Carbohydrates are a primary source of energy in the animal feed, which account for 40 to 80% dry matter. They are commonly divided into two major groups, namely, nonstructural carbohydrates (NSC) and the cell wall or structural carbohydrates. In the context of their transformation in the body of animals, the first group is composed of the mono-, di-, and oligosaccharides; low molecular weight fructosides; and starch, which are easily and almost fully digestible and, therefore, are referred to as readily hydrolyzed or easily digested.

The structural carbohydrates, along with lignin, compose the cell walls in plants; these include polymers (pectins, hemicelluloses, and cellulose), differing in degree of polymerization, type and shape of their monosaccharide components, bond types, etc. They are primarily of interest as a source of energy for ruminants, which are capable of partially digesting the structural carbohydrates due to their gut microbiota. Additionally, components of the cell wall are a main factor that governs digestibility and intake of the feed (forage) as well as the use of energy from digested substances by animals. Consideration should be also given to a physiological role of the structural carbohydrates, which ensures normal functioning of the rumen (rumination) and motor function of the digestive system.

Structural carbohydrates, together with lignin, commonly fall under the umbrella term fiber or roughage. In this context, roughage does not mean “cellulose” but rather serves as a term adopted in the assessment of animal feedstuffs based on a composition the structural components. It can be defined as a fraction of the forage, which is partially and slowly digestible, not fully available, and fills the major part of the digestive tract volume in ruminants.

Evaluation of the animal feedstuffs based on low digestible and slow carbohydrates and lignin was first performed using the Weende scheme by crude fiber (CF) content. As an indicator of digestibility and energy values of the animal feedstuffs, CF has been playing a critical role in forage quality research. It does not, however, represent a sum of structural or nondigestible substances (diet residues) and does not provide for a typification of the cell walls. The CF composition only includes cellulose, the variable part of lignin, and an insignificant portion of another structural carbohydrate, specifically, hemicellulose (HC). Thus, in forage plants, crude fiber accounts for as little as 19–33% of the composing lignin and 5–13% of HC [1]. At the same time, exact prediction of the intake and digestibility requires data on individual types of the foodstuff’s (forage) structural components. Additionally, the time- and labor-consuming partitioning into fractions or fractionation by the chemical composition, as well as the division based on the digestibility and nutritional values, are essential to determine a complete carbohydrate composition of the feed. Carbohydrate fractions have been commonly determined from one weighed portion (sample) of the feed using the successive acid hydrolysis. This technique, modified to various degrees, was used to intensively study the carbohydrate composition, as well as nutritional value of forage plants, feedstuffs, and rations, in the 1950s–1960s. The modifications include determining the carbohydrate composition of the plants according to Kizel, which has been employed in our country to date [2, 3].

Successive acid hydrolysis, however, is a technique that is too time- and labor consuming to be applied to the quality evaluation of feedstuffs in practice. For a solution of this problem, Van Soest proposed a scheme in the 1960s to analyze structural carbohydrates and lignin in the feedstuffs using detergent solutions. According to this scheme, neutral detergent dissolves the content of a cell while preserving the cells walls composed of hemicellulose, cellulose, and lignin, which are referred to as the neutral detergent fiber (NDF). Acid detergent removes the HC; the residue appears as a sum of cellulose (C) and lignin (L) or the acid detergent fiber (ADF). The ADF is used to determine the lignin (ADL) by means of its hydrolysis with a 72% sulfuric acid. The HC and C are calculated from the difference: HC = NDF – ADF; C = ADF – ADL.

Findings from the comparison between carbohydrates and lignin in the original forage and as components in the NDF and ADF compositions using the successive acid hydrolysis suggest that the detergent-based methods are not always specific for the structural carbohydrates and lignin. Notwithstanding, determining the NDF, ADF, and ADL levels and H and C from them has globally become a widespread C practice in the forage quality research.

Nationally, the successive acid hydrolysis is still largely in use since the detergent-based analysis has been gaining currency in the forage quality research only recently. The literature, however, offers a great amount of data obtained using the successive acid hydrolysis. It is essential to understand comparability of the results of these two methods for data generalization, comparison, and use in the ration formulation.

The goal of the study is to establish compatibility of the results of the evaluation of the structural carbohydrates and lignin obtained using these two different methods.

MATERIALS AND METHODS

The study used forage samples cultivated on sod-podzolic soil of the Central Experimental Station at the Williams Federal Research Center of Forage Production and Agroecology. The carbohydrate fractions were evaluated using the detergent analyses by the Van Soest scheme and the successive acid hydrolysis.

Several plots were arranged for sampling on the cross of the site. The samples were collected from the first crop in grasses (Bromus inermis, Festuca pratensis, and Phleum pretense) at the phases of boot, heading, and start of anthesis and legumes (Tifolium pratense and Medicago sativa) at the branching (tillering), budding, and start of anthesis. The cut herbage was dried at 60–65°С in a forced draft oven and grained to fit a screen with a 1 mm mesh size.

During the detergent analysis (method one), the NDF and ADF content was determined from individual weighed portions by boiling for 1 h in the neutral detergent solution (without amylase) or acid detergent, respectively. The HC content was calculated as the difference between NDF and ADF; C was calculated as the difference between ADF and ADL. The results were expressed as percentage of dry matter. The ADL was determined using the ADF residue, which was exposed to hydrolysis with a 72% sulfuric acid solution.

Successive acid hydrolysis (method two) consisted of the following stages:

(1) Removal from samples of water-soluble carbohydrates using the water extraction at 60°С for 2 h;

(2) Extraction of HC from the residue with a 2% sulfuric acid solution for 4 h on a boiling water bath;

(3) Suspending the residue in a 72% sulfuric acid solution (m/m) for 3 h at 20–23°С;

(4) Extraction of C by boiling the sample residues in a dilute sulfuric acid solution for 1 h;

(5) Rinsing the residue after the C extraction; subsequent drying and weighing for the lignin recovery.

After the neutralization, the HC and C hydrolysates were analyzed for a reducing sugar content using the permanganate Bertrand method. The obtained results were multiplied by a coefficient 0.9 and expressed as percentage of dry matter.

The NDF and ADF values were compared with a sum of C, HC, and L determined using the successive hydrolysis method and a sum of C and L, respectively.

RESULTS AND DISCUSSION

Using method 1, the cell wall content in grasses was by 6.3% higher than the sum (HC+C+L) determined using the successive acid hydrolysis (Table 1).

Table 1. Levels of the structural carbohydrates and lignin determined using different methods, % dry matter
Full size table

The difference between NDF and the sum (HC+C+L) was significant based on t0.95 confidence. A rather close positive correlation, however, was noted between the results of the two methods (Table 2). Overestimation in method one compared to method two appears to arise from crude protein and ash in the NDF composition, which are not factored in the successive hydrolysis. Protein content insoluble in the neutral NDICP detergent averaged 4.4% for nine samples; the ash was 1.5%. Subtracting from NDF the composing ash and NDICP produces a result close to a sum of C, HC, and L.

Table 2. Statistical parameters in the assessment of differences between the results obtained using the methods
Full size table

To the contrary, the ADF by method one is lower than the sum (C+L), which appears to be due to it comprising as little as 88.8% of C of the original herbage [4] and a lower ADL value compared with L. The ADICP is likewise present in ADF but in much smaller amounts than in NDF. Based on the results of our study, ADICP content averaged 1.2% for nine samples [5]. Using method 2, the C content from polysaccharides of the cell walls was insignificantly higher than based on the ADF–ADL difference (see Tables 1, 2). The difference, however, between the two methods proved to be insignificant.

As opposed to C, the content of the second polysaccharide of the cell walls, namely HC, was significantly higher using method one than by the successive acid hydrolysis (see Table 1). Additionally, the correlation coefficient noted for values of these parameters was lower than for others (see Table 2). Since the HC in method one appears as the difference between NDF and ADF, this situation can be partly attributed to an increased NDF and lower ADF content. There is also a possibility for elimination of part of the HC, e.g., uronic acids, during the water extraction from the sample employed in our experiment using method two before the HC extraction, which could have led to underestimation of this parameter value.

When using method 1 to determine HC, the result also includes the difference between the protein and ash insoluble in the neutral and acid detergents. Their contents are much higher in NDF than in ADF. The HC is a polysaccharide, which does not have protein and ash in its composition; therefore, it would be more accurate to determine its content based on the difference between NDF and ADF after subtracting from values of the latter the insoluble protein and ash contents. Other studies likewise report higher HC levels in grasses when determined based on the difference between NDF and ADF. This appears to be a reason why the HCs calculated from the difference between NDF and ADF were defined as “crude “and the following equation was developed to link the HC content determined using an independent (noncompartmental) technique (y) to the data obtained using the difference between NDF and ADF (x) [4]:

$$y = 0.98 + 0.73x.$$

Recomputation of the HC content using this equation produced the difference of as low as 1.8% between the methods for our data.

Legumes differ from grasses in lower HC and higher L levels as well as the occurrence of pectins. These peculiarities appear to have influenced the results of the considered methods.

In contrast to grasses, NDF values in legume herbage insignificantly differ from the sum (C+HC+L) obtained using the successive acid hydrolysis (see Table 1). It is associated with occurrence in the NDF composition of, e.g., alfalfa with as low as 83.7% of the C in the original herbage, while this parameter amounts to 96.3% in grasses. The same applies to protein, that is, 12.4 in NDF of legumes and 31.6% in NDF of grasses of protein in the original green matter [4]. Occurrence of pectins in a composition of the cell walls in legumes did not result in an increased NDF content compared to a sum of the HC, C, and L, since neutral detergent dissolves pectins and, thus, they do not become a part of the NDF composition while being contained in the cell walls.

Additionally, the legumes exhibited an insignificant difference between ADF level and the sum (C+L) determined by the successive acid hydrolysis. Acid detergent fails to eliminate the pectins composing the cell walls, but differences between the results of these two methods are insignificant due to the fact that only 88.7% of the C of the initial matter is preserved in the ADF composition.

When determined by calculating the difference between ADF and ADL, value of the C is higher (though not appreciably) than using the successive acid hydrolysis. This appears to be related to a lower ADL level compared to the L.

Data on mean HC content produced by different methods differed insignificantly (see t-confidence, Table 2). High error for difference between the means, however, was responsible for a low correlation coefficient between them.

Therefore, techniques for determining the structural carbohydrates (HC, C, and L) after Van Soest are not always specific of polysaccharides and lignin of the cell walls. But the physical and biological properties of the carbohydrates fraction, which affect the animal nutrition, are more important than their composition. A more advantageous method is the one that allows fractionation of the forage carbohydrates, which differ in degree of fermentation and digestibility and have an effect on their productivity. At the same time, the method is expected to be affordable in terms of time and labor intensity to be used in the feeding practice. According to stages of the analyses processes, the Van Soest techniques are more preferable; therefore, researchers have been focusing in recent decades on their various aspects as well as the potential to employ the obtained results in forage quality evaluation and utilization in the body of animals. This particularly applies to NDF, representing the total sum of all structural carbohydrates and lignin, which is among the critical indicators of the forage quality. The studies are concerned with the NDF levels in different forage items, its digestibility, the impact of various factors, and levels of its intake in cattle stocks [6–8].

Special attention is paid not only to the NDF intake levels in the ration but also to physically effective NDF, which is related to the forage particle sizes [9, 10]. This governs the forage intake, metabolism processes, and, ultimately, productivity of the animals.

A system of the NDF controlled by modeling has been proposed to integrate the forage physical and chemical parameters, since multiple other ration factors influence the processes occurring in the rumen in addition to NDF level and forage particle size [11].

Part of the NDF cannot be altogether digested by ruminants; after prolonged fermentation, content of the undigested NDF (uNDF) correlate with the nutritional value of the forage. Therefore, a method has been designed to determine this NDF in the routine analyses [12].

The proportion of lignin in NDF is used as an indicator of its digestibility, while more comprehensively explaining variations in the NDF digestibility and its relation with carbohydrates of the cell walls [13]. The type of lignin, depending on the technique for its determination, is also important. Thus, ADL correlated with the NDF digestibility more closely than Klason lignin, whose content was higher than the ADL [14]. The authors consider Klason lignin to be characterized by heterogeneity of the composition. Our experiment revealed higher lignin content using method two than the ADL, which appears to be additionally associated with the occurrence of different fractions in it.

Importantly, new studies emerged after determining the cell walls using neutral detergent had received a widespread use. Specifically, they determine carbohydrate fractions soluble in neutral detergent that represent a sum of the nonstructural carbohydrates and compounds contained in the cell walls. The studies focused on their composition and role in the animal diets.

Therefore, the content of the cell walls in grasses according to Van Soest (NDF), which is a sum of the hemicellulose, cellulose, and lignin, is significantly higher (by 6.89%) than the sum of the same compounds determined using successive acid hydrolysis. Subtraction from NDF of the composing insoluble protein and ash (4.4 and 1.5%, respectively), which do not belong to carbohydrates, produce the results closely similar between the methods.

Levels of ADF, cellulose, and lignin in grasses do not significantly differ between the methods. The mount of hemicellulose determined based on the difference between NDF and ADF is significantly higher than with the fractionation using the successive acid hydrolysis.

For legumes, the compared methods did not differ in the levels of NDF, ADF, cellulose, and hemicellulose. Concentration of the lignin based on the Kizel method, however, was significantly higher than with the Van Soest method.