Microgravity-induced constraints on melanin bioproduction: investigating E. coli metabolic responses aboard the international space station

TL;DR

This study shows that microgravity on the ISS reduces melanin production in E. coli, likely due to metabolic and stress-related changes.

Contribution

The study provides new insights into how microgravity affects microbial metabolism and bioproduction in space.

Findings

ISS-grown E. coli produced significantly less melanin than ground controls despite functional tyrosinase.

Microgravity caused elevated extracellular tyrosine and reduced bacterial viability in LSMMG experiments.

Proteomic and metabolomic analyses revealed oxidative stress and disrupted redox balance in microgravity.

Abstract

Space biomanufacturing using engineered microbes offers a sustainable approach for producing biomaterials, pharmaceuticals, and essential metabolites, critical for long-duration space missions. However, microgravity-induced physiological changes can alter microbial metabolism and biosynthetic efficiency. This study investigated the effects of microgravity on melanin biosynthesis in non-motile Escherichia coli aboard the International Space Station (ISS). Despite expressing functional tyrosinase, ISS-grown E. coli exhibited significantly lower melanin production than ground controls. Differential pulse voltammetry revealed high extracellular tyrosine in ISS samples, indicating inefficient substrate catalysis. Low Shear Modeled Microgravity (LSMMG) experiments in the Rotating Wall Vessel bioreactor confirmed reduced melanin production and bacterial viability. Proteomic profiling identified increased expression of membrane, transport, and stress-related proteins, while metabolomic analysis showed elevated trehalose and decreased glutathione, indicating oxidative stress and perturbed redox homeostasis. These findings highlight the impact of microgravity on microbial metabolism and provide insights for optimizing microbial biomanufacturing in extraterrestrial environments.

Figures

Overview of experimental setup, operations performed on the ISS and on ground and the post-flight data analysis.

The BRIC-PDFU hardware.A Opened BRIC-PDFU with five PDFUs and a DATA logger. B A 60 mm Petri plate is placed inside the chamber of the PDFU. Each PDFU has one fluid reservoir can be split into two (typically for a growth activation fluid and a chemical fixative to stop the growth prior to freezing). C The PDFU with 60 mm Petri plate inside is covered with the chamber lid. D Sealed BRIC-PDFU canister. E Canisters are frozen after the experiment is stopped.

Phenotypes of *E. coli* grown in petri dishes on the on ground (top) and ISS (bottom).A *E. coli* BL21(DE3) (pJV298) strain not containing *tyr1* gene; B The recombinant melanin-producing *E. coli* BL21(DE3) (pCB1005-Tyr1) constitutively expressing *tyr*1 gene.

A–C). We observed that the Ground samples supernatant had a reduced tyrosine-dependent peak current, while the ISS supernatant still had a significant tyrosine peak (Fig. [5C](#Fig5)). Compared controls performed without tyrosinase from the same experimental conditions, the tyrosinase-expressing ground sample depleted the tyrosine peak by 58.1% ( ± 13.2) while the ISS sample only reduced the tyrosine peak by 9.5% ( ± 4.6) (Supplementary Fig. 4D), indicating the decrease in tyrosine signal comes from changes to enzymatic consumption. Both samples contained DOPA, indicating tyrosinase activity i

Schematic diagram of the RWV bioreactor setup in both the LSMMG and Control 1 x *g* orientations.Fig. 7Bacterial Growth, Melanin Production, and Biofilm Formation in *E.coli* BL21 strains cultured in the RWV spaceflight-analogue cell culture system.A Bacterial cell density measured as log(CFU/mL) over time. B Melanin production quantified via OD490 absorption measurements from bacterial culture supernatants. Left: *E. coli* BL21(pCB1005-Tyr1), melanin-producing strain; Right: *E. coli* BL21(pJV298), negative control. C Comparison of cell pellets and supernatants collected every 3 hours between

Differentially abundant metabolites in recombinant *E. coli* TYR1 strain grown under ground control and ISS conditions.Each panel shows the relative quantitative intensity of an individual metabolite in ground control cultures (Control TYR1) and in samples grown aboard the International Space Station (ISS TYR1).

Full text

Introduction

The ability to harness microorganisms for biomanufacturing in space represents a transformative approach to sustaining long-term human missions beyond Earth^1–3^. Microbial systems offer a scalable and adaptive platform for producing essential biomolecules, materials, and therapeutics, reducing reliance on terrestrial resupply and enhancing mission sustainability^4–6^. This capability is particularly valuable for deep-space exploration, where resupply missions are impractical. By leveraging synthetic biology, microorganisms can be engineered to produce fuels, pharmaceuticals, food ingredients, and structural materials that are important to enable self-sufficiency in space habitats^7–9^. Notably, early pioneering studies began the transition from Earth-based industrial microbiology to the use of microorganisms in spaceflight biomanufacturing through the use of spaceflight analogue research to study secondary metabolite production^10^.

Unlike traditional manufacturing processes, microbial fermentation can operate in compact, closed-loop systems, making it well-suited for space habitats^5,11^. While studies onboard the ISS have demonstrated microbial synthesis of valuable compounds, such as antibiotics^12,13^, broad microbial biomanufacturing applications in space currently face both opportunities and challenges. Indeed, spaceflight imposes several constraints on microbial biomanufacturing. The microgravity environment alters microbial cellular behavior in unexpected ways, including global alterations in gene expression (such as those important for nutrient uptake and stress responses), and metabolite synthesis^4,14–17^, while limited fluid convection associated with this environment is predicted to reduce nutrient and oxygen diffusion, which could alter growth and productivity^4,15^. Additionally, biomolecule synthesis often requires a significant supply of substrates and feedstocks, yet transporting these molecules into microbial cells often introduces an added metabolic burden, which could be further exacerbated in spaceflight conditions^16,18,19^. Elevated radiation exposure in space also poses risks, generating oxidative stress that can lead to DNA damage and altered metabolism^20^. Addressing these challenges requires an essential first step: establishing on-demand synthetic biology production systems by characterizing the impact of space conditions on microbial host resiliency and their ability to consistently produce high yields of high-quality biomaterials or biomolecules.

Melanin, a secondary metabolite and multifunctional biopolymer, presents an ideal model system for evaluating microbial bioproduction in space. It is highly visible, enabling straightforward assessment of production efficiency, and is readily synthesized by microbial hosts^21–24^. Additionally, melanin has broad physicochemical properties, including radiation shielding, antioxidation, thermal stability, and metal chelation^25–28^, which make it an attractive potential biomaterial for space applications, from radiation protection to structural enhancement.

In this study, we engineered E. coli to express tyrosinase for melanin production and investigated its production when cultured under microgravity conditions aboard the ISS as compared to ground controls (Fig. 1). We conducted a series of biochemical, proteomic, and metabolomic analyses to evaluate the effects of spaceflight on microbial growth, melanin production, substrate utilization, protein expression and metabolic profiling. Key spaceflight results were validated using the microgravity analogue culture system, the Rotating Wall Vessel (RWV) bioreactor. By characterizing the impact of spaceflight and spaceflight analogue conditions on melanin biosynthesis, we aimed to identify key limitations imposed by the microgravity environment and propose strategies to enhance microbial bioproduction in space.Fig. 1. Overview of experimental setup, operations performed on the ISS and on ground and the post-flight data analysis.

Materials and Methods

Bacterial strains and growth medium for spaceflight and ground controls

The tyrosinase gene (tyr1) from Bacillus megaterium^29^ was cloned into the pCB1D5 plasmid containing the constitutive Anderson promoter J23106 and RBS B0032 as previously described in ref. ^22^. Melanin biosynthesis was conducted in the non-motile E. coli BL21 (DE3), E. coli (pCB1005-Tyr1) (melanin producing strain) and E. coli (pJV298) (negative control without tyr1 gene) were grown in modified M9 minimal medium consisting of minimal salts (40 mM K_2_HPO_4_, 10 mM NaH_2_PO_4_, 25 mM NH_4_Cl, and 2.5 mM Na_2_SO_4_), 25 mM NaCl, 1 mM MgSO_4_, 0.3 mM CaCl_2_, carbon sources (0.5% (w/v) glycerol and 0.5% (w/v) glucose), 0.2% (w/v) casamino acids, 0.2% (w/v) aspartate, 0.2X essential trace metals mix (50 mM FeCl_3_, 10 mM MnCl_2_, 10 mM ZnSO_4_, 2 mM CoCl_2_, 2 mM CuCl_2_, 2 mM NiCl_2_, 2 mM Na_2_MoO_4_, 2 mM Na_2_SeO_3_, and 2 mM H_3_BO_3_), 1 µM thiamine HCl, 2 g/L disodium tyrosine, 20 µM CuSO_4_ and 100 mg/mL chloramphenicol.

Biological Research in Canisters Petri dish Fixation Units (BRIC-PDFU) (Fig. 2) hardware was washed, leached, dried, greased and autoclaved before use. Overnight cultures of E. coli strains (control (WT) and melanin producing strain, (Tyr1)) were adjusted to 1 × 10^6^ CFU/mL and 500 μL was plated onto 60 mm Petri dishes containing 2 mL 1.5% agar. Plates were allowed to fully dry at room temperature prior to transfer into PDFU compartments. PDFUs were sealed and loaded into the BRIC hardware and kept at 4 °C until flight. Canister A housed five PDFUs containing E. coli control (WT) strains and one HOBO data logger while Canister B housed six PDFUs containing E. coli (Tyr1) strains. BRIC-PDFUs were kept four days at 4 °C before being launched to the ISS. Another set of BRIC-PDFUs as the ground control were maintained at the Kennedy Space Center by following the same integration protocol as the flight samples.Fig. 2. The BRIC-PDFU hardware.A Opened BRIC-PDFU with five PDFUs and a DATA logger. B A 60 mm Petri plate is placed inside the chamber of the PDFU. Each PDFU has one fluid reservoir can be split into two (typically for a growth activation fluid and a chemical fixative to stop the growth prior to freezing). C The PDFU with 60 mm Petri plate inside is covered with the chamber lid. D Sealed BRIC-PDFU canister. E Canisters are frozen after the experiment is stopped.

Experiment actuation on the ISS

One week after launch to the ISS, an astronaut initiated the experiment by injecting 8.5 mL of growth media into the plates through an external port, followed by incubation at 37 °C for 72 h under static conditions. BRIC-PDFUs were stored at −80 °C to halt the experiment and remained preserved until their return to Earth. Ground control experiments at the Kennedy Space Center (KSC) (Merritt Island, FL) were timed and identical to that on the ISS. Upon sample return to KSC, PDFUs from each compartment were removed, and Petri dishes were returned to −80 °C. Frozen samples were shipped to the US Naval Research Laboratory (NRL) (Washington, DC) for analysis. At their destination, samples were thawed, transferred to tubes, and stored at −80 °C until processing.

Differential Pulse Voltammetry (DPV) measurements

DPV measurements were performed on sample supernatants using a 1 mm gold standard electrode, a 4 cm platinum-wire counter electrode, and an Ag/AgCl reference electrode using a CHI 720E electrochemical analyzer (CH Instruments; Bee Cave, TX). Measurements were performed as previously described to approximate the concentrations of 3,4-dihydroxyphenylalanine (DOPA), and tyrosine^30,31^. After observing a peak associated with copper, we confirmed its presence by spiking in copper to samples and repeating the measurement. In each experiment, the potential was swept from 0 to 800 mV at a scan rate of 2 mV/s (1 mV step increment; every 0.5 s) with superimposed pulses (50 mV pulse amplitude; 0.2 s pulse period) to obtain data.

Protein analysis using SDS-PAGE

Frozen bacterial cultures (1 mL) were thawed, and cell suspensions normalized by adjusting the OD600 to 2.0. The normalized suspension (30 μL) was mixed with 10 μL of Bio-Rad 4X SDS-loading buffer under reducing conditions. The mixture was then boiled at 95 °C for 10 min and allowed to cool to room temperature. The entire prepared cell extract (40 μL) was loaded onto 4–15% SDS-PAGE gels and run at a constant voltage of 125 V for 50 min. Following electrophoresis, the gel was washed three times with 20 mL of Reverse Osmosis (RO) water for 5 min per wash, then stained in GelCode Blue stain for 1 h. The gel was subsequently de-stained in RO water for 1 h, followed by a water change and overnight de-staining. The final stained gel was visualized and imaged using a Bio-Rad imager. Gel was stored in distilled water at 4 °C prior to processing for proteomics.

Proteomics analysis

Stained SDS-PAGE gels were cut into 60 bands (10 per sample lane; Supplementary Fig. 1), and proteins were subjected to in-gel trypsin digestion. Gel pieces were washed, dehydrated with acetonitrile, reduced with dithiothreitol, alkylated with iodoacetamide, and digested overnight. Peptides were extracted using 2% formic acid in 50% acetonitrile, dried, and stored at −20 °C. Prior to LC-MS/MS, samples were reconstituted in 0.1% formic acid, and 3 µL of each was injected onto a C18 trap and analytical column (Thermo Scientific) using a 2 h gradient. Ultimate 3000 liquid chromatography instrument (Thermo, Sunnyvale, CA) coupled to Orbitrap Fusion Lumos mass spectrometer equipped with nanospray flex electrospray source was used for data analysis. MS/MS spectra were converted to Mascot generic files (mgf) format, merged by lane, and searched against the E. coli BL21 proteome using Mascot (Matrix Science, Inc., London, UK) with standard modifications. Spectrum counts were normalized using the standard Scaffold normalization procedure which normalizes the counts for each detected protein in a sample with respect to the total protein counts per analyzed sample to allow cross-sample comparison. Scaffold parameters were set to perform a t-test on the normalized weighted mass spectra assigned to the proteins for relative quantification. Results were validated and quantified in Scaffold with ≥2 peptides per protein identified, an estimated false discovery rate (FDR) of <1%, p-value <0.05, and fold change relative to ground samples. The quality assessment of samples for the proteomics workflow is described in Supplementary Method.

To survey the impact of the ISS environment on protein expression, we performed a comparative analysis of ISS and ground proteome profiles. As described above, proteins identified with p ≤ 0.05 between ISS and ground conditions were determined to be differentially modulated and subjected to subsequent protein sequence analyses. Functional annotations were inferred from similarities between UniProt. Functional enrichment analysis of differentially expressed proteins was performed with DAVID Functional Annotation tool (https://davidbioinformatics.nih.gov/tools.jsp)^32^. This approach identifies biological processes, molecular functions, and protein domains that are statistically overrepresented in the input protein set compared to the background proteome. Enrichment was evaluated using a modified Fisher’s exact test, and related annotation terms were grouped into functional clusters based on gene content overlap. Enrichment scores, defined as the –log₁₀ of the geometric mean p-value for terms in a cluster, were used to prioritize biologically coherent themes. Clusters with enrichment scores greater than 1.3 (equivalent to p <0.05) were considered significant.

Metabolomics analysis

The extraction solvent was prepared by adding 50 μL of a 10 mg/mL vanillic acid stock solution to 20 mL of ethyl acetate, resulting in a final vanillic acid concentration of 25 μg/mL in ethyl acetate. Metabolite extraction was performed on 1 mL ISS and ground E. coli cultures with 2 mL extraction solvent and vortexed. Samples were centrifuged at 10,000 × g for 2 min, and the supernatants were collected and set aside. Using a transfer pipette, two drops of 2 N HCl were added to each sample, followed by vortexing. Next, 2 mL of extraction solvent was added to each sample, and the mixture was centrifuged again under the same conditions. The resulting supernatants were collected and combined with the previously collected fractions. A rotary evaporator was set to 40 °C and spun at 177 rpm until all organic solvent was removed from each sample. Samples were reconstituted with methanol and syringe filtered and placed at −20 °C until injection into LC-MS.

Metabolite detection was performed on bacterial extracts using an U3000 Liquid Chromatograph (LC) system and a Diode Array Detector (DAD, Thermo Scientific, Sunnyvale, CA) coupled to a TripleTOF 5600 time-of-flight mass spectrometer (TOF-MS, AB Sciex, Framingham, MA). Sample extracts were transferred to 0.3 mL clear-snap 11 mm screw neck vials with bonded preslit PTFE/silicone septa (ThermoScientific, Langerwehe, Germany) for automatic injection into Phenomenex Luna 5 μm, C18(2), 100 Å, 100 × 2 mm LC column at 35 °C. The mobile phase consisted of solvent A (water + 0.1% formic acid) and solvent B (acetonitrile + 0.1% formic acid) at a flow rate of 0.3 mL/min and the pump pressure stabilized at ~46 bar. The gradient settings were 0–5 min 5% solvent B, 5–20 min solvent B equilibrating at 5%. The DAD monitoring was focused on 4 channels: 230, 254, 280, and 316 nm. The TOF-MS was set to negative electrospray mode with V-resolution, scanning from 100 to 1600 m/z. LC-UV-MS sample processing workflow is shown in Supplementary Fig. 2.

Chromatography and mass spectral data were processed with PeakView 2.2 software (Sciex, Framingham, MA). Single ion chromatograms were extracted based on the monoisotopic mass with a +/− 15 ppm window and peak areas were integrated using default software settings and normalized to the internal standard, vanillic acid (10 ng/mL). Identification of signature compounds was determined with LC-MS monoisotopic mass, retention times, and database comparison from EcoCyc^33^.

Bacterial strains and growth conditions in the spaceflight analogue RWV bioreactor

Single colonies of E. coli BL21(pJV298) (negative control) and E. coli BL21(pCB1005-Tyr1) (melanin-producing strain) were isolated from frozen stocks and separately inoculated into LB broth. Overnight cultures (750 µL per strain) were diluted 1:200 into 150 mL of fresh LB medium. Each diluted culture was then loaded into two disposable rotary wall vessel (RWV) bioreactors (50 mL capacity), one for the Low Shear Modeled Microgravity (LSMMG) condition and the other for the reoriented 1 x g (gravity) control, per strain. A total of four RWV bioreactors were mounted on rotators and incubated at 37 °C with continuous rotation at 25 rpm (Fig. 6). Samples were collected at T0, 3 h, 6 h, 9 h, 12 h, 15 h, 24 h, 27 h, 30 h, 33 h, 36 h, and 48 h. At each time point, bacterial cell density was assessed by measuring by both OD600 and serial dilution and plating for viable colony-forming unit CFU/mL.

Melanin production and biofilm formation during bacterial culture in the RWV

To quantify melanin amount, cultures were centrifuged at 7,000 rpm for 7 minutes, and the supernatant was analyzed using OD490. Photographs were also taken at each time point.

Statistical analysis was performed using R, employing a linear mixed-effects model (LMM) and ANOVA to evaluate the data. A linear mixed-effects model was fitted to account for repeated measurements over time: OD490 ∼ CFUlog × Condition × Hour + Strain + (1 | Strain) + (1 | Hour), where CFU_log_ (log-transformed bacterial density), Condition (1 × g vs. LSMMG), and Hour (time points) were included as fixed effects, while Strain and Hour were modeled as random effects to account for variability across strains and repeated measurements over time. Additionally, for strain-specific analysis, the data were filtered to include only the melanin-producing experimental strain, and the following model was fitted: OD490 ~ CFUlog × Condition × Hour + (1|Hour). This model focuses solely on the experimental strain, removing the negative control strain to assess melanin production under different conditions. ANOVA was performed to evaluate the significance of model terms.

Results

Melanin production from E. coli cells grown on the ISS and ground

Upon return to NRL, the melanin production of the bacterial strains flown on the ISS, as well as the ground controls, were characterized (Fig. 3). As expected, E. coli without the tyrosinase gene (E. coli BL21 (DE3) (pJV298)) did not produce any pigment (Fig. 3A). Notably, however, the melanin-producing E. coli strain (pCB1005-Tyr1) grown on the ISS synthesized significantly less melanin compared to the ground control (Fig. 3B), as visualized by the contrast in color (light brown vs black). Sequencing of tyr1 genes in pCB1005-Tyr1 plasmids from both ISS and ground control samples did not reveal any mutations (Supplementary Fig. 3). We considered two hypotheses that could contribute to the low yield of melanin production on the ISS: (1) lower tyrosinase gene expression, and therefore lower enzyme activity; (2) the precursor tyrosine was not effectively catalyzed by the tyrosinase due to substrate inaccessibility or altered redox conditions.Fig. 3. Phenotypes of E. coli grown in petri dishes on the on ground (top) and ISS (bottom).A E. coli BL21(DE3) (pJV298) strain not containing tyr1 gene; B The recombinant melanin-producing E. coli BL21(DE3) (pCB1005-Tyr1) constitutively expressing tyr1 gene.

Analysis of tyrosinase and tyrosine from the ISS samples

To test these hypotheses, we designed and performed two experiments. First, we examined tyrosinase production in each sample. Bacterial cells were normalized by cell densities and lysed by sonication. Soluble cell extracts were collected, and equivalent volumes were loaded onto an SDS-PAGE gel. Figure 4 shows that the concentration of tyrosinase in all samples was similar. It was also noted that cell lysates extracted from the frozen ISS samples with light brown color (Fig. 3B) quickly turned black (i.e., produced melanin) when sitting at room temperature for 10 minutes. We also observed that lysates from cells lacking the tyrosinase gene (Fig. 3A) did not turn black, consistent with our previous publication^22^. These results indicated that tyrosinase production in the E. coli strains was not affected by spaceflight and that the tyrosinase enzyme was still active within the cells at the conclusion of the experiment, retaining the ability to catalyze tyrosine into melanin upon return to Earth.Fig. 4SDS-PAGE protein expression analysis of soluble cell extracts prepared from E. coli BL21(DE3).Lane 1: E. coli BL21(DE3) only (negative control); Lane 2: the recombinant E. coli BL21(DE3) (pJV-Tyr1) (positive control). Expression of tyrosinase was induced by IPTG; Lane 3–5: the recombinant E. coli BL21(DE3) (pCB1005-Tyr1) from ground cultures; Lane 6–8: the recombinant E. coli BL21(DE3) (pCB1005-Tyr1) from ISS cultures. Tyrosinase was constitutively expressed in this strain.

Second, DPV measurements were performed to determine levels of tyrosine and DOPA, the intermediate molecule in the melanin biosynthesis pathway (Fig. 5A), within the supernatants of both the ISS and ground samples. This method collects current differences between a peak and trough voltage throughout a linear sweep of potential to identify current peaks that occur due to molecular electron transfer. Results are presented in Fig. 5B with each voltage window highlighted for melanin-associated molecules. These voltage windows were confirmed by comparison to control experiments without the tyrosinase gene and spiked media samples with the specific metabolite to be measured (Supplementary Fig. 4A–C). We observed that the Ground samples supernatant had a reduced tyrosine-dependent peak current, while the ISS supernatant still had a significant tyrosine peak (Fig. 5C). Compared controls performed without tyrosinase from the same experimental conditions, the tyrosinase-expressing ground sample depleted the tyrosine peak by 58.1% ( ± 13.2) while the ISS sample only reduced the tyrosine peak by 9.5% ( ± 4.6) (Supplementary Fig. 4D), indicating the decrease in tyrosine signal comes from changes to enzymatic consumption. Both samples contained DOPA, indicating tyrosinase activity in both samples (Fig. 5C). These findings suggest that the precursor tyrosine was not fully oxidized into melanin by the active tyrosinase in the ISS samples. Considering that precursor tyrosine is required to be actively transported into the bacterial cells, while tyrosinase was expressed intracellularly, we hypothesized that quiescent microgravity conditions on the ISS might decrease the transport and oxidation of tyrosine in the bacterial cells.Fig. 5. Differential pulse voltammetry measurements of ISS and Ground control samples.A Melanin synthesis pathway catalyzed by tyrosinase (Tyr1); (B) DPV measurement of tyrosinase expressing samples from the ground control (blue) and the ISS (green). Red-, green- and blue-shaded regions indicate voltage ranges where each redox-active molecule is expected to produce current. C Average current produced from the greatest peak current from the ground control (blue) and ISS (green) samples for each of the measured species. Error is standard deviation. Comparisons of DOPA – P-value of 0.0350, Tyrosine – P-value <0.0001, Cu – P-value <0.0001.

Analysis of melanin production from spaceflight analogue RWV cultures

We employed a Rotating Wall Vessel (RWV) bioreactor, spaceflight-analogue cell culture system, to confirm the differential melanin production observed between spaceflight and ground conditions and to investigate the conversion of tyrosine to DOPA and melanin (Fig. 6). Strains were cultured under low-shear modeled microgravity (LSMMG) or 1 x g condition^4^, with samples collected periodically to determine live bacterial cell density (CFU/mL) (Fig. 7A) and assess melanin production via OD490 (Fig. 7B). Unexpectedly, the viable counts of melanin-producing bacteria declined rapidly under LSMMG conditions, beginning at 15 hours, coinciding with the initial detection of melanin (Fig. 7A, left). By 48 hours, viable counts became undetectable. In contrast, under 1×g conditions, the same strain remained in stationary phase. The negative control strain, E. coli BL21(pJV298), which does not produce melanin, exhibited a slow and more gradual transition into the death phase compared to its melanin-producing strain (Fig. 7A, right).Fig. 6. Schematic diagram of the RWV bioreactor setup in both the LSMMG and Control 1 x g orientations.Fig. 7. Bacterial Growth, Melanin Production, and Biofilm Formation in *E.*coli BL21 strains cultured in the RWV spaceflight-analogue cell culture system.A Bacterial cell density measured as log(CFU/mL) over time. B Melanin production quantified via OD490 absorption measurements from bacterial culture supernatants. Left: E. coli BL21(pCB1005-Tyr1), melanin-producing strain; Right: E. coli BL21(pJV298), negative control. C Comparison of cell pellets and supernatants collected every 3 hours between 24 and 48 h. D Bacterial cells collected from membrane after 48 h of growth in the RWV. Top: RWV containing cultures; Middle: RWV after removing cultures; bottom: Membranes from RWV showing bacterial cells and melanin deposition. E Differential pulse voltammetry measurements of supernatants of E. coli BL21(pJV298) in the RWV. Comparisons of DOPA – P-value <0.0001, Tyrosine – P-value <0.001.

Nevertheless, findings from both strains indicate that LSMMG conditions accelerate the transition to the death phase for E. coli BL21 strains compared to 1 × g conditions, though the underlying mechanism remains unknown. We measured melanin level from supernatants by measuring optical density at 490 nm (Fig. 7B). For the melanin-producing strain, the first detectable signs of production appeared at 15 hours post-inoculation in the RWV, reaching the peak at 48 hours (Fig. 7B, left), Notably, the melanin level was significantly higher under 1 × g condition as compared to LSMMG (p<0.001), a finding consistent with the observation from the experiment conducted aboard the ISS (Fig. 3B). Specifically, the melanin levels in the supernatant under 1 x g remained high at all time points, exhibiting an exponential increase (Fig. 7B, left). The maximum OD490 ratio difference between 1 × g and LSMMG reached up to 5.5-fold, emphasizing the substantial melanin accumulation in the medium under 1 x g gravity condition. Although melanin was only marginally detectable in the supernatant of the melanin-producing bacteria under LSMMG conditions, the cell pellets consistently appeared black after 24 h of culture, indicating that melanin was synthesized and accumulated intracellularly (Fig. 7C). Notably, the size of the cell pellets remained relatively consistent, suggesting that cell death was not due to cell lysis.

We further analyzed the overall trend. Statistical analysis revealed that the condition (LSMMG vs. 1 × g) was the primary factor affecting melanin accumulation, with additional contributions from viable bacterial density (CFU/mL) and time (Hour) (Fig. 7A, B). After reaching its peak, bacterial density declined more sharply under LSMMG, whereas populations remained higher under 1 x g (Fig. 7A left). This growth suppression in LSMMG may have contributed to the reduced melanin levels observed in the supernatant as melanin accumulation (OD490) was positively correlated with bacterial density across all time points (p <0.001) (Fig. 7A, B). While both the control and melanin-producing strains showed a decline in bacterial density in LSMMG conditions following peak growth, the melanin-producing strain exhibited a more pronounced decrease (Fig. 7A). This suggests that early-stage defects in melanin biosynthesis and transport pathways may increase internal stress, promoting cell death under LSMMG conditions. This, in turn, could lead to reduced melanin release. Overall, our results highlight the significant impact of spaceflight-analogue condition (LSMMG vs. 1 × g) on bacterial growth dynamics as well as the melanin biosynthesis pathway and its associated transport processes. A significant density–condition–time interaction (p <0.001) further indicates that melanin accumulation follows distinct temporal patterns influenced by gravitational forces (Fig. 7A, B). The negative control E. coli BL21(pJV298) was confirmed to exhibit no pigment production under both LSMMG and 1 x g conditions (Fig. 7C right).

To assess biofilm formation, membranes from the interior of the RWVs were collected after culture removal at 48 h. Biofilms formed under the 1 × g condition were visibly thicker than those formed under LSMMG, along with the greater deposition of melanin polymer onto the membrane at 1 x g compared to the LSMMG condition (Fig. 7D). This result aligned with phenotypic observations from the spaceflight experiment conducted aboard the ISS (Fig. 3B).

We repeated the DPV experiment to measure the relative amounts of DOPA and tyrosine within the supernatants of both the 1 x g and LSMMG samples. The LSMMG samples exhibited significantly higher tyrosine and lower DOPA level than 1 x g samples, suggesting impaired conversion of tyrosine to DOPA by tyrosinase or reduced extracellular release of DOPA under the LSMMG condition (Fig. 7E). Correlated with the similar finding from the ISS samples (Fig. 5C), this result implied that, for reasons yet to be determined, the biochemical pathway involved in melanin synthesis and its associated transport process may be incomplete or less efficient under microgravity conditions, ultimately leading to reduced melanin levels in the media.

Proteomics analysis of melanin producing E. coli cells grown on the ISS and ground

To investigate the global protein-level response of melanin-producing E. coli to ISS conditions, we extracted total proteins from three biological replicates of ISS-grown and ground control samples and subjected them to LC-MS for proteomic analysis (Supplementary Table 1). Notably, tyrosinase (the lower band in Fig. 4) was identified as the second most abundant protein, with an average abundance of 1,078 in ISS samples and 1,493 in ground samples (Table 2). This result clearly indicates that the heterologous Tyr1 gene was highly expressed and efficiently translated under both conditions. The most abundant protein identified was (the upper band in Fig. 3 and Supplementary Fig. 1), a major E. coli porin, OmpF, that facilitates the diffusion of hydrophilic molecules ( <600 Da)^34^, including potentially tyrosine, with average abundance of 1,395 in ISS samples and 2,035 in ground samples. While both tyrosinase and OmpF showed lower trends in their abundance in ISS samples compared to the ground controls (Table 2), the differences were not statistically significant (p = 0.1 and 0.19, respectively), indicating that any impact of the microgravity environment on the translation of these two proteins at the time of profiling is likely modest. A comparative analysis of protein abundance between ISS and ground control samples identified 154 differentially expressed proteins (p <0.05, fold change > 1.5) in response to microgravity (Supplementary Table 1). Of these, 149 were upregulated, while only 5 were downregulated. We performed functional enrichment analysis on these differentially produced proteins using the DAVID bioinformatics tool. Clustering revealed four significantly enriched groups (enrichment score > 1.3, Table 1). The most significantly enriched cluster included TonB-dependent outer membrane receptors, which facilitate high-affinity uptake of scarce or large molecules^35^. The second cluster comprised fumarate lyase and L-aspartase family enzymes, indicating a shift toward anaerobic or microaerobic metabolism^36^ perhaps due to altered oxygen dynamics in the spacecraft culture system. The third cluster featured glutathione S-transferases, enzymes that detoxify reactive oxygen species^37^, suggesting activation of oxidative stress responses triggered by space environment. The fourth cluster was enriched in enoyl reductase domains associated with polyketide synthase-like proteins, possibly involved in membrane lipid remodeling or secondary metabolite production for cellular protection^38^. Collectively, these data suggest that E. coli responds to spaceflight-induced stress by enhancing nutrient transport, reconfiguring energy metabolism, activating antioxidant defenses, and modifying membrane composition.Table 1. Functionally enriched protein clusters in E. coli grown in the ISSEnrichment ClusterAccession NumberProtein NameEnrichment ScoreP-ValueFold ChangeProtein Abundance (Mean ± SD)ISS GroundOuter membrane nutrient receptors (TonB-dependent)A0A140NC22TonB-dependent receptor plug2.240.0095inf1.7 ± 0.60.0 ± 0.0A0A140NB30TonB-dependent siderophore receptor2.240.00281821.0 ± 5.01.0 ± 1.0A0A140NC63TonB-dependent receptor2.240.0347.12.7 ± 1.20.3 ± 0.6A0A140N564Signal recognition particle2.240.00473.825.7 ± 3.26.7 ± 4.5A0A140NCT3TonB-dependent siderophore receptor2.240.000131.861.0 ± 2.032.7 ± 2.9Anaerobic/microaerobic metabolism (fumarate/aspartase)A0A140NBG7Fumarate hydratase class II2.010.011199.3 ± 3.50.3 ± 0.6A0A140NCJ7Argininosuccinate lyase2.010.0341.749.0 ± 9.629.3 ± 4.5A0A140NGN0Aspartate ammonia-lyase2.010.0161.746.3 ± 4.027.7 ± 7.2Oxidative stress response (glutathione S-transferase)A0A140N5L4Glutathione transferase1.610.0148.13.0 ± 1.00.3 ± 0.6A0A140N7N1Glutathione S-transferase domain protein1.610.0374.17.0 ± 2.61.7 ± 1.5A0A140NBK8Glutathione S-transferase domain protein1.610.0432.711.0 ± 3.64.0 ± 1.0A0A140NAB8Glutaredoxin, GrxB family1.610.00492.211.7 ± 1.55.3 ± 1.2Redox/lipid metabolism with PKS-like domainsA0A140NBB5Alcohol dehydrogenase zinc-binding domain protein1.410.0028104.7 ± 1.20.3 ± 0.6A0A140NCB1Alcohol dehydrogenase zinc-binding domain protein1.410.00812.924.7 ± 0.68.3 ± 5.9A0A140N5R4Quinone oxidoreductase, YhdH/YhfP family1.410.0011.627.7 ± 1.518.0 ± 1.7A0A140N870Alcohol dehydrogenase GroES domain protein1.410.00950.560.3 ± 7.5117.0 ± 19.7

In addition to the enriched protein groups, several individual proteins exhibited changes in the ISS samples. Several transport-related proteins were significantly upregulated in ISS samples (Table 2), including a cationic amino acid ABC transporters, a periplasmic binding protein/LacI, and a taurine ABC transporter (absent in the ground control). This suggests that bacterial cells upregulated specific transport mechanisms to compensate for potential reduced transport efficiency in microgravity that could result in decreased amino acid uptake and potential nutrient stress. However, other possibilities also exist to explain these findings (alone or in combination with a potential transport mechanism), and are discussed below. Notably, among the five downregulated proteins in the ISS samples, one is a transporter protein: the membrane fusion protein (MFP) subunit, a key component of the tripartite efflux pump complex^39^. Several stress response proteins were also differentially produced in ISS samples. Universal stress protein E and a cyclic di-GMP-binding protein were upregulated by 3.5- and 8.1-fold, respectively, indicating a possible adaptive shift to biofilm formation in response to potential microgravity-induced nutrient limitations. Notably, the aerobic respiration control sensor protein and cytochrome d ubiquinol oxidase were upregulated 22-fold, suggesting that bacterial cells responded to potential oxygen limitation by stimulating proton motive forces for energy generation. Furthermore, key oxidative stress defense enzymes, including superoxide dismutase, NAD(P)H dehydrogenase, glutaredoxin were upregulated more than two-fold. Finally, DNA synthesis/repair proteins were produced at higher levels. DNA polymerase I was more than twice as abundant in ISS samples, DNA ligase was 3.9 times more abundant, and exodeoxyribonuclease III was uniquely detected in ISS samples. ATP-dependent RNA helicase RhlB, which facilitates the removal of damaged RNA under stress conditions, was upregulated 4-fold. Many of these proteins are under the regulation of the evolutionarily conserved protein, Hfq, a known global regulator of bacterial responses to spaceflight and LSMMG culture conditions^4,17^. These findings suggest that exposure to stressors like microgravity and cosmic radiation triggered oxidative stress in bacterial cells, leading to DNA and protein damage and subsequent impacts on cellular metabolism, viability, and nutrient uptake.Table 2. Selected differentially produced proteins in E. coli grown on the ISS relative to ground controls. Proteins were selected based on functional relevance to transport and membrane-associated processes, stress response, and DNA synthesisAccession numberProtein nameFold changeP valueProtein abundanceISSGroundTyrosinase0.70.11,0781,493Porin (OmpF)0.70.191,3952,035Transporter and membrane proteinsA0A140NHP9Periplasmic binding protein/LacI transcriptional regulator4.50.001812.00 ± 1.412.67 ± 1.25A0A140NBX8Cationic amino acid ABC transporter, periplasmic binding protein4.40.00511.00 ± 1.632.33 ± 1.25A0A140NCZ7Taurine ABC transporter, periplasmic binding proteininf0.0433.00 ± 1.410.00 ± 0.00A0A140NBE4Tol-Pal system protein TolQinf0.000114.00 ± 0.000.00 ± 0.00A0A140N9V3Cationic amino acid ABC transporter, periplasmic binding protein3.40.02811.00 ± 4.003.00 ± 1.00A0A140NCH5Efflux transporter, RND family, MFP subunit0.60.044.50 ± 0.487.00 ± 1.00Stress responseA0A140N3P8Aerobic respiration control sensor protein220.0257.33 ± 2.620.33 ± 0.47A0A140NA70Cytochrome d ubiquinol oxidase, subunit II4.50.00434.33 ± 0.471.00 ± 0.82A0A140N993NAD(P)H dehydrogenase (quinone)4.30.002422.00 ± 2.165.00 ± 2.94A0A140NEM4Superoxide dismutase3.70.0006151.00 ± 4.3214.00 ± 3.74A0A140N4Z5Cyclic di-GMP-binding protein8.1< 0.000109.67 ± 0.471.00 ± 0.00A0A140NC74Universal stress protein E3.50.01911.33 ± 2.623.00 ± 1.41A0A140NAB8Glutaredoxin, GrxB family2.20.004911.67 ± 1.255.33 ± 0.94DNA synthesisA0A140N7B9Exodeoxyribonuclease IIIinf0.00373.33 ± 0.470.00 ± 0.00A0A140N7D0DNA ligase3.90.00298.67 ± 0.472.00 ± 1.41A0A140NER9DNA polymerase I2.40.03724.67 ± 4.1910.33 ± 5.31A0A140SSA5ATP-dependent RNA helicase RhlB4.00.000184.00 ± 0.001.00 ± 0.00

Metabolomic analysis of melanin producing E. coli cells grown on the ISS and ground

To investigate the impact of ISS conditions on the metabolism of engineered melanin-producing E. coli and its potential effects on melanin biosynthesis, we developed a metabolomic analysis protocol to extract and analyze metabolites from ISS and ground control samples using HPLC and LC-MS^40^. Chromatographic profiles and MS data features revealed distinct metabolic signatures between the two conditions, with condition-dependent peaks identified (Supplementary Fig. 5), indicating alterations in metabolite synthesis in response to conditions on the ISS. We focused on a subset of metabolites that are well-documented in E. coli metabolism or associated with stress responses. In total, we identified nine metabolites that were differentially abundant (by a factor of more than two-fold) in ISS-flown samples using the EcoCyc database. Four of these were more abundant, and five were less abundant compared to the ground control (Fig. 8). Metabolites in the former group include α,α-trehalose, a well-known stress response molecule^41^, which was observed only in ISS samples, suggesting that E. coli enhanced its biosynthesis in response to flight on the ISS. In contrast, glutathione, one of the most abundant thiols in bacteria, which plays a critical role in protecting against osmotic and oxidative stress^42^, was significantly reduced in ISS samples, along with L-cysteine, a key precursor for glutathione biosynthesis. Two other associated metabolites exhibited opposite trends. 2-hydroxy-5-oxoproline, an intermediate in the glutathione cycle, and L-glutamate, a precursor for glutathione synthesis, were significantly upregulated. These findings suggest that redox homeostasis was perturbed under microgravity, potentially leading to increased consumption of glutathione and altered glutathione metabolism due to oxidative stress. Finally, amino acid metabolism was affected by spaceflight. L-phenylalanine was accumulated, while L-aspartate and L-proline were significantly reduced in ISS samples, indicating a shift in amino acid utilization and metabolic regulation under spaceflight conditions.Fig. 8. Differentially abundant metabolites in recombinant E. coli TYR1 strain grown under ground control and ISS conditions.Each panel shows the relative quantitative intensity of an individual metabolite in ground control cultures (Control TYR1) and in samples grown aboard the International Space Station (ISS TYR1).

Discussion

The objective of this study was to investigate the effects of the microgravity environment of spaceflight on biomolecule production from engineered microorganisms. Growth of E. coli producing the black pigment melanin, a secondary metabolite of economic importance, provides an excellent platform to characterize the impact of the microgravity environment on adaptation mechanisms of this model organism and to measure melanin bioproduction in this unique environment. The results of this study reveal that melanin production by engineered E. coli during culture on the ISS was significantly reduced compared to ground controls, despite the presence of. active tyrosinase enzyme. Our findings suggest that the microgravity environment may have induced alterations in molecular transport, redox environment and metabolic burden, rather than changes in enzyme expression or activity, which may have played a key role in limiting melanin biosynthesis. These results provide valuable insights into microbial adaptation to space conditions and highlight potential challenges for microbial biomanufacturing in extraterrestrial environments.

An intriguing hypothesis for the observed reduction in melanin production is that microgravity conditions negatively impacted the transport and/or conversion of tyrosine, the precursor required for melanin biosynthesis. Differential Pulse Voltammetry (DPV) analysis showed a significant accumulation of tyrosine in the culture supernatants of ISS-grown E. coli compared to ground controls, suggesting that extracellular tyrosine was not efficiently taken up by the cells. On Earth, gravity facilitates fluid mixing through sedimentation and density-driven convection. However, in microgravity environments, diffusion becomes the primary process driving the movement of nutrients and substrates that support growth and molecule production for non-motile microorganisms^15,43^. It is noteworthy that OmpF, a major porin protein that facilitates the diffusion of hydrophilic molecules across the bacterial membrane with minimal substrate specificity^44,45^, was potentially downregulated under ISS conditions. While tyrosine transport does not exclusively depend on OmpF, it can passively diffuse through OmpF pores, as its molecular weight (~ 181 Da) is well within the diffusion limit of OmpF (<600 Da)^46^. Similar to OmpF, the abundance of the expressed tyrosinase was modestly lower in ISS samples relative to ground controls (0.7-fold), although the difference did not reach statistical significance (p = 0.1). Given the limited sample size (three biological replicates in the proteomics analysis) and variability inherent to spaceflight studies, this trend may still reflect a biologically relevant effect of the space environment on heterologous gene expression or protein translation. Therefore, the potential reduction in both OmpF and Tyr1 expression under spaceflight conditions may have impacted efficiency of tyrosine diffusion and its subsequent catalysis, potentially contributing to the observed decrease in melanin production. It is hypothesized that non-motile cells (like E. coli BL21(DE3) in this study) might experience a substrate-deficient “depletion zone” surrounding themselves due to limited nutrient flow due to low fluid shear conditions, leading to altered growth rates and phenotypes in the space environment^4,14,43^. Thus, the possibility exists that such a depletion zone resulting from the microgravity environment of the ISS might have prevented the melanin-producing E. coli cells from effectively accessing and utilizing tyrosine in the growth medium. Proteomics data revealed significant upregulation of transport-related proteins, particularly TonB-dependent receptors and ABC transporters in ISS samples, suggesting that bacteria attempted to compensate for decreased transport efficiency or nutrient depletion by increasing expression of alternative uptake pathways. On the other hand, the membrane fusion protein (MFP), a key component of the efflux transporter responsible for expelling various substrates including antibiotics, toxins, and metabolic byproducts^39^, was found to be downregulated in the ISS samples, suggesting that bacterial cells may respond to spaceflight-induced stress by conserving metabolic energy, potentially impacting melanin biosynthesis. Tyrosine is mainly transported into bacterial cells from the environment through two transporters: AroP and TyrP, the activities of which are driven by the proton motive force^47^. However, these two proteins were not detected in the shot gun proteomics analysis, likely due to their low abundance and transmembrane nature. Therefore, it remains unclear whether the expression of these transporter proteins was affected by the spaceflight condition.

The microgravity simulation experiment using the RWV bioreactor revealed that melanin-producing E. coli not only synthesized significantly less melanin under LSMMG conditions but also exhibited a decline in colony-forming units over time compared to the 1 × g reoriented controls. Notably, the bacterial cell pellets under LSMMG appeared black and maintained consistent sizes, despite lower OD490 readings in the supernatant relative to 1 × g controls. This suggests that melanin synthesis still occurred under LSMMG but was retained intracellularly. The intracellular accumulation may have increased internal pressure, contributing to cellular stress and subsequent death. This hypothesis will require further validation using electronic microscopy and viability assays. While we were not able to assess cell viability aboard the ISS due to the operational constraints, we observed a similar reduction in melanin production, suggesting that the high metabolic burden imposed by microgravity may have also contributed to decreased bacterial fitness and overall reduction of melanin production in space.

As in any environment, bacteria cultured during spaceflight may adapt to anaerobic or microaerophilic conditions and alter their metabolism and other characteristics in response to reduced oxygen availability. This reduction could result from the absence of convection currents in microgravity, differences in aeration between flight and ground controls, and elevated oxidative stress^4,48–51^. These findings align with our proteomic analysis and indicate that ISS-grown melanin-producing E. coli activated a range of stress response pathways, including significant upregulation of anaerobic/microaerobic metabolic enzymes, aerobic respiration control sensor protein, universal stress protein E, DNA repair enzymes, and oxidative stress response proteins (Tables 1 and 2). These findings are in agreement with previous studies that have shown a wide range of microorganisms alter their expression of stress response proteins, DNA repair and metabolic enzymes in response to both spaceflight and analogue culture^4^. On the other hand, melanin biosynthesis requires oxygen to catalyze oxidation of substrate tyrosine and polymerization^52,53^. While it is possible that low oxygen transfer rate that could have occurred in the microgravity environment^48,54^ and elevated oxidative stress contributed to the observed decline in melanin synthesis and cell viability under ISS and LSMMG culture conditions, our current data do not conclusively support these possibilities.

Although numerous studies have investigated the effects of spaceflight on microbial metabolism^4,16,55,56^, reports on the effects on the metabolome remain limited, with previous reports primarily focusing on secondary metabolites from Aspergillus species^57–59^. To our knowledge, our study is the first to perform a metabolomic analysis identifying differentially synthesized metabolites in bacteria in response to ISS conditions. Notably, the results from this analysis (Fig. 8) align closely with our proteomics data (Tables 1 and 2), which indicate regulatory responses related to stress adaptation and amino acid metabolism.

For example, the induction of α,α-trehalose correlates with the upregulation of universal stress protein (UspE), suggesting the activation of bacterial stress response machinery to counteract osmotic stress under microgravity. Similarly, the differential production of metabolites associated with glutathione, a hallmark of the oxidative stress response, aligns with the upregulation of oxidative stress-related proteins, including superoxide dismutase, glutaredoxin, NAD(P)H dehydrogenase, and glutathione reductase. This consistency across metabolic and proteomic data suggests that redox homeostasis was perturbed, potentially triggering a shift in bacterial stress adaptation mechanisms.

Additionally, quantitative changes in key amino acids together with the upregulation of oxidative stress response proteins suggest that microgravity may disrupt intracellular amino acid metabolism. The sulfhydryl group (-SH) of cysteine is highly redox-active and forms the core of glutathione’s antioxidant function^60^, while proline, a precursor of glutamate, contributes to oxidative stress protection through maintenance of intracellular redox homeostasis^61^. Upregulation of oxidative stress proteins (e.g., superoxide dismutase, flavodoxin/ferredoxin–NADP reductase, and glutaredoxin) indicates that reactive oxygen species were generated under microgravity conditions. This oxidative pressure likely oxidized cysteine and promoted proline catabolism, thereby depleting intracellular pools of cysteine, proline, and glutathione, while indirectly increasing glutamate levels. Collectively, these metabolic shifts suggest that E. coli prioritized stress survival over secondary metabolite production, ultimately resulting in reduced melanin biosynthesis in the microgravity environment aboard the ISS. Although the current metabolomic analysis provides a meaningful initial insight into microbial metabolic adaptations in space, further comprehensive profiling and pathway-level analyses are needed to fully understand the broader metabolic shifts in E. coli under microgravity.

Along these lines, it is worth noting that a number of the proteins differentially regulated in ISS samples, including membrane transporters, stress-response proteins and biofilm-associated proteins, fall into functional categories known to be influenced by the global RNA chaperone protein Hfq. Hfq facilitates small RNA–mRNA interactions and has been implicated in regulating a diverse group of genes and proteins, including those important for membrane transport, stress adaptation, and metabolic reprogramming in E. coli and other bacteria^62–65^ Notably, previous studies have identified Hfq as a master response regulator of the low-fluid-shear response in diverse bacterial species cultured in spaceflight and/or low-shear modeled microgravity^4^, including Salmonella typhimurium^17^, Pseudomonas aeruginosa^49^, and Staphylococcus aureus^66^. While our data do not directly demonstrate Hfq involvement, its established role in microgravity (and microgravity analogue)-associated responses across multiple bacterial species suggests that it may also contribute to the regulatory shifts observed under ISS conditions.

While E. coli does not natively produce melanin, it has been genetically engineered to produce this compound from a single carbon source by introducing a tyrosinase-encoding gene in strains lacking the sugar phosphotransferase system and TyrR repressor, thereby utilizing intracellular tyrosine^67^. Of particular relevance to our study, Jung et al recently applied genetic engineering strategies to modulate Hfq expression levels in E. coli in order to amplify the growth rate of the organism and its production of L-tyrosine^68^, an important compound for development of economically valuable bioproducts, including melanin^69–72^. In addition, other bacteria related to E. coli have been shown to use Hfq to regulate their native melanin biosynthesis by affecting the availability of precursors, like pyomelanin.^73^

Taken together, the reduced melanin production observed in ISS-grown E. coli can be understood as the combined outcome of disrupted substrate transport, altered redox homeostasis, oxygen limitation, and increased metabolic burden under microgravity. In the absence of gravity-driven convection, nutrients such as tyrosine become diffusion-limited, creating local depletion zones around non-motile cells and hindering precursor uptake. Simultaneously, microgravity perturbs intracellular redox balance, evidenced by oxidative stress responses and shifts in amino acid metabolism, which redirects cellular resources toward stress survival rather than secondary metabolite synthesis. Reduced oxygen availability further constrains melanin biosynthesis, an oxygen-dependent process. These interacting constraints, amplified by the metabolic cost of heterologous tyrosinase expression and possible regulatory effects mediated by global stress-response pathways, collectively impose a systems-level bottleneck that limits metabolic flux through the melanin pathway during spaceflight.

Despite the insights gained from this study, several limitations should be acknowledged. First, operational constraints aboard the ISS prevented direct measurement of cell viability, limiting our ability to decouple reduced melanin production from potential decreases in bacterial fitness. Second, the proteomic analyses were performed with modest sample sizes (triplicates), which may limit statistical power and obscure subtle but biologically meaningful differences in protein abundance. Additionally, as with many spaceflight experiments, it remains challenging to attribute observed phenotypes exclusively to microgravity, given the simultaneous presence of other spaceflight-associated stressors such as radiation exposure, unique aeration conditions, and launch or re-entry forces. These inherent limitations highlight the need for future experiments with expanded replicates, improved on-orbit analytical capabilities, and refined ground-based analogues to more precisely dissect the mechanistic drivers of microbial responses in space.

Results from this study highlight a fundamental challenge for biomanufacturing in space, namely, the potential for microgravity to impair substrate transport and metabolic fluxes in non-motile bacteria. While microbial engineering efforts have primarily focused on optimizing gene expression and enzyme efficiency, our findings suggest that optimizing nutrient transport under space conditions is equally critical. However, it is important to mention that hypotheses other than nutrient depletion could also explain our results, in part or in whole, including mechanobiological responses to reduced fluid shear forces encountered in spaceflight and in the RWV bioreactor^4^. Several potential strategies could be employed to mitigate the effects of microgravity on bioproduction. For example, genetically engineering transport by over expression of high-affinity aromatic amino acid permease or engineering transport-independent metabolic pathways could improve substrate utilization. Introducing a secretion signal or expressing a translocation system to enable tyrosinase secretion or surface display^24^ could facilitate extracellular access to tyrosine and alleviate intracellular metabolic burden. Using motile bacteria or mechanical agitation as an approach to overcome the lack of convection currents in spaceflight may be an alternative approach that could be considered^74^. Using an adaptive evolution strategy^75^ to iteratively expose bacterial strains to simulated microgravity environments may also enable selection of melanin production efficient phenotypes. In addition, given the evolutionary conservation of Hfq across bacterial species and its central role in regulating metabolic and nutrient fluxes, it is plausible that a similar regulatory strategy as that used by could be applied to enhance endogenous tyrosine synthesis, thereby eliminating the need for supplementation in bacterial cultures and improving the production of melanin, as well as other economically valuable bioproducts.

In conclusion, this study integrates flight, analog, proteomic, and metabolomic data to identify key constraints on space biomanufacturing. Overcoming these challenges will be critical for long-duration missions aiming to establish self-sustaining microbial production systems, whether for biomaterials like melanin, pharmaceuticals, or life-support functions essential to future human space exploration.

Supplementary information

Supplementary Materials Supplementary Table 1