The sugar beet (Beta vulgaris) belongs to the Quenopodiaceae family and is a species that contains a high sucrose concentration. It is commercially cultivated for the production of sugar and by-products that its extraction generates add 10 % to the value of the harvest (Habeeb et al. 2017). One of these wastes is beet pulp (BP), which is an important by-product widely used in animal feeding, especially if it is destined for ruminants and rabbits, where it can represent more than 20 or 10 % of the daily ration, respectively (Omer et al. 2013, Münnich et al. 2017 and Delgado et al. 2018). So that the BP can be used for animal feeding, the knowledge of its nutritional value is essential. Due to the characteristics of its chemical composition, BP is rich in cellulose, hemicellulose and pectin, and poor in lignin and protein (Foster et al. 2001), but the beet variety, the conditions of its cultivation and the extraction process of the sugar can modify its composition.
In Spain there are two beet producing areas, one in the north that is harvested in winter, and another in the south that is harvested in summer, in both areas are cultivated several varieties of beet and both include dry and irrigated crops . This crop occupied in the 2013/2014 campaign, 26605 ha in the north of the country, with a production of 2.135 million tons of beet harvested in winter, and 8662 hectares in the south, with 749.5 thousand tons in summer (MAPAMA 2018).
Due to there are important differences between the two production areas, which could influence on the chemical composition of the BP, the objective of this study was to characterize the chemical composition of the granulated pulps that are commercialized, from one and another area.
A total of 12 samples of granulated beet pulp, collected directly from the warehouses of the sugar producing plants were analyzed at the beginning and at the end of the beet campaign, by sampling in different points of the warehouse. The chemical analyzes were performed by the official methods of the AOAC (2016) for dry matter (DM), total ash (TA), crude protein (CP), crude protein linked to neutral detergent fiber (CPNDF), crude fat (CF) , neutral detergent fiber corrected for ash and protein (NDF), acid detergent fiber corrected by ash (ADF) and acid detergent lignin (ADL). The starch content (SCRIE) and soluble fiber in neutral detergent (SFND) were also determined by the method described by Martínez- Vallespin et al. (2011). The hemicellulose content was estimated by difference between NDF and ADF and cellulose, by difference between ADF and ADL. The insoluble ashes correspond to the insoluble waste of the total ashes treated with hydrochloric acid and nitric acid, washed with 1N hydrochloric acid and water and new calcination (van Keulen and Young 1977). Likewise, the content of fatty acids was determined by gas chromatography (O'Fallon et al.2007) and the amino acid content by HPLC high performance liquid chromatography (Liu et al. 1995).
To carry out the statistical analysis of the results, the Statgraphics Centurion software (Statgraphics 2010) was used. The effect of the origin of BP, summer or winter, in its chemical composition was analyzed by analysis of variance (ANOVA). The means comparison was made with t Student's test, at the significance level of P <0.05. In addition, an analysis of the main components involved in the composition of the BP was performed.
Table 1 shows the comparison, in terms of the chemical composition and the contents of fatty acids and amino acids of the BP produced in winter and summer.
It was observed that there were differences for the variables insoluble ash, NDF, hemicellulose, SFND and starch (P <0.05). Although the obtained results were in the range of values reported, there was high variability (variation coefficients higher than 10 %), except for DM and CP; being the variability recorded for hemicellulose (21 %), ADL (67 %) and insoluble ash (33 %) especially highlighted.
The total ash content coincided with those reported by Volek et al. (2003), with values around 7 %, while other authors report values between 4.4-13.2 % (NRC 2000 and De Blas et al. 2010). The 54 % of total ashes were insoluble in the winter BP, while in summer they represented only 34 % (P <0.05).The inclusion of other by-products in BP, such as vinasse, may be at the origin of these differences.
Although the CP was relatively low and did not differ between groups, with an average of 8.8 %, it was in the range of reported values (7.4-15.6 %) (De Blas and Carabaño 1996, DePeters et al. 2000, NRC 2000 and Habeeb et al. 2017). More than a half of CP (5.15 %) of BP was linked to cell walls. The CF content was relatively low (1.2 %), and similar between the winter and summer BP, and in the range between 0.5 % (NRC 2000) and 2 % (Volek et al. 2003).
The pulps have a high content of fibers, soluble as insoluble, highly digestible (Habeeb et al. 2017). The NDF fraction showed differences between the two periods (P <0.001), being 16 % higher in the summer BP, which seems to be associated with a higher hemicellulose content (P <0.05). On the other hand, the ADF content did not show significant differences between the samples, with an average value of 22 %, lower than that obtained by other authors, which place them between 25.7 and 27.0 % (De Blas and Carabaño 1996, Volek et al. 2003 and De Blas et al. 2010).
Cellulose is an important fraction of the BP and the contents found in summer and winter are in the range of 13.1-27.5 %, reported by other authors (De Blas and Carabaño 1996, DePeters et al. 2000, and Pereira and González 2004).
The ADL content of the BP was similar at both seasons of the year. This was a very variable indicator. While De Blas et al. (2010) observed values close to 2 %, other studies refer variations from 2 to 6 % for samples from different sources (Pereira and González 2004), which may be related to the variety and the beet crop.
In correspondence with the differences found in the NDF content, the content in other complex carbohydrates more usable by the animals contained in the SFND fraction was higher (P <0.01) in the winter BP. On the other hand, the starch content was low in both seasons, although higher also in winter BP (P <0.01), which could be related to differences between the varieties of beet, or cultivation, or on the efficiency of the sugar extraction process between the factories, but that could also be due to the incorporation of molasses inside the BP before granulating, whose sugars (glucans) could be retained in this fraction during the analysis.
The BP contains little fat, and, therefore, the fatty acids are in proportions lower than 1 % of the dry matter, also presenting a high variability. The most representative fatty acids are linoleic, palmitic and oleic, with some differences between the collection time, in terms of palmitic acid, being its content higher in winter (P <0.05). The rest of fatty acids are at levels lower than 0.1 %.
The amino acid composition was similar between the BP of both groups. The most abundant amino acid was glutamic acid, followed by aspartic acid, leucine, lysine and valine, although all of them with very low values because the CP content of the BP was also very low. These values are in the range determined by De Blas et al. (2010) for lysine (0.54 %) and methionine (0.16 %).
In general terms, beet pulp is a by-product of high content of insoluble fibrous carbohydrates (cellulose and hemicelluloses) and carbohydrates soluble in neutral detergent, and with low lignin content, which makes it an excellent source of energy in those animals that have a good digestibility of this type of compounds (Cobos et al.1995 and Habeeb et al. 2017). In addition, the soluble fiber in neutral detergent is formed by soluble and insoluble pectins, soluble hemicelluloses (arabinoxylans and β-glucans), fructans and oligosaccharides (Martínez-Vallespín et al. 2011), which are given a very important function in the intestinal health of animals (Fishman et al. 2011), especially of monogastric animals (Carabaño et al. 2008 and Xiccato et al. 2011).
The analysis of the main components of the BP chemical composition shows that the main component consists of SFND, located on the left side of the graph and the insoluble fibrous components (NDF and ADF, on the right side, which explain 61.6 % of the variability (figure 1). On the other hand, the main component 2 shows that part of the variability could also be due to the ADF content and its lignification degree. These results highlight the importance of a detailed evaluation of the nature of the fibers contained in the BP.
It is concluded that beet pulp from the sugar industry in Spain, independent of the season, is a by- product that provides high fiber content, soluble as insoluble, although it is not an important source of BP, amino acids or fatty acids. However, it is necessary to characterize the chemical nature of these complex carbohydrates, since the pulps obtained in winter showed higher SFND content and lower in NDF.