Re: 우주 공간에서 발효 가능하다 2025년 논문!!

[문형철]

ArticleVolume 28, Issue 4112189April 18, 2025Open access

Food fermentation in space: Opportunities and challenges

Outline

Share

More

Highlights

•

We fermented a miso on board the International Space Station

•

Fermentation in space is possible with safe, successful results

•

The space environment also shapes fermentation differently: a “space terroir”

•

We lay a foundation for further interdisciplinary work on fermentation and space

Summary

Space exploration is expanding, which demands new technologies and enables new scientific questions. Food, as a bridge between disciplines, can bring these fundamental and applied goals together. Here, we investigate whether food fermentation in space is possible and if so, how it compares with fermentation on Earth. We fermented a miso, a traditional Japanese condiment, on the International Space Station over 30 days and compared it with two earthbound controls. Based on environmental metadata, shotgun metagenomics, whole-genome sequencing, untargeted metabolomics, colorimetry, and sensory analysis, we found that overall, the space miso is recognizable as a miso, indicating fermentation in space is possible. We also found some key microbiological and sensory differences in the space miso, which suggest distinctive features of the space environment. These findings can be harnessed to create more flavorful, nourishing foods for long-term space missions and invite further research questions across science, health, technology, and society.

국제우주정거장(ISS)에서 미소를 발효시켰습니다
•
우주에서의 발효는 안전하고 성공적인 결과를 얻을 수 있습니다
•
우주 환경은 발효를 독특하게 형성합니다: ‘우주 테루아’
•
발효와 우주 분야의 추가적인 다학제적 연구를 위한 기반을 마련했습니다.

요약
우주 탐사가 확장되면서 

새로운 기술이 요구되고 새로운 과학적 질문이 가능해지고 있습니다. 

음식은 학문 분야를 연결하는 다리 역할을 하며, 

이러한 근본적 및 응용 목표를 통합할 수 있습니다. 

본 연구에서는 우주에서의 음식 발효가 가능하며, 

가능하다면 지구에서의 발효와 어떻게 비교되는지 조사했습니다. 

우리는 국제우주정거장에서 30일 동안 전통적인 일본 양념인 미소를 발효시키고, 

두 개의 지상 대조군과 비교했습니다. 

환경 메타데이터, 샷건 메타게노믹스, 전체 유전체 시퀀싱, 비표적 대사체학, 색도계, 감각 분석을 기반으로, 

전체적으로 우주 미소가 미소로 인식될 수 있음을 확인했으며, 

이는 우주에서의 발효가 가능함을 나타냅니다. 

또한 우주 미소에서 미생물학적 및 감각적 차이를 발견했으며, 

이는 우주 환경의 독특한 특성을 시사합니다. 

이러한 결과는 장기 우주 임무를 위한 더 맛있고 영양가 있는 식품 개발에 활용될 수 있으며, 

과학, 건강, 기술, 사회 분야를 아우르는 추가 연구 질문을 제기합니다.

Graphical abstract

1.  

Introduction

Background

The International Space Station (ISS) is a laboratory and temporary home hosting a rotating crew of six international astronauts to conduct research for future space exploration. It orbits the Earth at a height of about 400 km, through what is known as Low Earth Orbit (LEO). It offers a unique environment for answering fundamental and applied questions, due to naturally occurring factors such as microgravity1 and increased radiation exposure2 and socio-technical factors including limited supply chain, isolation, and the imperative to form and reform a strong team across cultural differences. Food design and provisioning is one of the key challenges for sustaining life on the ISS and brings together these fundamental and applied questions.

The socio-technical complexity of space exploration puts significant constraints on food design. Safety, nutritional efficacy, weight, shelf-life, and cost are the main considerations. Entire departments are devoted to space food design and production: NASA’s Advanced Food Technology Project, for example, works on food for long duration missions such as a journey to Mars.3 Currently in the ISS, the highly engineered foods are mostly freeze-dried and reconstituted with recycled potable-quality water.4 Food and packaging waste is a great concern for the remote station where supply chain is limited.5 Aside from this preserved food, edible plants are being grown in the ISS with a Vegetable Production System called Veggie.6 In 2015, “Outredgeous red” Romaine became the first fresh food grown in the ISS to make its way onto the space menu.7 Yet plant growth time and astronaut labor make these experiments highly resource-consuming.8 Additional technical challenges for eating and preparing food in zero gravity environments include food “fly-aways” (when food floats away from the eater) and limited cooking equipment. A further socio-technical challenge is providing a broad enough range of food to accommodate the dietary and cultural needs of the six international astronauts the ISS hosts at a given time.

The complexity of space exploration also poses challenges to studying food in space. Sending biological materials to space for scientific experiments is costly and logistically challenging due to safety, storage, and transportation requirements. Studying fresh food samples in space is difficult since safety regulations restrict testing on human subjects (i.e., via consumption); most food needs to be tested on Earth first. There are also few data about sensory perception of food in space because historically the topic has not been prioritized by space agencies. Astronauts themselves have reported that in space their sense of taste and smell is reduced and that they prefer salty, spicy, and umami-rich foods.9,10

Fermentation can help address these technical, nutritional, and sensory challenges of space food design. Fermented foods are “foods made through desired microbial growth and enzymatic conversions of food components.”11 Recent decades have seen a revival of interest in fermentation traditions in many places in the world.12 This fermentation renaissance has been part of a growing interest in sustainable food practices and systems involving local production and the regeneration of biocultural diversity,13,14 and the rise of microbiome sciences revealing the importance of microbes to personal and planetary health.15,16

Although food fermentation has been practiced around the world for millennia,17,18,19 it has not yet entered the space environment. Since the fermentation process is shaped by its environment,20 new practices, flavors, and microbial communities may emerge as fermented foods migrate to outer space. This new research direction builds on the recent explosion of scientific interest in studying the microbiome of the ISS21,22,23,24,25 and the human microbiome in space.26 The ISS is a novel built environment in a larger non-terrestrial environment, but it is not hermetically sealed off from Earth nor is it in any way “sterile.” Astronauts from all over the world bring their microbes up to the ISS, which now has a distinct microbiome of its own that changes over time and shapes astronauts’ own microbiomes in turn.27 For these reasons, the ISS has become a rich site for microbiome research, both for health purposes and to learn about the fundamental dynamics of how microbiomes form and assemble in novel built environments.

Studying fermentation in space adds a new dimension to this research program, expanding interactions between human bodies and the built environment to include foods as a site of microbial exchange and testing the robustness of fermentation in novel extreme environments. The health benefits of fermented foods might also support astronaut health on future space missions. Though a few fermented and pasteurized products, such as kimchi and wine, have been sent to the ISS,28,29 and some fermentations have been modeled on Earth in “space-like” and “near-space” conditions,30,31 no actual process of food fermentation seems yet to have been carried out in space. The earlier spaceflights for already fermented and pasteurized products seemed to have nationalistic or commercial motivations, to promote cultural identity28 or increase a commodity’s market value; a kind of “space fetishism.”29 Here, we are interested in something else: using the uniqueness of the space environment to answer fundamental scientific and applied technical questions and to pose larger social questions relevant to space exploration and to life on Earth.

Therefore, in early March 2020, we sent a small container of miso-to-be up to the ISS. It stayed on board for 30 days to ferment, before returning to Earth as miso. With this experiment we had three main purposes: (1) to test the feasibility and robustness of fermentation in space; (2) to study how the space environment might shape microbial ecology, evolution, metabolism, and flavor in fermentation; and (3) to open up new multidisciplinary research directions across fundamental, applied, and social sciences. Although the experiment offers some new insights into miso, fermentation, and the space environment, the paper’s main contribution is methodological: to suggest how space fermentation can bring together fundamental science, health science, systems design, and social and cultural engagement in potentially groundbreaking ways.

Experimental design

Miso is a fermented, umami-rich paste from Japan, usually made from cooked soybeans, kōji (rice or barley fermented with the filamentous fungus Aspergillus oryzae), and salt (Figure 1 32). This food product was selected for this experiment for several reasons. The first is practical: its firm, solid structure meant there was a reduced risk of leakage that could potentially damage other experiments and equipment on the ISS, and the timeframe for a young miso fit the 30-day window we had for the experiment. The second is scientific: this experiment fits within a recent surge of interest in and research on miso in the scientific community,33,34,32,35 which allows for comparison and contextualization of our results. This work is beginning to show the diversity and uniqueness of miso microbial communities, which our study builds on. The third reason is sensory: miso is a strongly flavored, umami-rich product that can satisfy astronauts’ need for flavor—its salty and pungent experience can enliven the senses in the sensory-muting environment of microgravity, which also has implications for astronaut diet and health.10,36 The fourth reason is health-related: miso is highly nutritious37 and has multiple health benefits that could serve astronauts.33,38,39 The fifth reason is cultural: fermentation is a ubiquitous and ancient cooking technique that everyone has some relationship to, knowingly or not. Miso, a fermented food from East Asia, was selected to diversify cultural and culinary representation in space.

Figure 1 Miso appearance

Show full caption

The miso mixture was prepared using cooked soybeans, rice kōji, and salt. A high-kōji low-salt style of young miso (Figure 2A) was prepared to facilitate a faster fermentation appropriate for the 30-day period. The miso mixture was produced in Copenhagen, Denmark in a food-safe environment according to a HACCP plan,32,35 split into three portions and kept frozen until the start of the experiment. The three misos were packed into identical plastic containers under a flow hood and fermented on the ISS, in Cambridge, Massachusetts, USA, and in Copenhagen, Denmark (Figure 2A). While on the ISS, the space miso was contained in an environmental sensing box, which measured temperature, relative humidity, pressure, off-gassing, light, and radiation (Figure 2B). The Cambridge miso (CAM) was also kept in an identical sensing box. The Copenhagen (KBH) miso was not, as only two sensing boxes were built, but was kept in a cupboard of similar dimensions. This difference gave us the opportunity to see how the sensing box itself might impact the miso’s fermenting environment. Temperature and relative humidity for the Copenhagen miso were manually recorded daily. Following the fermentation of the misos for 30 days, various analyses were performed (Figure 2B). Further details about the production, fermentation, and analysis of the miso can be found in the STAR Methods.

Figure 2 Schematic representation of the experimental design

Show full caption

Sending experiments to space is costly, and space is at a premium. The little room we had we opted to fill with one maximal sample rather than three smaller ones, prioritizing successful fermentation over statistical robustness. This decision maximized the chance of the fermentation happening properly, though limits our ability to make conclusive claims about the similarities and differences among the misos, as we have no replicates to control for variation within the misos. We thus present our findings on the similarities and differences among the misos as exploratory, pointing the way for further studies.

Results and discussionThe space of space

The ISS is a unique space. Orbiting in LEO, it has distinct environmental conditions to those found on Earth. Two that are of particular relevance to this experiment are microgravity and increased radiation.40 The microgravity in LEO means that miso cannot be weighed down as it usually would, which might change how gas bubbles form and are released and the miso’s resulting density, how much oxygen is available, and how the microbial communities assemble and grow.41,42,43 Being outside Earth’s atmosphere also means the ISS is not as shielded from cosmic and solar radiation, which may also shape the microbial ecology and could potentially lead to higher mutation rates.44,45

To investigate the fermenting environment on the ISS and how it compared to the environments for the earthbound controls, the environmental sensing boxes collected metadata for temperature, relative humidity, pressure, and radiation (Figure 3). These data highlight the similarities and differences in environmental conditions between ISS and CAM. The manual temperature and relative humidity measurements for the KBH miso are also included in the graphs.

Figure 3 The environmental sensing box and its data

Show full caption

All temperatures presented statistical differences between the sites of fermentation (ANOVA, p < 0.05): on average (mean ± standard deviation [SD]), 36.3 ± 1.8°C for ISS, 23.1 ± 0.9°C for CAM, and 19.9 ± 0.5°C for KBH (for full data see Table S1.1). The higher temperature on the ISS may have been due to the sensing box being stowed tightly among other heat-generating equipment, raising the ambient temperature. The slightly higher temperature of the CAM miso supports the possibility that the electronics and enclosure of the sensing box generated and retained heat, raising the ambient temperature, as room temperature for the CAM sensing box, as for the KBH miso, was 20°. These differences in temperature are important, as they would likely affect microbial community formation and metabolism and sensory properties.

For relative humidity, there was a difference between the sensors: for Sensor 1, there was no statistical difference (ANOVA, p > 0.05) between sites (mean ± SD of 43.4 ± 1.9% for ISS, 42.8 ± 2.8% for CAM), whereas for Sensor 2, there was (mean ± SD: 52.6 ± 1.3% for ISS, 47.4 ± 3.2% for CAM, Tukey’s test, p < 0.05; for full data see Table S1.2). KBH had a mean relative humidity of 62.9 ± 3.1%. It is unclear why Sensor 1 should have recorded lower humidity than Sensor 2 at both sites. The higher relative humidity for KBH may simply have been because the ambient humidity in the room where the miso fermented was higher than in the CAM and ISS sensing boxes.

For pressure, the sensors recorded means of 98.0 ± 0.2 kPa for ISS and 102.0 ± 1.1 kPa for CAM, a statistically significant difference of 4.0 kPa (ANOVA, p < 0.05; for full data see Table S1.3). This difference was surprising, as air pressure on the ISS is supposedly held at 101.3 kPa, the same as sea level on Earth.46 Nonetheless, it is unlikely that this discrepancy is due to sensor malfunction, as both pressure sensors recorded the same results.

For radiation, the radiation rate on the ISS (mean of 67024 clicks/h) was 120 times higher than on Earth (mean of 559 clicks/h; for full data see Table S1.4).

Similarities and differences among the misos

Our analysis of the microbial communities, flavor compounds, and sensory properties of the misos show that (1) overall, the space miso is a miso, and (2) there seem to be differences between the misos that suggest a specific fermentation environment in space. For analysis, we took samples from three portions—the top, middle, and bottom of the jars—to investigate how the microbial communities and flavor compounds might also differ within each miso. Any apparent similarities and differences among the misos and their portions would require experimental replicates to confirm‎. We therefore present them here tentatively and as suggestions for further research.

Microbial communities

In analyzing the microbial communities of the misos, we investigated their taxonomic composition, the mutation of Aspergillus oryzae, and safety.

Taxonomic composition

We mapped the metagenomic reads against a genomic database (MetaPhlAn) and identified between 11 and 15 bacterial species per miso (18 in total across all samples) and A. oryzae as the only eukaryotic species, found in all samples (Table S2.1). To characterize the microbiota independently of the set reference genomes, we also conducted an analysis using the leuS marker gene (NCBI protein database) from the assembled metagenomes. This analysis yielded between 7 and 10 bacterial species per miso (14 in total across all samples) and A. oryzae as the only eukaryote (Table S2.2). For comparison, the only other study of miso ecology using metagenomic sequencing found between 19 and 48 species in six novel misos using MetaPhlAn and between 4 and 11 species using the leuS marker gene.32 So these misos are on the lower end of species richness for MetaPhlAn mapping and with comparable species richness for leuS marker gene analysis. The absence of non-Aspergillus eukaryotes is surprising, as salt-tolerant yeasts are a common feature of miso ecology.

The low richness for these misos could have been due to the relatively sanitized mode of production compared with traditional miso—ingredients were processed with gloved hands, mixed and packed into sterilized containers, and transferred only under flow hood. We did this to try to limit confounds of microbes from the producer’s body, locations of production, or sampling from influencing the ecology, to isolate the effect of the space environment. This meant there were no opportunities for microbes to enter the miso after it was made. It could also have been due to the freezing of the mixture after preparation and before fermentation, which would have killed some of the taxa in the mixture before they had a chance to grow. Freezing before fermentation is not part of traditional miso making, and this could explain the absence in these misos of many of the typical miso-associated taxa from the literature.

Some similarities emerged across the samples. Although the A. oryzae from the kōji was identified in all samples, only small proportions were detected and mainly in the ISS samples (Figure 4A). This relative absence may be due to a bias in DNA extraction methods that more easily break down prokaryotic cells.47 Even methods optimized for fungal extraction have been known to extract less of Aspergillus spp.48 The bottom portions of all three misos had the highest relative abundance of A. oryzae. We would expect to find A. oryzae, a strict aerobe, mostly at the top of the miso. However, this finding may rather be due to lower overall microbial abundance at the bottom of the miso, which would make the A. oryzae DNA seem more prominent. Among the bacterial species identified, the majority belong to the genus Staphylococcus. The Staphylococcus spp. detected in all misos belong to S. gallinarum (two different subspecies/strains), S. epidermidis, S. pasteuri, and S. warneri (Figure 4B).

Figure 4 Microbial composition of ISS, KBH, and CAM misos, and mutation rates and genomic distribution of variants for A. oryzae isolated from each

Show full caption

There is not so much literature on the microbial communities of miso (at least in English) to help contextualize our findings. What does exist is mostly based on culture-dependent techniques that cannot detect the communities’ full diversity,33 and the only studies that use culture-independent techniques have focused on novel misos (misos made with untraditional substrates, like yellow peas or lentils), not traditional ones.32,35 A recent review of miso microbial communities shows that multiple Staphylococcus spp. (S. gallinarum and S. kloosii) are commonly found in miso.33 Our misos suggest further Staphylococcus species might be added to this list, a finding supported by the recent culture-independent studies of novel misos,32,35 one of which also shows evident adaptation of human-skin-associated S. epidermidis to the miso niche.35 Other bacterial taxa often found in miso include Bacillus spp., Enterococcus spp., Lactobacillales, Tetragenococcus spp., and Weissella spp. None of these were found in our misos in significant amounts, which could be related to the relatively “clean” way they were produced and/or the freezing mentioned above.

Overall, we can observe that each miso presented a similar microbial composition throughout its top, middle, and bottom portions. There thus seems to be more variation among sample locations than among positions in the miso. Although this is not possible to confirm‎ statistically as we only have one miso per location, it is visually suggested in the bar plots (Figures 4A and 4B). Within the overall finding of the space miso being a recognizable miso comparable to the earthbound ones, this observation of potential differences among the samples leads us to consider the space miso’s apparent specificities.

In comparison to the Earth misos (KBH and CAM), the ISS miso presents a higher proportion of S. epidermidis and S. warneri. These species may have been favored by the higher temperatures on the ISS (36.3°C). Furthermore, one species, Bacillus velezensis, was only detected in the ISS samples and mainly in the top (Figure 4B). This species has been previously isolated from fermented soy foods such as meju and doenjang49,50,51 and is listed as safe in food by multiple national and supranational food safety authorities.52,53 B. velezensis is an aerobe, and many strains have an optimal growth temperature of 30°C–40°C,54,55,56 which could explain its presence in only the top portion of the ISS miso.

Studies have detected B. amyloliquefaciens,32,35,57 a closely related species to B. velezensis,58 in a variety of traditional misos and novel ones made with alternative plant substrates, which support the presence of multiple Bacillus spp. in miso.33 Other studies indicate that Bacillus amyloliquefaciens is used in the production and processing of kōji, the essential ingredient in miso production.59 This suggests that besides A. oryzae, certain Bacillus strains may be part of some spore starter cultures and may have entered into the misos this way. Another explanation for their presence in miso is that they occur as seed endophytes in pulses and their heat-resistant spores are not inactivated by cooking, allowing them to enter into the miso via the leguminous substrate.

A. oryzae mutation

To explore whether the space environment might also shape microbial evolution, we calculated the mutation rate of A. oryzae in the different environments. A. oryzae was selected because we knew it was in each miso and that it would grow on plates. We chose to isolate it from the bottom portion of each miso, because that portion contained the highest relative abundance of A. oryzae DNA, and from each portion of the space miso. One sample of A. oryzae was isolated from each of these five portions, and genomic analysis was performed to compare the samples to the reference strain grown from the same spore.

All samples presented more than 120 variants compared to the control reference strain, which were characterized by a combination of single- and multi-nucleotide polymorphisms (SNP and MNP, respectively) and insertion-deletion mutations (Indels; Figure 4C). These variants are potentially a result of the fermentation process, as suggested by some studies of Saccharomyces cerevisiae in wine fermentation, where the genetic diversity of some strains has been shown to change in response to stresses imposed during fermentation.60 Notably, the number of variants is highest in the ISS miso (Figure 4C), in all portions (Table S2.3), which could have been driven by the harsher space environment, especially increased exposure to radiation (Figure 3H) for which there is precedent in fungi.45 To confirm‎ this finding, triplicates should also be isolated from the KBH and CAM misos to ensure valid comparison and statistical significance.

An analysis of the distribution of these variants across the genome was also performed, revealing a clustering of variants in the eighth chromosome across all samples, and particularly in the ISS sample (Figure 4D). This finding is validated by the A. oryzae genomes isolated from the other two portions of the space miso (T and M; see Table S2.3). Possible explanations include that this chromosome is particularly vulnerable to damage and/or mutation, and/or that it has lower efficiency in DNA repair mechanisms than the other chromosomes. Functional genomic analysis could offer insight into this question.

Safety

As one of the key aims for exploring fermentation in space is to determine a product’s consumability, it was important to investigate the misos’ safety based on microbial species composition—here in particular with regard to the Staphylococcus spp. present in all the misos.

The most abundant Staphylococcus species detected in the misos are coagulase-negative Staphylococci (CoNS), which are occasionally found in fermented foods worldwide61 and may be part of normal human skin flora.62 Many CoNS are used as starters for cheese and meat fermentation to enhance color and flavor development63,64,65,66,67 and have been investigated for use as starters in soybean fermentation.68

Some CoNS may carry enterotoxin genes such as pyrogenic toxin superantigen (PTSAg) and exfoliative toxins.62 We searched for the presence of genes encoding for PTSAgs (sea-see, seg-sevu, selv, selx, sey, selz, sel26, sel27, and TSST1) and exfoliative toxins (eta, etb, etd). We detected only one virulence gene (sel26) in the top portion of the ISS sample, which corresponds to S. aureus present in low abundance in this sample (Table S2.4). It is important to highlight that the top portions of miso are usually removed and discarded before the rest of the miso is eaten, as we did here. This finding may be supportive of this practice. The other Staphylococci detected in higher abundance are not of food safety concern. No pathogenic species were detected in the other portions or samples. These provisional findings based on the metagenomic data suggest that the space miso is safe to eat. This conclusion would require culture-dependent methods to confirm‎, as some pathogens can pose risk even at levels lower than the detection threshold for sequencing. There is effectively no tolerance for the risk of food-borne illness in space,69 so further tests would be required to ensure safety for space-based fermentation.

Flavor

To further investigate the similarities and differences between the space and Earth misos as foods, and ultimately determine whether the space miso was indeed a miso, we also investigated the misos’ flavor chemistry and sensory qualities. For volatile aroma compounds and amino and organic acids, we analyzed all portions of the misos (top, middle, and bottom), to characterize the miso as a complete system. For sensory analysis, we removed and discarded the top portion and combined the middle and bottom portions, as would be done when preparing miso for consumption.

Volatile aroma compounds

Most of the compound classes (six out of seven) were present in all the misos (Figure 5), suggesting an overall aromatic comparability between the space miso and Earth misos. Previous studies of miso detected aldehydes, ketones, esters, and pyrazines as the main compound classes in miso,70 all of which are present here (Figure 5A), suggesting that these misos are also comparable to traditional ones. A more quantitative comparison with traditional misos may be further illuminating, though would require careful alignment of methodology to ensure validity.

Figure 5 Concentrations of volatile compounds

Show full caption

Most compound classes are present across all the misos (Figure 5A). The mean concentration of aroma compounds is significantly higher in the ISS miso (1020 ± 410 μg/L) than in KBH (177 ± 120 μg/L; it is also higher than in CAM, 332 ± 210 μg/L, though not significantly; Kruskal-Wallis, p < 0.05; Table S3.1). The concentration of aroma compounds seems to be positively correlated with the fermentation temperature (Figures 3B and 3C). In particular, the ISS miso contains significantly higher concentrations of esters and pyrazines—22.6 times as much esters and 5.8 times as much pyrazines in ISS than in KBH and 29.9 times as much pyrazines in ISS than in CAM (Kruskal-Wallis, p < 0.05; Table S3.1). Pyrazines are formed by the Maillard reaction between amino acids and reducing sugars,71 a reaction accelerated by heat. The higher temperature of the ISS miso could therefore have been responsible for the ISS miso’s higher levels of pyrazines.

The ISS miso also contains by far the highest concentration of the main ester found across most samples, the honey-like phenylacetic acid methyl ester—510 ± 190 μg/L, compared with 19 ± 32 μg/L in KBH and 190 ± 120 μg/L in CAM (Table S3.1). The difference between ISS and KBH is significant but not that between ISS and CAM or between KBH and CAM (Kruskal-Wallis, p < 0.05; Table S3.1). The one aromatic acid detected in the misos, 2-methyl-butanoic acid, has a cheesy aroma.72,73 It was found in the top and middle portions of the ISS miso, to a lesser degree in the top portion of KBH, and not in any of the other portions (Table S3.1) and correlates with the relative cheesiness of the misos from the sensory analysis (Figure 7A).

Certain patterns also appear in the portions of the miso across the different locations. In general, the top portions have more aroma compounds than the middle and bottom portions. This may be due to different mechanisms for pyrazines and esters, respectively, the two most abundant compound classes. It is possible that the top of the miso, exposed to the ambient heat in the air, could have had higher rates of Maillard reaction, yielding more pyrazines. We do not have temperature measurements of the miso surface to say for certain. Esters, meanwhile, can be formed through reactions between carboxylic acids and alcohols. These esterification reactions also increase with temperature.74 The presence of oxygen facilitates the oxidation of alcohols to aldehydes, which can further react with carboxylic acids to form esters. Thus the presence of oxygen could also have indirectly encouraged ester formation in the top portion.75 Meanwhile, the one phenolic compound—2-methoxy-4-vinylphenol, having a spicy, clove-like, roasted peanut aroma, common in buckwheat76—is found only in the middle and bottom portions of KBH and CAM and not in ISS. Its formation may have been favored by the cooler temperatures and anoxic conditions of these miso portions.

Free amino and organic acids

We analyzed free amino and organic acids to understand their contribution to the basic taste sensations, in particular those arising from microbial metabolism in miso: umami and sourness. Overall the free amino acid profile is similar among all misos (Figure 6A; Table S3.2). The most abundant free amino acid across all samples is glutamate, known for its characteristic umami taste, which is hydrolyzed by glutaminase from glutamine liberated from soy proteins in the fermentation process.77 Glutamate and aspartate have been found to be the most abundant free amino acids in different misos before, so our findings are in accordance with the literature.78

Figure 6 Free amino and organic acid compositions

Show full caption

There are three acids with statistically significant differences between some of the misos (Kruskal-Wallis, p < 0.05): the amino acids histidine and asparagine and the organic acid lactate (Figure 6). Lactate is a compound commonly found in many fermented foods, often produced by lactic acid bacteria and contributing to foods’ organoleptic properties and stability.79,80 Here, it was likely produced by the Staphylococcus spp., some of which are also known to produce lactic acid.81 The KBH miso had the most lactate, followed by CAM and ISS.

Histidine is more abundant in the Earth misos. Histidine is 6% of soybeans’ total amino acid composition.82 In soy sauce production, histidine has been observed to increase in the early stages of the fermentation, then decrease over time.83 We might mainly see free histidine as a result of initial degradation of plant protein, which would then decrease after uptake by some of the microbial community in the misos. Given that the misos were all started from the same batch with the same kōji, the difference in histidine concentration between the space and earth samples could be due to the different rates of substrate breakdown and metabolic activity from temperature differences and other environmental conditions increasing the fermentation rate in space.

Asparagine, meanwhile, is more abundant in the ISS miso than in KBH and CAM. Asparagine is fairly abundant in soybeans, with 10% of the total amino acid concentration,82 and is unlikely to be taken up and catabolized by A. oryzae as it enters late in the tricarboxylic acid cycle.83 Consequently its higher abundance in the ISS miso could indicate higher proteolytic activity, i.e. a more aged miso.

Sensory analysis

Visual observations indicated that all of the misos had significant white mold growth on the surface. Usually one covers a miso during fermentation (with a wooden board, plastic cling film, or other material) to minimize its exposure to oxygen and inhibit mold growth on the surface. Because of the microgravity conditions on the ISS, it was not possible to cover the space miso as one would usually do. This white mold growth was likely the A. oryzae. The space miso also had much more of a thick brown liquid on top than the Earth misos. This viscous liquid was tamari, a naturally occurring product that occurs as moisture is freed from the plant substrates during the fermentation and rises to the top of the miso. The greater presence of tamari in the ISS miso was likely due to increased rate of fermentation from higher temperatures and more disturbance during travel. The ISS miso was also darker than the Earth misos. This may also have been due to the higher fermentation temperature and possibly higher oxidation from being jostled more during transport. This combination of white mold growth, tamari, and oxidation on the surface supported our decision, based on miso tradition, to remove the top layer before the sensory analysis.

We later measured this difference in color and found the space miso was indeed statistically darker than the Earth misos (Tukey’s test, p < 0.01; Figure 7B; Table S4.7). The space miso’s darker color is linked to the greater production of pyrazines (Figure 5), which can be explained by Maillard reactions and/or Strecker degradation; the latter occurs in the relative absence of reducing sugars.84 The formation of pyrazines has also been described in other longer-fermented foods, such as parmesan cheese, where browning is sometimes also observed.84 Pyrazines are reported to display baked, roasted, and nutty flavor characteristics.85,86 In our sensory analysis, these aromas were statistically perceived as more preval‎ent in the space miso (Cochran’s Q test, p < 0.05; Figure 7A; Table S4.5).