Graphical abstract

   graphic file with name fx1.jpg
   [31]Open in a new tab

Highlights

     * •
       A muscle-on-a-chip platform launched to the International Space
       Station National Laboratory
     * •
       Muscle bioconstructs in microgravity underwent impaired
       regeneration
     * •
       Drug testing in this microgravity platform was feasible
     __________________________________________________________________

   Kim and colleagues examined the effect of microgravity in engineered
   muscle bioconstructs aboard the International Station National
   Laboratory. The transcriptional results show a metabolic shift toward
   lipid and fatty acid metabolism and increased apoptotic pathways in
   microgravity. This signature was partially inhibited by
   pro-regenerative drugs like insulin-like growth factor-1 and a
   15-hydroxyprostaglandin dehydrogenase inhibitor.

Introduction

   Representing approximately 40% of body weight, skeletal muscle is one
   of the most abundant tissues in the human body ([32]Reid and Fielding,
   2012). Skeletal muscle regenerates itself through satellite cells, a
   reservoir of quiescent muscle stem cells that can be activated with
   injury or disease and give rise to newly formed multi-nucleated
   myotubes and myofibers ([33]Mauro, 1961; [34]Morgan and Partridge,
   2003). As civilian space travel to the International Space Station and
   beyond will become increasingly more common in the next century, it is
   important to better understand how microgravity affects the
   regenerative capacity of muscle. Mice exposed to microgravity have
   shown reduced muscle regeneration, compared to those in normal gravity,
   as demonstrated by relatively fewer postmitotic nuclei within myofibers
   as an indicator of impaired de novo muscle fiber formation
   ([35]Radugina et al., 2018). Similarly, in vitro studies composed of
   murine myoblasts seeded within three-dimensional hydrogels in simulated
   microgravity also show reduced myogenesis, based on the formation of
   shorter and thinner myotubes with reduced fusion capacity ([36]Ren
   et al., 2024). These data support a larger body of literature showing
   that microgravity exerts profound effects on skeletal muscle ([37]Fitts
   et al., 2010; [38]Juhl et al., 2021; [39]Trappe, 2009).

   The identification of countermeasures against the effects of
   microgravity on muscle regeneration is an increasing priority for
   continued space travel. Developing drugs that counteract the effects of
   microgravity is a promising strategy, but no Food and Drug
   Administration (FDA)-approved drug currently exists, partially due to
   the limited access of participants to microgravity conditions. In
   contrast, a muscle-on-a-chip in microgravity platform can be performed
   in a facile manner in vitro and can allow for screening many drugs in
   parallel. Through the US National Science Foundation and the Center for
   the Advancement of Science in Space (CASIS), we were offered a unique
   opportunity to perform tissue engineering and mechanobiology research
   aboard the International Space Station National Laboratory (ISSNL).
   With the assistance of astronauts, we applied our expertise in the
   engineering of skeletal muscle ([40]Huang et al., 2006; [41]Nakayama
   et al., 2019) toward the study of muscle myogenesis using a
   muscle-on-a-chip platform. By performing multi-omics analyses, we show
   that the muscle-on-a-chip platform in microgravity mimics salient
   aspects of impaired muscle regeneration after merely 7 days in
   microgravity. Furthermore, we demonstrate the feasibility of drug
   screening in this microgravity platform by using two drugs that
   partially prevent the negative effects of microgravity in the
   muscle-on-a-chip bioconstructs.

Results

Engineering of muscle-on-a-chip system in microgravity

   To adapt engineered muscle for microgravity conditions aboard the
   ISSNL, we developed a muscle-on-a-chip platform to immobilize the
   engineered muscle onto rigid hydrophobic glass for culture within a
   flight-compatible custom-made bioreactor. Skeletal muscle constructs
   composed of primary human myotubes on parallel-aligned collagen
   nanofibrils were generated by a facile shear-based collagen extrusion
   assay. Taking advantage of collagen fibrillogenesis being controlled by
   both pH and shear, we sheared acidic monomeric collagen type I from a
   blunt needle onto a rigid glass substrate bathed in pH neutral buffer
   ([42]Figure 1A), forming strips of collagen with highly organized
   nanofibrils. The collagen strips were composed of nanofibrils that were
   <100 nm in diameter, and the strips were then bundled together
   ([43]Figures 1B and 1C). The anisotropic organization of the collagen
   fibrils provides topographical patterning cues to induce primary human
   skeletal muscle cells to organize along the direction of the
   nanofibrils ([44]Figure 1D) and subsequently to form aligned myotubes
   ([45]Figure 1E). The cell-seeded scaffolds were denoted as a
   “muscle-on-a-chip.”

Figure 1.

   [46]Figure 1
   [47]Open in a new tab

   Generation and characterization of biomimetic skeletal muscle-on-a-chip
   platform

   (A) Schematic depicting the fabrication of aligned collagen nanofibrils
   using shear-mediated extrusion of monomeric collagen into pH 7 buffer.

   (B) Gross image and schematic of anisotropic collagen strips deposited
   onto hydrophobic glass chips.

   (C) Scanning electron microscopy analysis shows highly organized
   anisotropic collagen fibrils.

   (D) Seeding of primary human skeletal muscle cells shows cell alignment
   along the direction of the collagen nanofibrils (denoted by arrow).

   (E) In induction media, the cells fuse to form parallel-aligned
   myotubes, as shown by myosin heavy-chain staining (green). Scale bars:
   (B) 5 mm, (C) 1 μm, (D–E) 100 μm.

   To adapt the muscle-on-a-chip platform to microgravity,
   microgravity-compatible bioreactors were custom-made by BioServe Space
   Technologies. Prior to launch, the flight hardware was assembled by
   securing the muscle-on-a-chip bioconstructs within the chamber, filling
   the chamber with medium, and sealing with an air-permeable membrane
   stabilized with a solid frame ([48]Figures 2A, [49]S1A, and S1B).
   During transit to the ISSNL on a Cygnus vehicle, the bioreactors were
   transported within gas-sealed space habitats that maintained
   temperature and CO[2] levels ([50]Figures 2 and [51]S1C). Upon arrival
   at the ISSNL, the space habitats housing the bioreactors were placed
   into the Space Automated Bioproduct Lab incubator and maintained at
   37°C and 5% CO[2]. The experiments were initiated (day 0) in
   microgravity with the syringe-mediated replacement of the media with
   induction media that supports myotube formation ([52]Figure 2B).
   Comparable bioreactors were assembled on Earth in normal gravity
   conditions, and the same experiments were performed in parallel to
   samples in microgravity.

Figure 2.

   [53]Figure 2
   [54]Open in a new tab

   Preparation of muscle-on-a-chip devices for microgravity studies and
   experimental overview

   (A) The overview of muscle-on-a-chip preparation consists of seeding
   human muscle cells onto immobilized aligned nanofibrillar collagen
   scaffolds followed by cell seeding. After one day, the cell-seeded
   scaffolds are placed into bioreactors and filled with growth media. The
   bioreactor is then secured with screws for transportation to the ISSNL.

   (B) The experimental overview consists of incubating the bioreactor
   during transit to the ISSNL in plate habitats that provide temperature
   maintenance during transport to the ISSNL or in normal gravity
   conditions on Earth. Upon docking at the ISSNL, microgravity
   experiments began (day 0) with media change into an induction media
   that supports myotube formation. In some experiments, drugs were added
   to the media on day 0 and again on day 3. On day 7, the conditioned
   media were extracted, and the samples were stabilized in RNAlater or
   fixed in paraformaldehyde for analysis on by RNA sequencing, proteomics
   analysis, or myotube formation.

Microgravity alters myogenesis and genetic program in skeletal
muscle-on-A-chip

   We first quantified myotube formation from healthy donor-derived
   engineered muscle after 7 days of exposure to microgravity. Our results
   showed reduced myotube length, width, and fusion index in microgravity,
   compared to gravity ([55]Figure S2; [56]Table S1). This finding of
   reduced myogenesis in microgravity was then corroborated by
   transcriptional analysis. Following 7 days of differentiation and
   myotube formation, the muscle-on-a-chip bioconstructs in both
   microgravity and gravity from two healthy donors were assessed by RNA
   sequencing (RNA-seq). We detected 58,735 genes, of which 25,914 passed
   the threshold for expression. Transcripts of low count, expressed in
   less than two samples, were excluded. Samples were clustered by the
   gravity condition ([57]Figure 3A), where 107 genes were upregulated and
   286 genes were downregulated in microgravity samples with a
   false-discovery rate (FDR) < 10% ([58]Figure 3B and [59]Table S2).
   Database for Annotation, Visualization and Integrated Discovery (DAVID)
   analysis classified 104 upregulated genes in microgravity into the
   Cellular Component Gene Ontology (GO-CC) term. The analysis shows
   significant enrichment in genes encoding proteins located in various
   subcellular structures ([60]Figure 3C). Among these genes, 14 of them
   were annotated under the “mitochondrion” GO-CC ([61]Figure 3D). For
   example, mitochondrially encoded cytochrome c oxidase III (MT-CO3) has
   previously been reported to increase in muscle stem cells during aging
   ([62]Perez et al., 2022), which is associated with impaired
   regeneration. Since muscle has high metabolic function, proper
   mitochondrial biogenesis is necessary for regeneration.

Figure 3.

   [63]Figure 3
   [64]Open in a new tab

   Transcriptional differences between microgravity and gravity control
   biomimetic skeletal muscle

   (A) Principal-component analysis of the transcriptome of biomimetic
   skeletal muscle samples after 7 days in microgravity or control
   conditions (gravity).

   (B) Heatmap shows all differentially expressed genes after 7 days in
   microgravity, compared to gravity (FDR 10%).

   (C) Gene Ontology (GO) enrichment terms of differentially upregulated
   genes in microgravity (p < 0.05).

   (D) Heatmap shows the expression of genes found in the mitochondrion GO
   enrichment term.

   (E) Pathway enrichment terms of differentially upregulated genes in
   microgravity (p < 0.05).

   (F) Pathway enrichment terms of differentially downregulated genes in
   microgravity (p < 0.05).

   (G) Heatmap shows expression of top 20 differentially downregulated
   genes in microgravity (n = 2 donors per group).

   Pathway enrichment analysis also supported changes in metabolic
   pathways, including genes implicated in cholesterol, lipid, and fatty
   acid metabolism ([65]Figure 3E). Among them, fatty acid desaturase 1
   (FADS1) and mitochondrial matrix Lon peptidase 1 (LONP1) were also
   implicated in cellular responses to muscle regeneration ([66]Huang
   et al., 2020). Additionally, the pro-apoptotic gene, Bcl-2 related
   ovarian killer (BOK), along with cell survival and
   differentiation-associated genes, serum/glucocorticoid regulated kinase
   1 (SGK1) and NADPH oxidase 4 (NOX4) ([67]Dehner et al., 2008;
   [68]Pedruzzi et al., 2004; [69]Yakovlev et al., 2004), was upregulated
   in microgravity. SGK1 and NOX4 are associated with muscle mass
   maintenance and myoblast differentiation, respectively
   ([70]Andres-Mateos et al., 2013; [71]Youm et al., 2019). In particular,
   SGK1 regulates muscle homeostasis by the downregulation of proteolysis
   and autophagy ([72]Andres-Mateos et al., 2013). NOX4-derived reactive
   oxygen species (ROS) positively regulate myotube formation ([73]Youm
   et al., 2019), which may be important for muscle regeneration.

   Pathway analysis of decreased genes in microgravity highlights
   significant changes in biological processes, including genes involved
   in cell adhesion and cell fate decision pathways, including Wnt and
   Notch signaling pathway ([74]Figure 3F). The top 20 downregulated genes
   included those involved in cell adhesion (protocadherin gamma C3
   [PCDHGC3], transforming growth factor β induced [TGFBI], neuroligin 2
   [NLGN2], dystonin [DST], laminin α2 [LAMA2], collagen type XII alpha 1
   chain [COL12A1]), Notch signaling pathway (NOTCH2), and inflammation
   (prostaglandin-endoperoxide synthase 1 [PTGS1/COX1]) ([75]Figure 3G).
   In particular, NOTCH2, which is an important marker of muscle stem cell
   self-renewal and stem cell fate decision ([76]Yartseva et al., 2020),
   was significantly downregulated in microgravity. Downregulation of
   NOTCH2 gene expression in satellite cells leads to a decrease in their
   ability to differentiate ([77]Fujimaki et al., 2018), suggesting an
   impairment in the balance between proliferation and differentiation of
   primary human skeletal muscle cells in microgravity. The downregulation
   of NOTCH2 concurs with our in vitro findings of a significantly lower
   fusion index in the engineered muscles in microgravity ([78]Figure S2).
   Additionally, extracellular matrix (ECM) remodeling is associated with
   muscle regeneration ([79]Calve et al., 2010), and we observed
   significant downregulation of ECMs such as COL12A1, LAMA2, and heparan
   sulfate proteoglycan 2 (HSPG2) genes. Together, this analysis suggested
   that microgravity conditions induced a cellular phenotype that is
   associated with hallmarks of impaired regeneration, including metabolic
   remodeling, imbalance between proliferation and differentiation, and
   decreased cell-cell and cell-ECM adhesion.

   To further examine the effects of microgravity, we performed proteomic
   analysis of 200 proteins from the conditioned media of bioconstructs
   cultured in microgravity or gravity conditions for 7 days. In comparing
   the microgravity versus gravity samples with FDR <25%, we identified 5
   proteins that were secreted in greater abundance and 4 proteins that
   were of reduced abundance. The proteins of greater abundance in
   microgravity include Eotaxin-3 (CCL26), C-X-C motif chemokine ligand 16
   (CXCL16), growth differentiation factor 15 (GDF-15), tumor necrosis
   factor superfamily member 14 (LIGHT/TNSF14), and pupoid fetus (PF). Of
   these proteins, CCL26 and CXCL16 are chemoattractants known to induce
   immune cell infiltration and are associated with chronic inflammation
   ([80]Scholz et al., 2007; [81]Yamada et al., 2019). In addition, GDF-15
   is a biomarker for mitochondrial dysfunction and cellular senescence
   ([82]Fujita et al., 2016). The proteins of reduced abundance in
   microgravity include bone morphogenetic protein 4 (BMP-4), Resistin,
   C-X-C motif chemokine ligand 12 (CXCL12/SDF-1b), interleukin-16
   (IL-16), and CD40. CXCL12, in particular, is an important player in the
   maintenance of muscle and myogenesis ([83]Gilbert et al., 2019;
   [84]Puchert et al., 2016).

   To determine whether the secreted proteomes were related to their RNA
   expression levels, we compared their relative expression between
   microgravity and gravity samples. We observed that some differential
   protein changes generally correlated with their corresponding RNA
   expression level changes ([85]Figure S3). For example, GDF-15 was
   consistently increased in microgravity conditions at both the protein
   and RNA levels ([86]Figure S3), whereas BMP-4 and SDF-1b (CXCL12)
   levels were consistently downregulated in microgravity at both the
   protein and RNA levels ([87]Figure S3). Since GDF-15 has previously
   been reported as a diagnostic biomarker for conditions associated with
   impaired muscle regeneration such as aging ([88]Ito et al., 2018;
   [89]Kim et al., 2020), we searched for potential interacting genes of
   GDF-15 in microgravity samples. The protein-protein interaction (PPI)
   network was constructed by the STRING database using 107 genes that
   were upregulated with microgravity, including GDF-15. The unconnected
   genes were then removed. The resulting PPI network contained 57 nodes
   and 135 edges ([90]Figure S4), where a node represents a gene and an
   edge represents the relationship between the two inter-connected genes.
   From this analysis, we identified early growth response factor 1 (EGR1)
   and tribbles homolog 3 (TRIB3) as the first neighbors of GDF-15. These
   results suggest that GDF-15 and its immediate neighbors may play an
   important role in microgravity-induced impairment in myogenesis.

Microgravity partially mimics genomic features of impaired muscle
regeneration in sarcopenia

   Since impaired muscle regeneration is associated with clinical
   conditions such as sarcopenia, we assessed whether the muscle-on-a-chip
   in microgravity platform mimics regeneration-related signaling pathways
   in clinical sarcopenia muscle samples. We compared the transcriptome of
   human muscle tissue obtained from either healthy donors (denoted as
   “Normal”; [n = 4]), or those with sarcopenia, as defined by accepted
   cutoffs for the relative skeletal muscle (RSM) index (n = 4,
   [91]Table S3; [92]Figure 4A). RNA-seq detected 58,735 genes, of which
   29,313 passed the threshold for expression. In particular, 48 genes
   were upregulated and 113 genes were downregulated in sarcopenia samples
   with an FDR <25% ([93]Figure 4B; [94]Table S4). Among the top
   upregulated genes were those involved in apoptotic signaling pathways
   in response to DNA damage (Apoptosis enhancing nuclease [AEN] and
   PHLDA3), ECM organization (TIMP1, COL1A1, and COL14A1), and the
   VEGFA/VEGFR2 signaling pathway (GIPC1, CTGF, GPC1, and MAP2K3).
   Interestingly, more genes were downregulated in sarcopenia and they
   included genes that are implicated in Wnt signaling (mutated in
   colorectal cancers [MCC]) and associated with rhabdomyosarcoma (FHL2)
   ([95]Figure 4C). Importantly, an increase in apoptosis or impairment in
   skeletal muscle cell function could contribute to the development or
   progression of sarcopenia ([96]Chung and Ng, 2006; [97]Leeuwenburgh,
   2003; [98]Siu et al., 2005), in which the upregulation of markers such
   as AEN and PHLDA3 is implicated in p53-dependent apoptosis
   ([99]Cruz-Jentoft and Sayer, 2019; [100]Kawase et al., 2008,
   [101]2009).

Figure 4.

   [102]Figure 4
   [103]Open in a new tab

   Transcriptional similarities between clinical sarcopenia muscle and
   engineered skeletal muscle in microgravity

   (A) Principal-component analysis of the transcriptome in healthy vs.
   sarcopenia clinical samples from both sexes.

   (B) Volcano plot shows differentially expressed genes in sarcopenia,
   compared to normal conditions (p < 0.05, log2 FC > 1.0). Genes involved
   in DNA damage response, extracellular matrix organization, VEGFA-VEGFR2
   signaling pathway, and Wnt signaling are labeled. Genes showing
   statistically significant expression (FDR 25%) are indicated in red.

   (C) Heatmap shows the genes involved in Wnt signaling,
   rhabdomyosarcoma, DNA damage response, extracellular matrix
   organization, and VEGFA-VEGFR2 signaling.

   (D–F) Enrichment plots for custom gene set enrichment analysis (GSEA).
   The plots report positive enrichment of Old_MuSC_UP gene set (D) and
   negative enrichment of Sarcopenia_UP and Sarcopenia-DOWN gene sets
   (E–F) in microgravity, compared to gravity samples. The enrichment plot
   shows the gene set names (top) and the running enrichment score. The
   red curve indicates positive enrichment, whereas the blue curve
   indicates negative enrichment scores. Black bars indicate the positions
   of the gene set on the rank-ordered list in GSEA, and the
   signal-to-noise metrics are shown at the bottom.

   We next established a custom gene set library to assess whether
   engineered muscle in microgravity mimics features of sarcopenia from
   clinical samples. In brief, we used the up- and downregulated gene list
   from sarcopenia versus normal samples and converted the list into a
   gene set enrichment analysis (GSEA) analysis-compatible format. We also
   analyzed publicly available datasets derived from old adult muscle
   myoblasts (Gene Expression Omnibus repository: [104]GSE52699
   ) and generated an enriched gene set list compared to young myoblasts
   (see [105]Supplemental Information). As shown in [106]Figure 4D, custom
   GSEA results indicate that genes upregulated in old human myoblasts
   (“Old_MuSC_UP” gene set) were over-represented in the muscle-on-a-chip
   bioconstruct in microgravity, as indicated by the positive running
   enrichment score (RES) ([107]Figure 4D). Although some of the
   sarcopenia-related genes were significantly decreased in gravity,
   compared to microgravity conditions, the enrichment plot for
   Sarcopenia_vs_Normal_UP shown in [108]Figure 4E suggests that the genes
   upregulated by sarcopenia were not enriched by microgravity. However,
   the genes downregulated by sarcopenia were also downregulated with
   microgravity exposure ([109]Figure 4F). These custom GSEA results
   indicate that the muscle-on-a-chip bioconstructs in microgravity had a
   negative enrichment of genes that were downregulated in sarcopenia
   (“Sarcopenia_VS_Normal_DOWN” gene set [[110]Figures 4E and 4F]). These
   data reveal a similarity in the transcriptional signature between the
   clinical sarcopenia samples and engineered muscle in microgravity
   samples, in which the genes that were downregulated in sarcopenia
   muscle tissue were also downregulated in the engineered muscle in
   microgravity conditions, compared to those of the gravity condition
   ([111]Figure 4F). Conversely, the genes that were upregulated in
   sarcopenia were not enriched in the microgravity condition compared to
   those of gravity conditions ([112]Figure 4E). Overall, the custom GSEA
   results suggest that space flight mimics in part the genomic signature
   of old adult myoblasts and sarcopenia.

Drug treatment partially prevents adverse effects of microgravity

   We next used the muscle-on-a-chip platform to perform proof-of-concept
   drug screening studies. Recently published studies showed that a
   15-PDGH inhibitor (15-PDGH-i) could stimulate myogenesis as well as
   abrogate sarcopenia in preclinical models through the regulation of
   prostaglandin E[2] signaling ([113]Ho et al., 2017; [114]Palla et al.,
   2021). Besides 15-PDGH-i, we also examined insulin-like growth factor-1
   (IGF-1) in the muscle-on-a-chip platform. The effects of IGF-1 in
   promoting muscle myogenesis and inhibition of muscle atrophy are well
   established ([115]Alcazar et al., 2020; [116]Ascenzi et al., 2019;
   [117]Ma et al., 2009; [118]Machida and Booth, 2004; [119]Yu et al.,
   2015). To examine the potential use of these two drugs to regulate
   myogenesis, the engineered skeletal muscle-on-a-chip bioconstructs were
   treated with either IGF-1 or 15-PDGH-i for 7 days in microgravity
   conditions, followed by RNA-seq analysis. Through principal-component
   analysis, we found that, in microgravity, the drug-treated groups were
   generally more transcriptionally similar to the samples in gravity,
   rather than their microgravity counterparts ([120]Figure 5A). Out of
   107 genes that were upregulated in microgravity compared to gravity
   counterparts, 49 genes showed similar expression levels to those of
   gravity conditions with IGF-1 and 15-PDGH-i treatment, suggesting that
   the drugs could prevent the transcriptional upregulation effects of
   microgravity. In addition, IGF-1 and 15-PDGH-i treatment prevented the
   transcriptional upregulation effects of microgravity in 6 and 14 genes,
   respectively ([121]Figure 5B).

Figure 5.

   [122]Figure 5
   [123]Open in a new tab

   IGF-1 and 15-PDGFH inhibitor treatment can partially prevent
   microgravity effect

   (A) Principal-component analysis of transcriptome in 7 days in
   microgravity (MG) or gravity control conditions in the presence of drug
   treatment (IGF-1 or 15-PDGH-i) in biomimetic skeletal muscle samples.

   (B) The 4-way Venn diagram shows the number of differentially expressed
   genes that were upregulated in microgravity, compared to gravity
   conditions, or whose differential expression could be either prevented
   by IGF-1 or 15-PDGH-i treatment (gray) or not prevented by drug
   treatment (IGF-1 in blue and 15-PDGH-i in orange).

   (C) 4-way Venn diagram showing the number of differentially expressed
   genes downregulated in microgravity, compared to gravity conditions, or
   whose differential expression could be either prevented by IGF-1 or
   15-PDGH-i treatment (gray) or not prevented by drug treatment (IGF-1 in
   blue and 15-PDGH-i in orange).

   (D) Heatmap showing genes implicated in lipid biosynthesis, metabolic
   pathways, and adipogenesis.

   (E) Heatmap showing genes implicated in cell adhesion, Notch signaling,
   and Wnt signaling (n = 2 donors per group).

   Of the 286 downregulated genes with microgravity, 200 genes showed
   similar expression levels to those of gravity conditions with IGF-1 and
   15-PDGH-i treatment, suggesting that the drugs could prevent the
   transcriptional downregulation effects of microgravity. In particular,
   IGF-1 and 15-PDGH-i treatment prevented the transcriptional
   downregulation effects of microgravity of 13 genes and 19 genes,
   respectively ([124]Figure 5C). As noted earlier, pathway enrichment
   analysis highlighted an enrichment of genes associated with metabolic
   pathways including lipid and fatty acid metabolism in microgravity
   samples. We found that the drug treatment prevented a shift
   in metabolic profiles and maintained normal expression of genes
   involved in lipid biosynthesis (DHCR7, FDPS, HACD1) and metabolic
   pathways (MT-CO3, NNMT, GOT1, INPP5K, DHCR7, DHRS4, PHGDH, MMAB, FAHD1)
   ([125]Figure 5D). Moreover, we also observed that many genes implicated
   in cell adhesion, the Notch signaling pathway (NOTCH1, NOTCH2, NOTCH3,
   SPEN), and the Wnt signaling pathway (MACF1, USP34, LRP6, RECK,
   KREMEN1, TCF7, TRABD2A) were downregulated with space flight but were
   similar in expression level to gravity samples upon drug treatment
   ([126]Figures 5E, [127]S5A, and S5B). Overall, the drug-treated samples
   displayed a signature that more closely resembled gravity samples,
   compared to their microgravity counterparts.

   Since the addition of IGF-1 and 15-PDGH-i prevented some effects of
   microgravity at the transcriptional level, we further assessed their
   effect on proteins found in conditioned media at 7 days after drug
   treatment. Interestingly, we found that IGF-1 and 15-PDGH-i treatment
   prevented the ability of microgravity to increase GDF-15 protein levels
   ([128]Figure S5C; [129]Table S5). Additionally, 15-PDGH-i treatment
   protected CXCL16 and PF4 protein levels from rising to levels, similar
   to those of microgravity samples. Despite the ability of the drugs to
   prevent an increase in protein levels to reach those of microgravity
   samples, none of the proteins that were decreased in microgravity
   compared to gravity conditions could be normalized to the levels
   associated with gravity samples by the addition of drugs
   ([130]Figure S5D). We further tested whether the first neighbors of
   GDF-15 (EGR1 and TRIB3) are affected with drug treatment. As described
   earlier, EGR1 was not affected by drug treatment, whereas IGF-1
   addition prevented TRIB3 level from reaching that of microgravity
   samples ([131]Figure S5E). Taken together, our multi-omics analysis
   demonstrates that microgravity induces in engineered muscle a signature
   that reflects salient aspects of impaired myogenesis, which could be
   partially prevented by IGF-1 or 15-PDGH-i.

Discussion

   The salient findings of the work are the following: (1) an engineered
   muscle-on-a-chip platform facilitates the study of microgravity effects
   on myotube formation and muscle-related biological processes; (2)
   microgravity mimics salient aspects of impaired regeneration, based on
   RNA-seq and proteomic analysis; (3) microgravity effects on engineered
   muscle can be prevented in part by IGF-1 and PDGH-i addition; and (4)
   drug screening in microgravity using an engineered muscle-on-a-chip
   platform is feasible. These findings suggest that aspects of impaired
   muscle myogenesis can be modeled in just one week of exposure in
   microgravity and that drug screening in microgravity may be a useful
   platform for identifying potential drug targets that regulate
   myogenesis.

   The study of the effects of microgravity on engineered muscle-on-a-chip
   is an emerging area of research. Our findings using engineered muscle
   in microgravity concur with Vandenburgh et al., whose engineered avian
   muscle aboard the Space Transportation Shuttle for 12 days showed
   evidence of reduced myofiber size and cross-sectional area
   ([132]Vandenburgh et al., 1999). Additionally, a larger body of work
   derived from in vitro cultured myoblasts in microgravity or simulated
   microgravity further supports the finding that microgravity abrogates
   myogenic differentiation and myotube size, owing to cellular senescence
   and epigenetic modifications ([133]Furukawa et al., 2018; [134]Parafati
   et al., 2023; [135]Takahashi et al., 2021). Our findings of impaired
   myotube formation in the presence of microgravity are in alignment with
   these studies.

   Previous studies have reported multiple effects of spaceflight on human
   physiology. Most recently, large-scale multi-omics, systems biology
   analytical approaches profiled 59 astronauts from the National
   Aeronautics and Space Administration (NASA)’s GeneLab to determine
   transcriptomic, proteomic, metabolomic, and epigenetic responses to
   spaceflight, which highlighted a significant enrichment for
   mitochondrial processes and DNA damage ([136]da Silveira et al., 2020).
   Alterations in mitochondrial activity and impaired lipid metabolism are
   associated with muscle atrophy ([137]Nikawa et al., 2004; [138]Vitry
   et al., 2022). In line with these reports, our engineered
   muscle-on-a-chip platform exposed to microgravity for 7 days also
   demonstrated changes in mitochondrial genes ([139]Figure 3C), a shift
   in metabolism toward lipid and fatty acid biosynthesis
   ([140]Figure 3E), and an increase in apoptotic processes (BOK, SGK1,
   and NOX4). The existence of mitochondrial dysfunction was further
   supported by proteomic studies reporting an increase in GDF-15. GDF-15
   is a biomarker of dysfunctional mitochondria and cellular senescence
   ([141]Fujita et al., 2016). Increased GDF-15 levels are associated with
   muscle atrophy in cancer models, and GDF-15 treatment was shown to
   induce muscle atrophy in C2C12 myotubes ([142]Zhang et al., 2022).
   GDF-15 neutralization could restore muscle mass and function
   ([143]Kim-Muller et al., 2023). More importantly, the addition of
   pro-regenerative drugs such as IGF-1 and 15-PGDH-i could partially
   prevent the effects of microgravity on our engineered muscle-on-a-chip
   platform.

   Notch signaling is major regulator of myogenesis ([144]Luo et al.,
   2005). The disruption in Notch signaling in aging muscle compromises
   new myofiber formation and impairs tissue regeneration as a potential
   contributing factor to sarcopenia ([145]Arthur and Cooley, 2012;
   [146]Carlson et al., 2008; [147]Conboy et al., 2003). Notch inhibition
   also contributes to the dysregulation of cellular quiescence and can
   perturb the quiescent state of the muscle stem cells ([148]Bjornson
   et al., 2012; [149]van Velthoven and Rando, 2019). Notch signaling can
   also regulate the stability of a tumor suppressor, p53. The Notch-p53
   axis declines with age and is detrimental to cell survival and
   expansion ([150]Liu et al., 2018). Our data are consistent with the
   literature in demonstrating a downregulation in Notch signaling in
   microgravity ([151]Figure 3F). The loss of such a signal is correlated
   with reduced differentiation potential, as demonstrated by the
   impairment in myotube formation.

   More genes were downregulated than upregulated in both microgravity
   versus gravityand the comparison of sarcopenia versus normal muscle
   samples. Microgravity samples were more sarcopenia like in terms of
   genes that are important for apoptosis and Wnt signaling. The genes
   involved in apoptosis were upregulated, whereas the genes involved in
   Wnt signaling were downregulated in both microgravity and sarcopenia
   samples. Although the cellular response to gamma radiation and
   apoptosis processes increased with microgravity but were not prevented
   with IGF-1 or 15-PGDH-i treatment, the gene expression of Wnt signaling
   components including Lrp6 was prevented with drug treatment
   ([152]Figure 5E). Lrp6 is one of the two co-receptors involved in
   canonical Wnt signaling in vertebrates and conditional Lrp6 knockout
   reduced fiber size in mice ([153]Gessler et al., 2022; [154]Ren et al.,
   2021), highlighting its importance in muscle mass maintenance.
   Interestingly, microgravity increased the expression level of TRIB3,
   which negatively regulates cell survival and inhibits myogenic
   differentiation ([155]Kato and Du, 2007; [156]Saleem and Biswas, 2017).
   TRIB3 was identified in the PPI network as one of the two immediate
   neighboring gene of GDF-15. IGF-1 treatment prevented the elevation of
   TRIB3 gene expression in microgravity. In a mouse model, Trib3 is
   implicated in both age-related and food deprivation-induced muscle
   fiber atrophy and fibrosis ([157]Choi et al., 2019; [158]Shang et al.,
   2020). Furthermore, our findings align with a recent report that also
   demonstrated differential expression of TRIB3 and PCK2 under
   microgravity conditions ([159]Parafati et al., 2023), further
   supporting the impact of microgravity on myogenesis and metabolism.

   We acknowledge the limitations associated with this study. Owing to
   constraints associated with space research, only one launch was
   possible. For this reason, the data were derived from two primary human
   muscle donors as a measure of variance in the dataset. Furthermore, we
   acknowledge that measurements of muscle contractility could be
   informative, but such a functional assay would be most appropriate for
   engineered muscles that are more mature, which would have further
   increased the duration of the studies. Finally, the findings from the
   data in microgravity do not preclude the possibility of other
   contributing factors such as ionizing radiation. Despite these
   limitations, we believe that the observed effects of microgravity on
   engineered muscle are compelling and provide an avenue for future drug
   screening explorations in space for identifying drug targets that can
   enhance muscle myogenesis.

   In conclusion, we show that engineered muscle-on-a-chip bioconstructs
   exposed to microgravity induced prominent changes to their
   transcriptome that mimic aspects of impaired myogenesis. GSEA
   demonstrated a shift in metabolic pathways toward lipid metabolism,
   along with downregulated Notch signaling pathways and the increased
   expression of apoptotic genes. Importantly, microgravity exposure to
   engineered muscle-on-a-chip bioconstructs resembled some similar
   features to that of clinical sarcopenia muscle. The effects of
   microgravity could be partly prevented by the addition of IGF-1 or a
   15-PGDH-i. Together, this transcriptomic and proteomics analyses of
   microgravity effects using a muscle-on-a-chip platform demonstrate that
   microgravity mimics aspects of impaired myogenesis. This work further
   highlights the utility of microgravity as a unique environment for drug
   discovery.

Experimental procedures

Resource availability

Lead contact

   Further information and requests for resources and reagents should be
   directed to and will be fulfilled by the lead contact, Ngan F. Huang
   (ngantina@stanford.edu).

Materials availability

   This study did not generate new unique reagents.

Data and code availability

   The accession number for the RNA-seq raw data reported in this paper is
   [160]GSE238215
   . The accession number will be accesible to readers upon publication.
   All data may be made available by request to the corresponding author.

Flight experiment in microgravity

   The experiments were initiated in microgravity with the exchange of
   media into induction media consisting of DMEM supplemented with 2%
   horse serum and 1% penicillin/streptomycin (day 0), according to
   [161]Figure 2B. The media were exchanged again on day 3. The viability
   of samples was qualitatively monitored using an on-orbit phase-contrast
   microscope ([162]Figure S1F). On day 7, the conditioned media within
   the bioreactors were harvested and stored at −80°C. The
   muscle-on-a-chip devices within each chamber were either treated with
   RNAlater stabilization agent (Fisher Scientific) and then stored at
   −80°C or fixed in 4% paraformaldehyde (Fisher Scientific) for 15 min
   and then stored at 4°C. Samples were maintained on the ISSNL at the
   indicated temperatures and then returned to Earth on SpaceX Cargo
   Resupply Service 23 about two months later. Semi-synchronous ground
   control experiments were performed on a separate set of
   muscle-on-a-chip bioconstructs from the same donors in identical
   bioreactors and syringe media exchange procedures. The bioreactors in
   normal gravity conditions were cultured in conventional incubators but
   were handled otherwise in the same manner and with the same time
   stamps, as in microgravity.

Statistical analysis

   Statistical analysis of transcriptional and proteomics data is
   described earlier or in the [163]supplemental experimental procedures.
   Where applicable, data are shown as mean ± standard deviation.
   Statistical significance was accepted at p < 0.05.

Acknowledgments