RESULTS:
Characterization of the overall spatial patterns of microbial community structure demonstrated that the factor that most contributed to variation among all samples was sampling site, in agreement with previous studies showing that the skin and nose microbiomes are more similar to each other than to those from feces or the mouth. In addition, samples collected from the same body site tended to cluster by crewmember. Interestingly, ISS environmental samples overlapped with specimens from the two skin sites and the nose microbiota. Further analysis showed that there was no significant difference among bacterial species that were present/absent in ISS microbial communities and inflight forehead and forearm skin microbiomes.

To determine the influence that the amount of time spent at the ISS has in any observed changes in astronaut microbiomes, investigators compared microbial profiles of microbiome samples collected at every inflight and postflight time points to all preflight time points. This analysis showed that compositional changes of the nose, skin, and gastrointestinal (GI) microbiomes were rapid and became evident by FD7. These changes persisted for at least six months until the end of the mission at the ISS. Furthermore, beta diversity changes did not significantly increase with the time astronauts spent in space, although preflight-inflight dissimilarity distances of the two skin sites and the nose microbiota showed a very modest upward trend associated with time spent inflight. In addition, compositional shifts in the skin and nose microbiomes persisted for at least 60 days after the astronauts returned to Earth. However, the composition of the GI microbiota became similar to preflight samples within two months of the astronauts’ return from the ISS.

To investigate the influence of the ISS as a contained and human built environment on the crewmembers' microbiomes, differential abundance analysis was performed between samples collected before, during, and after the mission to the ISS. This analysis identified 15 gastrointestinal genera whose abundance significantly changed in space. Ten out of the 15 genera belonged to the phylum Firmicutes and most belonged to the order Clostridiales. Among these taxonomic groups, there was a more than five-fold inflight reduction in Akkermansia and Ruminococcus, and an approximate 3-fold drop in Pseudobutyrivibrio and Fusicatenibacter. Most of these compositional changes reverted to preflight levels after astronauts returned to Earth, with the exemption of four genera of the phylum Firmicutes. Examination of skin samples also revealed changes in the relative abundance of several bacterial groups corresponding to 43 and 31 genera in the forearm and forehead, respectively. Noteworthy, skin microbial communities whose abundance decreased in space were mostly Gram-negative Proteobacteria. These groups included bacteria from the genus Acinetobacter, Cloacibacterium, and Pseudomonas. In contrast, most of the skin bacteria that became more abundant inflight belonged to the phylum Firmicutes, Bacteroidetes, and Actinobacteria, including bacteria of the genus Streptococcus, Staphylococcus, and Corynebacterium. Postflight samples showed a similar trend as inflight samples, with lower Proteobacteria and higher Firmicutes, Bacteroidetes, and Actinobacteria compared to Preflight skin. In addition, we observed a similar but milder response of the nares microbiota to the space environment with a significant drop in three genera of Gram-negative Betaproteobacteria, all of which were also reduced in skin. Likewise, nose inflight samples showed increases in five Gram-positive genera that also became more abundant on the skin of astronauts inflight. Many of these changes dissipated after the astronauts returned to Earth.

To gain insights into the potential impact of changes to the microbiome during space flight on immune functioning, changes to cytokine abundance in plasma were compared to changes in the composition of the GI microbiome. This analysis identified strong evidence of association between changes in astronauts’ GI microbiome and changes in cytokine profiles. Most notably, the abundance of bacteria of the genus Fusicatenibacter was negatively correlated with the concentration of pro-inflammatory cytokines. In addition, changes in bacteria of the genus Dorea were also negatively correlated with changes in the level of several cytokines, all of which were increased in space.

Given that the ISS microbiota resembled that of the astronauts’ skin, the investigators hypothesized temporal changes could be influenced by the arrival of new crewmembers to the ISS every three months. To test this hypothesis, they compared the microbial composition of the ISS samples from early and late time points to the skin microbiome of astronauts that traveled to the ISS at the beginning or end of the study. This analysis showed that while there were no differences in the species present/absent between skin and ISS samples at either early or late time points, early skin microbiome samples had a significantly different composition to late ISS microbiome samples and vice versa. Moreover, comparative analysis of taxonomic profiles showed that the skin and ISS samples collected by the time astronauts were leaving the ISS were more similar to each other than the ISS and skin samples collected at FD7, when crew members had just arrived at the ISS.

Reactivation and shedding of latent varicella zoster virus (VZV), Epstein bar virus (EBV), and Herpes simplex virus (HSV1) as well as diurnal salivary cortisol, alpha-amylase, and dehydroepiandrosterone (DHEA) were measured prospectively in 10 astronauts before, during, and after their mission at the ISS to assess astronauts’ stress. Two astronauts did not shed any virus in any of their samples collected during the study. VZV reactivation was detected in the saliva of four crewmembers. Except for one crewmember that showed VZV reactivation by L-60, no VZV was detected in saliva before flight. However, VZV was significantly reactivated in space, reaching a peak by FD90. After 30 days of the return to Earth, three of the astronauts became negative for VZV except for one astronaut that remained positive for up to 180 days. EBV and HSV1 were detected in the saliva of three and five astronauts, respectively, but no association was found between detection in saliva and space flight. No significant changes in levels of salivary cortisol were detected during the entire mission in any of the crewmembers. The inflight salivary concentration of alpha-amylase, however, was higher than preflight values reaching a maximum by the end of the mission in the ISS. DHEA showed an opposite trend, becoming less abundant by FD90 at the ISS.

To investigate whether the observed changes in the microbial composition of the astronauts’ GI microbiome have any consequence on its metabolic capacity investigators carried out metagenomic sequence analysis of the GI microbiome of six crewmembers during their missions to the ISS. Comparative analysis of the enzyme profiles predicted for each astronaut metagenome showed that, within subjects, inflight enzyme profiles tend to be different from preflight or postflight samples. Further analysis at the individual enzyme level identified 10 enzymatic functions that were differentially abundant between pre and inflight metagenomes, while only EC3.7.1.3 changed and became less abundant in postflight samples compared with preflight samples. In addition, investigators identified five metabolic pathways that were either over or underrepresented in inflight samples of the GI microbiome. One of these pathways, which consistently increased in space, corresponded to the pathway for the biosynthesis of polysaccharides, or lipopolysaccharides (LPS), which is a typical component of the outer membrane of gram negative bacteria. Taxonomic analysis of the genes participating in the LPS pathway showed that the observed changes are mostly driven by changes in the abundance of bacteria from the genus Bacteroides. The direction of the changes in gene abundance in the other four pathways identified were less consistent than the LPS pathway, with different GI metagenomes responding differently to space travel. One of these pathways was the one responsible for the biosynthesis of the bacterial flagellum. GI bacteria encoding for this pathway were enriched in three crew members and reduced in the metagenomes of the other three astronauts analyzed. Further taxonomic analysis showed that for all astronauts changes in this pathway were mostly caused by increases or decreases of bacteria of the genus Eubacterium and Roseburia, two of the most abundant motile bacteria found in the GI tract of healthy individuals. In agreement, the abundance of genes involved in bacterial chemotaxis was found to significantly change during spaceflight in a fashion similar to the related bacterial flagellum biosynthesis pathway. Further taxonomic analysis showed that the changes were mainly driven by alterations in the relative abundance of bacteria from the genus Eubacterium and Roseburia.