Single-cell multi-ome and immune profiles of the Inspiration4 crew reveal conserved, cell-type, and sex-specific responses to spaceflight

Abstract

Spaceflight induces an immune response in astronauts. To better characterize this effect, we generated single-cell, multi-ome, cell-free RNA (cfRNA), biochemical, and hematology data for the SpaceX Inspiration4 (I4) mission crew. We found that 18 cytokines/chemokines related to inflammation, aging, and muscle homeostasis changed after spaceflight. In I4 single-cell multi-omics data, we identified a "spaceflight signature" of gene expression characterized by enrichment in oxidative phosphorylation, UV response, immune function, and TCF21 pathways. We confirmed the presence of this signature in independent datasets, including the NASA Twins Study, the I4 skin spatial transcriptomics, and 817 NASA GeneLab mouse transcriptomes. Finally, we observed that (1) T cells showed an up-regulation of FOXP3, (2) MHC class I genes exhibited long-term suppression, and (3) infection-related immune pathways were associated with microbiome shifts. In summary, this study reveals conserved and distinct immune disruptions occurring and details a roadmap for potential countermeasures to preserve astronaut health.

Fig. 1. Immune-metabolic changes after 3-day spaceflight and recovery.

a Overview of I4 mission single cell GEX + ATAC, single cell TCR/BCR V(D)J repertoire, biochemical profiles (BCP) of 97 analytes, and complete blood count (CBC) of 15 analytes data collection and analysis, created with BioRender.com. b Heatmap of significantly changed biochemicals (cytokines, chemokines, and growth factors) in serum before spaceflight (Pre-flight: mean of L-92, L-44, L-3) and after spaceflight (Immediately Post-flight: R + 1, and Long-term Post-flight: R + 45, R + 82, R + 194). A significant increase in concentration is observed immediately after spaceflight (R + 1) in IL-1RA, IL-4, IL-5, IL-6, IL-7, IL-10, IL-27, MCP-1, TNFα, IP-10, ENA-78, CRP and Fractalkine. On the other hand, IL-9, IL-17E/IL-25, MIP-1α, MCP-2 and MCP-4 showed a significant decrease in their serum levels after spaceflight (R + 1). Wilcoxon-rank sum test (padj <0.05, two-sided). c GSEA of the ‘spaceflight signatures of the I4 astronauts’ (Hallmark, KEGG, and wikipathways: filtered with padj <0.05, GOBP and C2: top10 of positive and negative NES, padj < 0.05, padj calculated by fGSEA R package). d Overlap percentage of the GSEA pathways across the I4 immune cells (Fisher’s exact test, two-sided, padj < 0.05. Except for the Hallmark CD14 Mono: P value = 0.09). e Activity scores of top enriched motifs from pseudo-bulk PBMCs over time. Source data are provided as a Source Data file.

Fig. 2. Conserved and distinct spaceflight signatures across mission, species and mission duration.

a Log2 fold change heatmap of the “spaceflight signatures in mice” in 27 datasets with 10 different mouse tissues. Age (Day). Duration (Day). 1,288 up-regulated and 896 down-regulated genes. b The up-regulated genes (red) and down-regulated genes (blue) from the I4 data are shown in terms of percentage of overlap (y-axis) with the ‘spaceflight signatures in mice’. c GSEA analysis the I4 DEGs with the ‘spaceflight signatures in mice’. d Scatter plot of the -log10(padj)*sign(NES) of the ‘spaceflight signatures of the I4 astronauts’ and the ‘spaceflight signatures of mice’ GSEA pathways and the representative pathways. Pearson correlation (R) = 0.82. Slope: 0.69. Two-sided. The standard error should be used to create the band around the linear regression line. e Overlap percentage of the significantly enriched overlapped GSEA pathways (NASA Twins vs I4). f GSEA of PBMC and subpopulations at the immediately post-flight (R + 1) and long-term post-flights (R + 45 and R + 82) with up-regulated and down-regulated DEGs of skin spatial transcriptomics data (padj <0.05). OE Outer Epidermis, OD Outer Dermis, VA Vasculature) in skin biopsy data. The fgsea analysis employs a one-sided permutation-based test to determine the significance of gene set enrichment, with raw p values adjusted for multiple testing using the Benjamini-Hochberg procedure to control the false discovery rate (FDR). g Scatter plot of the −log10(padj)*sign(NES) of the ‘spaceflight signatures of the I4 astronauts’ and the I4 skin spatial transcriptomics GSEA pathways and the representative pathways. Pearson correlation (R) = 0.87. Slope: 0.85. Two-sided. The standard error should be used to create the band around the linear regression line. h The percentage of overlap of I4 DEGs and in vitro microgravity simulated DEGs. i MHC class I gene expression in the I4 immune cells. Source data are provided as a Source Data file.

Fig. 3. Immune cell sub-population changes after spaceflight and recovery profiles.

a Dot plots of previously reported spaceflight-associated CD14 monocyte markers (Top: gene expression, Bottom: ATAC derived gene expression). b GSEA of PBMC and subpopulations with the selected KEGG pathway significantly enriched with the over-representation analysis of the I4 immune cell DEGs (padj < 0.05). The fgsea analysis employs a one-sided permutation-based test to determine the significance of gene set enrichment, with raw p-values adjusted for multiple testing using the Benjamini-Hochberg procedure to control the false discovery rate (FDR). c Activity scores of TCF21 target genes in T, B, NK, monocyte, and dendritic cells. d Activity scores of FOXP3 target genes in T and Treg cells. e Heatmap of FOXP3 target genes in Treg cells. Color scale represents the normalized expression. f Dot plot of Treg markers and Treg activation markers in Treg cells (Left: gene expression. Right: ATAC derived gene expression). g Relative mRNA expression of Treg markers and Treg activation markers in T cells quantified by qPCR. Source data are provided as a Source Data file.

Fig. 4. Sex-dependent differences in response to spaceflight and recovery.

a Log2 Female to male DEGs ratio immediately post-flight (Top) and long-term post-flight (Bottom). The dotted line represents a Log2 ratio of 0. b Heatmap plot represents the overlap of up-regulated DEGs (Orange) and down-regulated DEGs (Purple) from females and males of PBMC and subpopulations. F Females, M Males. c Common and sex-specific HLA and CD expressions in immune cells, created with BioRender.com. Source data are provided as a Source Data file.

Fig. 5. The landscape of microbiome-immune associations.

a GSEA of all immune cells with the significantly enriched GO-BP pathways from the microbiome-associated immune cell DEGs related to immune function (padj <0.2). The fgsea analysis employs a one-sided permutation-based test to determine the significance of gene set enrichment, with raw p-values adjusted for multiple testing using the Benjamini-Hochberg procedure to control the false discovery rate (FDR). b We compared the associations reported in the main text to associations run on randomized data, computing the overlap therein at different stringency levels for controlling false positives. The three bars in each sub-panel correspond to the number of associations in the “real” (log-transformed) data versus randomized data and the overlap therein at different stringency levels in controlling for false positives. c The number of Bonferroni <0.05, positive significant microbiome associations by cell type. d The human genes, per cell type, with the greatest number of microbial associations that themselves had low or Bonferroni-significant p values (Two-sided). Each point in the plot bodies represent a different bacterial species (top) or viral genus (bottom). For each cell type, we ranked genes with non-zero LASSO coefficients first by the number of Bonferroni < 0.2 findings then by the total number of nominally associated (p value < 0.05) microbial features (bacteria or viruses). We report up to ten human genes per sub-panel. Source data are provided through the github link.