The influence of different storage conditions on Limnospira indica, a promising candidate for air revitalization in space

Summary

Cyanobacterium Limnospira indica is being explored for oxygen production and carbon dioxide removal from air in future space stations. Before activation in space, it undergoes a transport and storage phase lasting from one to several weeks, during which it must remain dormant. This study examines the effects of dormancy in dark and cold conditions on L. indica’s photosynthetic performance and biomass composition after storage. Results showed that storage negatively affects photosynthetic growth and biomass composition, but the impact depends on factors such as initial cell concentration, medium pH, cell pigment content, nutrient and gas availability, and storage duration. Storage was also tested under simulated microgravity conditions, but no adverse effects of reduced gravity were observed when healthy cultures were used.

Graphical abstract

Highlights

• •L. indica can be stored for 2 weeks at 4°C in the dark without biomass loss
• •The choice of a suitable inoculum is crucial for storage success
• •Gas and nutrient availability during storage strongly impact the storage outcome
• •Simulated microgravity does not have a detrimental effect on storage

Biological sciences; Microbiology; Applied microbiology; Space sciences

Introduction

Space agencies continue to develop more regenerative life support systems for space stations with the goal of increasing the autonomy of space travelers, reducing the payload costs of space missions, and increasing the sustainability of space exploration.^1,2 Many of those regenerative life support system projects plan to use photosynthesis as the primary process for air revitalization and for edible biomass production. Examples of such systems are MELiSSA,² Eu: CROPIS,³ and VEGGIE.⁴ In the ARTHROSPIRA-B and -C experiments, we tested the feasibility of oxygen and biomass production by the cyanobacterium Limnospira indica in space onboard the International Space Station (ISS). The ARTHROSPIRA-B and -C experiments used small prototype photobioreactors to investigate the influence of space flight on the growth, oxygen production, and molecular composition of L. indica.⁵

Storage of the photosynthetic microorganism in an inactive state (i.e., in dormancy) before its cultivation in space is inevitable for upload to space. But the available storage conditions, however, are strongly restricted by the launch circumstances. In most cases, the storage phase needs to be without active power consumption (thus no active cooling, no light, no aeration, no mixing), and it needs to span at least 1 week but preferably longer to accommodate launch delays. In addition, after arrival in space, the dormancy state should be reversible, in a fast and efficient way, with minimal resources. Thus, ideally, the upload of cells to space is done as dried biomass at ambient temperature in the dark, and stable over multiple weeks or months. Such launch storage conditions are, however, often not feasible for photosynthetic microorganisms. For example, the reviving of cyanobacterium Limnospira after drying has not yet been successful. In the ARTHROSPIRA-C (ArtC) experiment, the Limnospira indica cells are therefore stored and transported to space in a liquid suspension (without a gas phase) inside the bioreactors, where they are kept in the dark during a period of 7 days. Additionally, a passive cooling of the complete bioreactor to 4°C is implemented as it was found that storage in this condition without a gas phase at ambient temperature was difficult for longer than 1 week (Leys N., unpublished data). A storage of the cells in a liquid suspension in the tubing of the feeding loop of the bioreactor (no gas phase, static, in the dark, at 4°C), was indeed the most suitable storage conditions to allow fast revival and immediate bioprocess start without requiring additional hardware or crew intervention; and for providing highest probability for homogenous cell dispersion into the bioreactor culture chamber under microgravity conditions. Indeed, after storage, the cells can be directly pumped to and mixed with fresh medium in the culture chamber and illuminated and warmed up to 33°C in the bioreactor, to let the photosynthesis bioprocess begin. After an initial awakening from dormancy and proliferation to sufficient cell density in a batch incubation at 33°C and 45 μmol photons m⁻² s⁻¹ for one week, the cultures can then be further tested in space in four different semi-continuous bioprocess cycles of two weeks each. In these cycles, the light fluxes are 45 – 55 – 70 and 80 μmol photons m⁻² s⁻¹ at 33°C. Thus, storage conditions were selected in purpose of the bioprocess conditions to enable cultures to successful regrow and thrive steadily under these conditions, after storage.

Nevertheless, storage conditions still have a significant impact on the photosynthetic organism’s metabolism and efficiency of bioprocess revival. Therefore, we report here on our tests performed to further characterize the impact of the selected launch storage conditions, i.e., a liquid suspension in the dark at 4°C, on the post-storage growth and biomass composition of L. indica. Tests were performed with varying durations of storage, cell concentrations, medium pH, and nutrient and gas availability during storage (Table 1). Also, the impact of simulated microgravity and the hermetically closure of the ArtC bioreactor flight hardware on the outcome of the storage was briefly assessed (Table 1). All with the final goal to enable a successful upload of L. indica in a dormant state to space and an immediate activation of the photosynthesis bioprocess in the ArtC bioreactors in ISS. The following sections describe the tested methods to improve the outcome of a storage in liquid in dark and cold conditions inside the ArtC bioreactors.

Table 1.

Six different experiments were set up, each focused on one specific parameter possibly influencing the storage of Limnospira indica PCC8005 strain P3, in static cold and dark conditions

Experiment	Investigated parameter	Set-up
Experiment 1	Duration	1 and 2 weeks storage time, 50 mL Falcon tubes
Experiment 2	Cell to medium ratio (by dilution)	No dilution, 1:1 dilution and 2:1 dilution (culture: Zarrouk), 50 mL Falcon tubes
Experiment 3	Medium composition (by dilution, washing and addition of glucose)	Not washed, 1:1 dilution, washed + replenished with Zarrouk, washed + replenished with Zarrouk containing 1.5 g L−1 glucose, 50 mL Falcon tubes
Experiment 4	Gas availability (by headspace)	Different gas (ambient air) to liquid ratios were tested (0%, 25%, 50 and 75% gas phase), 50 mL Falcon tubes
Experiment 5	Simulated microgravity (μG)	Storage on horizontal RCCS (control) and on RPM (simulated μG)
Experiment 6	Hermetically closed advanced experiment container (AEC)	Storage inside the tubing of the ArtC space flight photobioreactor (PBR), inside the AEC

Results

Storage duration: The shorter the storage duration, the smaller the loss of cells

The differences between storage for 1 and 2 weeks in liquid without a gas phase, static, in the dark, and in the cold at 4°C were assessed (Experiment 1). The storage for 14 days has a much stronger effect than the storage for 7 days. The OD_770nm was the most affected parameter after both storage durations, and the pH was the least affected parameter (Figure 1). The significant decrease in dry weight after 14 days shows that many cells lysed, causing cell debris to pass through the filters used for the dry weight determination. The sedimentation experiment showed that the longer the storage, the slower the cells sediment (Figure 1D), which is indicative of lighter and shorter filaments. Filament fragmentation and cell degradation were indeed observed in microscopy after storage (Figure S2). The increased impact after 14 days was also seen in the flow cytometry data, represented by the FL3-H/FL4-H ratio and the percentage of P1 (Table S3 and Figure S11). The FL3-H/FL4-H ratio increased strongly over the storage time (Figure 1E). While the cultures stored for 1 week still showed an FL3-H/FL4-H ratio below 1 (0.58 ± 0.05), the cultures stored for 2 weeks reached a value of 2.30 ± 0.01. Prior to storage, the culture had an FL3-H/FL4-H ratio of 0.32 ± 0.01. Also, the fraction of long and highly pigmented cells (%P1) decreased over the storage time (Figure 1F). The fraction of P1 cells was very low (0.27% ± 0.01%) after 14 days, predicting a culture with a low photosynthetic capacity. The measured content ratios of phycocyanin per chlorophyll and allophycocyanin per chlorophyll revealed a significant decrease after 2 weeks of storage time (Figures 1C and 1D), confirming the flow cytometry data. Storage time also had a significant impact on post-storage proliferation in batch at a relatively low light intensity of 45 μmol photons m⁻² s⁻¹ at 33°C (Figure S1). The average biomass productivity (P_av) increased significantly with the increase in storage duration (Figure 1, Table). The μ_max after 1 and 2 weeks of storage was significantly higher than in the non-stored cultures Figure 1 table). This is most likely due to the lower cell numbers in the cultures after storage, allowing a higher light flux per cell (Figure S2). To calculate the μ_max of the cultures stored for 2 weeks, the first 2 days of the growth phase were not considered in the exponential regression as the spectrophotometer used was unable to produce reliable results in the low cell concentration range in the first days of the regrowth cultures (Figure S1).

Figure 1.

Comparison of cultures prior to storage (inoculum) and after storage when stored for 1 and 2 weeks (4°C, dark, static)

(A) Dry weight.

(B) OD770nm.

(C) pH.

(D) Sedimentation index at t = 60 min.

(E) FL3-H/FL4-H ratios.

(F) %P1.

(G) Phycocyanin to chlorophyll ratio.

(H) Allophycocyanin to chlorophyll ratio. Bottom: Post storage growth rates during exponential (μ_max) and linear (P_av) growth phase, values are shown as mean ± CI (95%), different letters in row represent significant differences (p < 0.05). Kruskal-Wallis test followed by Dunn’s multiple comparison was performed. All values are shown as mean ± SD (n ≥ 3).

Cell concentration during storage: Lower cell concentration and lower pH improve storage outcome

To improve survival over 2 weeks of storage, the cultures used in experiment 2 were harvested at a lower OD_770nm (1.59 vs. 2.11) and pH (10.83 vs. 11.45) compared to the culture used in experiment 1. In addition, part of the culture was diluted further to lower cell density with fresh medium prior to storage, i.e., dilutions of 2:1 and 1:1 for culture versus fresh medium were tested. In general, the lower cell concentration and pH of the cultures prior to storage had a positive impact on the storage outcome. The non-diluted cultures in experiment 2 were healthier after 14 days of storage (no gas phase, static, dark, at 4°C) when compared to the cultures previously described in experiment 1. The diluted cultures had very low decreases in all parameters, and the dry weight even slightly increased in the 2:1 and 1:1 diluted cultures when compared to the calculated dry weights of the cultures prior to storage (Figure 2A). For calculating the dry weight of the diluted cultures, a calibration curve previously obtained was used, because the dry weight is not changing linearly with the performed dilutions (Figure S4). The small increase in dry weight suggests that a (slow) metabolic activity can be assumed during the 2 weeks storage (static at 4°C in the dark without gas phase) in 2:1 and 1:1 diluted cultures containing fresh medium. The statistical analysis revealed a small but significant decrease of OD_770nm in the non-diluted and 2:1 diluted cultures (Figure 2B). The sedimentation behavior of the cultures stored at different dilutions was not impacted by the storage (Figure 2D). The data obtained by flow cytometry revealed that in the 2:1 and 1:1 diluted cultures, the FL3-H/FL4-H and percentage of highly viable cells (Figures 2E and 2F) remained similar to those in the culture prior to storage, which was not the case for the non-diluted cultures. The flow cytometry confirmed that the cultures with lower cell density and lower pH used in experiment 2 were much less affected than the cultures used in experiment 1. The measured phycocyanin per chlorophyll content ratio increased significantly in the 1:1 diluted cultures compared to the culture prior to storage, while the allophycocyanin to chlorophyll ratio was not affected by the storage (Figures 2G and 2H). The data of the biomass productivities of the regrowth cultures after storage showed that the diluted cultures had a steeper exponential growth phase than the undiluted cultures. All cultures showed efficient revival and photosynthetic regrowth after storage, and no delay in growth was detected (Figure S3).

Figure 2.

Comparison of cultures prior to storage (inoculum) and after storage when stored in different dilutions (4°C, dark, 2 weeks, static)

(A) Dry weight.

(B) OD770nm.

(C) pH.

(D) Sedimentation index at t = 60 min.

(E) FL3-H/FL4-H ratios, (F) %P1.

(G) Phycocyanin to chlorophyll ratio.

(H) Allophycocyanin to chlorophyll ratio.

(A, B, and C) Mann-Whitney-U tests; (D, E, F, G, and H) Kruskal-Wallis test followed by Dunn’s multiple comparison. Values are shown as mean ± SD (n ≥ 4). ND: no dilution, IN 2:1: inoculum after 2:1 dilution before storage, IN 1:1: inoculum after 1:1 dilution before storage; 2:1: 2:1 diluted culture after storage, 1:1: 1:1 diluted culture after storage. Bottom: Post storage growth rates during exponential (μ_max) and linear (P_av) growth phase, values are shown as mean ± CI (95%), different letters in row represent significant differences (p < 0.05).

Medium composition during storage: Fresh medium is beneficial for storage outcome

Next, it was tested if, instead of diluting, washing of the harvested cells with fresh medium could also be used to provide the cells with fresh medium prior to storage while maintaining a high cell concentration in the small volume (as mass and volume for upload to space is limited) (experiment 3). To rule out that washing had potentially a negative impact on the storage outcome, because it removes the excreted sugars from the medium and therefore limits the potential uptake of these during dark respiration, a condition of fresh medium with added sugar was also tested. A comparison was made between a 1:1 dilution and a washing treatment, and a washing treatment with medium supplemented with sugar (glucose). The dry weight and OD_770nm declined during storage, while the pH remained stable (Figure 3). Due to limited volume, the dry weight and OD_770nm were only measured before treatment and after the storage (not after washing treatment). Therefore, the decline in these parameters in the two washed conditions should be interpreted with caution, as it might have been influenced by the treatment, although it was confirmed that washing and centrifugation had no impact on dry weight measurements (Table S1). Nevertheless, the unwashed culture was clearly impacted the most by the storage (Figures 3E and 3F). The cells of the unwashed cultures sedimented significantly slower than the washed cultures with 1.5 g L⁻¹ glucose condition (Figure 3), indicating a stronger fragmentation of the trichomes into smaller pieces, which take longer to settle in the unwashed cultures, which was confirmed via light microscopy analysis (Figure S6). The percentage of P1 cells decreased, and the FL3-H/FL4-H ratio increased the strongest in the unwashed cultures, both signs of a decrease in photosynthetic capacity. Also, the amount of antenna pigments per chlorophyll decreased significantly in the unwashed cultures (Figures 3G and 3H). All other measured parameters did not show a significant change after storage. When revived, all cultures regrew, but the unwashed cultures took approximately 7 days to start growing (Figure S5). This is likely due to the lower initial cell count after storage. The growth rate during initial exponential growth (μ_max) was highest in the diluted cultures, while the highest growth rate during the following linear phase (P_av) was recorded by the unwashed cultures (Figure 3, bottom table).

Figure 3.

Comparison of cultures prior to storage (inoculum) and after storage, when stored after different treatments (unwashed, washed, washed + addition of 1.5 g L-1 glucose and 1:1 diluted; 4°C, dark, 2 weeks, static)

(A) Dry weight.

(B) OD770nm.

(C) pH.

(D) Sedimentation index at t = 60 min.

(E) FL3-H/FL4-H ratios.

(F) %P1>.

(G) Phycocyanin to chlorophyll ratio.

(H) Allophycocyanin to chlorophyll ratio.

(A and B) Kruskal-Wallis test followed by Dunn’s multiple comparison was performed to compare the inoculum with unwashed, washed, washed + Glu; Mann Whitney U tests were used to compare IN 1:1 diluted and 1:1 diluted; C: Mann-Whitney-U tests, and (D, E, F, G, and H) Kruskal-Wallis test followed by Dunn’s multiple comparison. Values are shown as mean ± SD (n ≥ 4). IN: Inoculum, Glu: 1.5 g L^-1 Glucose monohydrate. Bottom: Growth rates during exponential (μ_max) and linear (P_av) regrowth, values are shown as mean ± CI (95%), different letters in row represent significant differences (p < 0.05).

Air availability during storage: More air retains healthier cultures

Although it was not feasible in the ArtC flight experiment procedures and hardware setup to provide the L. indica cells with air during dormant storage and upload to space, the impact of air availability on revival after storage was nevertheless assessed (Experiment 4). The volume of available ambient air (“headspace” or “gas phase”) had a significant impact on the stored cultures, i.e., the more air available, the better the outcome after storage (Figure 4). The storage without gas phase (0% v/v air) resulted in a decline in OD_770nm of ca. 50% while a storage with 75% v/v gas phase only showed a decrease of 29%. The flow cytometry results showed that the FL3-H/FL4-H ratio and the percentage of P1 cells were affected in all cultures, but only some of these results were shown to be significant (Figures 4E and 4F). The FL3-H/FL4-H ratio was the highest in the cultures stored without ambient air (0%v/v air), and the %P1 was the lowest in this condition, indicating that the absence of air was disadvantageous for a successful storage. The sedimentation analysis showed that the culture stored with the biggest headspace had a very similar sedimentation velocity as the inoculum culture prior to storage, while the cultures with less air available showed slower sedimentation, indicating stronger fragmentation (Figure 4D). The pigment analysis showed that the cultures stored without headspace had a significantly lower antenna pigment to chlorophyll ratio than the inoculum, indicating a lower photosynthesis capacity (Figures 4G and 4H). All cultures showed efficient revival and photosynthetic regrowth after storage, and no delay in growth was detected (Figure S7).

Figure 4.

Comparison cultures prior to storage (inoculum) and after storage (4°C, dark, 2 weeks, static) when stored with different percentages of gas phase (0%, 25%, 50% and 75% vgas/vliq)

(A) Dry weight.

(B) OD770nm.

(C) pH.

(D) Sedimentation index at t = 60 min.

(E) FL3-H/FL4-H ratios, (F) %P1.

(G) Phycocyanin to chlorophyll ratio.

(H) Allophycocyanin to chlorophyll ratio. Kruskal-Wallis test followed by Dunn’s multiple comparison was performed. All values are shown as mean ± SD (n = 4). Bottom: Growth rates during exponential (μ_max) and linear (P_av) regrowth, values are shown as mean ± CI (95%), different letters in row represent significant differences (p < 0.05).

Modified gravity during storage: No negative impact of microgravity on storage outcome

To assess the potential impact of lowered gravity in space on the storage outcome, cultures were stored in cell culture bags (instead of tubes, as used in the previous test described above) on a random positioning machine (RPM), allowing for the simulation of microgravity (Experiment 5) (Figure S10). The storage (at 4°C in the dark) in the cell culture bags had only a small impact on the OD, pH and sedimentation (Figure 5). The flow cytometry parameters and pigment content of the cultures were not significantly affected. The cultures stored under simulated microgravity showed a decrease in sedimentation index at time point 60 min (Figure 5D), indicating that the trichomes stored in simulated microgravity are slightly shorter or lighter (e.g., contain more vacuoles or less glycogen) than the control cultures. Like in experiment 2, a slight increase in dry weight after storage (μG and control) was found, and the OD_770nm decreased slightly in the normal gravity (control) condition. The regrowth of the cultures was successful, and the growth rates did not show significant differences (Figure S8).

Figure 5.

Comparison of cultures prior to storage (inoculum) and after storage (4°C, dark, 2 weeks) when stored in cell culture bags under normal gravity (control) or simulated microgravity

(A) Dry weight.

(B) OD_770nm.

(C) pH.

(D) Sedimentation index at t = 60 min.

(E) FL3-H/FL4-H ratios.

(F) %P1.

(G) Phycocyanin to chlorophyll ratio.

(H) Allophycocyanin to chlorophyll ratio. Kruskal-Wallis test followed by Dunn’s multiple comparison were performed. All values are shown as mean ± SD (n = 4). Bottom: Growth rates during exponential (μ_max) and linear (P_av) regrowth, values are shown as mean ± CI (95%), different letters in row represent significant differences (p < 0.05).

Effect of precooling of hermetically closed space photobioreactor

Two storage tests could be performed inside the actual space flight photobioreactor (PBR) hardware set-up for the ARTHROSPIRA-C experiment to test the impact of the possible degassing of CO₂ and/or O₂ from the inoculum culture in the silicon tubing when the air volume inside the hermetically closed advanced experiment container (AEC) is cooled and pressure is lowered (Experiment 6). The OD_770nm and pH of the culture, which was stored for 14 days without precooling, i.e., cooling is done in a hermetically closed box conformation, decreased strongly (−80.4%), while the culture stored with precooling before closure of the AEC decreased less (−26.9%) (Table 2). Therefore, the storage condition with precooling in the open box configuration, before closure, was more favorable for the storage of the culture inside the space PBR hardware. It is hypothesized that degassing might occur with cooling in a hermetically closed box, which may have an impact on gas and medium composition and therefore may have a negative impact on the cells in the liquid suspension during storage (Figure S9). These tests could only be performed on one flight hardware PBR, and no statistical analysis could be performed; thus, the results should be interpreted with caution.

Table 2.

OD and pH values of cultures before and after storage (2 weeks, at 4°C, dark, static) when stored inside the ArtC space flight photobioreactor hardware, with or without precooling of the hardware before hermetically closing the advanced experiment container (AEC)

	Before storage (Inoculum)	After 14 days storage without AEC precooling	Decrease [%]
OD770nm	0.96	0.19	−80.4
pH	10.10	10.01	−0.9

	Before storage (Inoculum)	After 14 days storage with AEC precooling	Decrease [%]
OD770nm	1.11	0.81	−26.9
pH	10.33	10.11	−2.1

Characteristics of the culture prior to storage were indicative for the storage success

An additional correlation analysis was performed on the characteristics of the cultures harvested before storage and the outcome of each storage experiment. The cultures discussed in this correlation analysis were the reference cultures (i.e., Inoculum cultures) used in the first 4 storage tests described above. They represent different independent cultures, harvested for storage at slightly variable time points of their growth curve, and thus present some slight differences in cell concentration and culture medium composition. Inoculum cultures were harvested before the stationary phase, which starts between OD_770nm–2.2–2.5 for the culture medium and conditions used.⁶ Only for experiment 1, the culture was on the border of reaching the stationary phase with an OD_770nm of 2.11. The inoculum cultures used for storage in experiments 3 and 4 were harvested at OD_770nm of 1.28 ± 0.03 and 1.21 ± 0.03, which is halfway the exponential phase. Nevertheless, the dry weight of the culture was 1.9 g L⁻¹ in experiment 3 and 1.3 g L⁻¹ in experiment 4 (Table 3). The sedimentation index %SI was also found to be different between the different cultures, ranging from 27% (exp. 3) to 64% (exp. 1). Nevertheless, these reference cultures, were all stored in identical conditions, i.e., in closed 50 mL Falcon tubes at 4°C for 2 weeks without providing fresh medium, headspace, and mixing. Several parameters (i.e., DW, OD, pH, %SI, single cell analysis (%P1) and pigment ratios) which were measured in the cultures prior to storage were compared with an easy-to-assess parameter to define storage success: the difference [%] in OD_770nm after 2 weeks at 4°C in the dark (static). Several significant correlations were found. For example, a lower dry weight and pH, as well as a low percentage of P1 population, correlated with a less severe drop in OD_770nm after storage and thus a better storage outcome (Table 3). Additionally, also a low ratio of antenna pigments per chlorophyll prior to storage was beneficial for storage success.

Table 3.

Overview of the inocula cultures used in the first 4 experiments and the percentage of difference in OD770nm

Prior to storage culture characteristic	Exp. 3	Exp.1	Exp. 4	Exp. 2	Correlation significance
Dry weight (DW)	1.90	1.52	1.26	1.38	∗∗∗
OD770nm	1.28	2.11	1.21	1.54	ns
pH	10.9	11.5	10.6	10.8	∗
%SI (t = 60 min)	27	64	40	37	ns
FL3-H/FL4-H	0.31	0.32	0.27	0.36	ns
%P1	19.8	19.7	7.6	5.7	∗∗∗
Phycocyanin/chlorophyll	7.0	7.3	5.9	2.5	∗∗
Allophycocyanin/chlorophyll	2.2	2.0	2.0	1.8	∗∗
Post storage %Difference in OD770nm	−66.6	−62.8	−49.6	−6.5	–

Correlation analysis between the parameters and the resulting change in OD770nm was performed using Spearman correlation (right column). For experiments with different inocula per treatment, the untreated inocula of the control conditions were used (14 days at 4°C).

Discussion

The storage of dormant but fresh Limnospira (aka Arthrospira or spirulina) seed cultures, allowing fast revival and photosynthetic growth after storage, is a hardly explored field⁷ but essential for its application in space, where storage periods during upload to space cannot be avoided. Most studies available usually aim at preserving the bioactive compounds for use in e.g., supplements and do not aim to preserve the living organism (see for example^7,8,9). Nonetheless, there are exceptions such as the study of Fisher, Saban, Broddrick, and Settles,¹⁰ which investigated the storage of Arthrospira platensis NIES-39 at room temperature and with different carbon sources (glucose, glycerol, and acetate). They were able to show that an addition of glucose is beneficial for the regrowth of the cultures after dark storage, but also raised awareness of the higher risk of contamination when using external carbon sources.

This study showed that a photosynthetic axenic culture of L. indica PCC8005 P3 can be stored in liquid cell suspension in Zarrouk growth medium, static, at 4°C in the dark, for 14 days. However, the harvesting time influencing the characteristics of the inoculum culture was shown to be important. Cultures with a low biomass density (<1.4 g L⁻¹) in combination with a low pH (<11.0), low %P1 (<10%) and low antenna pigments to chlorophyll ratio (<6 for phycocyanin/chlorophyll and <2 for allophycocyanin/chlorophyll respectively) are preferably used to achieve a satisfactory outcome after storage (i.e., low decrease in cell density). Additionally, storage time limitation and several treatments prior to storage can be used to improve culture preservation, namely, the addition of fresh medium and sufficient ambient air availability (gas to liquid ratio above 3:1 or 75% gas phase).

The addition of fresh medium during storage is beneficial, as it has an impact on the pH and the nutrient supply in the culture. Inocula with high pH values prior to storage showed stronger cell concentration reduction after storage (e.g., pH 11.45 in experiment 1), which is most likely due to the higher energy requirements needed to maintain the proton gradient across the membranes. Many of the mechanisms bacteria use to survive alkaline pH values are energy-dependent, and energy is scarce in the dark and cold storage conditions used in this study. It has been reported that ATP levels decrease at cold temperature.¹¹ But several ATP-dependent antiporters (e.g., Na⁺/H⁺), deaminases, and other transport systems such as water channel proteins, were shown to be essential to maintain internal pH and ion homeostasis.¹² It has been reported that Arthrospira platensis (Gomont strain) also produces more antioxidants such as carotenoids under high pH conditions to prevent the reactive oxygen species (ROS) level from rising too high.¹³ At pH 7, pure phycocyanin extract is rather stable, and is not significantly degraded at 4°C by pure physicochemical effects. Khandual, Sanchez, Andrews, and de la Rosa¹⁴ showed a good stability up to 5 weeks. But this is likely not the case, inside a cell in a high pH medium. Therefore, lowering of the pH in the cell suspension before storage could help in preserving the antenna pigments in the cells, in addition to the benefit of the lower energy requirements for the cells to survive at lower pH. The pH in the cell suspension is likely to be maintained quite stable even over longer storage times, as the pH values in this study showed barely different values after the storage. It is important to keep in mind that the pH stability most likely comes from the buffering capacities of the Zarrouk medium and should not be used to assess storage success. The high pH is likely the main reason why the cells from undiluted and unwashed high density inocula showed a stronger impact of storage.

In this study, all the inocula cultures were grown at a constant but relatively low external light intensity of 30–45 μmol photons m⁻² s⁻¹, so the external light flux density was similar for all cultures. Nevertheless, the light availability for the cells inside the culture is not only dependent on the external light flux density, but also on the culture density because of shielding effects. L. indica contains phycobilisomes, also called light-harvesting antenna. These antennas contain the pigment-protein complexes phycocyanin and allophycocyanin¹⁵ which transfer the energy obtained from photons further to the photosystems II and I (PSII and PSI) reaction centers. The photosystems contain chlorophyll a, the primary electron donor in the electron transport chain.^16,17 L. indica does not contain other chlorophyll pigments; therefore, the term chlorophyll refers to chlorophyll a in the frame of this study. In our recent studies⁶ it was shown that a relatively low light flux intensity of 30 and 45 μmol photons m⁻² s⁻¹ at warm temperature (30°C–34°C) induces a higher phycocyanin and allophycocyanin concentration in the cells, while the chlorophyll content is staying more stable. Several growth experiments in our laboratory showed that cultures grown between 30 and 35 μmol photons m⁻² s⁻¹ have a relatively high %P1 of up to ∼20% (Table S3), because the P1 area in flow cytometry plots consists of cells with high amounts of antenna pigments (Figure S11), which are needed at low light availability. Even though these high pigment content characteristics are good to allow the cultures to growth at the light limiting conditions, it is not optimal for storage. Phycocyanin and allophycocyanin content, normalized over chlorophyll content, significantly decreased with increasing storage time. This might be due to the fact that phycocyanin and allophycocyanin are light-harvesting antenna proteins¹⁸ with fast turnover rates¹⁹ and under dark and cold conditions, they cannot be maintained. Under active photosynthetic growth conditions, the pigment proteins are actively produced and represent a large nitrogen stock. It has been reported that under cold stress, the genes for phycocyanin and allophycocyanin synthesis were downregulated in Arthrospira.²⁰ In dark respiration conditions, the cells could degrade phycocyanin and allophycocyanin to access nitrogen and prolong survival.^21,22 However, in the storage conditions used in this study, it is unlikely that NO₃⁻ and PO₄⁻ are limiting factors in the cell suspension, even in the non-diluted cultures, as the inocula cultures were harvested before the stationary phase. Nevertheless, a low (storage) temperature could potentially induce a nitrogen limitation response, because nitrate/nitrite transporters decrease in efficiency at lower temperatures,²³ and potentially could trigger a fast(er) pigment degradation in cold conditions. However, the inability to actively process the pigment debris due to the very slow metabolism at 4°C could also cause an ammonia toxicity. Under low energy and slow metabolism conditions, free ammonia from fast degrading pigments cannot be neutralized via its usual pathway, the ATP-dependent glutamine synthetase.^24,25 Ammonia has been shown to be irreversibly toxic for PSII in Arthrospira platensis and Chlorella vulgaris,²⁶ indicating that this higher ammonia concentration could induce a strong stress condition resulting in impaired recovery and cell death. The limiting light condition before the start of the storage, triggering a higher antenna pigment content, thus seems to put the culture already at risk of additional challenges (e.g., ammonia toxicity) during storage. A higher light flux (i.e., 45 to 60 μmol photons m⁻² s⁻¹) resulting in relatively lower % P1 (Table S3) would therefore be beneficial before a storage in dark and cold is started. Thus, growing a culture at a low or non-light-limiting light flux is a better strategy to prepare a suitable inoculum for storage when high-density cultures are needed.

In addition to the pH and nitrogen metabolism, the carbon metabolism plays an important role in survival during storage. During storage in the dark, the cells use dark aerobic respiration to generate energy, a process that uses sugars and O₂ to generate CO₂ and water while releasing ATP via the electron transport chain. The organic carbon could originate from the extracellular polymeric substances, which are excreted by L. indica PCC8005 during photosynthetic growth into the medium, free or associated with the cells.²⁷ In addition to taking up sugars from the surroundings, cyanobacteria can also use their internal carbon stocks (e.g., glycogen) via aerobic respiration in the dark.^28,29,30 To assess the role of free sugars in the medium to support the potential heterotrophic growth during storage, an additional storage experiment using Zarrouk supplemented with 1.5 g L⁻¹ glucose was conducted. The hypothesis was that additional glucose in the medium could increase the rate of aerobic respiration and could help during the regrowth period, as shown by Fisher, Saban, Broddrick, and Settles.¹⁰ The results indicated that additional glucose is not helping more than a simple dilution with fresh Zarrouk that did not contain sugars, predicting that the available sugar molecules from carbon stocks and EPS are sufficient to maintain the slow dark metabolism at 4°C. Since the addition of glucose did not have any effect on the storage success, the sugar availability was likely not the limiting factor of survival at 4°C in the dark. This enhanced the assumption that the oxygen availability is a more important factor for aerobic respiration in the dark during storage of the L. indicia photosynthetic cells. Which was also confirmed in the storage tests with different air availability. Experiment 4 showed that the headspace (gas phase above culture) clearly helps the cells to survive longer during cold storage. Most likely, the higher oxygen availability increases the rate of (slow) heterotrophic metabolism in the dark. This explains also why the cell concentration (OD) and dry weight are key factors for the selection of a suitable inoculum and storage success. In this study, the storage experiments (except experiment 5) were all conducted under normal gravity conditions (1 g) were causes the cells to sink to the bottom of the storage tube and to form a dense pellet. The thicker the pellet, the stronger the consumption rate and the lower the local concentration of the available O₂ and nutrients from the Zarrouk medium for dark aerobic respiration. Similar effects have already been shown for biofilms and marine sediments.^31,32

The sedimentation experiment after one and two weeks of storage at 4°C showed that the longer the storage, the slower the cells sediment. This can have various reasons. Firstly, due to fragmentation, the stored trichomes became shorter and therefore take longer to sediment. Secondly, the remaining healthy cells respired their internal C-stock molecules (i.e., glycogen) for survival during storage, causing lighter cells that sediment more slowly. Thirdly, many cyanobacteria, including L. indica produce gas vesicles to reach the top layer of the water in nature to obtain a higher light flux. Cells with gas vesicles sediment much more slowly due to their increased buoyancy.^33,34 The effect of the gas vesicles has most likely a less pronounced effect on the sedimentation compared to the fragmentation of the trichomes, as the metabolism for synthesis and assembly of new gas vesicles is probably slow and limited at 4°C.

The storage duration had an important impact on the storage outcome. The experiment comparing 1 and 2 weeks of storage time showed that storing high-density cultures (≥1.9 g L⁻¹), especially with high pH (≥11.45), results in unhealthy cultures. The longer the storage, the bigger the impact, and the longer the recovery, but the cultures regrew eventually to measurable cell concentrations. The μ_max after 1 and 2 weeks of storage was, however, significantly higher than the μ_max of the not stored cultures, but this can be explained by the higher light availability per cell due to lower cell numbers after storage, in the used batch culture set-up for the growth tests after storage. Additionally, the metabolic state of the inoculum played an important role (Table 3). In general, the regrowth experiments showed that even though some of the cultures seemed to be very unhealthy when dense cultures were stored at 4°C for 14 days, the regrowth was successful. If the 14 days limit for maximum storage duration is maintained, all cultures should be able to revive normally in space with just a maximum delay of approximately 7 days in conditions with a lower starting cell number.

Low shear simulated microgravity had no detrimental effect on the storage success. Nevertheless, the experiment strengthened the hypothesis that a high gas availability is beneficial, because the storage in cell culture bags provides an (almost) unlimited gas supply to the simulated microgravity as well as control cultures, which is most likely the main reason for the good survival compared to the cultures stored in tubes without headspace.

The culture stored with a precooling step in an open box configuration had a lower decrease in OD_770nm, showing that the precooling in open box configuration indeed circumvents the potential off-gassing caused by the temperature and pressure decline in closed box configuration. Off-gassing may potentially lead to the removal of dissolved CO2 from the cell culture suspension in the Zarrouk medium (leading to pH increase), and to the removal of dissolved O2, which is needed for the dark aerobic respiration. It must be kept in mind that the storage experiments inside the photobioreactors were done only once for each condition, due to the limited availability of the reactors. Nevertheless, this finding support again that fact the gas availability is an important factor for storage success.

All experiments in this study are part of the time-limited preparation of the ARTHROSPIRA-C space flight experiment. The overall aim was to secure a high number of healthy L. indica cells reaching the ISS, within the specific constraints of the ARTHROSPIRA-C space flight experiment procedures and hardware, i.e., a minimum storage time of 7 days, in liquid medium allowing fast revival by remote commanding without crew intervention. A sequential stepwise test approach was used in varying the storage conditions, allowing for the identification of some inoculum culture features and cell suspension conditions as more beneficial for storage outcome. But we recognize that a factorial experiment design testing more combinations of possibly influencing parameters (time, dilution, replenishment, pH buffering, nutrient addition, oxygen addition, simulated microgravity, etc.) should be used in future storage studies on cyanobacteria seed cultures, to identify and quantify in more detail the influence of each parameter.

Experiments in this study were performed in liquid suspension and at normal fridge temperatures (4°C), but for future space experiments, other possibilities could be investigated as well. For example, Syiem and Bhattacharjee³⁵ showed that several different cyanobacteria can be stored in dried agar cubes up to 3 years. A more recent study showed that cyanobacteria can be shipped in so-called cyano-gels (a mix of cyanobacterial strain and slow solidifying agarose gels.³⁶ Additionally, freezing of the biomass is often used to preserve biomass for analysis. In this case, cryoprotectant agents are needed to protect the freezing of cells from the formation of intracellular crystals and membrane destruction.^37,38 Our team developed a cryogenic method allowing us to freeze and thaw the L. indica strain used in this study, and other spirulina farm cultures, without loss of quality (https://www.sckcen.be/en/services/spirulina-cryopreservation). Such cyroprotectants need to be removed, however, to allow resuscitation, a step which unfortunately was not compatible with the ArtC space flight setup. All tests that were performed on frozen biomass without the addition of freezing reagents resulted in non-revivability (Mastroleo, unpublished data). In addition, freezing of the biomass during upload requires an active energy input, which was not feasible in the ArtC space flight plan. Although biomass frozen without cryoprotectants cannot be used as seed cultures, it has been used to assess the biochemical composition and nutritive value. For example, Papalia, Sidari and Panuccio⁸ used freezing (without medium) to test storage impact on the molecular composition of Arthrospira platensis and showed that the total protein content goes up, because the extraction works better. Similarly, the phycocyanin content increases, allophycocyanin content decreases, and chlorophyll content decreases, all significantly. These differences in extraction efficiency are important to keep in mind when freezing is, for example, used as a sample storage method for post-flight analysis. The pigment content is important for the nutritive value of L. indica, as the current planning of MELiSSA includes the biomass of L. indica as a food source for the space travelers.^2,39 It has also been shown that the pigments are bioactive compounds with possible beneficial effects on human health.^40,41

As a last remark, it must be pointed out that storage is not the only parameter that impacts the biology before the start of a space flight experiment. Also, vibrations and strong g-forces during launch or cosmic radiation can have a negative impact on the survival and revival of cells. In previous work in our lab, it was found that centrifugation and strong vibration can cause the long Limnospira trichomes to break into smaller pieces, which can have an additional negative effect on the regrowth after storage (data not shown). It was also shown that the launch itself, meaning the vibrations and g-forces during launch, can have a significant impact on eukaryotic and prokaryotic organisms (e.g., on primary bone marrow cells and planarians, respectively).^42,43 Another study mentions that the results from on-board controls (1 g centrifuge) and ground controls only give similar results in 50% of the experiments.⁴⁴ Thus, follow-up studies on vibrations and hyper gravity in combination with storage would be beneficial to gain a more complete insight into the impact on biological samples before the experiment starts onboard a space station.

In summary, these storage tests showed that it is feasible to store a culture of L. indica PCC8005 P3 with the right culture characteristics, for two weeks at 4°C in the dark with a low decrease in cell density and a successful and fast revival of photosynthetic growth. Our results show that dry weight, pH, %P1, and antenna pigments per chlorophyll ratios prior to storage are the key parameters to select the optimal culture for maximal revival success after storage. The dry weight should not exceed 1.4 g L⁻¹ and the pH should be below 11. The percentage of P1 cells should be between 4% and 10%. The phycocyanin content per chlorophyll ratio is optimal between 2 and 6, and the allophycocyanin content per chlorophyll ratio should not exceed 2. Moreover, to further improve the storage outcome, providing fresh medium by diluting or washing the culture prior to storage helps to prevent loss of biomass during the storage period. In addition, whenever feasible, a high gas availability (75% v_gas/v_liq) is recommended to secure the oxygen supply to the cells for respiration during the dark storage period. The article provides this insight on the influence of different storage conditions on L. indica to allow restart of photosynthetic growth in space. Additionally, it will help other researchers to design storage experiments for space missions.

Limitations of the study

The test (experiment 6) for the storage of the culture inside the space PBR hardware could only be performed once and on one flight hardware PBR, and no statistical analysis could be performed; thus, the results should be interpreted with caution.

Resource availability

Lead contact

Further information and requests for resources and information should be directed to and will be fulfilled by Dr. Natalie Leys, natalie.leys@sckcen.be.

Materials availability

The specific substrain Limnospira indica PCC8005 P3, which is used in these experiments, is maintained at SCK CEN's private live strain collection and can be requested from the lead contact.

Data and code availability

• •This study did not produce new datasets of a standardized datatype; and therefore, there were no data deposits in public databases. All raw datasets reported in this article and any additional information that is required to reanalyze are available from the lead contact upon request.
• •This article does not report original code. The software used for flow cytometry data analysis is commercially available from the vendor of the machine used (BD ACCURI C6 flow cytometer analysis software, from BD Biosciences). Statistical tests were performed using commercially available Graphpad Prism 9.0 software (Graphpad).
• •This article does not report any other original data or codes.

Acknowledgments

This article has been made possible through the MELiSSA project, the life support system program of ESA (https://www.esa.int/Enabling_Support/Space_Engineering_Technology/Melissa) and Belspo through the ARTEMiSS Prodex contract of the PhD grants of Jana Fahrion. We thank Ilse Coninx and Wietse Heylen for their help in the laboratory and for providing their previous results as anchor points for the presented study. Gabriele Ellena was supported via an Erasmus Traineeship travel grant to join SCK CEN as Master student of Prof. Arianna Mazzoli at the biology Department of the University of Napoli Federico II, in Italy.

Author contributions

J.F. wrote the first article draft. J.F. and G.E. conducted the experiments in the laboratory. J.F. and N.L. planned and designed the experiments. F.M., N.L., and C.G.D. contributed to reviewing, data interpretation, and editing.

Declaration of interests

The authors declare no competing interests.

STAR★Methods

Key resources table

REAGENT or RESOURCE	SOURCE	IDENTIFIER
Bacterial and virus strains
Limnospira indica PCC8005 P3	SCK CEN	N/A
Software and algorithms
BD ACCURI C6 flow cytometer analysis software	BD Biosciences	https://www.bdbiosciences.com/en-be/products/instruments/flow-cytometers/clinical-cell-analyzers/facscalibur/c6-plus-analysis-software-for-pc-or-mac.661083?tab=product_details
Graphpad Prism 9.0	Graphpad	https://www.graphpad.com/

Experimental model and study participant details

The test strain used, L. indica PCC8005 P3, has long straight trichomes (0.8 – 1.5 mm) and is identical to the strain that was chosen for the space flight experiment ArtC. It is easy to homogenize and hardly forms aggregates. The cells sediment quickly under terrestrial gravity conditions and thus, continuous mixing is needed to secure a homogenous suspension. The cyanobacterium L. indica strain PCC8005 subculture P3 was grown in Zarrouk’s medium (staring pH 9.5) as modified by Cogne, Lehmann, Dussap and Gros.⁴⁵ The cultures used in this series of experiments were grown in 250 mL Erlenmeyer flasks (culture volume 150 mL) at 30°C in Binder climate chambers under continuous light flux of 30-45 μmol photons m^-2 s^-1 (full PAR emitting fluorescent TL-D lamps). Air exchange was provided by equipping the Erlenmeyers with cotton stoppers wrapped in aluminum foil. The cultures were shaken at 120 rpm on horizontal shakers and illuminated continuously (no day/night cycle) before storage. All cultures were successfully checked for axenicity by inoculating 1 mL of culture in 10 mL LB medium and additionally, 1 mL culture in a 9 mL Zarrouk and 1 mL LB mixture, each in Greiner BIO-ONE culture flasks (50 mL). If there is no visible microbial growth after 3 days, the cultures were presumed to be axenic.

To monitor revival after storage, regrowth curves were obtained via optical density at 770nm (OD_770nm) and pH measurements (specifications in the next section). The biomass concentration [g L^-1] and OD_770nm have a linear correlation for L. indica PCC8005 P3.⁶ Hence, the slopes of the OD_770nm curves can be directly used to obtain the growth rates in the different phases. Exponential, linear and stationary phases were defined by exponential and linear regression trend lines using Excel 2016. The maximum growth rate μ_max [d^-1] was obtained by using a logarithmic conversion and linear regression of the exponential phase (approximately the first 4-5 days of growth). The average biomass productivity P_av [g L^-1 d^-1] represents the slope of the linear regression of the linear growth phase, starting at the endpoint of the exponential phase and ending at the transition point to stationary phase. The growth curves of the recovering cultures after storage can be found in the supplemental information.

Method details

Optical density and pH measurements

The OD_770nm was obtained using a NANOCOLOR® UV/VIS II spectrophotometer (MACHEREYNAGEL). For this, the cultures were put into 1 mL semi-micro cuvettes (Greiner BIO-ONE). Cultures above an OD_770nm of 0.6 were diluted to be in the optimal range of the spectrophotometer (OD_770nm between 0.1 and 0.6). The OD measurements of the regrowth culture after storage were also used to assess μ_max and P_av (see previous section). The pH of the cultures was assessed in 1 mL Eppendorf tubes (pure culture) using a KCl pH electrode (InLab®, Mettler Toledo).

Dry weight determination

The dry weight [g L^-1] was determined by putting 2 mL of culture on pre-weighed membrane disc filters (water wettable PTFE, Pall Laboratory, pore size 45μm, ⌀ 25mm) using a vacuum pump. The filters were dried for a minimum of 48 h at 60°C and weighed again afterwards. Per condition, a minimum of 6 filters were used to determine an average and standard deviation.

Storage conditions

In total, five different experiments were set up, each focused on one specific parameter possibly influencing the storage (Table 1). All inocula used were in the active linear growth phase of L. indica at the point of harvesting (OD_770nm were found to be between 1.15 (Experiment 5) and 2.11 (Experiment1)). In experiments 1 to 4, 50 mL Falcons filled with L. indica PCC8005 P3 cultures were wrapped in aluminum foil (darkness) and stored at 4°C. In experiment 5, PermaLife cell bags (Origen, PL70) were used on a horizontal rotary cell culture system (RCCS) (SYNTHECON) and a random positioning machine (RPM) (AIRBUS).⁴⁶ A graphic illustration of RCSS and RPM setup for the cultivation of L. indica can be found in Figure S10. Additionally, storage inside the science model and ground model photobioreactors of ArtC was conducted to obtain the best approximation of the conditions before launch (Experiment 6). In the first 4 experiments very similar controls (50 mL falcons filled to the top, no addition of medium, no gas phase available, dark) were used, and they were used to analyze the influence of the harvesting time and metabolic state of the inoculum on the outcome of the storage. All storage experiments discussed in this work were conducted in the dark. To ensure that there was no light available to the cultures, aluminum foil was used to cover all transparent flasks (Experiment 1-4). The cell culture bags in experiment 5 were also covered with aluminum foil. The storage experiments were conducted at 4°C, as this is the temperature used while bringing the L. indica cultures to ISS for the ArtC space flight experiment.

Experiment 1

The influence of different storage durations was tested on a high-density culture (starting OD_770nm = 2.11 ± 0.01) for 7 and 14 days. The 50 mL Falcon tubes were filled completely in these experiments.

Experiment 2

The influence of a dilution of the cultures prior to storage was tested. A culture with an OD_770nm of 1.59 ± 0.01 was divided in three parts: undiluted, 2:1 diluted (OD_770nm = 0.99 ± 0.08) and 1:1 diluted (OD_770nm = 0.81 ± 0.00). The dilution was performed before the start of the experiment, using modified Zarrouk medium: 1 part stayed undiluted (Inoculum in Figure 2), 1 part was diluted 2:1 (IN 2:1 in Figure 2) and the last part was diluted 1:1 (IN 1:1 in Figure 2). The dry weight, OD_770nm and pH of these cultures were assessed before the start of storage on the diluted cultures, because these parameters are dependent on biomass density. After 2 weeks, all cultures were measured again and compared to either only the inoculum (biomass-density independent parameters: %SI, FL3-H/FL4-H, %P1, pigment ratios) or compared to the diluted inocula (parameters dependent on biomass density: dry weight, OD_770nm, pH).

Experiment 3

To test the influence of washing and replenishing with fresh medium and a possible (slow) heterotrophic growth during the storage period, another storage experiment was set up. For this, four different storage conditions were tested. For the first condition, the fresh inoculum was washed 3 times with PBS and resuspended in fresh Zarrouk to remove all extracellular saccharides and waste products that might have accumulated in the spent medium, during preculture. For the second condition, the washed cells were supplemented with 1.5 g L^-1 glucose monohydrate in fresh Zarrouk. Additionally, an unwashed culture and a 1:1 diluted unwashed culture were stored as controls. To exclude a possible effect of centrifugation on the washed cultures, the unwashed and diluted cultures were centrifuged as well. The effects of dilution and centrifugation on the dry weight measurements can be found in the supplementary data.

Experiment 4

To assess the influence of a gas exchange between culture and ambient air, different gas phase to liquid ratios were investigated. The investigated percentages were 0% (no gas phase, control), 25% (12.5 mL air and 37.5 mL culture), 50% (25 mL air, 25 mL culture) and 75% (37.5 mL air, 12.5 mL culture). The 50 mL Falcon tubes used in this experiment were therefore filled to different volumes, while the area of possible gas exchange stayed the same.

Experiment 5

Since microgravity is one of the biggest environmental differences between Earth and space and storage of the cultures before start of a space flight experiment is partly performed onboard ISS, the influence of simulated microgravity on the storage was investigated. For this, a horizontal RCCS was used as a control (low-shear but normal gravity vector) and an RPM was used to simulate microgravity (low-shear randomized simulated microgravity) (Figure S10).⁴⁷

Experiment 6

The experiments inside the ArtC hardware (science model 2 reactor, SM2) were conducted with and without precooling, meaning that the hardware including the culture was either placed at 4°C and closed after acclimatization (with precooling) or closed before the transfer to the fridge (without precooling). The closure after precooling is ought to prevent degassing of CO₂ from the silicon tubings. When no precooling is performed, the drop in temperature in the closed air-tight hardware causes a decrease in pressure around the air permeable silicon tubings, and therefore a degassing from these tubings takes place. Due to the limited availability of the photobioreactors for these tests, statistical analysis of these results was not possible. Graphs on pressure, temperature and humidity over time with and without precooling can be found in the supplementary data (Figure S9).

Sedimentation analysis

The sedimentation behavior of the stored and control cultures was assessed. Equation 1 shows the so-called Sedimentation index (%SI) as described in Deschoenmaeker, Facchini, Leroy, Badri, Zhang and Wattiez.⁴⁸ %SI describes the ratio of a difference in OD₇₅₀ at time t_x versus the original OD₇₅₀ at time t₀ as a percentage. For all sedimentation experiments, the OD of the cultures was measured at 750 nm at several time points (t = 0, 10, 15, 30, 45, 60, 90 and 120 min). The closer %SI gets to 1, the stronger the sedimentation of the cells out of the suspension.

$$\%SI=\frac{{OD}_{750}\left(t_0\right)-{OD}_{750}\left(t_x\right)}{{OD}_{750}\left(t_0\right)}\ast 100$$ (Equation 1)

The cells are settling faster the larger size and the higher density (mainly because of the absence of gas vacuoles and carbon stocks) they exhibit. Strongly fragmented trichomes are sedimenting very slowly and a fully dead culture sediments almost instantaneously because all buoyancy is lost. The amount and size of internal vacuoles plays a role, because they increase the buoyancy of the cell.^34,48,49 For statistical analysis, the %SI at t = 60 min was used.

Flow cytometry analysis

Flow cytometry measurements were performed using an ACCURI C6 flow cytometer (BD Biosciences) equipped with a blue (488nm) and a red (640nm) laser. 100 μl of Limnospira culture were run in fast mode using the following thresholds: 10,000 in the forward scatter (FSC) and 800 in the fluorescent detector 4 (FL4-H), using the modified Zarrouk medium as blank background. The forward scatter (FSC) detector measures the light scattered in front of the sample when hit by the laser and gives information about the size of the cell or filament. The threshold on the value 10,000 allows to remove the background signal coming from cell-free medium during the analysis. FL3-H, associated to a 670 nm long pass filter and FL4-H, associated to a 675±12.5 nm filter, were used to detect pigments auto fluorescence, respectively chlorophyll a and antenna pigments (phycocyanin and allophycocyanin). The fluorescent detector 4 (FL4) specifically measures the light in the range 675±12.5 nm which is typical for detecting pigment autofluorescence, respectively chlorophyll a and antenna pigments (phycocyanin and allophycocyanin). The threshold on the value 800 allows to remove the background signal coming from cell-free medium during the analysis. The software used for flowcytometry data was provided by the vender of the machine used (BD ACCURI C6 flow cytometer analysis software, from BD Biosciences). Previous work in our laboratory reported the importance of a population (P1) within the total cell count cytogram, for the survival of a culture. This population shows as a distinct section visible in the flow cytograms of all healthy and revivable Limnospira indica cultures (Figure S11 and Table S2). Our previous research showed that when this P1 population is not visible in the flow cytogram anymore (e.g. after freezing and other harmful treatments), the culture will not regrow when transferred to fresh medium (F. Mastroleo, unpublished data). Additionally, all cultures with a P1 population were shown to be able to regrow. Thus, the abundance of this population can be used as a marker of revivable cultures. Additionally, the healthy cultures were shown to have a FL3 to FL4 ratio around 0.3. The higher this ratio gets, the worse the survival rate of the culture. Therefore, the percentage of P1 and the FL3-H/FL4-H ratio are used as measurements of survivability, with a %P1 of 0.0 and a FL3-H/FL4-H ratio of > 2 as indicators for an impaired survivability.

Pigment analysis

The protocol for pigment analysis used was slightly adapted from Badri, Monsieurs, Coninx, Wattiez and Leys⁵⁰ and was already described in our previous work.⁶ The protocol allowed to measure the concentrations of allophycocyanin, phycocyanin and chlorophyll spectrophotometrically. It was noticed that optical density as well as dry weight are highly influenced by the different storage conditions and are therefore not suitable for normalization purposes. Thus, the pigment content is reported here as ratios of the antenna pigments (allophycocyanin and phycocyanin) to chlorophyll ratio.

Quantifications and statistical analysis

Statistical tests were performed using Graphpad Prism 9.0 software (Graphpad). When three or more groups were compared, Kruskal-Wallis tests followed by Dunn’s multiple comparison tests were performed. Mann-Whitney U tests were used to compare 2 groups. Significant difference is assumed in this study as followed: ∗: p < 0.05, ∗∗: p < 0.01, ∗∗∗: p < 0.001, ∗∗∗∗: p < 0.0001. To obtain the specific growth rate (μ_max) and the corresponding 95% confidence intervals (CI) during the exponential growth phase, the OD_770nm values were converted using the natural logarithm followed by linear regression of the experiment time versus ln [biomass concentration (g/L)] (Excel 2016 data analysis tool pack). Statistical significance was investigated by comparing the mean μ_max values of one condition with the 95% confidence intervals (CIs) of the other conditions. The average biomass productivity during linear phase (P_av) and the corresponding 95% CIs were obtained similarly, but by directly using linear regression [OD_770nm = P_av x + C]. Correlation analysis between the parameters and the resulting change in OD_770nm was performed using Spearman correlation in Graphpad Prism 9.0 software.

Published: September 4, 2025