top of page


On a global basis, the use of proteases, NSPase and phytases saves the animal feed industry more than $8 billions per annum in nutritional input costs and contributes significantly to environmental sustainability for example in resolving planetary boundary pressures on phosphorus and nitrogen emissions. Together with their enzymes, manufacturers provide matrices of nutritional contribution to capture in the formulation process the value created by these enzymes.

Nowadays, probiotics, emulsifiers and even organic acids suppliers are now coming as well with matrices to help to optimize further feed costs.

However, as nutritionists increasingly require precision in formulation systems, it is imperative that manufacturers respond with clear, transparent guidelines on the way they develop their matrices, how variable these are, and factors that may contribute to this variance must be stated clearly.

Enzymes need the proper substrate to act. Phytases work on phytates, xylanase on arabinoxylans (constituent of hemicellulose), Glucanase on glucans (constituent of cellulose), pectinase on pectin, etc… Therefore, when using an enzyme, it is critical to check first that the diet contains the relevant substrates for the enzyme to provide the value mentioned in the matrix. If the substrates are insufficient, there is a risk that the enzyme will not provide sufficient nutrient to balance the nutritional requirement set by the nutritionist.

For phytases, we need to ensure the presence of phytic acids (or phytates). The most concentrated sources of phytates tend to be whole grains and beans. They are isolated in the aleurone layer in most grains, making it more concentrated in the brans.

Regarding NSP enzymes, soybean is rich in glucan and arabinoxylan and will therefore support the use of glucanase and xylanase. Rice bran and Corn bring significant amount of xylan to the diet and would justify the addition of xylanase.

Proteases will give the best return for ingredients with low protein digestibility as they provide sufficient long polypeptides for the enzymes to act. When using proteases on highly digestible ingredients as fish meal, we may not provide sufficient undigested protein for the enzyme to release the amino-acids claimed in the matrix. That is why, when using protease, we normally recommend nutritionist to provide some ingredients with low protein digestibility to ensure that the matrix contribution will be secured. Same conclusion applies with the percentage of protein. We will get a better value from protease when using them on high crude protein diets (carnivorous fish, broilers and piglets’ diets) rather than lower protein diets. The more substrate we offer to the enzymes to work, the safer will be the matrix value. The Mineral content

Many academical and field studies demonstrated the correlation between high level of calcium, zinc, copper and even iron and a reduction of phytase effect. Indeed, these minerals tend to complex with phytic acid in the small intestine and forming an insoluble complex or to bind to the active site of phytase enzyme. Once combined with these complexes, these minerals block the phytase from accessing the phytate substrate resulting thus in a reduction of its efficacy and a suboptimal supply of phosphorous. Therefore, one of the preconditions to the use of the phytase matrix is to control the level of calcium, zinc and copper in the diet. This is particularly important in the piglet and broiler diets where we often use high level of Zinc Oxide and Copper sulfate as growth promoter. For that reason, suppliers normally recommend to increase dosage of phytase and use rather a superdose when adding high level of Zinc Oxide and copper in the diet.

But instead of superdosing phytase, another strategy is actually to work first at reducing the level of Zinc Oxide and Copper Sulfate in the recipe. There is today on the market some low dosage Zinc Oxide and Copper source that can be used to maintain the propre growth promoting effect while securing the efficacy of the phytase without superdosing. The Performance level of diets The easiest way to confirm the efficacy of an additive is to add it to a control diet and compare the difference of performance. But that approach relies on the assumption that the control diet is not balanced in term of nutrient supplied to the animal. If all nutrients are already in excess, the additional nutrients provided by the additives will not be converted by the animal into performance.

To confirm the efficacy of an additive, we must first create a deficiency in one or several nutrients. The standard protocol could be to have one positive control diet, one negative control diet where we would reduce one or several nutrient and then the treatment group where we will add the additive on the top of the negative control group. The most accurate approach would actually be to incorporate the matrix of the additive into the formulation software, add some new ingredients to be targeted by this additive and ask the software to propose reformulation. The first learning we get from this strategy is that we will realize the additive is not economically sound for all diets. As mentioned above, the system will call for the protease only with the crude protein level exceed a certain level or for NSPase only when energy is required.

Once we get the adjusted recipe at an optimized cost, we can then validate it through performance trial in comparison with the control diet. We are looking at confirming that both diets offer the same performance and therefore that the value in the matrix can be validated as a strategy to reduce feed cost.

The Matrix development background It is quite easy to provide a matrix but it is actually very difficult to develop a sound and reliable one. It is therefore necessary to ask suppliers to provide with each matrix, the development background behind the numbers provided.

To develop a valid matrix, it takes many in-vitro, ex-vivo and in-vivo trials. The in-vitro and ex-vivo studies enable to establish the activity of the additives and issue the matrix value. Then, it is important to confirm and validate these values through performance trials in different situations as underlined in the paragraphs above. The suppliers need to confirm the details and the number of trials they use to validate their matrices and most importantly the context of the studies (mineral levels for phytase, level of substrate, protein level and nature of ingredients for protease, etc…) Synergism, additivity, and antagonism A common concern that nutritionists express when it comes to the use of matrices is the extent to which the nutrient delivery by one additive is synergistic, additive or antagonistic with the nutrient delivery of a second (or third) one. This is a very reasonable question but does not have an easy answer.

When comparing matrices from additives, we easily note that most of them supply energy and it is wise to ask if when using several additives in the formulation, each of them will be able to supply the quantity of energy promised without any cannibalization between them.

The answer starts from the suppliers who should confirm whether the matrix that they provide for their product has been developed together with the use of other additives in their experimental diets. In many parts of the US, SE Asia, or Latin America, phytase was the first enzyme to be used, and the NSPase and proteases needed to demonstrate sufficient incremental value generation in diets that contained appreciable phytase doses. However, in parts of Canada, Australia, New Zealand, and much of Europe, NSPase were there first. Phytase was the new comer and it has to prove its matrices in addition of xylanase or glucanase. As mono-component protease market has explicitly emerged in the past 10 years or so, there is now a reasonable expectation that the effect of this enzyme can be demonstrated in diets that contain both xylanase and phytase. Some probiotics are demonstrating as well some enzymatic effects that have been proved on the top of both endogenous and exogenous enzymes. Phytase does increase energy availability by freeing myoinositol from phytic acid. The energy coming from the use of protease has a different origin. It is either coming from the degradation of amino-acids or from the release of carbohydrate that were previously bound by proteins. NSPase contribute as well to the energy supply but through a different mechanism. Indeed, NSPase are solving the “cage effect” by creating some ‘holes’ in the vegetal cells and enable the other enzymes to access the starch that was initially locked inside these cells.

When considering these three different modes of action, we can easily imagine that these sources of energy are additives without cannibalizing each other and therefore the matrices of proteases, phytase and NSPase can be summed up.

But as mentioned above, it is safer to go through a validation process in your own diet before used matrices in your production diets. The matrices may need to be adjusted according the substrate used, the animal age, the level of minerals but most importantly, I recommend you to have a transparent discussion with your suppliers to understand the context that leads to the development of these matrices. It could be risky to use a matrix without a correct understanding of its background. Ideally, a matrix should be designed specifically for each diet.


bottom of page