Common mistakes using enzymes in poultry nutrition

ABOUT MATRICES OF FEED ENZYME PREPARATIONS (translated from Russian).
The feed contains known nutrients, the need for which has been established, so their concentration in feed is constantly monitored. However, they completely ignore the fact that they are present in the raw materials not in their pure form, but as part of complex structures. Scientific evidence points to the importance of the structure of food products, which significantly affects their nutritional properties as well as their health (Raeuber and Nikolaus, 1980; Aguilera, 2019). This is often associated with different productivity of animals consuming feed with the same content of controlled nutrients. The study of food structure has been developed in the field of raw material processing and human nutrition, but little attention is paid to its impact on animal productivity. Databases of computer programs used to develop compound feed recipes include information on the content of metabolic energy (ME) and digestible substances in individual ingredients. However, these databases do not contain data on the content of indigestible fractions of ingredients, although they are the basis for justifying the choice and use of exogenous enzymes. As a result, the content of indigestible substances, including substrates of exogenous enzymes, is not taken into account, which leads to their insufficiently justified choice. On average, 20-30% of substances are not digested in feed. With a successful selection of exogenous enzymes from this part, it is possible to increase the digestibility by no more than one third. Despite the progress in the creation of modern feed enzymes, about 70-80% of the undigested substances remain inaccessible for digestion. At the current level of scientific development, it is impossible to increase the hydrolysis of undigested substances even to 50%. The reasons for this inaccessibility remain unclear, and the explanations put forward are based on assumptions that have not been confirmed experimentally. When trying to account for indigestible compounds that would be available for the action of exogenous enzymes, limitations arise. Firstly, data on the digestibility coefficients of many substrates are not included in the traditional tables of feed composition. Secondly, many feed substrates cannot be determined in most laboratories. Thus, the fraction of non-starch polysaccharides (NPS) is not analyzed, the content of which varies greatly in feed ingredients (Choct and Annison, 1990). When studying the relationship between the nutritional value of food and its structural features, nutrition science uses the concept of the "food matrix" (Capuano et al., 2018; Zou et al., 2015). Attempts have been made to classify matrices characterizing food products based on macro- and microstructure and their physical properties, but clarity in this direction has not been achieved. In animal husbandry, interest in studying the structure of feed remains low. The effect of any biologically active substance can be expressed in a matrix that reflects the magnitude of the impact on the use of energy, protein (amino acids), and minerals, but this approach has not been widely used. At the beginning of the 80s, during the active spread of enzyme preparations, the concept of "matrix" was not used, although their effect was reflected in tables reflecting the impact on productivity, that is, in matrices.
Among specialists involved in animal feeding, the concept of matrix began to be used by enzyme suppliers, who introduced it into the environment of diet developers. Subsequently, the market of enzyme preparations turned into a "struggle of matrix values and numbers" (??????????, 2014). The matrices are set using various methods. Data obtained from the study of the effect of enzymes on the digestibility of nutrients is more often used, and less often on productivity. Each method has its own limitations, so if only data from the study of digestibility were used to develop the matrix, then the effect of additional absorbed substances on metabolism remains unaccounted for. The enzymes are active against specific available substrates. However, the instructions for the use of enzymes indicate only the type of diet, without carrying out the concentration of substrates against which the effect of enzymes will be manifested. This leads to unstable results of their effectiveness, since their effect depends on the concentration of the substrate, its accessibility to the enzyme, and the dose of the latter. Feed enzymes improve the digestibility of nutrients directly or by destroying those substances that reduce the digestion of nutrients, or by both methods (?????? ? ??????. 2021). The effectiveness of exogenous enzymes depends on the components and the initial nutritional value of the feed, respectively, these factors affect the magnitude of the digital values of the matrices. It is difficult to predict the effectiveness of enzymes in the rearing of young animals due to the active formation of the digestive system at an early age. Broiler farming enzyme suppliers usually suggest using a single matrix, which is recommended for a specific type of diet, regardless of age. However, even during the short life of broilers, the availability of feed nutrients increases significantly, and the return on exogenous enzymes will decrease accordingly. Thus, it was found that the availability of ME from wheat in 7-day-old chickens was 2,637 kcal/kg, at 21 days of age – 2,748 kcal/kg, at 35 days of age - 2,933 kcal/kg (Bedford, 1996). That is, the ME of wheat in the starter is 11% lower than when it is fed as part of the finisher. The value is quite significant, considering that under the influence of exogenous enzymes, the maximum achieved increase in total metabolic energy rarely exceeds 120 kcal/kg. Reference books usually cite the values of ME and available amino acids established in experiments on the digestibility of nutrients in chickens of a certain age. A similar approach is used to determine the matrix values of enzyme action. The omission of age-related digestive features reduces the reliability of nutritional values of developed diets, and on the other hand compromises the effectiveness of enzymes and their characterizing matrices (Olukosi et al., 2007; Babatunde et al., 2019). Using the example of phytase, it was shown that when it was included in poultry feed at doses of 125-2000 units/kg, the increase in phosphorus availability in the studied components varied significantly (Almeida et al., 2017). The analysis of the presented data allows us to note that with a higher initial digestibility of phosphorus in wheat compared with corn, the increase in phosphorus availability under the action of the maximum dose of phytase was higher in corn. The digestibility of phosphorus in wheat practically reached a maximum at a dose of 500 units/kg of phytase, whereas in corn it reached a maximum at a dose of 2000 units/kg. With the same initial digestibility of meal phosphorus, its maximum availability from soybean meal was noted when 250 units/kg of phytase were included in the diet, while from sunflower and canola meal it was at a dose of 2000 units/kg. It follows that in practical conditions, the dose of phytase that ensures maximum availability of phosphorus from the diet depends on its composition. It is important to note that the maximum dose of phytase reduced the availability of phosphorus from soybean meal compared to a dose of 250 units/kg. Enzyme suppliers recommend applying matrix values of enzymes to the entire diet, that is, a mixture of components – this facilitates calculations when optimizing diets. The data in Table 1 show that different raw material components are not equally susceptible to the action of the enzyme and its different doses, that is, phytase matrices in relation to different components should be expressed in different values.
Table 1.
The effect of different doses of phytase on the standardized digestibility
of phosphorus in the gastrointestinal tract from various types of raw materials, %.

Therefore, for more reliable calculations, the effect of enzymes on individual components should be taken into account and the data obtained should be summarized, but the summation method has not been developed.
Usually, suppliers of enzymes for a specific product offer a matrix with constant values regardless of age and differences in the composition of the starter and finisher. The world's leading experts in the field of studying the effects of feed enzymes indicate that enzyme suppliers should provide clear and transparent recommendations on the use of enzymes. (Cowieson, 2010). So, if the feed contains 1.4% of total lysine, of which 1.25% is digestible, then the content of undigested lysine will be 0.15%, and the effect of an exogenous enzyme on this part can give an increase in digestible lysine. Additional hydrolysis of substrates by exogenous enzymes depends on the content of substrates that pass through the gastrointestinal tract, remaining undigested.
In practical terms, diets can include 2-3 or more enzymes, the effect of each of them is reflected by its own matrix, however, the matrix of the final action of the enzymes remains unknown. To achieve maximum efficiency, the enzymes included in the feed must have different mechanisms of action aimed at splitting different substrates, in addition, it is necessary to make sure that their matrices are additive. Phytases are used to make more efficient use of phosphorus from plant feeds, while xylanases are used to increase energy availability. However, energy is not a substance, and its change (increase) is associated with an increase in the digestibility and absorption of carbohydrates, protein and fat, which release the energy contained in them during metabolism. When phytase and xylanase are used simultaneously, the availability of feed amino acids increases, although they do not have a proteolytic effect. The alleged mechanisms of their action in this direction have not been studied. However, when characterizing these enzymes, their matrices include an increase in the use of amino acids. In diet optimization programs, the feed database includes matrices of a number of enzymes, taking into account their effect on the complete diet independently of each other, that is, without taking into account their interaction. Experimentally, it was found that the digestibility of individual amino acids under the influence of phytase and xylanase when used together was lower than the sum of the values established by their individual use (Cowieson and Bedford, 2009). Phytase increased the digestibility of amino acids more actively than xylanase. The actual digestibility of the studied amino acids under the influence of enzymes did not change in proportion to their content in the diet. In addition, if phytase increased the availability of methionine, then its use in conjunction with xylanase inhibited this process. Xylanase reduced the effect of phytase on the availability of tryptophan and aspartic acid. This fact cannot be explained: is it due to the inhibitory effect of xylanase or a simultaneous decrease in the action of each enzyme. Nevertheless, the matrices of action of each enzyme separately contain data on increasing the digestibility of protein, which extends to all amino acids, which leads to misconceptions. It should be noted that this changes the ratio of digested amino acids, which balance the diet. This problem has not received sufficient coverage in the scientific literature, which limits the acceptance of an assessment of the action of enzymes based on matrices. The increase in the digestibility of feed nutrients under the influence of exogenous enzymes depends on their initial digestibility under the influence of their own enzymes. This is due to the fact that the added enzymes will act on the indigestible fraction. The higher it is, the higher the expected effect of the added enzymes. Summarizing the results of 19 studies showed that under the influence of xylanase, the availability of the undigested protein fraction increases by 16%. It follows from this that if the digestibility of the protein was 70%, then under the action of exogenous enzymes it will increase by 4,8% (30%×16%/100%), reaching 74.8%. However, if the digestibility before the inclusion of enzymes was 90%, then the digestibility will increase by only 1.6%, amounting to 91.6%. It follows from this that the use of an enzyme with the same matrix values against a background of diets with different digestibility will be accompanied by different effectiveness.
Determining the matrices of polyenzyme preparations is a difficult task to solve, since it is necessary to calculate how much additional nutrient intake provided by one enzyme is complemented by the use of a second and third enzymes. Recall that enzymes are added to feed to digest that part of the nutrients that are not digested by their own enzymes. For example, in feed containing 3733 kcal/kg of gross energy, the ME content is 75% or 2800 kcal/ kg, while the share of indigestible energy will be 933 kcal/kg. The matrix of the 1st enzyme indicates that its inclusion in the feed will increase the availability of ME energy by 110 kcal/kg (or 3.9%). The 2nd matrix indicates an increase in ME of 95 kcal/kg (3.4%), the 3rd matrix indicates an increase in ME of 72 kcal/kg (2.6%) and the 4th – 120 kcal/kg (4.3%). These values refer to the case when any one of the four enzymes is added to the feed. If we sum up the action of 4 enzymes, then the caloric content of the feed should increase to 110+95+72+120 = 397 kcal/kg. However, this calculation is not correct, since each enzyme acts on the remaining part of the indigestible energy. After the action of the 1st enzyme, 933-110= 823 kcal/kg will remain – 823×95/933 = 83.8 kcal/kg will be digested from it under the action of the second enzyme, as a result of indigestible energy, 823-83.8 = 739.2 kcal will remain. The 3rd enzyme will act on this part. Under its influence, 739.2×72/933 = 57 kcal/kg will be additionally digested. As a result of the action of the third enzyme, the amount of undigested energy will remain 739.2-57 = 682.2 kcal/kg. 87.7 kcal/kg will be digested under the action of the 4th enzyme. Under the action of four enzymes, the feed can be additionally digested, which will provide 110+83,8+57+87,7 = 338.5 kcal/kg. The above calculation is very simplified and represents a scheme, since in the body the added enzymes do not act in turn, but simultaneously and in different combinations, depending on the environment in which their action is manifested. Other little-known factors of interaction between enzymes are not taken into account. The purpose of this example is to prove that the effect of each enzyme individually is higher than in the presence of other enzymes. In the above calculation, it is conventionally assumed that each subsequent enzyme acts on the initial undigested part, however, the first enzyme will cleave the most accessible part of the substances, and less accessible substrates will remain for the second and subsequent enzymes. How will their effectiveness change? In scientific terms, the question has not yet been answered. Apparently, in multi-enzyme preparations, the most effective combination would be the simultaneous use of carbohydrase and phytase.
The corn-soy diet of the positive control group was balanced in terms of nutritional content in accordance with the recommendations. Feed for chickens of negative control contained less nutritious components, so the level of ME and available phosphorus was lowered, which led to a significant decrease in chick growth. The inclusion of carbohydrates and proteases in the feed did not change the digestibility of protein, phosphorus and ME, but increased the live weight and efficiency of feed use compared with the results of negative control. However, only one phytase also increased the growth of chickens, as did carbohydraz+protease, although the level of ME in the feed decreased. The simultaneous inclusion of carbohydrates, proteases and phytases in the feed did not affect the ME of the feed, compared with the negative control, but the maximum growth of chickens and a decrease in feed consumption for growth were noted (Cowieson and Adeola, 2005). The data presented indicate that the indicators of the effect of various enzymes on the digestibility of nutrients are not always consistent with their effect on the growth and efficiency of feed use. The effect of exogenous enzymes in the body is not limited to their effect on the digestibility of nutrients, therefore, substantiating the effectiveness of enzymes with different activities in one preparation is a difficult task. In the absence of results confirming the need for each component in the composition of the multi-enzyme complex, there is no evidence confirming their need in the composition of the drug (Masey O'Neill et al., 2014), that is, recommendations for the simultaneous use of several enzymes should be accompanied by experimental confirmation. To justify the effectiveness of the use of enzymes, suppliers must provide clear recommendations on the use of the products offered. The enzyme matrix, proposed without justification, will look like a table of random numbers. It is necessary to refer to the test reports in which the matrix values of the enzyme were established. Research should reflect not only the most successful results, but also the less successful ones, with an explanation of the reasons for their discrepancy. In addition, strategies should be outlined to ensure that the use of enzymes does not cause undesirable effects (Bedford and Cowieson, 2020). Well-chosen enzymes increase the digestibility of nutrients (Cowieson et al., 2017; Wu et al., 2019; Ravindran et al., 2017), the results of which are used to calculate the matrix values of enzymes. However, data on nutrient digestibility is a secondary metabolic indicator of a possible increase in productivity (Bedford, 2008). Due to the fact that the effectiveness of the use of enzymes in practical conditions is influenced by many factors, some of which experts are aware of, but cannot control or control them, therefore, the decision to use feed enzymes must be confirmed in specific production tests. In conclusion, we note that on the information portal of the Federal Service for Veterinary and Phytosanitary Surveillance (Rosselkhoznadzor) "Irena" (https://galen . vetrf.ru ), where the characteristics of enzyme preparations approved for use are posted, none of the instructions contain their matrix values, which means that the matrices are not validated because they do not meet the requirements of the regulator.

Dear Piotr, thank you very much for your point of view. It helps me use enzymes correctly, depending on the available substrates contained in the feed. I think that attention should also be paid to the methods of feed enzymes application, which have historically been used in different countries, this is 1) "over the top" of complete feed (for maximum absorption of nutrients) and also 2) "down specification" to reduce the cost of feed. The risk factor from using of the matrix values in the first case will be lower than in the second. The desire of some manufacturers to achieve maximum enzyme matrix values increases the degree of commercial risk. A good methodical technique for determining the real effect of enzymes is model experiments on the effect of enzymes on grain mixtures. I conduct such experiments in the laboratory of our enterprise. In this way, we can compare the effectiveness of various enzymes on various feed rations.