Autecology and Phytochemistry of Genus Amaranthus in the Nile Delta, Egypt

Attention should be paid to increase our knowledge of the best conditions for propagation of economic plants. In this connection, the importance of studying plants in their natural habitats, the effect of each habitat factor upon growth, establishment and distribution must be emphasized.

Many investigators studied the main active constituents of several species belonging to family Amaranthaceae. Nodeide et al. (1996) reported that the green leaves of Amaranthus viridis were rich for water, energy, fats, proteins, minerals, amino acids and carotenoid. In some species of genus Amaranthus, sixteen phenolic acids were identified by Sokolowska-Wozniak (1996). Two comarins and three flavonoids were isolated from Amaranthus paniculatus by Bratoeff et al. (1997). Singh and Whitehead (1996) mentioned that Amaranthus species are commonly utilized as vegetable and consumed in Africa, China, India and Italy.

Jale et al. (1999) mentioned that grain amaranth was used as a partial substitute for barley in diets fermented in artificial rum. Syamdaya and Naidu (1999) studied the nutritive value of amaranth to sheep. One can expect the prime importance of the individuals belonging to this family as a source of substances that can be use for several industrial, medicinal and fodder purposes.

The present study aims at the description of the weed communities associated with the studied plant species in their natural habitats by using multivariate techniques of classification and ordination, analysis of soil samples to determine the variables controlling the distribution and abundance of the identified weed communities, seed germination under different environmental factors and physiochemical investigation to detect the main active constituents and amino acids in the studied plant species.

In the present study, ten localities (sites) were chosen in three governorates of the Nile Delta region (Fig. 1). These governorates are Kafr El-Sheikh, El-Dakahlyia and Damietta. After regular visits to the different sites, forty stands representing the apparent physiognomic variations in the vegetation and environmental features were used for sampling vegetation of the different habitat types supporting the growth of Amaranthus graecizans, A. lividus and A. viridis. The stands were distributed as follows: 7 stands in canal banks, 7 stands in orchards and 26 stands in cultivated lands. The sampling processes have been carried out during the years 2004-2006.

Fig. 1:	Map of the Nile Delta region showing different localities (sites) as indicated by (●) in the study area

The density and plant cover of each species have been estimated in each stand using quadrat of 5 m². The relative values of density and cover were calculated for each species and summed up to give an estimate of its importance value (IV) in each stand, which is out of 200. The Nomenclature and identification of the species was according to Tackholm (1974) and Boulos (1999-2005).

Soil sample was collected from each stand at a depth of 0-25 cm for physical and chemical analyses. Soil texture was determined using the hydrometer method, while the water-holding capacity was estimated using the Hilgard-Pan box method of Piper (1947). Oxidizable organic carbon was estimated using the Walkely and Black rapid titration method (Black, 1979). The percentage of calcium carbonate was determined by addition of 100 mL 1 N HCl to 5 g soils and the excess of acid titrated against 1 N NaOH. Soil salinity (EC) and soil reaction (pH) were estimated in 1-5 water extract using the conductivity and pH meters, respectively. Chloride was determined by titration against N/35.5 silver nitrate, while sulphate was estimated gravimetrically using 5% barium chloride. Estimation of carbonate and bicarbonate were carried out by titration against 0.1 N HCl. The cations Na⁺, K⁺ and Ca⁺⁺ in the soil solution were estimated using flame photometer as described by Allen et al. (1974).

Two trends of multivariate analysis of vegetation were applied, namely classification and ordination. Both trends have their merits in helping to understand the vegetation and environmental phenomena. Two-Way Indicator Species Analysis (TWINSPAN-a FORTRAN Program) was used for classification (Hill, 1979; Gauch and Wittaker, 1981), while the ordination techniques applied were the Detrended Correspondence Analysis (DCA) and Canonical Correspondence Analysis (CCA) using CANOCO- a FORTRAN Program (Ter Braak, 1986, 1988). The relationships between the vegetation gradients and the environmental variables can be indicated on the ordination diagram produced by canonical correspondence analysis (CCA biplot), on which points represent species and arrows represent environmental variables. The statistical treatments applied in the present study were according to Snedecor and Cochoran (1968) and Anonymous (1993).

Germination experiments were conducted to find out the effect of salinity levels, light and dark, temperature and water spray (humidity) on the rate of seed germination of the three different Amaranthus species. For the first three experiments, germination was tested in equal sized Petri-dishes (13 cm) containing double layered filter paper moistened with distilled water or with different test solution. For each treatment, one hundred seeds were sown in each dish and two replicates Petri dishes were used. In case of water spray experiment, equal sized pots (14 cm height and 14 cm diameter) were filled with clean sand and one hundred seeds also sown at 0.5 cm depth.

Concerning the phytochemical analysis, the plant samples were handly cleaned, separated into roots, stems and leaves, air-dried, ground to fine powder and kept in a well stopper vessels to be ready for different phytochemical investigations. The mean values of moisture, ash, water-soluble ash, acid-insoluble ash and total lipid content were investigated according to AOAC (1970) methodology. Soluble sugars, glucose, sucrose, polysaccharides and total nitrogen content were estimated according to Naguib (1963, 1964). The protein content was determined colorimetrically as described by Waslein (1975). The preliminary phytochemical screening was carried out following the methods described by Wall et al. (1964), Claus (1967) and Markham (1982). Hundred grams of each plant powder was subjected to extraction with successive solvents using AOAC (1970) methodology. The macro and microelements were determined by atomic absorption spectrophotometer using the methods described by Allen et al. (1974). The identification and quantitative determination of amino acid in the plant powders were carried out using amino acid analyzer (Model, LC 3000) as described by Moore and Stein (1958).

Vegetation analysis
Classification of stands: The dendrogram obtained from cluster analysis based on the importance values of 65 species recorded in 40 sampled stands in the study area indicated the distinction of four vegetation groups (Fig. 2, Table 1). Group A comprises 12 stands codominated by Amaranthus graecizans (IV = 37.70) and Portulaca oleracea (IV = 29.42). The important species in this group include Sonchus oleraceus (IV = 13.99), Cyprus rotundus (IV = 13.98) and Dactyloctenium aegyptium (IV = 11.72). Group B includes 17 stands codominated by Amaranthus lividus (IV = 34.42) and Cynodon dactylon (IV = 26.29). In this group, the important species are numerous such as: Sorghum vigratum (IV = 18.51), Cyprus rotundus (IV = 14.42), Ammi majus (IV = 14.31), Convolvulus arvensis (IV = 13.87) and Bidens pilosa (IV = 10.25). Group C includes 9 stands codominated by Alternanthera sessilis (IV = 36.53) and Echinchloa crus-galli (IV = 27.15). The important species in this group are Eclipta alba (IV = 18.29) and Phyla nodiflora (IV = 10.25), group D consists of 2 stands codominated by Aster squamatus (IV = 40.18), Conyza bonariensis (IV = 27.67) and Paspalum distichum (IV = 31.04). The important species in this group comprise Bassia indica (IV = 26.65), Phragmites australis (IV = 25.94), Pluchea dioscoridis (IV = 15.50) and Alternanthera sessilis (IV = 13.75).

Fig. 2:	The dendrogram resulting from cluster analysis of 40 sampled stands representing habitat types of some Amaranthus species. The dashed line denotes the level at which the dendrogram yields four distinct vegetation groups

Ordination of stands: The ordination of the sampled stands which obtained by detrended correspondence analysis (Fig. 3) indicated that, the vegetation groups yielded by cluster analysis are markedly distinguishable and having a clear pattern of segregation on the first and second axes of the ordination planes. Group A codominated by Amaranthus graecizans and Portulaca oleracea is separated at the central part of the DCA diagram. Group B codominated by Amaranthus lividus and Cynodon dactylon is segregated at the left side of the ordination diagram. On the other hand, group C codominated by Alternanthera sessilis and Echinochloa crus-galli is segregated at the right side of the DCA diagram. It is clear that, groups B and C are separately segregated at both sides of group A, where these three groups (A, B and C) are distinctly located on the positive and negative sides of the first and second axes of DCA diagram. However, group D codominated by Aster squamatus, Conyza bonariensis and Paspalum distichum is separated at the upper most right positive side of DCA diagram.

Vegetation-soil relationships
Soil analysis: It has been found that, most of the soil characteristics showed a little variation between the different groups of the sampled stands. The soil texture is mainly formed of coarse fraction (sand) and partly of fine fractions (silt and clay). The mean values of water-holding capacity and soil porosity are obviously comparable in all groups. The mean values of calcium carbonate content are higher in groups C (10.89%) and B (7.88%) than in groups A (5.32%) and D (3.75%), while those of organic carbon content are higher in groups A (0.33%), B (0.29%) and D (0.26%) than in group C (0.14%). The pH values indicated that, the soil reaction is neutral or slightly alkaline and it ranged between 7.38 in group A and 7.80 in group D (Table 2). The Electrical Conductivity (EC), chloride and sulphate attained higher mean values in groups C and D than in groups B and A. The soluble bicarbonate is detected in traces. The concentration of extractable cations: Na⁺, K⁺ and Ca⁺⁺ attained their highest mean values in group D (1465.00, 494.25 and 111.70 ppm, respectively).

Table 1:	Mean value and coefficient of variation of the importance value of species in the vegetation groups resulting from cluster analysis of the sampled stands

Fig. 3:	Detrended Correspondence Analysis (DCA) ordination of the 40 sampled stands with four cluster groups

Table 2:	Mean value and standard error (±) of the different soil variables in the sampled stands representing the four vegetation groups obtained by cluster analysis in the habitat types of Amaranthus species
WHC: Water-Holding Capacity, EC: Electrical Conductivity

The correlation between vegetation and soil variables: The relationship between vegetation and edaphic variables is indicated on the ordination diagram produced by Canonical Correspondence Analysis (CCA) of the biplot of species-environment as shown in Fig. 4. It is obvious that, the values of clay, bicarbonate, porosity, sodium cation, sand fraction, water-holding capacity, organic matter, soil reaction (pH) and electrical conductivity are the most effective soil variables which showed a distinct significant correlations with the first and second axes of the CCA biplot diagram.

Seed germination: The seed germination capacity of Amaranthus species is investigated under different levels of salinity, light and dark, temperature and water spray (Table 3). The effect of different salinity levels on the seed germination of the three studied Amaranthus species showed that, the rate of germination is reached its highest values of 97% with distilled water treatment. When the low salinity levels of 0.02, 0.03 and 0.04 M NaCl solution are used, the percentages of germination attained 92, 77 and 57% for A. graecizans, 86, 80 and 57% for A. lividus and 94, 80 and 68% for A. viridis, respectively. But at salinity levels of 0.1, 0.2, 0.3, 0.4 and 0.5 M NaCl solutions, the percentages of germination decreased gradually and the minimum rate of germination at 0.5 M NaCl solution attained 9% for A. graecizans and 5% for both A. lividus and A. viridis. The results obtained from the effect of light and darkness on seed germination of the studied species showed that, the highest values of germination attained 90, 98 and 70% under continuous light for A. graecizans, A. lividus and A. viridis, respectively. The minimum numbers of germinated seeds are 65% for A. graecizans and 58% for both A. lividus and A. viridis. The seeds of the investigated plant species had the capacity of germination between 20-40°C for both A. lividus and A. viridis and 25-40°C for A. graecizans. It is evident that, the optimum temperature for the seed germination of the three selected species are 35°C (49%), 40°C (98%) and 30°C (78%) for A. graecizans, A. lividus and A. viridus, respectively. It is also obvious that, the decreased amount of water spray is badly affected on the rate of seed germination of the three plant species. In case of A. graecizans, seed germination is started at 10 mm water spray being 34%. At 5 and 10 mm water spray, both A. lividus and A. viridis seeds are failed to germinate and started at 15 mm water spray being 23 and 46%, respectively. At the highest level of applied water spray (saturated), the germination percentages reached 75, 90 and 96% for the seeds of A. graecizans, A. lividus and A. viridis, respectively.

Fig. 4:	Canonical Correspondence Analysis (CCA) ordination diagram with soil variables represented by arrows. The indicator and preferential species are abbreviated to the first three letters of each of the genus and species

Table 3:	No. of germinated seed of Amaranthus species under different levels of temperature, water spray (humidity), salinity and light/dark

Table 4:	Mean value of chemical constituents in different organs of Amaranthus species

Table 5:	A preliminary phytochemical screening of active constituents of the different organs of Amaranthus species
+ve: Present, -ve: Absent

Phytochemical analysis
Determination of chemical constituents: The data analysis showed that, A. lividus contained a relatively high percentage of the mean values of moisture content (9.51%), ash content (20.67%), water-soluble ash (11.17%), total protein (214.8 mg/100 g dry wt.) and total lipid (13.73%). A. graecizans contained a relatively high percentage of the mean values of acid insoluble ash (2.45%) and total carbohydrates (196.5 mg/100 g dry wt.). The highest mean value of total nitrogen content (271.14 mg/100 g dry wt.) is recorded in A. viridis (Table 4).

Preliminary phytochemical screening: The presence of alkaloids, carbohydrates, flavonoids, sterols and tannins in all organs of the studied species. Saponins is detected only in the leaves of both A. lividus and A. viridis as well as in the stems of A. viridis. Sulphates are recorded in all organs of the studied species except in the leaves of A. graecizans. Chlorides are recorded in all investigated plant organs except in the leaves and roots of A. graecizans (Table 5).

Table 6:	Extraction of the different fractions of Amaranthus species with successive organic solvents

Table 7:	Mean value of amino acid concentrations (μg mg<-1) in Amaranthus species

Fig. 5:	Variation in cation concentrations (K+, Mn++, Cu++, Na+, Mg++, Fe++, Ca++, Zn++ and Cd++) of root, stem and leaf of Amaranthus species, AG: Amaranthus graecizans, AV: Amaranthus viridis, AL: Amaranthus lividus

Extraction with successive solvents: The results indicated that, the leaves of A. lividus attained a relatively high percentages of total extractives being 96.45 g%, while the lowest one (15.25 g%) is recorded in the roots of A. graecizans (Table 6).

Elementary analysis: It is clear that, the highest values of potassium ion concentration (112.6 mg/100 g dry wt.), iron (198.5 mg/100 g dry wt.), copper (1.17 mg/100 g dry wt.) and cadmium (0.19 mg/100 g dry wt.) are recorded in A. graecizans. The sodium ion concentration (276.74 mg/100 g dry wt.), calcium (93.14 mg/100 g dry wt.), magnesium (3.34 mg/100 g dry wt.), manganese (0.45 mg/100 g dry wt.) and zinc (2.03 mg/100 g dry wt.) are recorded in A. lividus (Fig. 5).

Amino acids investigation: The data obtained from the amino acids investigations are shown in Table 7. Fifteen amino acids are detected in each of the studied species, namely: aspartic, threonine, serine, glutamic, proline, glycine, alanine, valine, leucine, isoleucine, phenylalanine, tyrosine, histidine, lysine and arginine, in addition to cystine which is detected only in A. graecizans.

Amaranthus is a cosmopolitan genus comprises almost 65 species, distributed in the tropical, subtropical and warm regions of the world (Boulos, 1999). In the present study, the chosen species, namely: Amaranthus graecizans, A. lividus and A. viridis have high medicinal and nutritive values (El-Morsy, 2001). The habitat types supporting the growth of these plants are mainly ruderal habitats including orchards, cultivated lands and canal banks, which predominate in the agricultural areas of the Nile Delta region. According to the map of the world distribution of the arid regions (UNESCO, 1977), in the Nile Delta, summer is warm with an average temperature ranges between 20 and 30°C, while winter is mild with an average temperature ranges between 10 and 20°C. Most of rainfall occurs during winter.

The weed vegetation is classified by cluster analysis into four groups, each group comprises a number of stands which are similar in their vegetation and characterized by dominant and/or codominant species as well as by a number of indicator and/or preferential species. The recognized groups are: Group A is codominated by Amaranthus graecizans and Portulaca oleracea, group B is codominated by Amaranthus lividus and Cynodon dactylon, group C is codominated by Alternanthera sessilis and Echinochloa crus-galli and group D is codominated by Aster squamatus, Conyza bonariensis and Paspalum distichum. These groups may be related to alliance of Digitarietalia sanguinalis described by Zohary (1973). The associations of weed vegetation recognized in the present study might be similar to those described by El-Fahar (1989), El-Ashri (1996) and Omar (2006). The ordination of the sampled stands by DCA indicated that, group A (Amaranthus graecizans and Portulaca oleracea) and group C (Alternanthera sessilis and Echinochloa crus-galli) are more closely related to each other than group B (Amaranthus lividus and Cynodon dactylon) and group D (Aster squamatus, Conyza bonariensis and Paspalum distichum). This may be due to the distinct similarities of the floristic composition in these vegetation groups. The application of CCA biplot between the vegetation groups and soil variables indicated that, fine fraction (clay), bicarbonate, porosity and sodium ions are the most effective soil variables controlling the distribution and richness of the weed vegetation in the study area, followed by coarse fraction (sand), salinity (EC), water-holding capacity and soil reaction (pH). These findings are in accordance with those of Mashaly and Awad (2003) and Omar (2006).

With regard to seed germination, it is denoted that, Amaranthus graecizans is more salt tolerant than the other two species, also A. viridis is more sensitive for salinity than A. lividus. The seed germination showed distinct sensitivity to continuous darkness, while in continuous light, the seeds attained their highest values of germination. These observations, may give an indication that these species are long-day plants. The seeds had the capacity to germinate between wide ranges of temperature. This may explain why these species prefer to flourish at early and mid-summer. The percentage of seed germination of the studied species increased with rise of water spray or humidity level.

Dealing with the phytochemical investigation, it is notable that the leaves of the studied species are usually higher in their chemical constituents than the stems and roots. The highest value of total protein content was observed in A. viridis, that of total lipid in A. lividus and that of total carbohydrates in A. gracezans. It may be concluded from the results of the preliminary phytochemical screening that Amaranthus plants may be used as sources of potentially useful products. and that they deserve further investigation to explore the nature of these products. The results of amino acids investigation provide evidence that the genus Amaranthus being rich in proteins with essential amino acids like lysine which is considered as a good candidate for food supply both as grain crop and as vegetable. In this connection, Amarantin as storage protein of Amaranthus was isolated by Romero et al. (1996). Sena et al. (1998) reported that, A. viridis being an excellent source of protein. The phytochemical results in the present study, seemed to be comparable with those obtained by Raja et al. (1997), El-Morsy (2001) and Omar (2006). Consequently, the selected plant species appeared a promising weeds as a renewable natural resources and raw materials for different uses in industrial, food, forage and pharmaceutical purposes.