Spatiotemporal expression and control of haemoglobin in space

Spaceflight is associated with space anaemia, characterized by reduced red blood cell mass during missions on the ISS that generally resolves months after return. The mechanisms underlying this anaemia remain unclear, as does the regulation of haemoglobin gene switching under space conditions. Given the globin loci’s well-characterized developmental switching and regulation by specific transcription factors and miRNAs, the authors investigated how microgravity and spaceflight alter spatiotemporal expression of globin genes and their regulators. Using multi-omic datasets from the NASA Twins Study, the JAXA Cell-free Epigenome (CFE) study, and the Inspiration4 mission, they aimed to determine whether spaceflight perturbs erythropoiesis and induces an adult-to-fetal haemoglobin switch, and how these effects vary across timepoints (pre-, in-, and post-flight) and blood cell fractions.

Prior work has documented space anaemia and red blood cell mass reductions during long-duration spaceflight and post-flight recovery. Haemoglobinopathies (e.g., sickle cell disease, β-thalassaemia) highlight clinical relevance of disrupted globin switching. Erythropoiesis is governed by transcription factors including GATA-1/2, LMO2, FOG, c-MYB, TAL1, BCL11A, RUNX1, PU.1, and KLF1. Space omics resources (e.g., NASA GeneLab; NASA Twins Study) have linked spaceflight to broad molecular changes and guided countermeasure strategies. These foundations motivate probing globin loci regulation and the fetal-to-adult switch in microgravity.

Multi-cohort, multi-omic analysis across astronaut datasets and blood compartments.

• NASA Twins Study: Two male twins; flight twin spent 340 days on ISS; ground twin remained on Earth. Blood collected in CPT tubes; cell separation at 1800 × g for 20 min at room temperature on ISS and ground. Fractions included plasma, PBMCs, and lymphocyte-depleted (LD). Single-cell profiling with BD Rhapsody (custom RNA and epitope panel); sample tagging, AbSeq staining, cell capture beads, cDNA synthesis; targeted cDNA amplification (10 cycles), separation of mRNA and AbSeq by SPRIselect, further mRNA amplification (15 cycles), indexing (8 cycles). QC by Bioanalyzer and Qubit; sequencing on NovaSeq 6000 (mean ~13,244 mRNA reads/cell and ~11,507 AbSeq reads/cell; total ~243,751 reads/cell). Processing via Seven Bridges using BD Rhapsody pipeline 1.4 Beta; downstream analysis in Seurat 3.2.0 with scaling/normalization and differential expression using MAST; curated globin gene heatmaps with pheatmap 1.0.12.
• NASA Twins miRNA-Seq: Small RNA libraries from 50 ng total RNA prepared with NEBNext Multiplex Small RNA Library Prep for Illumina (adaptor/RT primer 4× diluted; 17 PCR cycles; no size selection); dual indexing; sequenced on Illumina NextSeq (1×50 bp). Pre-processing with miRTraces; mapping against MirGeneDB via miRDeep2 quantifier; normalization to RPM; ANOVA p<0.2 per cell type to identify significant miRNAs; flight-only comparisons excluded ground/ambient return; heatmaps via pheatmap.
• JAXA CFE study (OSD-530): Plasma cell-free RNA-seq from 6 astronauts before, during, and after ~6-month ISS missions. Mean normalized expression values computed per timepoint. Differential expression by DESeq2; volcano plots (log2FC cutoff 1.5, p<0.05); PCA via scikit-learn; visualization with Matplotlib; curated globin genes and trans-acting factors summarized in heatmaps (pheatmap).
• Inspiration4: Blood collected pre-flight (L-92, L-44, L-3) and post-flight (R+1, R+45, R+82). PBMC single-cell RNA-seq using 10x Genomics Chromium Next GEM Single Cell 5' v2; subpopulations annotated using Azimuth; Seurat used for normalization and average expression. Plasma cfRNA isolated with Norgen kit; DNase treatment; library prep with Takara SMARTer Stranded Total RNA-Seq v3-Pico; QC via Qubit and Agilent Fragment Analyzer; Illumina NextSeq 2000 paired-end 150 bp (~26M reads/sample). PCA assessed separation by flight status using genes of interest (globin loci and trans-acting factors). Analytical focus: Differential expression of globin genes (alpha and beta loci), erythropoietic regulators, and globin-switch transcriptional repressors/promoters across pre-, in-, and post-flight; comparison across cell fractions (e.g., LD, CPT, PBMCs) and cfRNA; volcano plots, PCA variance explained; integration with physiological parameters.

• Global patterns: Spaceflight is associated with down-regulation of adult globin gene expression in flight, with recovery and often overshoot up-regulation post-flight; multiple TFs and miRNAs implicated in the globin switch are perturbed in a direction consistent with increased fetal haemoglobin (HbF) expression in flight.
• JAXA cohort (cfRNA, n=6): • EPO significantly down-regulated in flight vs pre-flight, then up-regulated post-flight vs in-flight (DESeq2; volcano plots with log2FC cutoff 1.5, p<0.05). • HBA1, HBB, HBD significantly up-regulated post-flight vs in-flight; HBA1 also significantly up-regulated post-flight vs pre-flight. ALAS2 significantly up-regulated post-flight vs both pre-flight and in-flight. HBE1 significantly up-regulated in flight vs pre-flight. KLF1 significantly up-regulated post-flight vs in-flight. • Alpha locus (HBA1/2): down-regulated in-flight (5–60 days after launch), then up-regulated immediately upon return, exceeding pre-flight levels. • Beta locus (HBB): reduced in-flight, increased post-flight above pre-flight; HBG2 up-regulated in-flight; HBE1 up-regulated in-flight. • Trans-acting factors: MBD2 down in-flight, returning post-flight; Sin3A down in-flight; NR2C2 decreased in-flight; CHD4 down in-flight, recovered post-flight; KLF1 slight down in-flight and significant up post-flight; MYB down in-flight, increased post-flight; EIF2AK1 up in-flight; ATF4 down post-flight vs in-flight; ACVR1B down in-flight, up post-flight; ARID1B down in-flight; TAL1 up in-flight and down post-flight; LMO2 increased post-flight; NFE2 slight in-flight increase, down post-flight. • PCA: Using trans-acting factors, separation between pre-, in-, and post-flight groups with PC1 and PC2 explaining ~68.3% variance; using globin genes, separation along PC1 which explained ~99% of variance.
• NASA Twins Study: • In-flight down-regulation of globin genes across relevant cellular fractions (e.g., CD4, CD8, LD, CPT), with post-flight recovery/up-regulation in several cell types (e.g., HBA1/2 increased in CD19 post-flight; HBB up in CD8 post-flight vs ground twin). HBG1/2 and HBB down in-flight in CD4/CD8 vs pre- and post-flight. • TFs indicative of HbF shift: BCL11A down-regulated in-flight (CD4), up post-flight; Sin3A down in-flight (LD); NR2C2 decreased in-flight (LD); CHD4 up-regulated in-flight (contrast to JAXA); KLF1 down in-flight, up post-flight (CD8); TAL1 up in-flight, down post-flight (LD); LDB1 up in-flight (LD). miRNAs in erythroid fractions (LD, CPT) mostly down in-flight and higher post-flight, except miR-150-3p and miR-23a-5p; CD4 showed higher pre-flight than post-flight.
• Inspiration4: • PBMCs: Post-flight up-regulation of HBA1/2 and HBB immediately after return, then return to baseline thereafter; HBG2 up-regulated post-flight vs pre-flight. • cfRNA heatmap shows overall up-regulation in many trans-acting factors post-flight; PCA of cfRNA genes of interest shows separation by flight status with first two PCs accounting for ~60% variance.
• Physiological integration: Observed gene expression changes align with space anaemia phenotype (reduced RBC counts in-flight; recovery post-flight). Patterns suggest combined contributions of haemolysis and reduced erythropoiesis during flight.
• Overall, multiple repressors of HbF (e.g., BCL11A, SOX6, ZBTB7A, NR2C2, Sin3A) show in-flight down-regulation (cohort-dependent), while factors promoting erythropoiesis and haem biosynthesis (e.g., ALAS2, EIF2AK1, TAL1, NFE2) show dynamic regulation consistent with compensatory responses and post-flight recovery.

Findings across three astronaut cohorts support that spaceflight induces a reversible suppression of adult globin gene expression and activates components of the developmental globin switch favoring fetal haemoglobin. In-flight down-regulation of adult globin genes (HBA1/2, HBB) accompanied by in-flight up-regulation of fetal/embryonic globins (HBG2, HBE1) in certain cohorts, and down-regulation of HbF repressors (BCL11A, SOX6, ZBTB7A, NR2C2, Sin3A), are consistent with an adult-to-fetal shift. The recovery and overshoot of adult globin and erythropoietic gene expression post-flight correlate with known post-landing recovery from space anaemia. Transcription factor dynamics (e.g., KLF1, CHD4, TAL1, LMO2, NFE2) and haem synthesis/transport genes (ALAS2, FLVCR1) reflect both in-flight compensatory responses and post-flight restoration of erythropoiesis. PCA across cfRNA and cellular datasets reveals clear separation by flight status, indicating robust spatiotemporal transcriptional signatures related to haemoglobin regulation. Together, results suggest space anaemia involves both increased haemolysis and decreased erythropoiesis, with microgravity-associated stressors eliciting regulatory programs that can transiently favor HbF, analogous to ameliorating patterns seen in haemoglobinopathies. These insights inform potential biomarkers and targets for countermeasures during long-duration missions.

The study integrates multi-omic datasets (NASA Twins, JAXA CFE, Inspiration4) to characterize spatiotemporal regulation of haemoglobin and erythropoiesis in spaceflight. Key contributions include evidence of in-flight suppression of adult globin expression with concurrent activation of fetal/embryonic globin programs and coordinated changes in canonical globin-switch regulators and erythropoietic pathways, followed by post-flight recovery and overshoot. The consistent separation of samples by flight status across cohorts and assays underscores robust spaceflight effects on the haemoglobin regulatory network. Future work should include direct quantification of fetal haemoglobin levels in-flight and post-flight, extended mission-duration sampling to refine temporal dynamics, and exploration of targeted countermeasures leveraging identified pathways (e.g., modulation of BCL11A/KLF1 networks, haem biosynthesis, and haem export).

The authors note limitations related to heterogeneous and limited exposure durations across cohorts and a focus on specific time windows, constraining temporal resolution of the full dynamic response. Some datasets lack in-flight sampling (e.g., Inspiration4 PBMCs only pre- and post-flight), and differences in sample types (cellular fractions vs plasma cfRNA) complicate direct comparisons. In the NASA Twins miRNA analysis, substantial inter-individual baseline differences led to excluding ground twin values for certain comparisons.