The impact of different agroecological conditions on the nutritional composition of quinoa seeds

Plant material, experimental design and locations

Quinoa (Chenopodium quinoa Willd.) seeds of cultivars Regalona (registered variety of BAER, Chile), Salcedo-INIA (experimental station of Illpa—Puno, Peru, Mujica, Izquierdo & Marathee, 2001) and Titicaca (generously provided by Dr. Jacobsen of Copenhagen University) were selected to evaluate their agronomic potential and seed nutritional traits at three locations with different agroecological conditions: El Pobo (Teruel, Spain), Arequipa (Peru) and Río Hurtado (Chile). The field experiment at El Pobo (40.50°N, 0.86°W, 1,399 m a.s.l.) (Spain) was carried out under rainfed conditions between May and October 2016 within a range of temperatures between 24.1 and 4.7 °C on average (registering a maximum of 32.3 °C and a minimum of −2.8 °C) and 194 mm total precipitation during the mentioned period. The field trial at Río Hurtado (30.3°S, 70.6°W, 1,462 m a.s.l.) (Chile), was carried out between November 2015 and April 2016 within a range of temperatures between 11 and 25 °C on average (registering a maximum of 34 °C and a minimum of 3.7 °C). The total precipitation during that period was 150 mm and the irrigation was applied using drip lines that released 4 L m⁻¹ h⁻¹ according to Martinez et al. (2009). In Arequipa (16°S, 71°W, 2,355 m a.s.l.) the experiment was carried out under irrigation between March and July 2016 with an average temperature of 14 °C (registering a maximum of 28.8 °C and a minimum of 4.2 °C) and 15.3 mm total precipitation.

The soil type in Spain (a clay-silty-loam soil) presented a pH of 7.9, 4.8% organic matter, 3 dS m⁻¹ of electrical conductivity (EC) of the saturated paste and phosporous (P) and potassium (K) equivalent to 37 and 438 mg kg⁻¹, respectively. The soil type in Chile (a loamy alluvial Entisol) presented a pH of 7.8, 7.7% of organic matter, 2.6 dS m⁻¹of EC and content of P and exchangeable K equivalent to 49.96 and 237 mg kg⁻¹, respectively. The soil type in Arequipa (Peru) (a loam soil) presented 4.89% of organic matter, 2.25 dS m⁻¹ of EC, a pH of 6.95 and a content of 39.31 mg kg⁻¹ of P and 624.96 mg kg⁻¹ of K.

The experimental design consisted in randomized blocks (8 m² per block, 4 m × 2 m, L × W) with 4 replications in each location using the three varieties (Regalona, Salcedo-INIA and Titicaca). Each block was composed of 4 rows of 4 m in length (row spacing  = 50 cm). Seeds were directly germinated in the soil with a sowing density of 10 kg/ha between 1 and 2.5 cm depth.

Quantitative multi-elemental analysis

Quantitative multi-elemental analysis by inductively coupled plasma (ICP) spectrometry was used to determine total content of calcium (Ca), magnesium (Mg), iron (Fe), sodium (Na), zinc (Zn), P and K contents in C. quinoa seeds. Seed samples were firstly grinded to a fine powder. The samples were then submitted to the Interdepartmental Investigation Service Laboratory at the Universidad Autónoma de Madrid (SIdI-UAM, Madrid, Spain). Samples were digested in a microwave oven and subsequently analyzed using the equipment ICP-MS NexION 300XX (Perkin Elmer Inc., Hopkinton, MA, USA).

Total phytate content

Total phytate content was determined using Myo-Inositol Hexakisphosphate Determination method. This method involved acid extraction of myo-inositol phosphates from 0.5 g of flour (ground seeds) in 20 mL of HCl 0.66 M with vigorously stirring at room temperature overnight followed by treatment with a phytase and alkaline phosphatase (K-PHYT 07/11; Megazyme, Bray, Wicklow, Ireland). The total phosphate released was proportional to myo-inositol hexakisphosphate in non-processed seeds. It was measured using a colorimetric method with ammonium molybdate reactive to form 12-molybdophosphoric acid, which was subsequently reduced under acidic conditions to molybdenum blue. The amount of molybdenum blue formed in the reaction was proportional to the amount of free phosphate present in the sample and was measured at 655 nm using a spectrophotometer SPECTROstar nano (BMG LabTech GmbH, Ortenberg, Germany). Phosphorus was quantified interpolating from a calibration curve using standards of known phosphorus concentration. The samples were done by triplicate and the results were expressed in grams of phytic acid per 100 g of seeds in dry matter.

Protein content

The Kjeldahl method with a conversion factor of 6.25 by AOAC method no 960.52 (Association of Official Analytical Chemists International, 2016) was employed to quantify the total crude protein content of the quinoa seed samples. All determinations were done in triplicate.

Amino acid quantification

Liquid chromatography mass spectrometry (LC/MS) was used to determine free amino acid content of C. quinoa seeds. Seed samples were grinded to a fine powder. Free amino acid was extracted as described previously by Hacham, Avraham & Amir (2002). One hundred fifty mg of seed powder was homogenized in 400 µL water:chloroform:methanol (3:5:12 v/v) and centrifuged at 14,000 rpm for 2 min. This step was repeated twice and both supernatants were combined and mixed with 200 µL chloroform and 300 µL of water. The resulting mixture was centrifuged at 14,000 rpm for 2 min. The supernatant (corresponding to the water:methanol phase) was subjected to speed-vac to dry and resuspended in 100 µL miliQ H₂O.

Free amino acid extracts were submitted to the Chromatography Laboratory at the SIdI-UAM (Spain) for analysis. Amino acid determination was carried out using HPLC-MS with an Agilent system detector composed by an 1,100 series HPLC coupled to a single 6,120 Quadrupole. For the chromatographic separation, 5 µL were injected in an ACE 5 AQ (250 × 4.6 mm, 5 µm) thermostated column at 30 °C with 0.4 mL/min flow rate and binary gradient elution. The elution was performed in H₂O with 0.1% formic acid (v/v) as eluent “A” and acetonitrile (ACN) with 0.1% formic acid (v/v) as eluent “B”. The gradient program was as follows for eluent B: 0 min, 0%; 30 min, 100%; 35 min; 100%; 36 min, 0%; 55 min, 0%. The ionization parameters were as follows: positive atmospheric pressure chemical ionization (APcI+), fragmentor voltage 40 V, capillary voltage 2.0 kV, charging voltage 2.0 kV, nebulizer pressure 20 psig, drying gas flow 5 L/min at 350 °C, vaporizer temperature 250 °C, and corona current 4 µA. Data was recorded scanning from 50 to 250 Da.

Ferric reducing antioxidant power (FRAP) assay

The FRAP assay was used to determine the antioxidant capacity of seed samples. The procedure described by Benzie & Strain (1996) was used with some modifications. Briefly, 30 µL sample aliquots were mixed with 90 µL of distilled water and 900 µL of freshly prepared FRAP reagent at 37 °C (2.5 mL of a 10 mmol/L 2,4,6-tripyridyl-s-triazine (TPTZ) solution in 40 mmol/L HCl with 2.5 mL of 20 mmol/L FeCl₃ and 25 mL of 0.3 mol/ L acetate buffer at a pH of 3.6). The absorbance of the reaction mixture was measured spectrophotometrically (atomic absorption spectroscopy (PinAAcle 900F FL HSN, WinLab32 software; Perkin Elmer Inc., Hopkinton, MA, USA) at 593 nm following incubation at 37 °C for 2 h. FRAP concentration was calculated from a calibration curve obtained by linear regression and the results are expressed in Trolox equivalents (mmol TE 100g⁻¹). The reference used was the synthetic antioxidant Trolox at a concentration of 100 to 1500 mmol in 80% methanol solution, which was tested under the same conditions.

Fiber and saponin determination

For fiber determination samples were submitted to the SERBILAC laboratory (Universidad Nacional de San Agustín de Arequipa, Peru). Fiber content was determined following the protocol described in AOAC Methods 2016 (Association of Official Analytical Chemists International, 2016). Samples were submitted to the Laboratory of quality control (Universidad Nacional del Altiplano Puno, Peru). Total saponin content was determined spectrophotometrically at 528 nm using an SQ-2802 single beam scanning spectrophotometer (UNICO) (Lozano et al., 2012). The concentration of saponin was read off from a standard curve of different concentrations of saponin (Calbiochem, CAS 8047-15-2, Darmstadt, Germany) (from 0.5 to 7.5 mg/ L) dissolved in an aqueous solution.

Statistical analysis

Results are presented as Mean value ± Standard Deviation or Error. Pairwise comparisons were done by using Student t-test at a probability level of 5% (P < 0.05). Multiple comparisons were done by one-way analysis of variance (ANOVA) followed by Duncan or Tukey HSD post-hocs to analyze the quantitative data at a probability level of 5% (P < 0.05). The JMP^® (ver.11.0) statistical package (SAS Institute) and the Free Software R were used for the statistical analyses.

Variations in agronomical traits of Chenopodium quinoa cultivars under different agroecological conditions

Among the three quinoa varieties studied, differences were found in terms of seed yield when consider total yield and seed weight per plant (Table 1). Although a larger total yield (Kg/Ha) was achieved by Salcedo-INIA grown in Peru followed by Titicaca grown in Chile, this was due to a larger plant density in these two varieties and not because of a better variety performance as revealed by the seed weight per plant (that was the smallest in Salcedo-INIA from Peru). Also, the harvest index parameter values did not differ among the varieties tested. On the contrary, Titicaca in Spain yield less seeds than Regalona (t-Student, P < 0.05) and Regalona showed the largest seed weight per plant among cultivars and locations. Besides, Titicaca and Regalona in Peru were unable to yield seeds and Salcedo-INIA in Spain showed important seed losses due to seed dehiscence detected at harvesting related to defects in the timing of maturity, the uniformity of maturity and the lack drydown of plant at seed maturity.

Most of the morphological traits varied with the location and among cultivars (Table 1). Salcedo-INIA cultivar showed the smallest plants in Chile but the biggest plants in Spain. Salcedo-INIA presented the largest stem diameter and panicle length in Peru and Spain while the stem diameter of Regalona was the largest in Chile. Salcedo-INIA showed the biggest panicle length among varieties in the different locations. Plant weight did not differ among locations but varieties as shown in Table 1.

Regarding the analysis of phenological traits, days to flowering and days to maturity were evaluated. Titicaca showed the shortest time to flowering and to maturity at both locations (Chile and Spain), followed by Regalona. Salcedo-INIA presented the largest time to flowering and maturity (approximately 95 days and 184, respectively) in Spain and Chile while reduced the days to flowering and maturity in Peru.

Mineral composition and phytate content in C. quinoa seeds

Quantitative multi-elemental analysis was performed aiming to assess differences in the seed mineral composition among cultivars and locations due to distinct agroecological conditions. When analyzing the effect of the location in each genotype (Figs. 1A–1C), it was observed that Regalona seeds stored larger amounts of mineral nutrients in Chile with exception of P. Salcedo-INIA showed larger quantities of Mg, Fe, Ca and Zn in Peru, while in Chile presented the largest amount of P and the lowest of Na. Titicaca in Spain had larger amounts of Ca, K, P and Na but less amount of Fe and Zn comparing the same variety grown in Chile.

Differences among varieties were also found in each location (Figs. 1C and 1D). In Chile, Regalona cultivar stored the highest amounts of Ca followed by Salcedo-INIA. Salcedo-INIA presented the largest amounts of Fe and K. Titicaca in Chile showed the lowest contents in Ca, P, Mg and Na compared to the other two cultivars. In Spain, Titicaca showed the highest level of K. Salcedo-INIA grown in Spain showed the highest amounts of P and Mg but smallest of Fe, Ca and Na. All the cultivars grown in Spain or Chile showed similar Zn contents.

Overall, a larger accumulation of Mg and Fe tend to appear in Chile. Zn contents changed with the location but remains unchanged within cultivars. Generally, the type of cultivar and location affects the content of certain minerals, indicating that both factors (variety and location) are determinants of the mineral composition of quinoa seeds.

Phytic acid is considered an important seed component conditioning the nutritional properties of seeds (Lott et al., 2000). The analysis of phytate content in quinoa seeds (Fig. 2) revealed that Regalona seeds from Spain showed the largest phytate content, followed by the Salcedo-INIA and Titicaca seeds from this country. Titicaca seeds from Chile presented the smallest phytate content. The fact that no differences were found among cultivars in the same location but a given genotype vary among locations suggests that phytic acid content in quinoa seeds might be determined by environmental factors and not by the type of cultivar.

Total Protein content and free amino acid profile of quinoa seeds obtained from Chile, Peru and Spain

The range of total protein content was found between 14 and 17% among the cultivars and the locations analyzed (Fig. 3). No differences were found when comparing among varieties in a particular location. Consistently, seeds from Chile showed a higher total protein content comparing to Spain or Peru. These results might suggest that agroecological conditions can influence on the protein content of the seeds.

Regarding amino acid composition, the most abundant amino acids found in the quinoa seeds analyzed in the present study were arginine and glutamic acid (Fig. 4 and Table S2). Asparagine, glutamic acid, glutamine, histidine, glycine, hydroxyproline, serine and threonine and remained unchanged in all cultivars and locations. The rest of the amino acids quantified showed differences among cultivars and/or locations. For instance, Regalona seeds grown in Spain showed higher contents of arginine, aspartic acid, lysine and methionine compared to the Chilean Regalona seeds. Spanish Salcedo-INIA seeds showed larger contents of aspartic acid, isoleucine, leucine and valine compared to the Peruvian and Chilean Salcedo-INIA seeds. Arginine and phenylalanine showed higher contents in Salcedo-INIA seeds obtained from Spain when compared with the seeds from Chile but no differences were detected in the contents of these two amino acids between Spain and Peru.

Noteworthy, an elevated amount of tryptophan was found in Salcedo-INIA from Peru, amount that was superior to any of the cultivars analyzed. In Spain, the amount of lysine in Regalona and valine in Salcedo-INIA were higher when compared to other cultivars or locations. In contrast, the content of amino acids in Titicaca did not change significantly among cultivars nor locations.

When analyzing the amino acid content by location it was observed that the varieties grown in Chile did not present differences in their amino acid content. However, inter cultivar differences were observed in the amino acid contents in the Spanish quinoa seeds for alanine, asparagine, isoleucine, lysine and valine.

Antioxidant properties of quinoa seeds in different cultivars and locations

Antioxidant capacity varied among cultivars and locations ranging from 0.75 to 0.15 mmol TE 100 g⁻¹ (Fig. 5). Titicaca and Regalona seeds from Chile presented the highest antioxidant activity measured as FRAP, while Salcedo-INIA from Peru presented the lowest values. The big differences observed between the antioxidant properties of Regalona at the two locations studied did not appear in Titicaca nor Salcedo-INIA (differences among locations in these two cultivars were smaller) what would indicate that changes among cultivars appear when evaluating the effect of the agroecological conditions on the antioxidant properties.

Fiber and saponin contents

Although no differences were found in the saponin content (Fig. 6A) among the cultivars or locations, Titicaca seeds from Chile showed larger fiber contents compared to Regalona seeds obtained from Spain (Fig. 6B). These results might indicate that fiber might be more susceptible to variations associated with changes in agroecological conditions despite no big differences were found among cultivars and locations.