Urease activity: Delta pH Method vs European Method determined on soybean meal samples and correlations with trypsin inhibitors

Dear Dr. Wiseman, you are absolutely right in that urease activity does not measure overprocessing of soybean meal. It was never its objective. Caskey and Knapp [Method for determining inadequately heated soybean meal. Ind. Eng. Chem. Anal. Ed. 16: 640-641 (1944)] developed the technique to determine underprocessing of soybean meal fed to ruminants which at the same time were fed urea in the feed. Obviously, high levels of urease will generate excess ammonia and subsequent toxicity. That was the initial objective of the urease activity test. But later, Bird et al. [Urease activity and other chemical criteria as indicators of inadequate heating of soybean oil meal. J. Asso. Official Agr. Chem. 30:354-364 (1947)] demonstrated that the very same technique was adequate to establish if SOYBEAN MEAL fed to poultry was underheated or adequately processed. While the Caskey and Knapp (1944) paper established a DIRECT relationship between urease and adequate heating in the presence of urea, the Bird at al. (1947) paper established an INDIRECT relationship between antinutritional factors (ANF) in SOYBEAN MEAL, urease and heating in the absence of urea since we don't feed urea to chickens. Let's remember that that was in the 1940s when ANF such as trypsin inhibitors were just being discovered and no analytical methodology was available (the methods by Kakade et al. were not published until 1969-1974). Consequently, the urease activity test became "the test" for SOYBEAN MEAL quality since the end of the 1940s. Although overprocessing of soybean meal was already an issue at the time (in fact, Caskey and Knapp explicitly expressed in their paper that the urease activity test was not useful to test for overprocessing) there was no in vitro method available to detect it. Again, you are right, urease activity is not useful to detect overprocessing, but it is useful to determine UNDERPROCESSING of SOYBEAN MEAL because there is a high statistical correlation between urease activity and trypsin inhibitors in soybean meal particularly soybean meal obtained by the solvent-extraction process (yes, lectins are also part of the heat-labile ANF, but once trypsin inhibitors are significantly reduced, lectins are almost gone). Such a correlation (urease activity measured by the pH-rise method and trypsin inhibitors measured in mg/g in SOYBEAN MEAL) is well documented (see Ruiz, 2012 Arkansas Nutrition Conference Proceedings review) and given the fact that it is more expensive and time consuming to measure trypsin inhibitors as a routine quality control procedure, the urease activity test is a practical shortcut. The reason Ruiz et al. presented a poster at the 2019 European Symposium on Poultry Nutrition was to document that even though some European laboratories measure urease activity by a method DIFFERENT than the pH-rise method [AOCS Official Method Ba 9-58, AOCS (2011) which is based on Caskey and Knapp (1944)] the correlation with trypsin inhibitors is still valid also with the European Method [Détermination de l’activité uréasique de produits dérivés du soja. Journal Officiel des Communautés Européennes N° L 155/36 (1971)] for SOYBEAN MEAL. Finally, the reason I have capitalized SOYBEAN MEAL throughout this discussion is because the correlation between urease activity and trypsin inhibitors does not apply to full-fat soybeans. But that is a subject for a separate discussion!

Mr. Aslam, the discussion on urease activity and its correlation with trypsin inhibitors applies only to SOYBEAN MEAL. Because full-fat soybeans (FFSB) are generated through different methods such as roasted soybean seeds, it is not possible to establish a general correlation across the different methods. In the case of SOYBEAN MEAL which is generated worldwide mostly by the solvent extraction method, there is a general high correlation between urease activity and trypsin inhibitors. The solvent extraction method is quite standard with two key steps (flaking and desolventizer-toaster) that make the exposure of both trypsin inhibitors and urease very similar in different processing plants. In contrast, FFSB processes are quite different with very different designs, therefore the exposure of trypsin inhibitors and urease in the beans is not the same from one FFSB process to another which has a different design. Even for a given method, let's say wet-extrusion, there are different designs for wet-extrusion with different retention times, with or without pre-conditioning etc., etc. so, the exposure of trypsin inhibitors and urease to heat is different in different extruders. Now, that said, for a GIVEN SPECIFIC MACHINE, your roasted soybean processor for instance, in which you always do the same process, at the same temperature, at the same retention time, lot after lot, yes, a relationship between urease activity and trypsin inhibitors may be established, but is valid for that SPECIFIC equipment only! If your neighbor has similar equipment BUT works with different retention time, and different temperature most likely your correlation is not going to apply to your neighbor's. And finally, the reference is TRYPSIN INHIBITORS. If you use phenol red (a semi-quantitative urease quick measurement) and decide that once you don't detect urease your FFSB is ready, is OK to be formulated in a poultry feed, not necessarily. You also need to measure trypsin inhibitors and need to be below 2 mg/g of FFSB (the lower the better!) before you can assert that your FFSB is properly done. Trypsin inhibitors command the process, not urease activity.

For those interested in the differences between soybean meal and full fat soy that Dr. Ruiz mentions, you can see here his explanation: https://en.engormix.com/MA-feed-machinery/videos/important-differences-between-soybean-meal-full-fat-soy-t55221.htm