Seeds of two Chilean ecotypes of C. quinoa (Willd.) from contrasting agro-ecological origins were used in the experiments: Socaire from the Chilean Altiplano (Socaire, 23°35’31.58” S, 67°53’17.69” W, and 3,500 m.a.s.l.) and the Lowland/Southern coastal Faro ecotype (Chillan, 36°35’ 43.2” S, 72°04’ 39.9” W and 140 m.a.s.l.). Seeds of Socaire were collected in a private field (23°34′S 67°54′W, 3,500 m.a.s.l.). The soil in the zone is basic (pH between 7.8 and 8.8) and the total percent and available N were 0.09% and 46 mg/kg, respectively. Soil and climate characterizations of the zone can be found in García et al. (2007). Faro seeds were collected in Chillan at “El Nogal Experimental Station” (36°35’ 43.2” S, 72°04’ 39.9” W and 140 m.a.s.l.). The pH of the soil was approximately 4.5, and the total percent and available N were 0.01% and 44 mg/kg, respectively. Soil and climate characterizations in this station are found in Fischer et al. (2013) and Stolpe (2006). Seeds of both ecotypes (Socaire and Faro) were collected in summer (February) in each location (in Socaire soils and in Chillan soils, respectively). Seeds of both ecotypes are included in the National Seed Bank of Chile managed by the Genetic Resources section of the National Institute of Agriculture Research (http://163.247.128.32/gringlobal/search.aspx, INIA-Intihuasi Vicuña, Chile).

In order to compare the performance of both ecotypes under NO 3 − and NH₄ ⁺, we established an optimal N concentration for plant growth. We set up biomass curves under the supply of different amounts of N, using NO 3 − as a referential condition because it is generally the preferred N source for plants without toxic effects. Germinated seeds of both ecotypes with similar lengths of emerged radicle were transplanted into 700-ml pots containing sand:perlite (1:1), and six concentrations of NO 3 − (using KNO₃ salt) and NH₄ ⁺ [using (NH₄)₂SO₄ salt] were tested: 0.0, 5.0, 10, 20, 40, and 100 mM. Ten pots (with three plants each) were submitted to each N concentration of NO 3 − or NH₄ ⁺. The conditions of the growing chamber and the irrigation solution were the same as explained below. Thirty-day-old plants were collected for total biomass determination.

Socaire and Faro plants reached maximum biomass at 20 mM of N ( Supplementary Figure 1 ). Therefore, this amount was used for the physiological experiments comparing N sources. Plants were germinated and grown in pots supplied with 20 mM of NO 3 − or NH₄ ⁺ using the following growing conditions: light intensity of 575 µmol/m²/s, 21°C/19°C day/night, 16 h light/8 h dark photoperiod, and 75% relative humidity. Plants were supplied once with MS 407 nutrient medium described by Murashige and Skoog (1962), and consisting of the following: 0.30 mM MgSO₄.7H₂O, 0.22 mM CaCl₂, 0.62 mM KH₂PO₄, 12.7 mM KCl, 0.05 µM KI, 1.00 µM H₃BO₃, 1.32 µM MnSO₄.4H₂O, 0.30 µM ZnSO₄.7H₂O, 0.01 µM Na₂MoO₄.2H₂O, 0.001 µM CuSO₄.5H₂O, 0.001 µM CoCl₂.6H₂O, 0.51 µM Na₂.EDTA, 0.50 µM FeSO₄.7H₂O, 2.78 µM inositol, 0.02 µM nicotinic acid, 0.01 µM pyridoxine HCl, 0.001 µM thiamine-HCl, and 0.13 µM Glycine. KNO₃ and (NH₄)₂SO₄ varied according to the treatment. The pH was set at 5.8.

Thirty pots of 700 ml (three plants per pot) were used for each N source and ecotype. The experiment was run in a completely randomized design and additional plants were grown to prevent the bordering effect. Juvenile plants were collected 30 days after sowing at midday. Harvested plants for biochemical analysis were fractionated into belowground and aboveground (leaves and biomass). The root samples were rinsed in distilled water and dried with a paper towel. The tissue samples were frozen in liquid N and then freeze-dried. Tissue was ground to a fine powder in liquid N using a mortar and pestle and stored in Falcon tubes at −80°C. Measurements were performed belowground and aboveground, except for pigments that were performed in leaves.

Images of plants and leaves submitted to the different treatments were taken with the Scanner Epson Perfection V850 Pro Photo (Epson Corporation, San Jose, CA). The image acquisition parameter was set to “high” accuracy (600 dpi; image size 18 MB). Leaf area was measured through image analysis using the ImageJ software (NIMH, Bethesda, Maryland, USA). Above and belowground biomasses were determined by drying the tissues at 60°C for 48 h till constant weight.

N and C elements were measured in the whole plant from the different N treatments and ecotypes (n = 3). Briefly, plant tissue was oven-dried at 60°C for 48 h and ground. A subsample of 5 mg was weighed and stored in plastic vials. Samples were measured using the Elemental Combustion System CHNS-O (Costech Analytical Technologies Inc., Valentia, USA). C and N are reported as the percentage of elements per dry matter.

We used above- and belowground tissue from each ecotype and N source (n = 6). Changes in total soluble sugars (TSS), starch, and total nonstructural carbon (NSC) in aerial and roots were assayed following the method of Marquis et al. (1997). Total soluble sugars were separated and extracted with methanol/chloroform/water according to and determined by colorimetry using 2% phenol and sulfuric acid at 490 nm according to Dickinson (1979) and Chow et al (2004). The insoluble fraction that contains starch was hydrolyzed to glucose overnight using a sodium acetate buffer and amyloglucosidase (Sigma-Aldrich 10115, St. Louis, MO, USA) at 45°C and then measured with a phenol–sulfuric acid reaction as in Marquis et al. (1997).

Total soluble protein and ammonium contents were determined in the above- and belowground tissue of the two quinoa ecotypes studied. NH₄ ⁺ was determined according to Forster (1995). Absorbance was measured at 660 nm in a spectrophotometer (Infinite 200 Pro, Tecan, Männedorf, Switzerland). We used the TCA/acetone procedure for protein extraction (Wang et al., 2008). The Quick Start Bradford Assay kit (Bradford, 1976) was used for protein quantification using BSA as the standard protein (n = 4), according to manufacturer instructions (Bio-Rad, Hercules, CA, USA).

Chlorophylls a and b were extracted from leaves of plants of all treatments (n = 6) in 80% of acetone overnight and centrifuged at 12,000g for 10 min. The content of chlorophylls was determined at 664 and 647 nm following the method described by Lichtenthaler and Buschmann (2001).

Betacyanins were extracted from leaves (n = 6) in water/methanol and spectrophotometrically determined at 536 nm. The betacyanin content of the plant aqueous extracts was estimated according to Abderrahim et al. (2015).

The lipid peroxidation in the above- and belowground tissue of the two quinoa ecotypes studied (n = 5) was determined in vitro by estimating the formation of malondialdehyde (MDA) according to the method described by Ortega-Villasante et al. (2005). Frozen leaf tissue (0.1–0.2 g) was homogenized in 1 ml of TCA–TBA–HCl reagent [15% (w/v)] trichloroacetic acid, 0.37% (w/v) 2-thiobarbituric acid, 0.25 M HCl, and 0.01% butylated hydroxytoluene. After homogenization, samples were incubated at 90°C for 30 min and centrifuged at 12,000 g for 10 min. Absorbance was measured at 535 nm and 600 nm.

Gas exchange measurements of net photosynthesis (A_N) were performed in leaves from both ecotypes and N sources in six plants per treatment, using a portable photosynthesis system (Li-6400XT, LI-COR Inc., Lincoln, NE, USA) equipped with a light source (6200-02B LED, Li-Cor).

Light curves were run to determine the light saturation intensity at which plants reached maximum photosynthesis. A_N rates were measured at mid-morning (between 11 a.m. and 1 p.m.) with the gas exchange previously stabilized. Conditions in the leaf chamber were as follows: block temperature of 25°C, 1,500 µmol photon m⁻² s⁻¹, an air flow of 300 (mol s⁻¹), and a CO₂ concentration (Ca) of 400 mol mol⁻¹. Relative humidity ranged between 45% and 50% and VPD ranged between 1.6 and 1.9 kPa. A_N data were normalized by the area of leaves (n = 6).

For respiratory measurements, leaves of quinoa plants, which were grown at the University of the Balearic Islands under similar growing conditions, were placed in a 3-ml stainless-steel closed cuvette and maintained at a constant temperature of 25°C. Air samples were sequentially removed from the cuvette and fed into the mass spectrometer (Delta XPlus; Thermo LCC, Bremen, Germany). Changes in the ¹⁸O/¹⁶O ratios and O₂ concentration were obtained to calculate the oxygen-isotope fractionation and the electron partitioning to the AOP (τ_a), allowing calculations of the in vivo activities of AOP and COP as described in Del-Saz et al. (2017). End point fractionation values of the AOP (Δ_a) were determined in leaves with a solution of 25 mM potassium cyanide (KCN) for 20 min. A value of 32.8 ± 0.69‰ (n = 3) was obtained in leaves of the Faro ecotype. Owing to a limitation in the number of Socaire plants, we assumed a similar value of Δ_c of 32.8‰. We also assumed a value of 20.0‰ for the endpoint fractionation values of the COP (Δ_c) as this has been shown to be constant in most leaves and species examined (Ribas-Carbo et al., 2005). Values presented are the mean of one measurement in four plants per ecotype that were performed from 9 a.m. to 5 p.m. during five consecutive days.

For the determination of sufficient N source supply, we used three-way ANOVA (level of significance p < 0.05) using ecotype, source of N, and concentration as factors. Data from the effects of N source ( NO 3 − or NH₄ ⁺) on the different quinoa ecotypes (Socaire and Faro) were analyzed by two-way ANOVA. Tukey HSD test was used to identify means with significant differences (level of significance p < 0.05).

The total dry biomass of both ecotypes enhanced with increased N supply, until reaching mean maximum biomass values at 20 mM N ( Supplementary Figure 1 ). Consistently with previous results published by Pinto-Irish et al. (2020), we considered 20 mM N as optimal for growing conditions (because both ecotypes reached the highest biomass either under NO 3 − or NH₄ ⁺). At higher concentrations (40 and 100 mM), plant biomass decreased under NO 3 − , and mortality increased under NH₄ ⁺; for this reason, these data were omitted in Supplementary Figure 1 . Considering that NO 3 − is generally the preferred N source without deleterious effects, our results focused on the comparison of the ecotypes under NH₄ ⁺ regarding their performance under NO 3 − .

The two-way ANOVA reveals interactions between ecotypes (E) and nitrogen (N) source (E × N) in aboveground, belowground, and total leaves biomass ( Figure 1 and Table 1 ). Ecotypes presented similar aboveground biomass under NO 3 − ( Figures 1A, E ); however, under NH₄ ⁺ supply, Socaire showed a 72% reduction in plant biomass, while Faro maintained similar values to those observed under NO 3 − (p < 0.031) ( Figures 1A, E ). Regarding belowground biomass, ecotypes exhibited a large contrasting response under NH₄ ⁺ (E × N), with a 80% reduction in Socaire under NH₄ ⁺ compared to NO 3 − , while a twofold increase (p < 0.001) was observed in Faro ( Figure 1C ). Both the total leaf area and total leaf biomass per plant (sum of all leaves) ( Figures 1B, D, E ) correlated well with aboveground biomass. No differences in leaf area or leaf biomass were observed between N sources in Faro.

The %C, %N, and the C:N ratio were significantly affected by E × N interaction ( Table 1 ). Both ecotypes presented similar %C values under NO 3 − supply; however, their content was significantly reduced under NH₄ ⁺ supply (p < 0.001) ( Figure 2A ).

Under both N sources, Socaire displayed an enhanced level of %N compared to Faro ( Figure 2B ). A significant increase in %N was observed in Socaire plants under NH₄ ⁺ compared to NO 3 − conditions. Contrastingly, Faro showed a similar %N under both N sources. Under, the %N in Socaire was twice as high as in Faro.

The C:N ratio was higher in Faro than in Socaire at both conditions. However, both ecotypes showed a significant decrease in the C:N ratio under NH₄ ⁺, with the lowest values in Socaire ( Figure 2C ).

The two-way ANOVA revealed that NH₄ ⁺ content belowground was affected by E × N interaction (p < 0.001). The NH₄ ⁺ content in roots increased by 30% in Faro ecotype under NH₄ ⁺ treatment compared to NH₄ ⁺, while it was maintained in Socaire ( Figure 3 ). Aboveground NH₄ ⁺ was not affected by N sources and differences were based on E (p < 0.001).

The two-way ANOVA reveals that protein content at above- and belowground (p < 0.001) tissues was affected by E × N interaction ( Table 1 and Figures 3A, B ). Under NO 3 − , protein level was three times higher in Socaire than in Faro (in both above- and belowground tissues). Under NH₄ ⁺, Faro showed a fourfold increase in protein content in both shoot and root, while Socaire showed an increase of 33% in protein content only aboveground.

A non-significant E × N interaction was observed in the storage of carbohydrates aboveground (normalized by dry weight) (p > 0.05); however, significant E × N interaction was observed for TSS, starch, and NSC belowground ( Table 1 and Figure 4 ).

In aboveground tissues (stem plus leaves), TSS tended to increase under NH₄ ⁺ in both ecotypes and significant differences were observed in Socaire ( Figure 4A ). Starch only depended on E aboveground (p < 0.001) ( Figure 3B ). In belowground tissues, Faro showed higher TSS and starch contents than Socaire under NO 3 − conditions ( Figures 4D, E ). Under NO 3 − , Faro showed the highest values of NSC in both above- and belowground tissues. However, a significant decrease in belowground NSC was observed under NH₄ ⁺, while NSC values in the aboveground tissues reached similar values to those observed in plants under NO 3 − ( Figures 4C, F ). On the other hand, Socaire displays similar NSC values under both sources of N ( Figure 4 ). These results changed drastically when they are interpreted as organs ( Supplementary Figure 2 ), where strong reductions were observed in NSC of Socaire under NH₄ ⁺ at both aboveground tissues and roots.

A significant E × N interaction was observed for chlorophylls, betacyanins, and MDA (p < 0.001). Chlorophylls a and b increased in Faro under NH₄ ⁺ compared to NO 3 − , while no significant changes were observed in Socaire ( Figures 5A, B ). Regarding betacyanins, an increase of 30% was observed in Socaire under NH₄ ⁺ compared to NO₃, displaying significantly higher values than those observed in Faro ( Figure 5C ).

An increase of 50% in belowground MDA content was observed in the Socaire ecotype under NH₄ ⁺ compared to NO 3 − ( Figure 6 ), while similar values were observed aboveground in comparison to Faro under NO 3 − and NH₄ ⁺. Faro maintained MDA levels under both N sources ( Figure 6D ).

Both Faro and Socaire plants displayed similar A_N rates in leaves (normalized by area) under NO 3 − . Under NH₄ ⁺, A_N was approximately 50% higher in Socaire than in Faro (p < 0.05) ( Figure 7 ).

Oxygen consumption and electron partitioning to the COP and AOP under NO 3 − and NH₄ ⁺A significant E × N interaction was observed in total O₂ uptake (V _t) (p < 0.05) and in the electron partitioning to the AOP (τ_a) (p < 0.05). There was a significant reduction (by 30%) of τ_a in Faro under NH₄ ⁺ compared to NO 3 − , together with a significant increase (by 80%) in V _t via COP ( Figure 8 ). No respiratory changes were observed in Socaire (E × N, p < 0.05) ( Figure 8 ).