Introduction

In Mexico, 454 ha of purslane (Portulaca oleracea L.) are sown under irrigation and dry land farming conditions. In the states of Morelos, Baja California, and Mexico City, the average yields are 19.18, 10.13, and 9.47 t ha^-1 (^{SIAP, 2014}), respectively. Producers in Mexico City have achieved yields up to 40 t ha^-1 (according to personal communication with local producers).

Researches about nutrient concentration in plants establish the basis of the crops’ fertilization. Concentration can be adjusted during the crop cycle and the amount of fertilizer to be used can be optimized (^{Greenwood, 1983}). As a result of its contribution to human nutrition, determining the nutrient concentration in plants is also useful (^{FNB/FNIC, 2001}).

The content of N, P, K, Ca, and Mg varies between wild purslane, ornamental purslane, stages of growth, weather conditions, fertilization, and harvest date (^{Teixeira and Carvalho, 2009}; ^{Kaymak, 2013}; ^{Alam et al., 2014}; ^{Petropoulos et al., 2015}). In Mexico, the studies have focused on the ethnobotanical description, while studies about nutrient concentration in the above-ground biomass are infrequent. The objective of this research was to determine the concentrations of N, P, K, Ca, Mg, Na, S, Mn, Fe, Cu, Zn, and B in the above-ground biomass of purslane in three dates of commercial harvest, according to the amount of N, P, and K applied.

Materials and Methods

The study was carried out in the Experimental Field of Colegio de Postgraduados, Campus Montecillo, Estado de Mexico (19° 28’ 2.07” N, 98° 54’ 1.71” W, at a 2215-m altitude), in July and August, 2014. The zone has a temperate semi-arid climate (Bs 1 kwe (w) (i’) g), with summer rains (July to September), a 625-mm average annual precipitation, and 22 °C average temperatures (^{García, 1973}). It has a Mollic Ustifluvent soil type (^{Soil Survey Staff, 2010}), with a 150-cm depth, and a silt loam texture (Ap horizon). Rainfall in June, July and August, 2014, was 135.2, 142.5, and 134.1 mm, respectively. The average daily maximum temperatures in July and August varied from 19 to 30 °C, and the minimum temperatures varied from 6 to 11 °C.

Results and Discussion

The regression models of the nutrient concentrations showed variable effects depending on the nutrient applied and the harvest date. In contrast, none showed effect due to the variation of organic matter, P, or K in the soil.

The water content in the above-ground biomass of purslane increased with the N applied, from 94.2 to 95.3 % in harvest 1 and from 94.2 to 95.8 % in harvest 2; the water content in harvest 3 increased from 92.7 to 94.9, with 0 and 300 kg N ha^-1 (Table 2, Figure 1).

Figure 1 Water content and concentration of N, P, and K in the above-ground biomass of purslane (Portulaca oleracea L.), based on the amount of N, P, and K applied, and harvest date. C₁: harvest 1 to 27 days after emergence (dae); C₂: harvest 2 to 34 dae; and C₃: harvest 3 to 42 dae. 

The water content in above-ground biomass of purslane was similar to the results of ^{Kaymak (2013)} and ^{Lara et al. (2011)}. ^{Karkanis and Petropoulos (2017)} reported differences between purslane cultivars from Iran and Greece, without significant variations in harvest time: 26 das (i.e., 91 to 93 %) and 45 das (i.e., 90 to 94 %). ^{Uddin et al. (2012a)} observed that water content decreased as the harvest time was extended, from 90 to 74 % at 15 and 60 das. The water content in the above-ground biomass of purslane decreased under salt stress conditions, as a result of adding NaCl to the nutrient solution (^{Teixeira and Carvallho, 2009}; Kafi and Rahimi, 2011; ^{Uddin et al., 2012b}). In contrast, the water content in the tissue was not affected by water stress (drought) or water saturation in the soil (Uddin et al., 2017).

The concentration of N in above-ground biomass increased when N and P were applied -and by the interaction between both factors- from 25.2 to 40.8 mg N g^-1 with 300 kg N ha^-1. The maximum concentration was 50.0 mg g^-1 with 90 kg P₂O₅ ha^-1, and 300 kg N ha^-1 (Table 2, Figure 1).

These values matched those obtained by ^{Kaymak (2013)} (24 mg N g^-1, applying 150 kg N ha^-1), Viana et al. (2015) (28.6 mg N g^-1), and ^{Graifenberg et al. (2003)} at 124 das under hydroponic conditions (28 and 32 mg N g^-1), with 0 and 3 g L^-1 of NaCl in the nutrient solution. ^{Uddin et al. (2017)} observed that water stress affected the concentration of N in purslane -10 saturated days followed by field capacity condition, with 15.7 mg N g^-1- compared to continuous flooding -with 31.4 mg N g^-1. ^{Fontana et al. (2006)} reported concentrations of 6.3 and 7.8 mg g^-1, with 12 and 36 mmol N L^-1 in the nutrient solution, respectively. ^{Alam et al. (2014)} obtained maximum concentrations of 118 and 184 mg N g^-1 in two genotypes of Malayan purslane.

The N and P applied increased the content of P: starting from 3.28 mg g^-1 (in 0.77 mg P g^-1 with 300 kg N ha^-1), and from 1.01 (90 kg P₂O₅ ha^-1) (Table 2, Figure 1).

Nitrogen fertilizer modified the P content in purslane, with values of 2.4, 2.5, 2.7, and 2.9 mg P g^-1 with ammonium sulfate, ammonium nitrate, urea, and calcium ammonium nitrate, respectively (^{Kaymak, 2013}). Under saline conditions, the concentration of P in above-ground biomass increased, from 4.7 to 5.8 mg P g^-1 with 0 and 3 g NaCl L^-1 (^{Graifenberg et al., 2003}), respectively. Concentrations of P in purslane from Brazil and Malaysia were higher than those of our study: 5.3 mg g^-1 (Viana et al., 2015), and 8.3 and 14.1 mg g^-1 (^{Alam et al., 2014}).

The concentration of K in the above-ground biomass increased with the N and K applied, from 12.1 to 13.5 mg g^-1 with 95 kg N ha^-1, and up to 14.9 mg g^-1 with 50 kg K₂O ha^-1. After reaching these maximum values, the concentration of N and K decreased. During harvest 3, K concentration decreased by 1.43 mg g^-1 (Table 2, Figure 1).

According to the bibliography, concentrations of K in purslane are higher than those of this study: 22.5 and 37.9 mg K g^-1 in Turkey; 53.7 mg K g^-1 in Brazil; and 478 and 586 mg K g^-1 in Malaysia (^{Kiliç et al., 2008}; ^{Kaymak, 2013}). In addition, ^{Uddin et al. (2012a)} observed 50, 51, and 59 mg K g^-1 increases according to the harvest date at 30, 45, and 60 das, respectively. In hydroponic systems, adding NaCl to the nutrient solution decreased concentrations of K of 29.1 and 43.7 mg g^-1 fresh weight (FW), with 0.8 and 24.2 dS m^-1, respectively (^{Teixeira and Carvalho, 2009}). In the nutrient solution of floating tray systems, aeration modified K concentrations, with levels from 3.8 to 4.5 mg K kg FW and from 3.2 to 4.3 mg K kg FW; meanwhile, the lack of aeration modified the levels from 2.5 to 4.0 mg K kg FW (^{Lara et al., 2011}).

The concentration of S in the above-ground biomass decreased in harvest 2 and 3, by 1.34 and 1.55 mg g^-1. In contrast, applying doses of N of up to 170 kg N ha^-1 increased the content of S in all crops and reached 3.92, 2.59, and 2.37 mg S g^-1 in harvest 1, 2, and 3, respectively (Table 3, Figure 2). Viana et al. (2015) obtained similar concentrations of S (2.8 mg S g^-1); in contrast, ^{Kaymak (2013)} determined lower concentrations (0.47 to 0.55 mg S g^-1).

Figure 2 Concentrations of S, Ca, Mg, and Na in the above-ground biomass of purslane (Portulaca oleracea L.), based on the amount of N and K applied, and harvest date. C₁: harvest 1 to 27 days after emergence (dae); C₂: harvest 2 to 34 dae; and C₃: harvest 3 to 42 dae. 

In harvest 1, the concentration of Ca in the above-ground biomass decreased from 9.86 to 8.90 mg g^-1 when 120 kg K₂O ha^-1 was applied; at the same time, in harvest 2 and 3, the decrease in the concentration of Ca was 0.19 and 0.55 mg g^-1 (Table 3, Figure 2).

In harvest 1, the concentration of Mg in the above-ground biomass decreased, from 15.67 to 13.34 mg Mg g^-1 with 120 kg K₂O ha^-1; in contrast, in harvest 2 and 3, the concentration increased by 0.55 and 0.63 mg Mg g^-1 (Table 3, Figure 2).

N (300 kg N ha^-1) and K (115 kg K₂O ha^-1) decreased Na, from 4.67 to 3.29 mg g^-1 and 3.38 mg g^-1 in the above-ground biomass; in contrast, harvest 2 and 3 increased the concentration by 0.39 and 0.20 mg Na g^-1 (Table 3, Figure 2).

^{Alam et al. (2014)} determined Ca (37 and 43 mg g^-1) and Mg (46 and 49 mg g^-1) concentrations, which are higher than those found in our study; Viana et al. (2015) also obtained more Ca (14 mg g^-1) but less Mg (8.4 mg g^-1). ^{Kaymak (2013)} observed that the concentrations of Ca (12 to 15 mg g^-1) and Mg (1.4 to 1.5 mg g^-1) in purslane depend on the source nitrogenous fertilizer that is used (ammonium sulfate, ammonium nitrate, urea, and calcium ammonium nitrate). ^{Uddin et al. (2012a)} reported that the concentration of Ca (64.6 to 77.9 mg g^-1) and Mg (51.7 to 59.4 mg g^-1) in purslane increased at 15 and 60 das, while the concentration of Na (6.3 to 8.2 mg g^-1) decreased depending on the harvest date. ^{Teixeira and Carvalho (2009)} observed that the presence of NaCl (0.8 to 24 dS m^-1) in the nutrient solution decreased the concentration of Ca (44.7 to 87.37 mg g^-1) and increased the content of Na (14.2 to 36.9 mg g^-1) in purslane stems and leaves, while concentrations were higher in spring than in summer. ^{Petropoulos et al. (2015)} reported higher content of Ca and Mg in purslane from Iran (0.21 to 0.23 mg Ca g^-1 FW and 0.14 to 0.17 mg Mg g^-1 FW) than in purslane from Greece (0.15 to 0.17 mg Ca g^-1 FW and 0.12 mg Mg g^-1 FW). ^{Egea-Gilabert et al. (2014)} observed different concentrations of Na among purslane genotypes from Spain, with variations ranging from 0.55 to 0.79 mg g^-1 FW. ^{Lara et al. (2011)} indicated that the concentration of Na increased under anaerobic conditions when the nutrient solution was placed in floating trays.

The concentration of Mn in the above-ground biomass increased as a result of applying N (from 54.9 to 65.7 mg Mn kg^-1 with 300 kg N ha^-1) in harvest 1; likewise, an increase was observed in harvest 2 and 3 (66.6 and 64.6 mg Mn kg^-1) (Table 4, Figure 3).

Figure 3 Concentrations of Mn, Fe, Cu, Zn, and B in the above-ground biomass of purslane (Portulaca oleracea L.), based on the amount of N applied, and harvest date. C₁: harvest 1 to 27 days after emergence (dae); C₂: harvest 2 to 34 dae; and C₃: harvest 3 to 42 dae. 

The concentration of Fe increased in harvest 1, from 613.6 to 769.8 mg Fe kg^-1, in the above-ground biomass with 190 kg N ha^-1. In harvest 2 and 3, the increase reached 549.8 and 703.2 mg Fe kg^-1 (Table 4 and Figure 3).

Doses of 300 kg N ha^-1 increased the concentration of Cu in the above-ground biomass, from 11.3 to 15.3 mg Cu kg^-1. Harvest 2 and 3 increased by 1.5 and 2.4 mg Cu kg^-1 (Table 4, Figure 3).

In the above-ground biomass of harvest 1, 300 kg N ha^-1 increased the concentration of Zn from 31.5 to 54.6 mg Zn kg^-1; however, in harvest 2 and 3, Zn decreased by 19.3 and 18.1 mg Zn kg^-1 (Table 4, Figure 3).

With 165 kg N ha^-1, the concentration of B also increased, from 31.9 to 33.7 mg B kg^-1 in harvest 1 and 3, respectively; however, the concentration decreased in harvest 2, by 4.4 mg B kg^-1 (Table 4, Figure 3).

Viana et al. (2015) observed 22, 188, 14, 126, and 41 mg of Mn, Fe, Cu, Zn, and B per g of purslane from Brazil. Alam et al. (2015) determined 1.6, 1.0, and 1.4 mg of Mn, Fe, and Zn per g of purslane from Malaysia. In addition, ^{Petropoulos et al. (2015)} observed significant differences in the concentration of Zn (2.1 to 3.3 mg kg^-1 FW), Fe (1.6 to 23.4 mg kg^-1 FW), and Mn (4.5 to 5.9 mg kg^-1 FW) in purslane genotypes from Greece and Iran. ^{Kaymak (2013)} observed significant differences as a result of the source of nitrogen fertilizer (ammonium sulfate, ammonium nitrate, urea, and calcium ammonium nitrate) in B (8.9 to 9.7 mg kg^-1), Cu (8.3 to 11.4 mg kg^-1), Fe (184 to 255 mg kg^-1), and Zn (16 to 20 mg kg^-1); but Mn (15 mg kg^-1) was not modified. Concentrations lower than those mentioned above were obtained in our study. ^{Uddin et al. (2012a)} reported that the concentration of Fe (12.2 to 14.6 mg g^-1) and Zn (8.4 to 10.4 mg g^-1) increased between 15 and 60 das.

The recommended daily intake for humans is 34 to 56 g of protein, 0.7 to 1.2 g of P, 4.5 to 4.7 g of K, 1.0 to 1.3 g of Ca, 0.30 to 0.42 g of Mg, 1.2 to 1.5 g of Na, 1.3 to 2.6 mg of Mn, 8 to 18 mg of Fe, 0.7 to 0.9 mg of Cu, 8 to 12 mg of Zn, with a maximum limit of 2.3 g of Na, 10 mg of Cu, 40 mg of Zn, and 17 mg of B (^{FNB/FNIC, 2001}). The results of this study indicate that -depending on the application of N, P, K, and the harvest date-, 5 g of dried purslane or 87 to 119 g of fresh purslane can provide 0.8 to 1.6 g of protein (equivalent to 25 and 50 mg N g^-1), 19 to 23 mg of P, 36 to 77 mg of K, 41 to 49 mg of Ca, 66 to 81 mg of Mg, 10 to 20 mg of Na, 0.3 to 0.4 mg of Mn, 3.1 to 7.4 mg of Fe, 0.1 mg of Cu, 0.2 to 0.3 mg of Zn, and 0.1 to 0.2 mg of B. Since it includes all these components, purslane can be considered as a nutritious alternative.

Conclusions

Applying N increases the water content and the concentration of N, P, S, Mn, Fe, Cu, Zn, and B, and decreases the concentration of K and Na in the above-ground biomass. Applying P increases the concentration of P. High doses of K decrease the concentration of K, Ca, Mg, and Na. The content of N, K, Na, Mn, Fe, and Cu in late harvests is higher, while water, S, Ca, Mg, Zn, and B content decreases. Purslane is a nutritious alternative, as a result of its high content of N protein, Mg, Mn, and Fe, and its low content of Na, Cu, Zn, and B.