Since the year 2000, humans have been continuously present in space on the International Space Station (ISS). Due to its relative proximity to Earth and regular space flights, replenishment of resources such as food and water is ensured to the crew onboard ISS. However, future plans for long term human spaceflight beyond the low Earth orbit or establishment of colonies with a larger crew, will bring critical challenges connected to resupply and waste management [1
]. To reduce the need for replenishment, transport mass and costs, different concepts for Bioregenerative Life Support System (BLSS) are developed for future in situ food production in space. Higher plants are foreseen to be an essential part of such systems [2
]. As reviewed by Wheeler [5
], ground demonstrations and plant research for BLSS have been performed by the major governmental space agencies for the past half century. In addition to ground-based research, crop cultivation experiments under space conditions with reduced gravity are required [1
]. Reduced gravity is expected to influence plant physiology, nutrient uptake and thereby growth speed and potentially nutritional value in space grown crops [3
]. On the ISS, scientific work and technology demonstrations can be performed with fractional gravities including microgravity and simulated Moon and Mars gravity using research facilities with centrifuges [8
In the EU Horizon 2020 TIME SCALE project, an advanced crop cultivation system prototype was developed for imminent use on an existing centrifuge on the ISS (Figure 1
). The crop cultivation concept developed comprises a system to facilitate both technology demonstration and research on algae or plants in fractional gravity. Cultivation can be performed with or without substrate; i.e., deep water culture, which allows pure nutrient research unaffected by soil properties. The system contains two growth chambers per centrifuge rotor with nutrient solution volumes as large as allowed by the centrifuge diameter of 600 mm (Figure A1
). Each growth chamber is connected to independent systems that can monitor plant health, provide light, recycle water, and manage nutrient solution electric conductivity (EC) and pH. A multi-ion (NO3−
) sensor monitoring system (CleanGrow, Wolverhampton, United Kingdom) is able to accurately detect dynamics in macro nutrient uptake. The crop cultivation chambers are interchangeable with algae cultivation chambers [10
]. All other parameters required for optimal cultivation (temperature, humidity, and CO2
) are ensured by integrating the system in an incubator on ground or on the ISS.
Restricted growth volumes, as used in our crop cultivation system (Figure A1
), can result in a wide range of physiological and morphological responses [12
]. This effect is undesirable because it will confound the fractional gravity effects assessed on the ISS. Moreover, nutrient supply is an important environmental signal that strongly affects root development [13
] and shoot gas exchange rates [14
]. As gas exchange in micro gravity is severely hindered by the lack of buoyant thermal convection [18
], any potential regulatory effects of nutrient concentration on plant water fluxes could be used to enhance transpiration in space crop cultivation systems. The effects of transpiration on nutrient acquisition are well documented (reviewed by Tibbitts [19
]): transpiration drives the mass flow of nutrients from the soil to the roots [14
], and aids in translocation of nutrients within the plant [20
]. In turn, nutrient availability can influence transpiration [14
]. Although the regulating mechanisms are not clear, nitrogen (N) is among the elements proposed to have a role in regulating plant water fluxes [14
]. Plant fertilization with low or restricted N, causing N limitation but not deficiency, has induced increased stomatal conductance and transpiration in maize [16
] and bean [15
] as compared to deficient and supra optimal concentrations. According to Wilkinson, Bacon and Davies [16
] the observed response is stronger in well-watered plants, pointing in the direction of interaction between N concentration and stomatal conductivity.
The current paper aims to address three aspects of plant (Lactuca sativa) production in confined and closed loop cultivation systems:
The effect of a restricted rooting volume was studied by comparing a small (0.6 L) and a large (3.5 L) root container. We hypothesize that as longs as the conditions in both containers are similar there will be no effect of the root container.
The effects of a limited amount of nutrient solution were tested by comparing a 3.4 L nutrient solution for the cultivation of two lettuce heads to plants which receive an unlimited supply of fresh nutrient solution. We hypothesize that plants in both systems will be similar.
The effects of nitrate concentration on stomatal conductance, transpiration and nitrate uptake in intact lettuce was studied by growing plants on different nitrate concentrations; causing growth limitation but no morphological deficiency symptoms. To look for variations throughout the diel cycle, conductance and transpiration was measured during both dark and light conditions. We hypothesize that nitrate concentration has a regulating effect on plant water fluxes and that the relation between nitrate concentration and transpiration can be represented by a “bell curve” as described by Wilkinson, Bacon and Davies [16
]. That is, when nitrate is supplied in a concentration range between 0 and 30 mM plant responses will gradually increase until reaching an “optimum concentration” at which transpiration peaks and then declines as nitrate concentrations becomes supra optimal.
2. Materials and Methods
Three experiments were performed: the first two experiments determined the effect of restricted rooting- and nutrient solution volumes (Figure A2
), and based on this a third experiment was performed to assess plant responses to various nitrate nutrient solution concentrations.
2.1. Plant Material and Growth Conditions
Lettuce (Lactuca sativa
, cv. Cecilia RZ butterhead) seeds, from Rijk Zwaan Nederland B.V. De Lier, The Netherlands, were sown in round seed holders filled with vermiculite and water. All plants were cultured in climate chambers at Wageningen University under a photoperiod of 16 h. CO2
concentration, temperature, and relative humidity (RH) were controlled and recorded by a “Hoogendoorn®
climate control system”. Temperature was set to 24/19 °C (day/night), relative humidity to 75% and CO2
concentration to 400 ppm (ambient). Light was provided by fluorescent-tubes (T5-36W, Philips, Eindhoven, The Netherlands), average light intensity at plant height was 335 µmol·m−2
PAR (Photosynthetic Active Radiation). Harvested plants were directly stored in a cooling box and root and shoot fresh weights, leaf area, leaf number and both rooting volume and length were determined within two hours after harvest. To obtain dry weights (biomass), shoots and roots were dried separately at 70 °C until constant weight (max. 4 days). The leaf area (cm2
) was measured using a Li-Cor-3100 (Li-Cor Biosciences, Lincoln, NE, USA) and a flatbed scanner for early growth stages. In the first two experiments, to get an impression of relative differences in stress or nutrient shortages the Dualex ScientificTM
was used. The DualexTM
provides relative estimates of concentrations of chlorophyll, flavonoids, and anthocyanins (http://www.force-a.com/en/publications
). Plant nitrogen content was determined by a LECO element analyzer at the Chemisch Biologisch Laboratorium Bodem (CBLB) lab of the Wageningen University.
2.2. The Effects of a Restricted Rooting- and Nutrient Solution Volume
In the first experiment the effects of rooting volume per se were determined and in a second experiment the effects of a limited nutrient solution were determined. For both experiments the nutrient solution had an EC of 1.65 dS·m−1 and was composed of the following ions in mmol L−1: NO3− 9, NH4 1.5, P-H2PO4− 1.5, K+ 5.5, Ca2+ 3, Mg2+ 1, SO42− 1.5, Cl− 1.5, Si 0.5, and in µmol L−1: Fe 28.1, B 47, Cu 1, Zn 6.4, Mn 1.5, Mo 0.7. EC and pH were measured both in the main solution tank and in the growth units. EC was maintained daily by adding small amounts of deionized water when the EC increased due to evapotranspiration and pH was maintained in the range of 5.6–6.0 with citric acid (0.1 M) and K2HCO3 (0.1 M) added to the main solution tank when necessary.
For the root volume experiment two container types were used: small root containers, inner dimensions 105 × 105 × 75 mm; i.e., similar size as in the space crop cultivation concept, which were filled with 0.6 L nutrient solution; and large root containers, inner dimensions: 265 × 165 × 115, which were filled with 3.5 L nutrient solution. Ten blocks containing one replicate of each small and large root container were distributed over the climate room. The system was setup such that all root containers received the same nutrient solution from a 100 liter nutrient tank. Solution was pumped (‘Eheim Universal’ 1200 L/h, EHEIM GmbH & Co. KG, Germany) via a distribution tube into the root containers. By adjusting the inlet valves of each container, the flow rate of the nutrient solution was set to approximately 21 L h−1
. The drain tube in container was located at the upper part of the containers. The drain was collected into a main drainage pipe and returned to the main nutrient solution tank by gravity, creating a closed nutrient solution loop (Figure A2
). An air pump (28 L min−1
) pumped air to a main PVC distribution pipe Ø 4 cm from which air out flow to each root container was controlled per container. This insured the equal distribution of air to maintain O2
saturation levels in the root zones of all the units. The 100 L nutrient solution tank was refreshed weekly and as a check samples of the solution were taken every 5 days and sent to the lab facilities of Eurofins Agro NL, Wageningen, the Netherlands, for ion concentration analysis. EC, pH and dissolved oxygen where measured with a calibrated Orion Star™ A329 pH/ISE/Conductivity/Dissolved Oxygen sensor. Four plants were harvested after: 10, 15, 20, 25, 30 and 35 days. Fresh and dry mass of both root and shoot, leaf area, root length and root volume were obtained. DualexTM
measurements were performed on both sides of two leaves per plant just before harvest.
In the second experiment on nutrient solution volume the small root containers (0.6 L) were used. Five blocks containing two root containers each were distributed over the climate room. The two containers were either connected to a container with 3.5 L nutrients solution that was not refreshed; i.e., limited nutrient solution treatment, or to a main system with 100 L that was refreshed weekly, ‘unlimited’ nutrient solution treatment. Environmental conditions for both experiments were similar except for CO2 concentration that was 1000 ppm in the nutrient solution experiment opposed to 400 ppm in the rooting volume experiment.
For both the first and second experiment treatment means per plant measurements, i.e., root length, root volume, leaf area and both root and shoot dry and fresh mass were compared per time point using ANOVA for a randomized block design using five blocks (‘aov’ function in R version 3.5.1.). Assessment for significant differences between means was done using Fishers Least Significant Difference (LSD) test (p ≤ 0.05).
2.3. Plant Responses to Various Nitrate Concentrations
Two experimental runs were performed to determine the effects of nitrate (NO3
) concentrations on lettuce biomass, nitrogen content, stomatal conductance and transpiration. In the first run (hereafter referred to as Run 1), plants grew in five different nutrient solutions with NO3
concentrations ranging from 2.5–30 mM (Table 1
). Preliminary analysis of Run 1 indicated a curvilinear response to NO3
concentration, with a decline in stomatal conductance (gs
) and transpiration (E) mainly under dark conditions. Based on this, a second run (Run 2) including more replicates and lower NO3
concentrations than 2.5 mM, was performed to further explore the plant responses in the lower concentration range and improve data resolution. Following gas exchange measurements on day 12 (Run 1) and day 8 (Run 2), plants were harvested and measured for root length, leaf number, leaf area, fresh biomass of shoots, and roots separately.
2.3.1. Nutrient solution formulation for nitrate treatments
Based on the root volume experiment, the nutrient solution of the breeding phase and used as starting point for the different nitrate treatments was composed of the following ions in mmol L−1
7, Si 0.5, and in µmol L−1 :
Fe 28.1, B 47, Cu 1, Zn 6.4, Mn 1.5, Mo 0.7. To allow for the various N concentrations while keeping EC constant, EC was kept at 3.3 dS·m−1
by substituting NO3−
). At seedling emergence, four days after sowing (DAS) plants were exposed to a 1/3rd (1.1 dS·m−1
) nutrient solution, six DAS to a 2/3rd (2.2 dS·m−1
) and eight DAS to a full strength (3.3 dS·m−1
) nutrient solution. At 16 DAS, plants with similar size and dimension were transferred to the experimental root containers (265 × 165 × 115 mm, same as large root container in the root volume experiment) prefilled with the different nutrient solution treatments (3.5 L in each container, filled to 35mm under the lid). Nutrient solution composition was determined before onset of treatments (without plants) and at experiment end (after harvest). Samples of 50 ml were taken from each growth pot and analyzed for all essential plant nutrients (Eurofins®
2.3.2. Stomatal Conductance and Transpiration Rate Measurements
After exposure to the different nitrate concentration treatments (Table 1
), stomatal conductance and transpiration were measured during the light and the dark period with a LI-6400 Portable Photosynthesis System (Li-Cor Biosciences, Lincoln, NE, USA) equipped with a leaf chamber fluorometer (area = 2 cm2
) and controlled LED source (90% Red + 10% Blue). Three hours after start of the dark or light period five measurements per leaf, one every five seconds were recorded. In order to reduce the variance between plants the distal part of the last fully expanded leaf blade, avoiding the leaf vein, were measured. Rates of night-time transpiration were measured with the same climatic settings as the in the cultivation chamber and a PPFD of 0 µmol m−2
. Low intensity green light (<0.07 µmol m−2
at plant level) from a fluorescent tube was used as working light during dark measurements. All plants in one block were measured consecutively.
In Run 1, analyses showed an inaccuracy in nutrient solution mixing and all treatments were replaced prolonging the experiment with four days. This affected plant size at the time of measurements and its` potential impact on the results are included in Section 3
. Consequently, the plants in Run 1 were measured on day 11 (dark) and 12 (light) after onset of treatments, while in Run 2 plants were measured on day 7 (dark) and 8 (light) after onset of treatments. As suggested by I. Matimati (pers. comm. January 2017), Run 2 included additional measurements 24 h after onset of treatments (in dark only) to look for plant initial responses to nitrate treatments.
2.3.3. Statistical Set-Up and Analysis
Both experiment runs were set up as complete randomized block designs; comprising 5 treatments, one plant per treatment and 6 blocks (replicas) in Run 1, and 7 treatments and 8 blocks in Run 2. To analyze the effects of nitrate concentration on transpiration (E) and stomatal conductance (gs
), we ran linear mixed effects regression models (LMERs) from the lmer function in package lme4 for R
. The five data points per leaf sample from the LiCor 6400 photosynthesis system (E and gs
) were aggregated using the median. LMERs were run with NO3
concentration as a continuous variable (log-transformed and adding 0.1 to avoid log of 0). First, we built a global model for model selection based on the Akaike Information Criteria corrected for sample size (AICc; Burnham [26
], Table A2
). The global model included the main effects of experimental run (two-level factor) and light conditions (two-level factor) and their two-ways interactions with both a linear and a quadratic term for NO3
. Plant ID nested within Block was included as random intercept effects. In the model selection, all possible subsets of this global model were included, keeping the random structure constant. A similar approach was used to evaluate effects of nitrate concentration on plant biomass and root and shoot tissue N content, except that the factor Light Conditions (and hence Plant ID) was not included as measurements were performed only once at harvest. The residuals were tested for normality and homogeneity of the distribution using Shapiro–Wilk test and QQ-plots. All analyses were performed in R version 3.3.2 (R core Team, 2016).
For space plant research both the confined environment and mass restriction favor recycling of all resources. Soilless cultivation systems provide effective control in terms of nutrient solution monitoring, adjusting, and recycling. The deep water culture with limited root- (0.6 L) and nutrient solution volume (3.4 L per two plants) used in this study, provided stable and reliable plant growth and high biomass production over a period of at least 24 days. Nitrate concentrations as low as 1.25 mM did not reduce biomass and plant N content.
As expected, the absence of nitrate in the nutrient solution results in low transpiration (E) and conductance (gs). When moving from zero to 1.25 mM the increase in E and gs was much steeper in light than in dark. At concentrations above 1.25 mM, no response was detected in light, while night time gs and night E decreased in response to nitrate concentrations above 5 mM.