The growing awareness of the relevance of food composition for human health has increased the interest of the inclusion of high proportions of fruits and vegetables in diets. To reach the objective of more balanced diets, an increased consumption of legumes, which constitutes a sustainable source of essential nutrients, particularly low-cost protein, is of special relevance. However, the consumption of legumes also entails some constraints that need to be addressed to avoid a deleterious impact on consumers' wellbeing and health. The value of legumes as a source of nutrients depends on a plethora of factors, including genetic characteristics, agro-climatic conditions, and postharvest management that modulate the dietary effect of edible seeds and vegetative material. Thus, more comprehensive information regarding composition, especially their nutritional and anti-nutritional compounds, digestibility, and alternative processing procedures is essential. These were the challenges to write this review, which focusses on the nutritional and anti-nutritional composition of Vigna unguiculata L. Walp, an emerging crop all over the world intended to provide a rational support for the development of valuable foods and feeds of increased commercial value. © 2016 Society of Chemical Industry.

As doutrinas apresentadas são da exclusiva responsabilidade do autor.

No abstract

This study aimed to assess the use of very long-chain fatty acids (VLCFA) as an alternative or a complement to n-alkane markers for estimation of diet composition of goats fed 7 different diets, composed of different proportions of herbaceous (Lolium perenne and Trifolium repens) and heathland woody species (Erica umbellata, Erica cinerea, Calluna vulgaris, Erica arborea, and Ulex gallii), in a metabolism study. Diet composition was estimated from VLCFA (i.e., C(22) to C(34)) and alkane (i.e., C(25) to C(33)) concentrations in diet and feces, by least squares procedures. For all plant species VLCFA concentrations were greater than their alkane concentrations, especially for the herbaceous species and U. gallii. In general, fecal recovery of both markers was incomplete and increased in a curvilinear (P < 0.001) fashion with carbon-chain length. The plants comprising the diets had a significant effect (P < 0.001) on fecal recovery of VLCFA and alkanes. Diet composition estimates based on VLCFA alone were less accurate (P = 0.013) than those obtained using alkanes alone. Combination of VLCFA and alkane data resulted in the most accurate (P < 0.05) estimates, indicating an increase on the discriminatory power among plant species. Use of uncorrected fecal marker concentrations provided the poorest estimates of diet composition, and use of individual recovery data and mean fecal recovery data of the dietary treatment yielded the most accurate ones. Results obtained in this study show that VLCFA have potential to be used as diet composition markers along with alkanes.

No abstract

The effect of n-alkane faecal recovery on the accuracy of diet composition estimates, when increasing the number of diet components, was studied. Seven dietary treatments, composed of different proportions of herbaceous (Lolium perenne and Trifolium repens) and woody species (Erica umbellata, Erica cinerea, Calluna vulgaris, Erica arborea and Ulex gallii) were offered to 28 goats housed in metabolism crates. Diet composition was estimated from alkane concentrations in diet and faeces, with and without correction for incomplete faecal recoveries, using least-squares procedures and was compared with the known diet proportions. There were no significant differences between measured proportions of dietary components and those estimated with alkanes when applying the faecal recovery corrections. In contrast, the proportions calculated without faecal recovery correction differed significantly (p < 0.05) from the actual proportions, overestimating feeds with higher content of long-chain n-alkanes. The diet composition affected significantly the faecal recovery of alkanes if there were significant differences on in vivo digestibility. The n-alkane faecal recoveries decreased as the diet digestibility increased. The results obtained show that the application of the alkane methodology for grazing animals should be preceded by calculation of the actual alkane faecal recoveries for each type of vegetation community and experimental conditions.

An experiment was conducted to assess the potential of long-chain alcohols (LCOH), in alternative or combined with alkanes and long-chain fatty acids (LCFA), as faecal markers to estimate the diet composition of goats grazing heathland vegetation with associated improved pastures. A total of seven diets were offered across the grazing season. The diets were composed of mixtures of herb species (Lolium perenne and Trifolium repens) and woody species (Erica spp., Calluna vulgaris and Ulex gallii) in an attempt to simulate diet selection of goats on these complex vegetation areas. The diet composition was estimated using LCOH markers alone or combined with alkanes, LCFA and alkanes+LCFA, by least square optimization procedures. The data showed large differences between plant species in their LCOH profile. Generally, plant species showed higher LCOH concentrations than those of alkanes and lower than LCFA markers. Faecal recovery of LCOH was incomplete and increased in a linear manner (P < 0.001) with the carbon-chain length, and was influenced by diet composition and its digestibility. The diet composition estimates based on LCOH alone were more accurate (P < 0.05) than those using alkanes or LCFA alone. Results showed that the combination of LCOH with alkanes, LCFA and alkanes+LCFA resulted in more accurate (P < 0.05) estimates of diet composition, indicating that LCOH provided different discriminatory information to that of alkanes and LCFA, helping in the discrimination of the plant species used in this experiment. Results indicate that correction of faecal LCOH concentrations to incomplete faecal recovery is necessary to obtain more accurate estimates of diet composition. Nevertheless, it seemed that the use of a less accurate recovery correction on LCOH markers had a lower impact on the accuracy of estimates than that of alkanes and LCFA.

The present study aimed to evaluate the reliability of the n-alkane technique for estimating herbage intake in beef cattle receiving two different feeding levels (low and high) and to test the effect of different faeces sampling procedures (total faeces samples or rectal grab samples at 8 h intervals) on the n-alkane faecal recoveries and hence on intake estimates. Two consecutive experiments were performed with 11 non-lactating beef cows of "Asturiana de los Valles" breed, fed with lucerne hay at two feeding levels: 1.0 or 1.7 kg DM/100 kg body weight (BW). Animals received a daily dose of paper pellets containing C 24 , C 32 and C 36 n-alkanes as external markers with the purpose of using different n-alkane pairs of adjacent chain length (C 23 /C 24 , C 25 /C 24 , C 31 /C 32 , C 33 /C 32 , C 35 /C 36 ) for intake calculations. A 5-day equilibrium period was sufficient for external marker concentrations and n-alkane pair ratios to reach steady state in faeces. There was no effect of different sampling times (every 8 h) on the faecal excretion of n-alkane pairs C 31 /C 32 and C 33 /C 32 . Grab samples obtained at the time of dosing (0830 h) gave the best estimate of n-alkane pair ratios in total faeces collection. There was a general trend of increasing n-alkane faecal recoveries with increased chain length, although n-alkanes C 30 , C 31 and C 33 showed lower faecal recoveries than expected. n-Alkane recoveries were in general lower at the low feeding level, where an increase in individual variability of natural n-alkane recoveries was observed. At both feeding levels, the n-alkane pairs C 23 /C 24 and C 35 /C 36 gave accurate estimates of the treatment average intake, although the natural n-alkane (C 23 or C 35 ) faecal recoveries showed high individual variability, which could be in part due to analytical bias caused by their extremely low concentration in the diet and faeces. However, the n-alkane pairs C 31 /C 32 and C 33 /C 32 gave the greatest deviations of intake estimates, due to the high discrepancy between the faecal recoveries of natural and dosed n-alkanes. These results demonstrate that in beef cattle, feeding level may have some influence on the relative faecal recoveries of the n-alkane pair used for intake calculations. This effect, together with the high individual variability of n-alkane recoveries, especially under low feeding level, may produce significant deviations of the individual intake estimates.