MATERIALS AND METHODS

Seed samples: The Trifolium species used in this study were T. alexandrinum L., T. spumosum L., T. resupinatum L., T. lappaceum L., T. campastre Scrheb., T. angustifolium L. and T. purpureum Loisel. Matured seeds of these species were collected from various locations in Adana (Turkey) between June and August 2009.

Oil extraction and preparation of Fatty Acid Methyl Esters (FAME): Impurities were removed from the seeds and the cleaned seeds were ground into powder using a ball mill. Lipids were extracted with hexane/isopropanol 2 v/v (Hara and Radin, 1978). The lipid extracts were centrifuged at 10.0 g for 5 min and filtered; then the solvent was removed on a rotary evaporator at 40°C.

Capillary GLC: Fatty acids in the lipid extracts were converted into methyl esters by means of 2% sulphuric acid (v/v) in methanol (Christie, 1990). The fatty acid methyl esters were extracted with hexane. Then the methyl esters were separated and quantified by gas chromatography and flame ionization detection (Schimadzu GC, 17 Ver.3) coupled to a glass GC 10 software computing recorder. Chromatography was performed with capillary column (25 m in length) and 0.25 mm in diameter, Permabound 25, Machery-Nagel, Germany using nitrogen as carrier gas (flow rate 0.8 mL min^-1). The temperatures of the column, detector and injector valve were 130-220 and 240-280°C, respectively. Identification of the individual method was performed by frequent comparison with authentic standard mixtures that were analyzed under the same conditions.

Determination of protein and tannin contents: The extracts were prepared in 70% aqueous acetone (v/v). Total phenols were measured by using the Folin-Ciocalteu method and the results are expressed as tannic acid (Merck; Darmstadt, Germany) equivalent. Protein precipitation capacity of tannins was measured by the method of Makkar et al. (1995). For the protein precipitation assay, seed extracts were prepared in 50% aqueous methanol as presence of acetone intereferes with these assays. Nitrogen content (Crude protein = Nx6.25) in the seeds was determined by Kjeldahl method (AOAC, 1990). Protein and tannin analyses were carried out in triplicate.

Statistical analysis: The experimental design was completely randomized design with 3 replications. Data were analyzed SAS packet program. Biplot analysis was done to explain between fatty acids and Trifolium species by the method of Yan and Kang (2003).

RESULTS AND DISCUSSION

In this study, the total protein amount, fatty acid composition and tannin contents of some Trifolium species from Turkey were determined. The results of the fatty acid analysis are shown in Table 1 and Fig. 1 and total protein and tannin contents in Table 2.

The fatty acid composition of the seeds: The fatty acid composition of some plants used as feed crops from Fabaceae family showed different saturated and unsaturated fatty acid concentrations. The main components in the seed oils of Trifolium are linoleic, oleic, palmitic and linolenic acids. In all Trifolium studied found linoleic acid as major component (47.16-60.82%). Oleic acid content was at the highest level in T. spumosum (23.19%), but found to be at the lowest level in T. lappaceum (13.58%) and T. alexandrianum (12.86%). Linoleic acid was found in the greatest proportion in the seed oil. Linoleic acid content was at the highest level in T. lappaceum (60.82%), but found to be at the lowest level in T. alexandrianum (47.16%). The seed oils of all the species were richer in linoleic than linolenic acid contents. Linolenic acid was also detected at low level in T. campestre (4.0%) (Table 1).

Table 1:	Fatty acid composition of some Trifolium species from Turkey
14:0: myristic acid, 16:0: palmitic acid, 16:1Δ9: palmitoleic acid, 17:0: margaric acid, 18:0: stearic acid, 18:1Δ9: oleic acid, 18:2Δ9,12: linoleic acid, 18:3Δ9,12, 15: linolenic acid, 20:0: arachidic acid, 22:0: behenic acid, 24:0: lignoceric acid, TSFA: Total saturated fatty acid, TUSFA: Total unsaturated fatty acid

Fig. 1:	The biplot of 7 Trifolium species for fatty acid composition. Fatty acid: M1: Myristic acid, P1: Palmitic acid, P2: Palmitoleic acid, M2: Margaric acid, S: Stearic acid, O: Oleic acid, L1: Linoleic acid, L2: Linolenic acid, A: Arachidic acid, B: Behenic acid, L3: Lignoceric acid)

Table 2:	Total protein (%) and tannin contents (%) of some Trifolium species from Turkey

The oil of the five Trifolium species (T. balansae, T. stellatum, T. nigrescens subsp. petrisavii, T. constantinopolitanum and T. resupinatum var. resupinatum) was reported a similar fatty acid profile. The major fatty acids were linolenic acid (16.6-31.1%), linoleic acid (5-11.3%) and palmitic acid (11.1-18.3%) (Sabudak et al., 2009).

Linolenic acid was found to be the predominant free fatty acid and triacontanol was determinate as the principal free fatty alcohol in T. repens. The hydrocarbons of C₂₉ and C₃₁ were observed in the largest amount in T. repens (Body, 1974; Maffei, 1996). Also, linolenic acid (61.5%) was found the major component in T. resupinatum (Nazir and Shah, 1990).

Behenic acid (22:0) was not found (T. spumosum) or it was in very low amounts in the legume seed oils studied, T. alexandrianum (0.99%), T. resupinatum (0.25%), T. lappaceum (0.82%), T. campestre (0.44%), T. angustifolium (0.26%) and T. purpureum (0.29%). The low amounts behenic acid in legume seed oils is important because of some researchers have indicated that oils with have high levels of behenic acid may be difficult for digestive enzymes in humans and animals (Hilditch and Williams, 1964; Balogun and Fetuga, 1985). Despite its low bioavailability compared with oleic acid, behenic acid is a cholesterol-raising fatty acid in humans and therefore not a suitable for palmitic acid in manufactured triacylglycerols (Cater and Denke, 2001).

We have used the biplot to compare Trifolium species on the basis of fatty acid compositions and to identify species or groups of Trifolium that are particularly good in certain aspects (Fig. 1).

The GT biplot (Fig. 1) displays 63% of the information in the standardized data of the 7 Trifolium species for 11 fatty acids, which is partially presented in Table 1. This biplot can be visualized from two perspectives. First, it shows the associations among the traits across the 7 Trifolium species: (1) a positive correlation (acute angle) between linolenic acid and myristic acid; between arachidic acid, behenic acid, stearic acid and palmitoleic acid; between margaric acid, linoleic acid and palmitic acid (2) a negative correlation (obtuse angle) between oleic acid and behenic acid, stearic acid and palmitoleic acid and (3) a negative correlation (obtuse angle) between linolenic acid and : linolenic acid. Similar relationships among some legumes fatty acid composition of were reported by Kokten et al. (2010). Second, it shows the trait profiles of the Trifolium species, particularly those that are placed farther away from the biplot origin. For example, it shows that T. angustifolium had extremely high linolenic acid and high myristic acid but low palmitic acid, linoleic acid and margaric acid; T. spumosum had extremely high oleic acid but lower levels for other fatty acits and T. alexandrinum had high behenic acid, T. resupinatum had highest palmitic acid, T. campestre had higher linoleic acid (Table 1, Fig. 1).

Total Saturated Fatty Acid (TSFA) of Trifolium species were between 18.5 and 24.4%. T. spumosum has lowest level of saturated acid and T. resupinatum has highest amount of saturated fatty acid concentration (Table 1). On the other hand, the unsaturated fatty acid composition of Trifolium species were determined as high levels reported as other family members of Fabaceae (Bagci et al., 2004a; Masoud, 2009; Bakoglu et al., 2010), Lamiaceae (Bagci et al., 2004b; Bagci, 2007), Boraginaceae (Bagci et al., 2004c), Apiaceae (Bena et al., 1998) family patterns. T. angustifolium has highest level of unsaturated fatty acid (80.3%) and also T. spumosum (80.0%), T. purpureum (79.0%), T. campestre (78.7%), T. lappaceum (77.3%), T. resupinatum (75.4%) and T. alexandrianum (71.1%), respectively. Medicago sativa and M. lupiluna (Fabaceae) has 83.46 and 78.55% unsaturated fatty acid concentrations in their seed oils (Bakoglu et al., 2010).

Total protein amounts of the seeds: Total protein amounts of feed crops studied were between 23.57-36.60% in T. spumosum and T. angustifolium and 28.13, 31.93, 33.63, 35.00 and 36.17% in T. lappaceum, T. resupinatum, T. campastre, T. alexandrinum and T. purpureum, respectively. In the Bakoglu et al. (2009) study, the protein levels of T. pratense and T. aureum were determined which belong to seeds. According to the authors, the protein levels studied were found between 26.1 and 28.6% (Table 2). Many previous studies have reported the changes in protein contents for other forage legumes (Ukhun and Ifebigh, 1988; Bena et al., 1998; Djemel et al., 2005; Maheri-Sis et al., 2007; Bakoglu et al., 2009; Ciftci and Bagci, 2009).

Tannin contents of the seeds: Tannin amounts of feed crops studied were between 0.77-1.73% in Trifolium spumosum and Trifolium lappaceum and 0.87, 1.00, 1.07, 1.37 and 1.43% in, T. alexandrinum, T. resupinatum, T. purpureum, T. angustifolium and T. campastre, respectively (Table 2). Tannins may play a major role in limiting proteolysis in some legumes during ensiling (Albrecht and Muck, 1991), but it would appear that an enzyme, possibly a polyphenol oxidase, may act to inhibit proteolysis in red clover (Jones et al., 1995; Hatfield and Muck, 1999) and may account for the Crude Protein (CP) in red clover (T. pratense) being less degradable than that in Lucerne, both in the silo and in the rumen. Grain legumes are potential sources of energy and micronutrients but their use is still limited because of in certainty about the amount and the effect of antinutritional factors they may contain (Wiryawan and Dingle, 1999).

CONCLUSION

The oil contents of the studied legumes belonging to the Trifolium genus, showed quantitative differences but the seed oils showed uniform fatty acid composition. The results of Trifolium patterns in this study, its chemotaxonomy are noteworthy. Total protein contents have found high level for studied of Trifolium species. By whatever mechanism protein-tannin binding occurs, this process has a negative impact on herbivore nutrition. Tannins can inactivate herbivore digestive enzymes and create complex aggregates of tannins and plant proteins that are difficult to digest. The results revealed that the seed oils of Trifolium species studied with a substantial amount of very long chain fatty acids might have attracted attention because of their value of nutritional, industrial and renewable resources.