Nanobioreactor detection of space-associated hematopoietic stem and progenitor cell aging

Abstract

Human hematopoietic stem and progenitor cell (HSPC) fitness declines following exposure to stressors that reduce survival, dormancy, telomere maintenance, and self-renewal, thereby accelerating aging. While previous National Aeronautics and Space Administration (NASA) research revealed immune dysfunction in low-earth orbit (LEO), the impact of spaceflight on human HSPC aging had not been studied. To study HSPC aging, our NASA-supported Integrated Space Stem Cell Orbital Research (ISSCOR) team developed bone marrow niche nanobioreactors with lentiviral bicistronic fluorescent, ubiquitination-based cell-cycle indicator (FUCCI2BL) reporter for real-time HSPC tracking in artificial intelligence (AI)-driven CubeLabs. In month-long International Space Station (ISS) missions (SpX-24, SpX-25, SpX-26, and SpX-27) compared with ground controls, FUCCI2BL reporter, whole-genome and transcriptome sequencing, and cytokine arrays demonstrated cell-cycle, inflammatory cytokine, mitochondrial gene, human repetitive element, and apolipoprotein B mRNA editing enzyme, catalytic polypeptide-like 3 (APOBEC3) deregulation together with clonal hematopoietic mutations. Furthermore, HSPC functionally organized multi-omics aging (HSPC-FOMA) analyses revealed reduced telomere maintenance, adenosine deaminase acting on RNA1 (ADAR1) p150 self-renewal gene expression, and replating capacity indicative of space-associated HSPC aging that may limit long-duration spaceflight.

Keywords: ADAR1; APOBEC3; aging; cell cycle; clonal hematopoiesis; dark genome; hematopoietic stem and progenitor cells; low-earth orbit; nanobioreactor; repetitive elements.

Figure 1.. Nanobioreactor detection of HSPC fitness

(A) Experimental design of HSPCs seeded in 3D nanobioreactors (top) for culture on SpaceX CRS (SpX-CRS) missions to the ISS. Upon return, the cells underwent colony-forming survival and self-renewal assays, as well as whole-genome and whole-transcriptome sequencing. Furthermore, they were co-cultured, underwent colony-forming survival and self-renewal assays, and were subsequently analyzed using NanoString. In four independent NASA SpaceX CRS (SpX-24, SpX-25, SpX-26, and SpX-27) missions, ranging in duration from 32 to 45 days, nanobioreactors containing CD34+ HSPCs and a CD34− feeder layer from paired and unpaired samples were flown (Table 1). (B) The CubeLab integrated system is equipped with a thermal management system, a microscope imaging system known as TangoScope, and fluid routing systems to transport media to and from cells. (C) FUCCI2BL-transduced HSPCs imaged within the nanobioreactor. mVenus-hGem indicates the S/G2/M phase and mCherry-hCdt1 indicates the G1 phase of the cell cycle. mVenus+/mCherry+ cells indicate the G1/S phase. (D) Representative fluorescent images acquired during the SpX-24 mission. The mission lasted for 34 days, while the cells were cultured for a total period of 42 days in the nanobioreactor. Images were taken at the same X-Y-Z position within each nanobioreactor to capture the cell-cycle transit of ABM CD34+ HSPCs. Images were taken at 20×. (E and F) Representative Volocity fluorescent intensity quantification was measured in paired SAK-109 spaceflight and ground nanobioreactors over the course of the SpX-24 mission. The mission lasted for 34 days, while the cells were cultured for a total period of 42 days in the nanobioreactor. Fluorescence was quantified in mCherry- and mVenus-expressing CD34+ HSPCs and measured as total surface area (μm²). (G–L) Colony-formation and replating assay of spaceflight and ground-returned samples after the SpX-24 mission. The mission lasted for 34 days, while the cells were cultured for a total period of 42 days in the nanobioreactor. (G) Representative images of primary colonies (SAK-109)–HSPC survival assay. (H) Representative images of secondary colonies (SAK-109)–HSPC self-renewal assay. (I and J) Representative analysis of human bone marrow-derived cells (SAK-109) upon return of mission SpX-24 cultured in parallel in spaceflight and ground nanobioreactors for a duration of 42 days. Subsequently, 10,000 CD34+ cells were subjected to clonogenic assays, as described in (A) and STAR Methods. (I) In the survival assay (primary colonies; G), the graph demonstrates the absolute number of differentiated and multi-lineage (p = 0.002) colonies. (J) In the self-renewal assay (secondary colonies; H), the graph depicts the % of replated cells. Each bar represents mean ± standard error for triplicate conditions. Statistical analysis included Student’s t test and one-way ANOVA, including all pairwise multiple comparison procedures (Holm-Sidak method). (K and L) Summarized bar graph of four individual paired ABM samples (SAK-109 p = 0.003, SAK-252 p < 0.001, SAK-290 p < 0.001, and SAK-291 p = 0.001) cultured in parallel in spaceflight and ground nanobioreactors for 42, 60, 42, and 42 days, respectively, then subjected to clonogenic assays, as described in (A) and STAR Methods. (K) In the survival assay, the graph demonstrates the absolute total number of colonies. (L) In the self-renewal assay, the graphs depict the % of replated cells. Each bar represents mean ± standard error for triplicate conditions. Statistical analysis included Student’s t test and one-way ANOVA, including all pairwise multiple comparison procedures (Holm-Sidak method). See also Figure S1.

Figure 2.. HSPC-FOMA analyses

(A) Schematic of telomere length estimation from WGS data derived from ground and spaceflight HSPC nanobioreactors. Raw FASTQ files were downloaded within Triton shared computational cluster environment. Next, the GRCh38.d1.vd1 reference sequence was used for the alignment of raw reads. Subsequently, following GATK4 best practice, Telomerecat was employed for estimation of telomere length. (B) Bar graphs illustrating the Telomerecat-estimated overall telomere length (in base pairs [bp]) of cells that are derived from paired spaceflight and ground samples of SAK-109, SAK-252, SAK-290, and SAK-291 upon their return. (C) Bar plots of mean telomere length estimates (in bp) from the four replicates shown in (B). Bar plots represent the mean ± SEM. Statistics reported using the Wilcoxon signed-rank test with p = 0.125. (D) Summary of top expression differences (measured by limma p value) for telomere-associated pathway genes curated from the Reactome pathways HAS-157579, HSA-171319, and HSA-171306. (E) GSEA results for comparison of spaceflight (SAK-066, SAK-109, and SAK-252) and ground (SAK-072, SAK-109, and SAK-252) RNA-seq samples using Reactome pathways as an ontological database. NES, normalized enrichment score. (F) Bar plots of mitochondrial copy-number estimates from the four replicates shown in (B). (G) (Top) Waterfall plot of top 30 genes (by logFC) in the significantly enriched (by GSEA) MitoCarta 3.0 s-tier class categories. All genes are increased in the spaceflight samples. (Bottom) Heatmap of all expressed genes in GSEA-enriched MitoCarta 3.0 top class categories. The heatmap dendrogram shows that spaceflight and ground samples separate, and genes are primarily increased in the spaceflight samples. (H) GSEA enrichment plot for the comparison of ground and spaceflight RNA-seq samples shows the distribution of the enrichment score (ES, green line) and the position of each gene in the set across the gene ranks for mitochondrial translation (positive NES). Boxplot of the significantly enriched GSEA pathway classes for mitochondrial translation showing the top 5 genes and corresponding log fold changes. (I) GSEA enrichment plot for the comparison of ground and spaceflight RNA-seq samples shows the distribution of the ES (green line) and the position of each gene in the set across the gene ranks for cell-cycle checkpoints (negative NES). Boxplot of the significantly enriched GSEA pathway classes for cell-cycle checkpoints showing the top 5 genes and corresponding log fold changes. (J) GSEA enrichment plot for the comparison of ground and spaceflight RNA-seq samples shows the distribution of the ES (green line) and the position of each gene in the set across the gene ranks for interferon signaling (negative NES). Boxplot of the significantly enriched GSEA pathway classes for interferon signaling showing the top 5 genes and corresponding log fold changes. See also Figure S2.

Figure 3.. Space-associated mutagenesis patterns

(A) Bar plots showing the total number of somatic mutations acquired in spaceflight in comparison to its ground control. The y axis reflects the number of total mutations measured in somatic mutation counts. The x axis corresponds to the four replicated experiments. (B) Bar plots showing the total amount of somatic SBS mutations acquired in spaceflight in comparison to their ground control. y axis reflects the amounts of total SBS, measured in somatic mutation counts. x axis corresponds to the four replicated experiments. (C) Bar plots showing the total amount of somatic insertion and deletion (indel) mutations acquired in spaceflight compared with its ground control. The y axis reflects the amounts of total small indels, measured in somatic mutation counts. The x axis corresponds to the four replicated experiments. (D) (Top) Oncoplot displaying CH mutated genes acquired in naive CD34+, FUCCI2BL-transduced CD34+, spaceflight nanobioreactor, and ground nanobioreactor. Sample size of n = 3, SAK-252, SAK-290, and SAK-291, from SpX-26 and SpX-27 missions. (Bottom) Table summarizing CH-associated mutations induced during spaceflight for n = 3 samples, SAK-252, SAK-290, and SAK-291. (E) Patterns of SBS for the samples are shown using the SBS96 classification scheme on the x axis. The y axis is scaled differently in each plot to optimally show each mutational pattern, with the y axis reflecting the number of mutations for the respective mutational scheme. The data from returned spaceflight samples are normalized to returned ground data. (F) Patterns of small indels are shown using the ID-83 classification scheme on the x axis. The y axis is scaled differently in each plot to optimally show each mutational pattern, with the y axis reflecting the number of mutations for the respective mutational scheme. The data from returned spaceflight samples are normalized to returned ground data. (G) Bar plots showing total number of somatic mutations acquired in APOBEC3C-containing pCDH vector-transduced cells. The y axis reflects the number of total mutations measured in somatic mutation counts. The x axis corresponds to the three replicated experiments. (H) Bar plots showing the total amount of somatic SBS mutations acquired in APOBEC3C-containing pCDH vector-transduced cells. The y axis reflects the amount of total SBS, measured in somatic mutation counts. The x axis corresponds to the four replicated experiments. (I) Bar plots showing the total amount of somatic indel mutations acquired in APOBEC3C-containing pCDH vector-transduced cells. The y axis reflects the amounts of total small indels, measured in somatic mutation counts. The x axis corresponds to the four replicated experiments. (J) Patterns of SBS for the samples from APOBEC3C-containing pCDH vector-transduced HSPCs are shown using the SBS96 classification scheme on the x axes. y axes are scaled differently in each plot to optimally show each mutational pattern, with the y axes reflecting the number of mutations for the respective mutational scheme. (K) Patterns of small indels for samples from APOBEC3C-containing pCDH vector-transduced cells are shown using the ID-83 classification scheme on the x axis. The y axis is scaled differently in each plot to optimally show each mutational pattern, with the y axis reflecting the number of mutations for the respective mutational scheme. (L) Total radiation, as a percentage of total dose, exposure during the total duration of SpX missions. Total radiation exposure for n = 4 samples (SAK476, SAK477, SAK478, and SAK479) given in one single dose. (M) Bar plots showing the total number of somatic mutations acquired in samples exposed to a single dose of radiation. The y axis reflects the number of total mutations measured in somatic mutation counts. The x axis corresponds to the four individual paired samples. (N) Bar plots showing the total number of somatic SBS mutations acquired in samples exposed to a single dose of radiation. y axis reflects the amounts of total SBS, measured in somatic mutation counts. The x axis corresponds to the four individual paired samples. (O) Bar plots showing the total number of somatic indel mutations acquired samples exposed to a single dose of radiation. The y axis reflects the amounts of total small indels, measured in somatic mutation counts. The x axis corresponds to the four individual paired samples. (P) Patterns of SBS for the samples are shown using the SBS96 classification scheme on the x axis. The y axis is scaled differently in each plot to optimally show each mutational pattern, with the y axis reflecting the number of mutations for the respective mutational scheme. The data from irradiated samples are normalized to non-irradiated data. (Q) Patterns of small indels are shown using the ID-83 classification scheme on the x axis. The y axis is scaled differently in each plot to optimally show each mutational pattern, with the y axis reflecting the number of mutations for the respective mutational scheme. The data from irradiated samples are normalized to non-irradiated data. See also Figure S3.

Figure 4.. Space-associated base deaminase and repetitive element deregulation

(A) Expression of the ratio of ADARp150 and ADARp110 log-counts per million (CPM) values in spaceflight and ground samples upon return. Student’s t test was performed to determine significance with p = 0.036. (B) Summary of detected RNA editing events normalized by library size and sequencing run for spaceflight and ground samples upon return. Student’s t test was performed to determine significance with p = 0.007. (C) Expression of detectable APOBEC3 genes measured by logCPM in spaceflight and ground samples upon return. Student’s t test was performed to determine significance. APOBEC3A p value: 0.37, APOBEC3B p value: 0.76, APOBEC3C p value: 0.16, APOBEC3F p value: 0.48, APOBEC3G p value: 0.17. (D) Heatmap of repetitive elements with p < 0.01 in the comparison of spaceflight and ground samples upon return. (E) Volcano plot visualizing LINE expression levels in ground and spaceflight samples upon return. The y axis represents the negative log of the p value (−log p value) where the base is e, and the x axis is the log fold change (logFC). The red dots represent LINEs with p < 0.05, and the black line represents the p value threshold of 0.05. (F) Volcano plot of LINEs expressed in spaceflight vs. ground RNA-seq samples for LINEs that are differentially expressed in the APOBEC3C overexpression vs. pCDH comparison. The y axis represents the negative log of the p value (−log p value) where the base is e, and the x axis is the logFC. Red dots represent repetitive elements with p < 0.05, and the black line represents the p value threshold of 0.05. (G) Summarized data of survival and self-renewal for SAK-290 and SAK-291 (SpX-27, 42 days in space [F] or ground [G] nanobioreactors). These cells were additionally co-cultured for 2 weeks with SAK290 or SAK291 primary autologous stroma (grown terrestrially from the CD34-negative fraction of their bone marrow as described in STAR Methods) or with a control human bone marrow stromal cell line HS-27A, also maintained. These data were compared with the clonogenic results of the original naive CD34+ cells (no stroma). Basal colony formation of original naive untreated cells before the spaceflight was considered to be 100%, and individual values were calculated as % of change. Data presented show the mean ± SD for both samples. Statistical analysis included Student’s t test and one-way ANOVA, including all pairwise multiple comparison procedures (Holm-Sidak method). (H) Volcano plot of gene expression difference between spaceflight and ground stem cells co-cultured with the fibroblast cell line HS-27A. LogFC is shown on the x axis with the negative log of the nominal p value on the y axis. Red dots represent genes with nominal p < 0.05. Dashed lines represent cutoffs for logFC > 0.5 (x axis) and p < 0.05 (y axis). (I) Volcano plot of gene expression differences between spaceflight and ground HSPCs co-cultured with matched autologous stroma. LogFC is shown on the x axis with the negative log of the nominal p value on the y axis. Red dots with gene labels represent genes with a nominal p < 0.05. Dashed lines represent cutoffs for logFC > 0.5 (x axis) and p < 0.05 (y axis). The top 12 genes, ranked by p value, are labeled. (J) Pie chart of the distribution of significantly upregulated and downregulated genes in the comparison of autologous stroma and HS-27A (NanoString). (K) Bar graph of the logFC changes in the NanoString expression differences between normal-ABM (nABM) co-cultured with the primary autologous stroma and the control human bone marrow cell line HS-27A. Significance was determined using the Benjamini-Hochberg method for estimating false discovery rate (FDR) with a p value of <0.05. See also Figure S4.