Nutritional anti-nutritional chemical composition and antioxidant activities of the leaves of the sea cliff dwelling species Limonium spathulatum (Desf.) Kuntze

This work explored the nutritional and antioxidant properties of the leaves of the halophytic species Limonium spathulatum (Desf.) Kuntze from Tunisian sea cliffs. Furthermore, the analysis of the total phenolics and flavonoids contents and their individual compounds using high-performance liquid chromatography coupled with electrospray ionization mass spectrometry (HPLC-ESI-MS/MS) were also studied. L. spathulatum leaves had high levels of moisture, ash, neutral detergent fiber, and acid detergent fiber, but low concentrations of crude protein, crude fat and acid detergent lignin. It contained low carbohydrates levels, and low energetic values. The most abundant macroelements were Cl, Na and Ca while the microelements detected in the highest levels were Fe and Zn. No relevant α-amylase inhibition was observed, and no toxic metals (Pb and Cd) and phytic acid were detected. The ethanol and the hydroethanolic extracts had the highest capacity to scavenge free radicals, to chelate iron and copper and to inhibit lipid peroxidation. The same samples were also the most active towards oxidative haemolysis. These extracts contained high total phenolic and flavonoid contents. HPLC analysis, performed on ethanolic extracts identified 58 individual compounds known for their high antioxidant actvitiy including hydroxybenzoic acids (gallic, syringic acids), hydroxycinnamic acids (caffeic, coumaric, ferulic acids) and flavonoids (catechin, epigallocatechin gallate and naringin).In conclusion, the leaves of Tunisian accession of L. spathulatum were good source of minerals and fibers useful in the human diet for attaining nutritional sufficiency. The high in vitro and ex vitro antioxidant activities associated with high favonoids contents and compounds suggest the possibility to use the extracts of L. spathulatum in herbal products with the aim of improving general health and well-being, and/or as food additives for preventing lipid oxidation of lipid-rich foods.


Introduction
The Mediterranean basin is considered one of the world's biodiversity hotspots due to its high variety of plant species and endemism's Petropoulos et al., 2018;Bolaric et al., 2021;Hasanbegovic et al., 2021;Curadi et al., 2022) The Limonium genus (Plumbaginaceae) includes approximately 370 species of perennial herbs and shrubs belonging to a particular type of halophytes,'recretohalophytes', that can secrete salt from their leaves through salt bladders and salt glands, as a mechanism of adaptation to high salinity conditions (Yuan et al., 2016;Gonzaĺez-Orenga et al., 2021).
Having in mind the high importance of single-country endemic plants as sources of high added value products (Shelef et al., 2017;Sefi et al., 2021), this work focused on the species L. spathulatum (Desf.) kuntze which grow wild in the sea cliffs of Tunisia ( Figure 1). Despite the traditional uses and potential commercial applications of several Limonium species, information regarding L. spathulatum is limited and refers to the phenolic composition and antioxidant, anti-alzheimer, antidiabetic, and anti-inflammatory in vitro properties of organic extracts extracts from aerial parts collected from plants in Algeria (Mazouz et al., 2020), mineral, phenolic, carotenoids and vitamins contents, in vitro antioxidant properties, erythrocytes cellular antioxidant activity (CAA-RBC) and oxidative hemolysis protection of methanol extracts from plants collected in Tunisia (Souid et al., 2019).
This work aimed to explore the use of the coastal L. spathulatum leaves in the food industry either as food and as a source of bioactive herbal products. For that purpose, leaves were collected in Tunisian sea cliffs and profiled firstly for their nutritional and anti-nutritional properties. The in vitro and ex vivo antioxidant properties and the total levels of phenolics and flavonoids of food grade leaf extracts were also determined. Furthermore, analysis of individual phenolics and flavonoids compounds was carried out by HPLC-ESI-MS/MS.

Plant material and extracts preparation
Leaves of L. spathulatum were collected in March of 2019 from flowering adult plants growing in coastal areas of Tabarka in Tunis (Tunisia) (coordinates: 36°57'23" N 8°45'28.5" E). The taxonomical classification was performed by the botanist Dr.
Abderrazek Smaoui (Center of Biotechnology of Borj Cedria, Tunisia) and a voucher specimen is kept in the herbarium of the Laboratory of Extremophile Plants (voucher code LPEH01). Depending on the analysis, two drying methods were used. For the nutritional analysis, samples were lyophilized, ground in liquid nitrogen, and stored at -20°C. For the preparation of the extracts, leaves were dried at 37°C for one week, milled and stored in the dark at 4°C. For extract's preparation, dried powder was mixed with ethanol (100 % and 50 %, w/w) and water (1:40, w/w), and extracted overnight, at room temperature (RT) with stirring. The extracts were then filtered (Whatman paper no. 4), and dried in a rotary evaporator under reduced pressure at 40°C. The water extracts were freeze dried. The resulting dried extracts were weighed, dissolved in the corresponding solvent at the concentration of 50 mg/mL, and stored at −20°C until analysis.

Nutritional properties 2.3.1 Proximate composition
Moisture was determined as the difference of the weight of the fresh leaves before and after drying at 90°C for 2 d. Ash was determined by incineration of dried biomass at 500°C in a muffle furnace for 7 h. Crude protein content was estimated by the Kjeldahl method and was obtained by multiplying by 6.25 the evaluated nitrogen. Crude fat was determined by a modified protocol of the Bligh and Dyer method (Bligh and Dyer, 1959). Total sugar content was determined using the Anthrone method of Yemm and Willis (1954), while neutral detergent fibre (NDF), acid detergent fibre (ADF) and acid detergent lignin (ADL) were determined in agreement with the International Organization for Standardization (ISO) directives 16472:2006, 13906:2008 and 13906:2008, respectively). Metabolizable energy (ME) was calculated using the Atwater specific factor for vegetables (FAO, 2003) according to the following equation: ME (kcal) = 2.44 × (g protein) + 3.57 × (g carbohydrate) + 8.37 × (g lipid).

Minerals
Dried leaf samples were ground into fine powder. 10 mg of leaf powder were mixed in sulfuric acid (H 2 SO 4 , 1N) for 1 h at 80°C to extract the different minerals (Zorrig et al., 2010). The extract samples were prepared by filtration with a 0.45 µm presyringe filter. Sodium (Na), potassium (K) and calcium (Ca) were assayed by flame emission photometry. Iron (Fe), zinc (Zn), magnesium (Mg), cadmium (Cd) and lead (Pb) were determined through atomic absorption spectrophotometry. Different standard solutions were used : 0-20 µg/ml for Na, K, Ca, Mg and Fe, 0-2 µg/ml for Zn, Cd and Pb, Phosphorous (P) was measured by spectrophotometry at 430 nm. Chloride (Cl) was determined by chloride analyzer model 926. Iodine determination was performed according to the European S t a n d a r d E N 1 5 1 1 1 : 2 0 0 7 . B r i e fl y , d r i e d s a m p l e s (approximately 100 mg) were weighed directly in borosilicate glass tubes (16×125 mm) to which ultrapure water (> 18.2 MW. cm at 25°C) and TMAH (25 wt. % in H 2 O) were added. The glass tubes were capped and placed in a drying oven adjusted to 90 ± 3°C. Iodine was analyzed by inductively coupled plasma mass spectrometry (ICP-MS) using an iCAP TM Q instrument (Thermo Fisher Scientific, Bremen, Germany). The elemental isotope 127 I was monitored for quantitative purposes. The elemental isotope 125 Te was used as internal standard (IS).

Anti-nutritional properties and toxic factors
Trypsin inhibition was assessed by the method of (Bacon et al., 1995) adapted to 96-well microplates. In brief, samples (60 µL at 1 mg/mL), were mixed with the enzyme (60 µL; 0.02 mg/ mL of bovine in 0.001 M of HCL) and incubated in the dark, for 15 min at 41°C. Then, 150 µL of the substrate solution (BAPNA in 20 mM CaCl 2 and 50 mM Tris-HCl pH 8.2), were added and incubated for 10 min, at RT. The reaction was stopped by adding 30 µL of 30 % acetic acid, and the absorbance was measured at 410 nm. Results were expressed as inhibition (%) relative to a blank containing the solvent of the extraction. Inhibition towards a-amylase was evaluated by the method described by (Xiao et al., 2006) using extracts at the concentration ranging from 0.009 to 5 mg/mL. The results were expressed as inhibition (%) relative to a blank containing the solvent of the extraction. The phytic acid content of the extracts was determined according to the protocol described by (Lorenz et al., 2007), in extracts at the concentration of 150 mg/mL. Results were calculated in relation to a calibration curve made with different concentrations of phytic acid.

Determination of in vitro antioxidant activity by radical based methods
The radical scavenging activity (RSA) of the extracts was tested towards DPPH and ABTS according to the methods described previously (Rodrigues et al., 2015). Leaf samples (22 µL, at concentrations ranging from 0.009 to 5 mg/mL) were mixed with 200 µL of DPPH solution (120 µM) in methanol in 96-well microplates, and incubated in darkness at RT for 30 min. The absorbance was measured at 517 nm (EZ read 400, Biochrom). For RSA determination on ABTS radical, a stock solution of ABTS•+ (7.4 mM) was diluted with ethanol to obtain an absorbance of at least 0.7 at 734 nm (EZ read 400, Biochrom). The samples (10 µL at concentrations between 0.009 and 5 mg/ mL) were mixed in 96-well microplates with 190 µL of ABTS•+ solution. After an incubation for 6 min, the absorbance was measured at 734 nm (EZ read 400, Biochrom). RSA was expressed as percentage relative to the negative control containing the corresponding solvent, and as half-maximal effective concentration (EC 50 values, mg/mL) when possible. Butylated hydroxytoluene (BHT) was used as a positive control at concentrations up to 1 mg/mL.

Determination of in vitro antioxidant activity by metal-based methods
The ferric reducing antioxidant power (FRAP), the metal chelating activity on copper (CCA) and iron (ICA) were determined according to previously described protocols (Rodrigues et al., 2015). FRAP determines the ability of the extracts to reduce Fe3+. Samples (50 µL at concentrations from 0.009 to 5 mg/mL), distilled water (50 µL) and 1% potassium ferricyanide (50 µL) were mixed and incubated at 50°C for 20 min. Then, 50 µL of 10% trichloroacetic acid (w/v) and ferric chloride solution (0.1 %, w/v) were added, and absorbance was measured at 700 nm (EZ read 400, Biochrom).
The ICA chelating activity was determined by measuring the formation of the Fe 2+ ferrozine complex according to (Rodrigues et al., 2015). 30 µl of the samples were mixed with 200 µL of dH 2 0 and 30 µL of a FeCl 2 solution (0.1 mg/mL in water) in 96well microplates. After 30 min, 12.5 µL of ferrozine solution (40 mM in water) was added. Aborbance was measured at 562 nm using a microplate reader (EZ read 400, Biochrom).
EDTA (1 mg/ml) was used as the positive control. For all the above mentionned methods, increased absorbance of the reaction mixture indicated increased reducing power. Results were expressed as (%) of inhibition, relative to the positive control, (FRAP) and to the negative control (CCA and ICA) and as EC 50 values.

Determination of ex vivo antioxidant activity
The ex vivo antioxidant activity of the extracts were evaluated by their ability to inhibit lipid peroxidation of porcine brain cells by the thiobarbituric acid reactive substances (TBARS) assay, and by the oxidative haemolysis inhibition assay (OxHLIA), using a sheep erythrocyte solution and AAPH as a free radical generator, according to the methods described in . For TBARS assay, a porcine brain cell solution (1:2, w/v; 100 µL) was incubated with 200 µL of sample or trolox, 100 µL of FeSO4 (10 µM) and 100 µL of ascorbic acid (0.1 mM) at 37°C for 1 h. Then, 500 µL of trichloroacetic acid (28 % w/v) and 380 µL of thiobarbituric acid (TBA; 2 % w/v) were added and the mixture was heated at 80°C for 20 min. After centrifugation, the color intensity of the malondialdehyde (MDA)-TBA complexes formed in the system was measured at 532 nm.

Total phenolic (TPC) and flavonoid (TFC) contents
The TPC and TFC were determined in the extracts at the concentration of 5 mg/mL. TPC was determined by the Folin-Ciocalteu (F-C) assay, and TFC by the aluminum chloride colorimetric method adapted to 96-well microplates. In brief, the extracts (5 µl at a concentration of 5 mg/ml) were mixed with 100 µl of tenfold diluted F-C reagent and incubated at RT for 10 min. Subsequently, 100 µ l of Na 2 CO 3 (75 g/1, w/v) were added and the absorbance was measured on a microplate reader (EZ read 400, Biochrom) at 725 nm after a 90 min incubation period at RT. TPC was expressed as gallic acid equivalents (GAE) in milligrams per gram of dry extract using a calibration curve plotted from gallic acid standard solutions (0 -2 mg ml -1).
The total flavonoid content (TFC) of the extracts was estimated by the aluminium chloride (AlCl 3 ) colorimetric method according to (Akrout et al., 2011). 1 ml of diluted sample was mixed with 1 ml of 2% aluminium trichloride (AlCl 3 ) methanolic solution. After incubation at room temperature for 15 min, the absorbance of the reaction mixture was measured at 430 nm with a microplate reader (EZ read 400, Biochrom). Results were expressed as milligrams of quercetin equivalents per gram of dried sample (mg QE/g DW) using a calibration curve produced with quercetin concentrations between 0.01 and 2.5 mg/mL.

2.9
High-performance liquid chromatography coupled with electrospray ionization mass spectrometry (HPLC-ESI-MS/MS) analysis of phenolic and flavonoid compounds.
The chemical composition of the extracts was determined using a Dionex Ultimate 3000RS UHPLC instrument. Samples were filtered (0.22 mm PTFE filter membrane, Labex Ltd, Hungary) before HPLC analysis, and injected onto a Thermo Accucore C18 (100 mm x 2.1, mm i. d., 2.6 mm) column thermostated at 25°C (± 1°C). The solvents used were water (A) and methanol (B), acidified with 0.1% formic acid, and the flow rate was maintained at 0.2 mL/min. A gradient elution was used: 5% B (0-3 min), a linear gradient increasing from 5% B to 100% (3-43 min), 100% B (43-61 min), a linear gradient decreasing from 100% B to 5% (61-62 min) and 5% B (62-70min). The column was coupled with a Thermo Q-Exactive Orbitrap mass spectrometer (Thermo Scientific, USA) equipped with electrospray ionization source. Spectra were recorded in positive and negative-ion mode, respectively. The trace finder 3.1 (Thermo Scientific, USA) software was applied for target screening. Most of the compounds were identified based on previously published work or data found in the literature. The exact molecular mass, isotopic pattern, characteristic fragment ions and retention time were always used to identify the molecules.

Statistical analysis
Experiments were conducted at least in triplicate and results were expressed as mean ± standard deviation (SD). Differences in significance (p< 0.05) were evaluated by one-way analysis of variance (ANOVA), pursued by the Tukey HSD test. Statistical analyses were performed using XLStat2014 ® . The EC 50 values were determined by sigmoidal fitting of the data in the GraphPad Prism v. 5.0 software.
The presence of antinutritional and toxic factors in the extracts was evaluated in terms of trypsin and amylase inhibition, and levels of phytic acid (Table 3). A high trypsin inhibition was observed with the water extract (82.8%), followed by the hydroethanolic (75.1%) and ethanol (72%) extracts. No relevant a-amylase inhibition was observed, and no phytic acid was detected.

Antioxidant properties
The antioxidant potential of the extracts was evaluated by five in vitro methods, namely two radical-based assays (RSA on DPPH and ABTS radicals), and three metal-related methods (FRAP and metal chelation of iron and copper). As can be seen in Table 4, the ethanol and the hydroethanolic extract had the highest capacity to scavenge free radicals, with EC 50 values of 0.04 and 0.08 mg/mL for DPPH and 0.10 and 0.05 mg/mL for ABTS, respectively. For those extracts, the EC 50 values were similar or even lower than those obtained with the positive control (BHT, 0.11 and 0.141 mg/mL for the DPPH and ABTS assays, respectively). Samples had no capacity to chelate iron, but exhibited significant copper chelating properties, and again, the best results were obtained with the ethanol and hydroethanolic extracts, with similar EC 50 values (0.48 mg/mL). Samples also had the capacity to chelate iron, with the ethanol and hydroethanolic samples exhibiting the lowest EC 50 value (0.04 mg/mL).
To gain further knowledge on the antioxidant properties of the extracts, samples were tested by two ex vivo antioxidant assays, which allowed to evaluate their capacity to inhibit lipid peroxidation (by the TBARS formation) and oxidative haemolysis (OxHLIA) (Figure 2). The hydroethanolic and the ethanol extracts displayed the highest capacity to inhibit lipid peroxidation, with EC 50 values of 126 and 247 mg/mL, respectively. The same samples were also the most active towards oxidative haemolysis, with EC 50 values of 138 and 146 mg/mL for the ethanol and the hydroethanolic extract, respectively.

Total phenolic and flavonoid quantification and HPLC identification
The total levels of phenolics (TPC) and flavonoids (TFC) were quantified in the extracts, and results are shown in (Figure 3). The TPC peaked in the water (334.85 mg GAE/g, dw) and hydroethanolic extracts (324.0 mg GAE/g, dw), followed by the ethanol extract (251.7 mg GAE/g, dw). In the contrary, the ethanol extract had the highest level of flavonoids (49.3 mg QE/g,), followed by the hydroethanolic (19.8 mg GAE/g, dw) and the water (11.6 mg GAE/g, dw) extracts.
To gain a deeper knowledge on the individual chemical components of the extracts, an analysis was made by HPLC-ESI-MS/MS, and results are summarized in Table 5. The ethanolic extract was used for this HPLC analysis because of its high antioxidant activities. HPLC analysis identified 58 individual compounds (Table 5) including mainly hydroxybenzoic acids (gallic, syringic acids), hydroxycinnamic acids (caffeic, coumaric, ferulic acids) and flavonoids (catechin, epigallocatechin gallate and naringin).

Discussion
This study appraised the nutritional profile of L. spathulatum leaves aiming to evaluate its suitability for human consumption. Its moisture level was like the values reported for other halophytes species, such as Polygonum maritimum L. cultivated with saline water containing up to 100 mM of sodium chloride (NaCl) (sea knotgrass, 70 -80%; Rodrigues et al., 2019) and L. algarvense Erben cultivated in greenhouse conditions and irrigated with freshwater (79.8%; Rodrigues et al., 2020). However, moisture was lower than the values reported for edible halophytes characterized by its succulence, such as Sarcocornia and Salicornia species, which moisture levels are usually higher than 85% (Custodio et al., 2021), and of some common vegetables, including Lactuca sativa L. (lettuce, 94.7%) (Custodio et al., 2021;USDA, 2021). A high moisture content is usually related to a higher tendency for food spoilage, as observed for example in lettuce (Barg et al., 2008;Kyere et al., 2020), therefore having a high influence on the product shelf life and in the consumers' acceptance of a product. Therefore, L. spathulatum with a lower moisture level than other common edible succulent halophytes may result in a greater consumer acceptability.
The ash content of a plant biomass is related to its total mineral level. Halophytes thrive in saline conditions, have a high  capacity to absorb and retain minerals without toxic effects to the plant, and therefore, usually have higher ash contents than glycophyte plants (Borah et al., 2008;Dıáz et al., 2013). The ash content of L. spathulatum similar to that of the halophyte Cladium mariscus L. (Pohl.) It was however lower than the ash levels of related species, including L. axillare (Forssk.) Kuntze (Al-Easa, 2003) and L. pruinosum (L.) Chaz (El-Amier and Ejgholi), and also than other edible halophytes, including Sarcocornia and Salicornia Custodio et al., 2021). The ash level of L. spathulatum was however higher than that of lettuce cultivated in hydroponics and in the soil (Lei and Engeseth, 2021). Such differences may be dependant on the species and/or on the mineral level of the soils from which the plants were collected. Halophytes usually have a high content in dietary fibre (Dıáz et al., 2013). In this work, NDF was determined to estimate the quantities of fibres including cellulose, hemicellulose, and lignin, and also cutin (Dhingra et al., 2012). While being normally used to appraise feed quality, NDF is considered a valuable tool to estimate the insoluble portion of dietary fibre in food (McDougall et al., 2009;Dhingra et al., 2012). The level of NDF of L. spathulatum leaves is higher than that reported for other vegetables, including Lens culinaris Medik (McDougall et al., 2009;Dhingra et al., 2012), and other edible halophytes, such as Sarcocornia perennis subsp. alpini (Mill.)and Salicornia ramosissima J.Woods (Barreira et al., 2017). It was however lower than Bassia hyssopipifolia (Pall.) Kuntze (Dıáz et al., 2013). Our results suggest that L. spathulatum is a good source of fiber, which has relevant health advantages including prevention of cardiovascular diseases and diabetes, besides contributing to weight loss, due to its low caloric content (Whelton et al., 2005;Yao et al., 2014).
The crude protein of L. spathulatum was as expected low but higher than that of L. axillare, Sarcocornia and Salicornia (Custodio et al., 2021), and C. mariscus Oliveira-Alves et al., 2021). It was however lower than other Limonium species, such as L. pruinosum and other common vegetables, including lettuce and spinach (USDA, 2021), thus suggesting that the consumption of L. spathulatum can contribute to a higher input of protein that these latter species.
Similar to protein, the crude fat content of L. spathulatum was also low, and lower than the levels detected in other Limonium species, such as L. pruinosum (0.92%) and L. axillare, and also than other edible halophytes, including S. perennis perennis and S. perennis alpini (Akyol et al., 2020), and some common vegetables, such as raw lettuce (Lactuca sativa var. logifolia and spinach (USDA, 2021). Moreover, L. spathulatum also had low levels of carbohydrates level, which resulted in a low energetic value (33.7kcal/100 g, dw, corresponding to 7.49 kcal/100 g, fw), lower than the values reported for common vegetables, includings lettuce (20 kcal/100 g, fw), spinach (27 kcal/100 g, fw) (USDA, 2021) and Salicornia bigelovii (3.8 MJ kg −1 , dw, corresponding to 20.17 kcal/100 g, fw) (Dıáz et al., 2013). Such a low energy value, combined with the low-fat and carbohydrates content, suggests that consuming L. spathulatum leaves can contribute to weight loss, and therefore, to prevent relevant non communicable diseases.
Dietary minerals have vital roles in the human body, including bone formation and muscle function (Gharibzahedi and Jafari, 2017), and can be obtained from different food Youssef et al. 10.3389/fpls.2022.979343 sources, including vegetables, fruits, and animal products.
Halophytes have a high capacity to accumulate minerals without toxicity and are therefore indicated as very interesting sources of such elements. In this work, the most abundant macroelements detected in L. spathulatum leaves were Cl -, Ca and Na, while the most abundant microelements were Fe and Zn. Although Clwas previously considered harmful to conventional crops due to its impairment effects on nitrate (NO -3 ) nutrition and consequent crop yield reduction, new findings show its beneficial properties, including improvement of the overall plant growth, tissue water balance, plant water relations, photosynthetic performance, and water-use efficiency (Raven, 2016;Rosales Miguel et al., 2020). Most glycophytes contain 1 -20 mg Cl − g (dw) (Marschner, 2011), while in halophytes Cl − is only toxic at concentrations higher than 50 mg/g (dw) (Geilfus, 2018), which is a higher value than that detected in L. spathulatum.
The Na content of L. spathulatum leaves were lower than the level detected in the same species collected in different locations, in Tunisia (Souid et al., 2019), and than the values reported for different edible halophytes, such as Sarcocornia and Salicornia species (Custodio et al., 2021). It was however higher than the levels detected in the leaves of drought-resistant amaranth (Sarker et al., 2022a), A. tricolor (Sarker and Oba, 2020a) and the leaves of Cladium. mariscus Oliveira-Alves et al., 2021), and in the range of the levels reported for common green vegetables, including (Kim et al., 2016) and seaweed (El-Said and El-Sikaily, 2012). According to the World Health Organization (WHO), the Na daily intake should not exceed 2 g. Therefore, to achieve the maximum daily intake of Na it would be necessary to consume as much as 553.08 g of fresh leaves of L. spathulatum.
The Ca concentration detected in L. spathulatum was higher than those of the leaves of danta (Sarker et al., 2022b), A. lividus (Sarker et al., 2022c), stem amaranth (Sarker et al., 2022d), Salicornia perennis, S. ambigua, and S. neii (Bertin et al., 2014;Riquelme et al., 2016;Barreira et al., 2017), but lower than the Ca level S. fruticosa (Castañeda-Loaiza et al., 2020a). Limonium spathulatum leaves can be considered good source of Ca when compared with vegetables considered rich sources of this element, such as kale, (USDA, 2021). The daily recommended dietary allowances (RDA) for Ca are age and country dependent (Rose and Strombom, 2019), and usually peak in the adolescence (1300 mg) and in the elderly (1000 -1200 mg) (Rose and Strombom, 2019). The consumption of 100 g of fresh L. spathulatum leaves would cover 38 and 29% of the RDA for the elderly and adolescents, respectively. The intake of vegetables rich in Ca is especially important in vegetarians and vegans, where no dairy products are consumed. While absorption of Ca from vegetables is often better than from dairy products, bioavailability issues may arise related with the oxalate levels of plant tissues, since Ca absorption is inversely proportional to the oxalic acid content of the food (Rose and Strombom, 2019). Therefore, future studies should consider determining the oxalate levels of L. spathulatum leaves.
Iron was the major micro element in L. spathulatum, in similar or lower levels than those detected in Sarcocornia species (Riquelme et al., 2016;Barreira et al., 2017). It was however higher than and in Fe rich vegetables, such as parsley (Petroselinum crispum (Mill.) Fuss) (USDA, 2021). Therefore, consuming 84 g and 191 g of fresh L. spathulatum could contribute to fulfill the recommended daily Fe intake of 8 -18 mg/day for adults (Trumbo et al., 2001). The Zn levels of L. spathulatum were in the range than those in different Sarcocornia and Salicornia species (Custodio et al., 2021). These were however higher than and spinach (USDA, 2021). The consumption of 1.4 and 1.9 g of fresh L. spathulatum could contribute to fulfill the recommended daily Zn intake of 8 -11 mg/day for adults (Trumbo et al., 2001).
The iodine level of L. spathulatum was lower than that found in some edible halophytes, such as Crithmum. maritimum, grown in a hydroponic system (Sarroua et al., 2019) and Inula crithmoides L. cultivated in a controlled environments under irrigation with different salinities (Zurayk and Baalbaki, 1996). It was however higher than lettuce and asparagus (Asparagus Youssef et al. 10.3389/fpls.2022.979343 Frontiers in Plant Science frontiersin.org officinalis L.) (WHO, 2018), and therefore, could be an interesting source of iodine, when compared with common vegetables, especially for pregnant woman.
Halophytes can accumulate toxic metals, including Pb and Cd, when growing in contaminated soils (Caetano et al., 2008). However, the accumulation of such elements generally occurs in the roots, since its translocation to aboveground organs is limited, as observed in different halophytic species, such as S. fruticosa, S. ramosissima and A. macrostachyum (Caetano et al., 2008;Moreira da Silva, 2008;Redondo-Goḿez et al., 2010). In this work, Pb and Cd, were not detected in the leaves of L. spathulatum. Some other molecules exhibit toxicity and/or antinutrient activity may be present in halophytes. This is the case of tannins, phytic acid, trypsin and alpha-amylase inhibitors which are considered antinutritional factors since they might interfere with the bioavailability and/or digestibility of some nutrients, including proteins and minerals (Samtiya et al., 2020). In this work, the extracts of L. spathulatum were phytic acid free and presented a high capacity to inhibit trypsin, but reduced aamylase inhibition, when tested at 1 mg/mL.
In this work, the antioxidant potential of L. spathulatum leaves was evaluated by different in vitro methods, covering different mechanisms of action, namely those involving free radicals and metal ions. The ethanol and the hydroethanolic extracts had in general the highest capacity to scavenge free radicals when compared to water extracts, thus suggesting that such extracts contain primary antioxidant compounds with the capacity to neutralize free radicals and prevent the initiation and propagation of oxidative chain reactions (Loganayaki and Manian, 2010). Such activity was similar or higher than that of the tested standard (BHT), which is one of the most used synthetic antioxidant additives to food stuffs In general, Limonium species are acknowledged as sources of strong antioxidants. For example, a free radical scavenging activity guided fractionation of a methanol root extract and obtained fractions of L. brasiliense Kuntze resulted in the isolation of five active antioxidant compounds, namely gallic acid, epigallocatechin 3-O-gallate, epigallocatechin, gallocatechin and myricetin 3-O-a-rhamnoside (myricitrin) (Murray et al., 2004). Myricitrin exhibits relevant antioxidant properties, with stronger free radical scavenging activity than other flavonol rhamnosides or quercetin (Wu et al., 2008); all detected in the L. spathulatum extracts. Methanol leaf extracts of L. algarvense also had a strong capacity to scavenge the DPPH radical, with an EC 50 value of 0.54 mg/mL (Rodrigues et al., 2015), although less effective than L. spathulatum.
The strong antioxidant potential of L. spathulatum is most probably related with its high content in polyphenolic compounds, since such molecules are recognized antioxidant agents (Granato et al., 2018;Stankovićet al., 2019).
Since a high antioxidant activity was obtained in the in vitro assays, L. spathulatum was evaluated for the first time for their ability to reduce lipid peroxidation in porcine brain cell membranes (TBARS) and oxidative hemolysis of sheep erythrocytes (OxHLIA). Such assays are appropriate ex vivo models for evaluating inhibition of lipid peroxidation by the presence of antioxidants (Takebayashi et al., 2009;Takebayashi et al., 2012). Similar to the observed in the free radical and metalbased assays, the upmost activity was observed after the application of the hydroethanolic and ethanol extracts, which may be related with the highest levels of polyphenolics and flavonoids detected in such extracts, as stated before. A relevant inhibition of lipidic peroxidation was also detected in a water extract from leaves of L. algarvense (Rodrigues et al., 2015). Lipids are highly vulnerable to peroxidation, which is linked with the onset of several degenerative disorders, including cardiovascular (Gianazza et al., 2021) and neurodegenerative diseases (Angelova et al., 2021). In addition, lipid peroxidation alters the composition, structure, and function of the lipids present in cellular membranes, that may result in DNA and proteins damage. The use of natural products from limonium species such as L. spathulatum capable to decrease cellular lipid peroxidation is therefore considered an important therapeutical tool to prevent the occurrence of degenerative and chronic disorders linked to oxidative stress. There is an increasing interest in the use of these natural extracts to improve foodstuff stability (Da . The high activity detected in the ethanolic extract may be related with its higher level of total polyphenolic compounds, while the activity of the ethanol extract is most probably related with its richness in flavonoids. This hypothesis was conformed by the study of TPC and TFC and the identification of their individual compunds in the most active ethanol extracts of L. spathulatum. TPC of L. spathulatum leaves of all extracts were greater than the leaves of drought-tolerant leafy vegetable amaranth (Sarker and Oba, 2020b), Amaranthus gangeticus (Sarker and Oba, 2020a). Such levels are higher when compared to other medicinal halophytes species with confirmed pharmacological properties such as Limoniastrum monopetalum (L.) Boiss, Trabelsi et al., 2012), Tamarix gallica L. and Mesembryanthemum edule L. (syn. Carpobrotus edulis L.) (Ksouri et al., 2008), and also higher than the levels detected in water extracts made from different medicinal herbs and spices, Rosmarinus officinalis L., Salvia officinalis L., Thymus vulgaris L. and Origanum vulgare L. (Ulewicz-Magulska and Wesolowski, 2019). The TPC of L. spathulatum was similar than that detected in a methanol extract of the same species from Algeria (Mazouz et al., 2020), but higher than that detected in a ethanol extract from L. boitardii (Sefi et al., 2021), and of a methanol extract from leaves from L. algarvense (Rodrigues et al., 2015). In plants, phenols are responsible for pigmentation (Sarker and Oba, 2020a;Sarker and Oba, 2021) and astringency, serve as protective agents against abiotic (e.g.,UV light), and biotic (e.g., parasites and insects) stress (Caleja et al., 2017;Durazzo et al., 2019). Such molecules also have important human health implications, since they exhibit relevant health improvement properties, including antioxidant, anti-diabetic, antiinflammatory and anti-tumor (Albuquerque et al., 2020;Diasa et al., 2021).
More interestingly, flavonoids peaked in the ethanol extract, similar to the total flavonoids found in a methanol extract from L. algarvense (Rodrigues et al., 2015), but in lower amounts than those detected in a hydroethanolic leaf extract from L. boitardii (Sefi et al., 2021). Such differences are highly dependent on several factors, includings the type of extraction used, plant species, as well as biotic and abiotic stresses (Do et al., 2014;Karoune et al., 2015;Cujic et al., 2016;Bakhouche et al., 2021). Flavonoids exhibit important biological properties potentially associated with multiple health benefits to the antioxidant system of the human body. They are also considered as an important element in dietary supplements, pharmaceutical, medicinal and commercial applications. (Panche et al., 2016;Castañeda-Loaiza et al, 2020b).
In conclusion, the leaves of L. spathulatum collected from Tunisian sea cliffs were good source of minerals and fibers useful in the human diet for attaining nutritional sufficiency. The high in vitro and ex vivo antioxidant activities associated with high phenolics and favonoids contents and compounds suggest the possibility to use extracts of L. spathulatum in herbal products with the aim of improving general health and well-being, and/or as food additives for preventing lipid oxidation of lipidrich foods.

Data availability statement
The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.