of genes for defense mechanisms (COG V) in ISS genomes compared to their Earth counterparts (Fig. 1a). Notably, E. ludwigii from ISS revealed an elevated number of defense-associated genes (ata, cas1, cas3f, cas5f, cas6f, cas7f, cas8f, csa3, gmrSD, hicB, higA, hsdM, mazE, mazF, mcrB, mcrC, PP2433, relE, res, tad, and vapI), which were not present in the Earth genome. In case of inorganic ion transport and metabolism (COG P), the ISS genomes of P. conspicua (248 genes), E. ludwigii (242 genes), and E. coli (242 genes) exhibited higher gene counts than their Earth counterparts (Fig. 1b). In the COG J subcategory, K. quasipneumoniae from the ISS had an average of 313 genes, which was higher compared to the Earth genomes. Similarly, S. hominis (242 genes) and E. cancerogenus (291 genes) showed higher gene counts in ISS genomes (Fig. 1c).

Fig. 1.

Column plots showing COG counts in Earth and ISS genomes for the functional subcategories of (a) defense mechanisms (COG V), (b) inorganic ion transport and metabolism (COG P), (c) translation, ribosomal structure and biogenesis (COG J), (d) replication, recombination and repair (COG L), (e) signal transduction mechanisms (COG T), and (f) posttranslational modification, protein turnover, chaperones (COG O). ISS genomes with higher COG counts compared to Earth counterparts in each category are marked with red boxes. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

Among all species analyzed, S. hominis (123 genes), P. conspicua (173 genes), and S. aureus (116 genes) from the ISS showed higher COG counts for COG L subcategory compared to their Earth counterparts (Fig. 1d). In signal transduction mechanisms subcategory (COG T), E. cancerogenus (202 genes), E. coli (181 genes), and P. conspicua (179 genes) exhibited elevated gene counts in the ISS environment (Fig. 1e). In the posttranslational modification, protein turnover, chaperones subcategory (COG O), P. conspicua (151 genes) and S. hominis (87 genes) showed a high average COG count in ISS environment (Fig. 1f).

3.4. Species-specific functional gene profile

Further analysis of the functional annotation revealed several unique genes that were specific to bacterial genomes from ISS, distributed across diverse COG functional categories. In A. pitti, 23 unique genes were identified, covering 13 COG categories, primarily associated with the subcategory mobilome: prophages, transposons (COG X), as well as COG G and COG M. E. bugandensis harbored 41 unique genes, predominantly associated with COG X, including genes such as beeE, gp18, gp36, gpD, gpG, gpI, gpV, gpX, lmaA, PA5055, pri-rep1, socA, xtmA, ymfN. E. cancerogenus exhibited 152 unique genes across 21 categories, with the highest being associated with general function prediction (COG R), COG V, and COG G. E. ludwigii had 95 unique genes, predominantly involved in COG V, COG X, COG R and COG M. In E. coli, 21 unique genes were identified, mainly linked to COG V, COG G, and COG T.

A total of 12 unique genes in K. aerogenes were mainly related to COG P and COG L. For K. pneumoniae, 23 unique genes were identified, primarily assigned to intracellular trafficking, secretion, and vesicular transport (COG U), COG X, and COG K. Among the 39 unique genes found in K. piersonii, many were involved in COG X and included genes namely, gp18, gpA1, gpG, insA, insB, insO, IS5, jayE, rve, xtmA, ykfC. A high number of 111 unique genes were identified in P. conspicua, mainly linked to COG X, COG J, and COG U. Among the 37 unique genes identified in S. haemolyticus, most of them were associated with COG L, COG M, and COG X. The genome of S. hominis had 31 unique genes, largely associated with mobilome including genes namely, IS5, rep, socA, xtmA, ykfC, yqbO. Genes identified as specific to ISS genomes are tabulated in Table S5a-k.

3.5. Phylogenetic analysis

The phylogenetic tree based on the 16S rRNA sequence revealed distinct clustering patterns that largely align with genus-level taxonomy. Within the Staphylococcus clade, numerous 16S rRNA sequences of S. hominis, S. haemolyticus, and S. aureus from the ISS formed a coherent cluster with other ISS and terrestrial strains, implying potential clonal expansion and persistence in the space environment. Similarly, ISS-derived K. quasipneumoniae and K. pneumoniae strains clustered closely with terrestrial genomes, reflecting a higher degree of similarity. In contrast, the Enterobacter and Pantoea clades displayed more diffuse clustering. Particularly, ISS strains such as E. cancerogenus (IF2SW-B1), E. ludwigii (IF2SW-P2), and P. conspicua (IF5SW-P1) were more phylogenetically distant from their Earth counterparts possibly due to the lower ANI scores (87–91%). This divergence may indicate either longer evolutionary separation or adaptive genomic changes following exposure to the unique selective pressures aboard the ISS (Fig. 2).

Fig. 2.

Phylogenetic tree of bacterial 16S rRNA sequences from Earth and ISS groups, with Earth strains presented in green font and ISS strains presented in blue font. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

3.6. ARG profiling

Resistome profiling of the genomes revealed the distribution of resistance genes in the ISS and Earth genomes. A total of 815 ARGs were identified across all genomes, encompassing 144 unique genes (Table S6). The Earth genomes accounted for 435 ARGs, comprising of 136 unique genes, whereas ISS genomes harbored 380 genes, including 81 unique genes. Gene distribution of the identified ARGs in the dataset is shown in Fig. 3. The most predominant ARGs across most genomes in both environments were those encoding CRP (cAMP receptor protein) and OmpA (outer membrane protein A). Other frequently occurring genes included oqxB, hns, acrB, cpxA, emrR, marA, and msbA.

Fig. 3.

Heatmap illustrating the distribution profiles of identified antimicrobial resistance genes (ARGs) across bacterial genomes from Earth (green font) and ISS (blue font). The X-axis represents the identified resistance genes, while the Y-axis denotes the corresponding bacterial genomes. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

Analysis of ARGs classified based on drug classes revealed that the most commonly affected antibiotic classes in both Earth and ISS genomes were fluoroquinolones, beta-lactams (including penicillin and cephalosporins), and tetracyclines. Notably, ARGs conferring resistance to fluoroquinolone were highly prevalent in both the Earth (13.4%) and ISS (13.7%) environments, suggesting strong selective pressure for this class (Fig. 4a and 4b). Classification of ARGs based on resistance mechanisms revealed antibiotic efflux as the most dominant strategy in both Earth (278 ARGs) and ISS (258 ARGs) genomes, accounting for more than half of all mechanisms identified. This was followed by antibiotic inactivation, with 70 and 47 ARGs identified in Earth and ISS genomes, respectively, contributing to this mechanism. Other mechanisms such as target alteration and reduced permeability were similarly represented in both environments (Fig. 4c and 4d).

Fig. 4.

Depicts the distribution of ARGs categorized based on drug classes affected and resistance mechanisms (a) Distribution of ARGs based on antibiotic classes in Earth genomes (b) Distribution of ARGs based on antibiotic classes in ISS genomes, (c) Proportion of resistance mechanisms exhibited by ARGs in Earth genomes, and (d) Proportion of resistance mechanisms exhibited by ARGs in ISS genomes.

3.7. Species-wise comparison of ARG profiles

Species-level comparison of ARG distribution in Earth and ISS genomes shown distinct patterns of resistance distribution. In general, Earth genomes exhibited a higher ARG count compared to their ISS counterparts. Particularly, K. pneumoniae, A. pittii, and E. coli harbored substantially higher ARG counts in Earth genomes. Conversely, S. aureus exhibited slightly higher ARG burden in ISS genomes compared to Earth counterparts. Notably, ISS-derived S. aureus genomes carried PC1, ant(9)-Ia, and ermA genes, which were absent in Earth counterparts. For K. aerogenes, both Earth and ISS genomes comprised of an equal number of ARGs, with an identical gene composition, indicating a conserved resistome across the environments. However, several species shown enrichment of specific ARGs in ISS-derived genomes. E. cancerogenus from the ISS harbored ACT-146, TolC, oqxA, and oqxB genes, which were absent in the Earth counterpart. Similarly, ISS genome of P. conspicua carried the kpnH gene, and K. piersonii from the ISS uniquely possessed emrB gene, none of which were detected in the Earth genomes.

Upon comparing resistance by drug class, ISS-derived genomes of E. cancerogenus, K. piersonii, and P. conspicua exhibited a slightly higher prevalence of genes conferring resistance to fluoroquinolones. Increases in macrolide resistance genes were also observed in ISS-derived S. aureus, P. conspicua, K. quasipneumoniae, and E. cancerogenus. Similarly, resistance genes associated with penicillin beta-lactams were marginally more abundant in ISS genomes of S. aureus, P. conspicua, and E. cancerogenus. Resistance to peptide antibiotics was more in ISS-derived P. conspicua and E. cancerogenus. Aminoglycoside resistance genes were enriched in K. quasipneumoniae and E. cancerogenus from the ISS. Furthermore, ISS genomes of E. ludwigii and E. cancerogenus demonstrated elevated ARGs conferring resistance to glycylcyclines, tetracyclines, phenicols, and rifamycins. Of particular note, the ISS-derived E. cancerogenus genome harbored additional ARGs associated with resistance to other drug classes as well, including aminocoumarins, cephalosporins, carbapenems, and disinfectants/antiseptics, suggesting enhanced multidrug resistance potential in the microgravity environment. It is observed that ARGs responsible for both antibiotic efflux and reduced permeability to antibiotic were found only in the genomes of Enterobacteriaceae family of both the environments.

3.8. VF profiling

The observation from VF profile comparison were interesting and significant. A total of 2,234 VFs were identified across the genomes, with 334 unique virulence genes (Table S7). Of these, the Earth genomes accounted for 1,170 VFs, comprising all 334 unique genes, while the ISS genomes contained 1,064 VFs, representing 284 unique genes. A. pittii, K. pneumoniae, and S. aureus genomes from both environments exhibited a greater number of VFs compared to the other species included in the study. Heatmaps illustrating the distribution of identified VFs across the ISS and Earth genomes classified based on the virulence mechanisms were depicted in Fig. 5, Fig. 6, Fig. 7. The most commonly detected VFs across both environments included ompA, fur, rcsB, rpoS, gndA, acrB, and entA, which are associated with membrane integrity, regulatory responses, stress tolerance, and nutrient acquisition.

Fig. 5.

Heatmap illustrating the virulence factors (VFs) across bacterial genomes from Earth (green font) and ISS (blue font), X-axis representing virulence genes and Y-axis denoting the genomes. The virulence genes represented are responsible for the mechanism of host interaction and colonization (adherence, biofilm, motility, invasion). (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

Fig. 6.

Heatmap illustrating the virulence factors (VFs) across bacterial genomes from Earth (green font) and ISS (blue font), X-axis representing virulence genes and Y-axis denoting the genomes. The virulence genes represented are responsible for the mechanism of host damage and immune modulation (exotoxin, exoenzyme, effector delivery system, immune modulation). (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

Fig. 7.

Heatmap illustrating the virulence factors (VFs) across bacterial genomes from Earth (green font) and ISS (blue font), X-axis representing virulence genes and Y-axis denoting the genomes. The virulence genes represented are responsible for the mechanism of survival and fitness factors (antimicrobial activity/competitive advantage, nutritional/metabolic factor, regulation). (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

Although the overall VF count was higher in Earth-derived genomes, several ISS genomes exhibited a modest but consistent increase in VF abundance relative to their Earth counterparts. In particular, ISS-derived genomes of P. conspicua, E. ludwigii, and K. pneumoniae demonstrated slightly elevated VF profiles, suggesting potential environmental adaptation or selection.

Functional categorization of VFs revealed variation in the distribution of certain virulence mechanisms in ISS genomes. Exotoxin production, biofilm formation, exoenzyme activity, regulatory functions, and antimicrobial activity/competitive advantage mechanisms exhibited a proportionally higher representation among total ISS-associated virulence genes compared to those from Earth genomes (Fig. 8). While these proportional increases do not directly imply increased virulence, they underscore the importance of monitoring microbial functional shifts during extended spaceflight missions.

Fig. 8.

Bar plot illustrating the distribution of virulence factors (VFs) categorized based on the virulence mechanisms in Earth and ISS genomes.

3.9. Species-wise comparison of VF profiles

A species-level comparison of VFs between Earth and ISS genomes revealed notable differences in VF composition, suggesting niche-specific adaptations. ISS genome of P. conspicua harbored three unique VFs, hcp/tssD, tssG, and vipB/tssC, associated with Type VI secretion system (T6SS) components, which were absent in the Earth counterpart. Similarly, E. ludwigii genome from ISS origin carried unique VFs including acrA, entS, fepG, phoP, and tssF, indicating enhanced efflux, siderophore transport, and regulatory capabilities. K. pneumoniae ISS genome exhibited a set of 11 exclusive VFs, fyuA/psn, irp1, irp2, ybtA, ybtE, ybtP, ybtQ, ybtS, ybtT, ybtU, and ybtX, all of which are components of the yersiniabactin siderophore system, suggesting enhanced metal ion acquisition potential in the ISS environment. These findings indicate a potentially enhanced capacity for persistence, immune modulation, and environmental adaptation in ISS conditions.

4. Discussion

The disparities between terrestrial and ISS environments might engender a slow but steady shift in the selection and prevalence of pathogenic determinants, especially of AMR and virulence. Temperature, nutrient availability, high antibiotic consumption, sanitation practices, and microbial interactions influence the resistome and virulence profiles in the environmental as well as clinical pathogens.25, 26, 27

The comparative analysis of ARGs between ISS and Earth-derived bacterial genomes revealed a complex landscape of resistance adaptation shaped by environmental pressures. While Earth genomes exhibited a greater total number of ARGs and a higher number of unique genes, ISS genomes displayed notable similarities in resistance profiles and mechanisms. The predominance of efflux pump-mediated resistance in both groups aligns with global trends in resistance evolution, reflecting a conserved survival strategy across diverse conditions. Besides, there exist predominant β-lactam resistance uniform to both the counterparts. The prominent β-lactam resistance genes in our work corroborate to a previous metagenomics study of ISS isolates.²⁸

The controlled environment in a spacecraft presents unique stressors, including high-energy ionizing radiation, the near-perfect vacuum, microgravity, and nutrient limitations which profoundly affect the physiology and genetic evolution of the microorganisms. The high levels of radiation in space can induce mutations in the microbial genomes, potentially driving the rise of enhanced antibiotic resistance traits. Microgravity alters the growth dynamics and bacterial interactions, which can also drive unique adaptive mechanisms among ISS isolates. Limited nutrient availability in space further compounds these challenges, forcing organisms to develop survival strategies that may include different resistance mechanisms when compared to their Earth counterparts.29, 30 In the current study, VF profiling further emphasized the functional divergence of space borne microbes. Despite Earth genomes harboring a larger total number of VFs, certain ISS genomes, particularly those of P. conspicua, E. ludwigii, and K. pneumoniae, carried unique virulence determinants.³¹ These included genes involved in metal ion acquisition and secretion systems, indicating a shift towards attaining an extra-edge towards survival and propagation, which is distinct from the earth pathotypes. While these differences do not equate to increased pathogenic potential, they underscore the importance of monitoring functional shifts in microbial traits during extended space missions.

Functional profiling of the genomes revealed that E. ludwigii and E. cancerogenus from ISS exhibited noticeably higher numbers of genes associated with defense mechanisms compared to their Earth counterparts. Notably, the ISS-derived E. ludwigii harbored 21 unique genes, predominantly related to CRISPR-Cas systems, which were absent in terrestrial genomes. Similarly, E. cancerogenus from the ISS had 17 defense-related genes that were not identified in the genome of Earth. Furthermore, ISS-derived genomes of E. bugandensis, E. ludwigii, K. piersonii, P. conspicua displayed an elevated number of genes associated with the mobilome, with each genome containing over 10 unique genes not present in their Earth-derived strains.³²

The genes gp18 and gp36, identified as phage tail proteins and Mu-like prophage proteins, respectively, were detected in the majority of the ISS-derived Gram-negative species, including A. pittii, E. bugandensis, E. cancerogenus, E. ludwigii, and P. conspicua. These genes belong to functional category COG X, which includes mobilome elements such as prophages and transposons. Their distribution across diverse taxa suggests a common genetic feature associated with prophages within these microbial populations. A previous study showed that prophage regions in ISS-associated microbes encode traits that enhance survival under extreme environments, including AMR, increased virulence, and dormancy.³³ This highlights the potential role of these phage-related elements in microbial adaptation to spaceflight conditions.

These insights are key for developing effective strategies to maintain a secure environment for the astronauts, including measures for sterilization, microbial surveillance, and AMR mitigation. To contextualize our findings, the broader implications of how terrestrial microbes adapt and thrive in the extreme conditions of space-bound vehicular environment should be explored. Previous research has shown that spaceflight conditions can induce physiological changes in bacteria, including altered growth rates, biofilm formation, and resistance to antibiotics.1, 34 Besides, the potential dissemination of resistance and virulence traits through HGT and the role of mobile genetic elements (MGEs) in confined environments like the ISS requires more in-depth exploration.

The synergistic impacts of abiotic factors such as temperature and nutrient availability alongside acute stressors of the space environment, contribute to the distinct adaptations of pathogenic bacteria at ISS. ​Thus, understanding the evolution and distribution of the microbial communities at ISS is crucial, since it will guide mitigation strategies against potential outbreaks in long-term space missions, and any unforeseen future risks of AMR contaminations from inbound space passengers.

Though our study provides insights into the virulence and resistance determinants in the ISS-originated bacterial genomes, it has some limitations. The relatively small sample size of ISS genomes may limit the generalizability of our findings to the broader diversity of microbial species in the space environment. However, to capture the genomic diversity of Earth genomes, we employed an ANI-based screening approach to select representative strains closely related to the ISS genomes, enabling a focused yet meaningful comparative analysis. Additionally, while our study provides a genomic comparison, it does not delve into the functional mechanisms that may underlie microbial adaptation to space conditions. Future studies incorporating comprehensive pan-genome analyses, along with transcriptomic or phenotypic assessments, will be crucial to uncover the core and accessory genomic features and their potential functional roles in microbial survival and evolution in space.

5. Conclusion

In conclusion, the genomic analysis of microorganisms from ISS revealed a significant presence of resistance genes, conferring resistance to various antibiotic classes, and hyper-virulence factors that may enhance their pathogenicity. The adaptation of certain ISS bacterial species, with increases in genes associated with defense mechanisms highlights the unique and stressful conditions that promote their survival in space. Our research highlights the need for periodic monitoring of virulence and AMR posed by microbial communities in space station and it is pertinent. By deepening our understanding of microbial adaptations to extreme environments in space station, we can better equip ourselves to address future challenges in human space exploration and to prevent the emergence of hypervirulent antibiotic resistance strains and their spread.

CRediT authorship contribution statement

Sara Pearl: Writing – original draft, Methodology, Formal analysis, Data curation. Hithesh Kumar: Writing – original draft, Methodology, Formal analysis, Data curation. Santhiya Vijayakumar: Writing – original draft, Methodology, Formal analysis, Data curation. Soumya Basu: Writing – original draft, Methodology, Formal analysis, Data curation. Sudha Ramaiah: Writing – review & editing, Validation, Supervision, Funding acquisition. Anand Anbarasu: Writing – review & editing, Validation, Project administration, Funding acquisition, Conceptualization.

Funding

This work was supported by the Indian Council of Medical Research (ICMR), New Delhi [IRIS ID: 2021-11889; AMR/Adhoc/290/2022-ECD-II].

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Appendix A. Supplementary data

The following are the Supplementary data to this article:

Data availability

The data supporting the findings of this study are available within the paper and its supplementary information files.