Nielsen, S.S., M.A. Belury, K.P. Nickel, and C.A. Mitchell. 1996. Plant nutrient composition altered with controlled environments for future space life-support systems. p. 624-532. In: J. Janick (ed.), Progress in new crops. ASHS Press, Arlington, VA.

Plant Nutrient Composition Altered with Controlled Environments for Future Space Life-Support Systems*

S. Suzanne Nielsen, Martha A. Belury, Kwangok P. Nickel, and Cary A. Mitchell

1. METHODOLOGY
1. CELSS Candidate Crops
2. Analysis of Plant Materials
3. Calcium Bioavailability Study Using Rat
2. RESULTS
1. Proximate Composition of CELSS Candidate
2. Nitrogen Allocation
3. Calcium and Phosphorus Content
4. Calcium Bioavailability
3. SUMMARY
4. REFERENCES
5. Table 1
6. Table 2
7. Table 3
8. Table 4
9. Table 5
10. Table 6

Future manned outposts on the Moon and Mars will rely on self-sustaining life-support systems that utilize plants to revitalize atmosphere, recycle water, and provide edible biomass for vegetarian diets without resupply from Earth (Mitchell 1993). The crops selected for such Controlled Ecological Life-Support Systems (CELSS) most likely will be grown hydroponically under controlled environment conditions (Mitchell 1994).

There is increasing evidence that the nutritional content of food crops is modified by controlled environments (CE) relative to soil culture in the field (McKeehen et al. 1996 a,b). This finding raises questions regarding diet composition, food safety, bioavailability, and palatability for limited vegetarian diets.

Calcium metabolism during long-term space flight is of great concern due to bone demineralization caused by a weightless environment (Schneider et al. 1989). Since the first generation of CELSS diets likely will be composed of plant-derived sources, it is important that such diets provide sufficient levels of available calcium. Calcium bioavailability can be affected by dietary levels of phytate and oxalate, which form insoluble complexes of calcium salts. Pure vegetarian diets may contain high amounts of these antinutrients (Havala and Dwyer 1993). Because the risk of renal stone formation is high due to increased urinary calcium excretion during space flight (Lane et al. 1993), calcium oxalate stone formation could be minimized by controlling the level of oxalate in the diet.

The present study compared the macronutrient content of the edible portion for nine CELSS candidate crop species grown in controlled environments or in the field, investigated the mineral content and the partitioning of N into components of nutritional significance in plant biomass as a function of growth environment, and investigated the calcium bioavailability of CELSS diets using a rat model.

METHODOLOGY

CELSS Candidate Crops

Select cultivars of nine CELSS candidate crops [soybean (Glycine max (L.) 'Hoyt'), cowpea (Vigna unguiculata (L.) 'IT87D-941-1'), peanut (Arachis hypogaea (L.) 'Ga Red'), wheat (Triticum aestivum (L.) 'Yecora Rojo'), rice (Oryza sativa (L.) 'Ai-Nan-Tsao'), canola (Brassica napus (L.) 'Westar'), potato (Solanum tuberosum (L.) 'Norland'), sweet potato (Ipomoea batatus (L.) Lam 'Tu-82-155'), and lettuce (Lactuca sativa (L.) 'Waldmann's Green')] were grown under controlled environment (CE) conditions, and the same cultivars of all but cowpea and peanut were grown under field conditions (see Table 1). The typically edible part of each plant was dried and ground prior to analysis or incorporated into animal diets.

Analysis of Plant Material

Fat, ash, carbohydrate, total N, protein N, total nonprotein N (NPN), nitrate N, and nonnitrate NPN were determined as described by McKeehen (1994) for seeds or grain of CE and field-grown soybeans, wheat, rice, and canola, for tubers of CE and field potato, for leaves of CE and field lettuce, and for seeds of CE cowpea. For all other CE and field samples (values taken from the literature or by personal communication) and for all handbook (Haytowitz and Matthews 1986) values, protein content presumably was determined by the standard Kjeldahl procedure, fat and ash contents presumably were determined by standard AOAC (Association of Official Analytical Chemists International) methods (AOAC 1990), and carbohydrate content was calculated by difference [Carbohydrate = 100% - (% protein + % fat + % ash)]. Mineral analysis [calcium (Ca) and phosphorus (P)] were performed by Inductively Coupled Plasma-Atomic Emission Spectroscopy (ICP-AES), as described by McKeehen (1994). Energy content for all crops was calculated as: kcal/100g dry matter = (g protein/100g ◊ 4 kcal/g) + (g carbohydrate/100g ◊ 4 kcal/g) + (g fat/100g ◊ 9 kcal/g).

Calcium Bioavailability Study Using Rats

Diets formulated with CELSS candidate crops were used to study calcium bioavailability in a rat model system. Calcium bioavailability in CELSS diets was evaluated by the femur 45Ca uptake method (Weaver et al. 1987), which has been validated for the absorption of calcium in many previous studies (Buchowski et al. 1989; Koo et al. 1993).

Forty male Sprague-Dawley rats (100-125 g) (Harlan Sprague-Dawley, Indianapolis, IN) were housed in stainless steel cages with a reverse 12 h light-dark cycle. Animals were randomly assigned to groups of 8 rats each.

All the crops except kale, broccoli, mustard greens, bokchoy, and canola oil were grown in the semi-controlled environment of the Purdue University Horticulture greenhouses. Vegetables were freeze-dried and diets were prepared in a powdered form. The composition of CELSS diets is shown in Table 3. Diets 1 and 2 contained lower levels of calcium compared with the control diet. Diet 3 was formulated with calcium-rich vegetables (kale, bokchoy, broccoli, and mustard greens) (Tables 3 and 4). Because calcium is most efficiently absorbed at low levels (Heaney at al. 1975), all diets were adjusted to contain similar levels of calcium. Therefore, calcium as CaCO₃ was supplemented in Diets 1 and 2. Rats were given free access to the semi-purified control diet [(American Institute of Nutrition (AIN)-76A] and deionized water for one week. They were meal-trained by fasting for 12 h each day for 3 d before receiving test meals labeled with ⁴⁵Ca. Four g of three CELSS diets and the control diet were extrinsically labeled with 7 µCi⁴⁵Ca as a tracer and test meals were fed to rats. One group of rats was injected intraperitoneally (IP) with 7 µCi⁴⁵Ca in 300 µl 0.9% saline 2 h after receiving an unlabeled 4 g of the control diet. It was assumed that the radioactivity of an IP dose that reached the femur represented 100% calcium absorption. Rats were allowed to consume test meals for a 4-h period and the remaining food was collected for counting of radioactivity in order to calculate the exact amount of labeled food intake. All rats were sacrificed with CO₂ 24 h after administration of ⁴⁵Ca and the femurs were removed. Femurs were dried in a vacuum oven at 70°C overnight, dissolved with 3 ml concentrated HNO₃ in 25 ml volumetric flasks, and diluted with double deionized water. A 1 ml aliquot was counted for radioactivity by liquid scintillation spectroscopy. Absorption of ⁴⁵Ca from the test meal was calculated from the ratio of the percent ⁴⁵Ca dose accumulated in the femur compared to that in the femur of rats receiving the IP injection.

RESULTS

Proximate Composition of CELSS Candidate Crops

Nitrogen determination presumably was the only difference between assays of samples according to methods reported by McKeehen (1994) used for most CE and field samples of CELSS crops, compared to the assays used to obtain all other CE, field, and handbook values. The standard Kjeldahl procedure (AOAC 1990) is commonly used for protein analysis, but it measures both protein N and some, but not all, of the nitrate N. This method alone is inappropriate to determine the true protein content of plant materials that have a significant fraction of N as nitrate (Goyal and Hafez 1990). McKeehen (1994) used a total N procedure by Goyal and Hafez (1990), with a predigestion procedure to include NO3- N in the Kjeldahl digestion. To differentiate protein N from total N, McKeehen (1994) determined protein N content after precipitating protein in 6% trichloroacetic acid (TCA) (Bensadoun and Weinstein 1976), and then measuring N content of the washed pellet by the Kjeldahl method (AOAC 1990).

Protein content values obtained under CE were higher than field and handbook values of the typical edible portion of CELSS candidate crops (Table 1). Some handbook values are considerably lower than CE or field values, likely in part because of differences in the N analysis methods described herein. The higher total N content in CE conditions compared to field conditions is attributed to luxuriant uptake of N by hydroponically grown plants. Fat and ash contents were relatively unaffected by environment. For most species, carbohydrate content values tended to be lower under CE than field or handbook values. Specific energy content was relatively unchanged by different growth environments.

Nitrogen Allocation within CELSS Candidate Crops

Protein content calculated from true total N for the typical edible portion of CELSS candidates was higher under CE than under field conditions (Table 2). This total N included a significant amount of NPN. Total NPN was highest for soybean seeds (CE and field) and potato tubers (CE). This NPN did not include nitrate in seeds or tubers, but nitrate was significant in CE-grown leaf lettuce. Nitrate is known to accumulate in the vegetative material of rice, wheat, white potato, soybeans, and cowpeas when grown under CE conditions (McKeehen et al. 1996a, Nielsen 1995).

Calcium and Phosphorus Content

The Ca/P ratios of edible plant parts differed for CE compared to field and handbook values (Table 5). This ratio ideally is maintained at a ratio of 1:1 in the diet for proper calcium absorption and retention in humans (Stare and Williams, 1984). The Ca/P ratios of all seeds, grains, and tubers were quite low due to low Ca levels in the typical edible parts of most plants. Especially for potato under CE, Ca content was reduced and P content increased, to reduce the Ca/P ratio. Although the Ca content is high in green, leafy vegetables such as lettuce, the higher P content of lettuce leaves under CE compared to field conditions led to a Ca/P ratio under CE half that of under field conditions.

Calcium Bioavailability

The average Ca absorption by rats fed CELSS diets was slightly higher than that of the control diet (Table 6). However, there was no significant difference (p = 0.34; ANOVA) in Ca absorption between control and CELSS diets in spite of different levels of phytate, oxalate, and total dietary fiber (Nickel et al. 1996). Calcium supplementation of Diets 1 and 2 showed Ca absorption to exceed that of the control diet. These data support earlier work suggesting that Ca supplementation of CELSS diets may be necessary for optimal Ca absorption in rats (Saha et al. 1996). It is notable that Diet 3 containing Ca-rich vegetables showed high bioavailability of Ca.

SUMMARY

Hydroponic growth of food crops under CE conditions represents an opportunity to manipulate and control the composition of edible plant parts. The predictability of biomass composition for crops grown under standardized, modified conditions represents a form of "nutritional value added" in preparing balanced vegetarian diets for crews of long duration space missions. Hydroponic culture conditions combined with optimizing environments enhanced the protein content of the edible portions of soybean, wheat, rice, potato, and lettuce. Since legumes and cereals constitute the staple basis of vegetarian diets, amounts of both types of crops grown in CE would have to be reduced when balancing dietary protein for composition and amount. Because protein and carbohydrate vary reciprocally but fat is relatively unaffected by growth environment, caloric density is similar across growth environments for each crop. CE conditions enhanced the true protein content of soybeans and rice in particular. For the starch crops rice, wheat, and potato, total nitrogen (N) and non-protein N (NPN) contents of plant biomass generally increased under CE/hydroponic conditions relative to the field, especially for leafy plant parts and roots. The nature of the NPN in seeds and tubers is yet uncharacterized in terms of diet safety and nutrition. Vegetative plant material is known to accumulate nitrate and other NPN (Aldrich 1980). Potentially toxic NO₃^- accumulated in vegetative tissues of CE-grown plants (e.g., lettuce) but was excluded from seeds and tubers. For lettuce leaves and radish roots, CE/hydroponic culture generally increased total N, protein N, and NO₃^- . While nitrogen-containing compounds are vital components for optimal crop yield, excess dietary nitrogen may be toxic to humans. Controlled environments enhanced phosphorous content of leaf lettuce, potato tuber, and rice grain. Controlled environments decreased the calcium/phosphorus ratio of lettuce leaves and white potato tubers. A calcium bioavailability study in rats suggested that calcium supplementation of diets from CE plants may be necessary for optimal calcium absorption. Bioavailability of minerals and macronutrients are issues deserving attention in CELSS vegan diets, and the degree to which controlled environments enhance or retard content and bioavailability require further research.

REFERENCES

• Aldrich, S.R. 1980. Protein, nitrates, and nitrites in foods and feeds, p. 97-110. In: Nitrogen in relation to food, environment, and energy. Agr. Expt. Sta., Univ. Illinois, Urbana-Champaign.
• AOAC. 1990. Official methods of analysis. 15th ed. Association of Official Analytical Chemists, Arlington, VA.
• Belitz, H.D., and W. Grosch (ed.). 1987. Food chemistry. Translation from 2nd German ed. Springer Verlag, Berlin, Germany.
• Bensadoun, A., and D. Weinstein. 1976. Assay of protein in the presence of interfering materials. Anal. Biochem. 70:241-250.
• Buchowski, M.S., K.C. Sowizral, F.W. Lengemann, D.W Campen, and D.D. Miller. 1989. A comparison of extrinsic tracer methods for estimating calcium bioavailability to rats from dairy foods. J. Nutr. 119:228-234.
• Goyal, S.S., and A.A. Hafez. 1990. Quantitative reduction and inclusion of plant tissue nitrate in Kjeldahl digestion. Agron. J. 82:571-576.
• Havala, S. and J. Dwyer. 1993. Positon of the American Dietetic Association: Vegetarian diets. J. Am. Diet. Assoc. 93:1317-1319.
• Haytowitz, D.B. and R.H. Matthews. 1986. Composition of foods. Ag. Handb. 8. USDA, Washington DC.
• Heaney, P.R., P.D. Sarille, and R.R Recker. 1975. Calcium absorption as a function of calcium intake. J. Lab. Clin. Med. 85:881-890.
• Koo, J., C. M. Weaver, M.J. Neylan, and G.D. Miller. 1993. Isotopic tracer techniques for assessing calcium absorption in rats. J. Nutr. Biochem. 4:72-76.
• Lane, H.W., A. D. LeBlanc, L. Putcha, and P.A. Whitson. 1993. Nutrition and human physiological adaptations to space flight. Am. J. Clin. Nutr. 58:583-588.
• McKeehen, J.D. 1994. Nutrient content of select controlled ecological life-support system candidate species grown under field and controlled environment conditions. M.S. Thesis. Purdue Univ., West Lafayette, IN.
• McKeehen, J.D., C.A. Mitchell, R. M. Wheeler, B. Bugbee, and S.S. Nielsen. 1996a. Excess nutrients in hydroponic solutions alter nutrient content of rice, wheat, and potato. Adv. Space. Res. 18(4/5):73-83.
• McKeehen, J.D., D.J. Smart, C.L. Mackowiak, R.M. Wheeler, and S.S. Nielsen. 1996b. Effect of CO₂ levels on nutrient content of lettuce and radish. Adv. Space Res. 18(4/5):85-92.
• Mitchell, C. 1993. The role of bioregenerative life-support systems in a manned future in space. Trans. Kans. Acad. Sci. 96(1-2):87-92.
• Mitchell, C. 1994. Bioregenerative life-support systems. Am. J. Clin. Nutr. 60:820S-824S.
• Nickel, K.P., Nielsen, S.S., Mitchell, C.A., and Belury, M.A. 1996. Calcium bioavailability of vegan diets for a controlled ecological life-support system. FASEB J. 10:A232.
• Nielsen, S.S., W.E. Brandt, and B.B. Singh. 1993. Genetic variability for nutritional composition and cooking time of improved cowpea lines. Crop Sci. 33:469-472.
• Nielsen, S.S., L.J. Jurgonski, D.J. Smart, and B. Burgbee. 1996. Controlled environments alter nutrient content of soybeans. Paper presented at Council on Space Research (COSPAR) Meeting, Birmingham, England, July 15, 1996.
• Ohler, T.A. 1994. Evaluation of cowpea (Vigna unguiculata (L.) Walp) as a candidate species for inclusion in controlled ecological life-support systems. M.S. Thesis, Purdue Univ., West Lafayette, IN.
• Saha, P.R., and P.R. Trumbo. 1996. The nutritional adequacy of limited vegan diet for a controlled ecological life support system. Adv. Space Res. 18(4/5):63-72.
• Schneider, V.S., A. LeBlanc, and R.C. Rambaut. 1989. Bone and mineral metabolism, p. 214-221 In: Space Physiology and Medicine, A.E. Nicogossian, C.L. Huntoon, and S.L. Pool (eds.), Lea & Fibiger, Philadelphia, PA.
• Stephens, S.D. 1995. Effect of growing conditions on the proximate and fatty acid compositions of three Brassica napus varieties and determination of the storage stability of screw-pressed extracted oils and meals from soybean, peanut and canola. M.S. Thesis, Purdue Univ., West Lafayette, IN.
• Weaver, C. M., B.R. Martin, J.S. Ebner, and C.A. Kruger. 1987. Oxalic acid decreases calcium absorption in rats. J. Nutr. 117:1903-1906.

*Paper No. 15006 of Purdue University Agricultural Research Programs. Research supported in part by NASA grant NAGW-2329

Table 1. Proximate composition and energy content (calculated) of edible portion of some CELSS candidate species, as de-determined for edible portions of plants grown under controlled environments (CE), in the field, or as reported in USDA Handbook No. 8 (Hdbk).

Last update July 9, 1997 aw