*Dr. William A. Dudley-Cash is a poultry and fish nutritionist and has his own consulting firm in Kamuela, Hawaii. To expedite answers to questions concerning this article, please direct inquiries to Feedstuffs Bottom Line of Nutrition, 7900 International Drive, Suite 650, Bloomington, Minn. 55425; or e-mail [email protected].

THE measurement of the solubility of protein in a potassium hydroxide solution is a widely accepted test of the quality of soybean meal.

Evans and St. John demonstrated that increasing the time of heating soybean meal in an autoclave resulted in a decrease in the amount of protein that was soluble in a 0.2% solution of potassium hydroxide. This is the basis for the use of potassium hydroxide protein solubility (KOHPS) to evaluate soybean meal quality.

A large number of research papers have been published documenting the use of KOHPS to evaluate soybean meal quality and examining factors (like particle size) that have a significant effect on the laboratory results.

Parsons et al. (1991) documented that growth performance was highly correlated with the protein solubility of soybean meal samples for both broilers and pigs in diets where soybean meal was the sole source of protein.

There is an obvious need in the commercial industry for a test to evaluate the quality of full-fat soybean meal. As a logical extension of the effective use of KOHPS in evaluating soybean meal, KOHPS has become the default test for estimating the quality of full-fat soybeans.

Ruiz and Parsons (2015) concluded that there was little or no technical information documenting the use of the KOHPS test in evaluating full-fat soybeans. Ruiz and Parsons presented a paper (abstract 218) at the Poultry Science Assn. annual meeting suggesting that protein solubility in potassium hydroxide is not correlated to poultry digestible lysine in full-fat soybeans. The objective of the research was to evaluate the correlation between KOHPS in commercial samples of full-fat soybean meal and the in vivo measurement of digestible lysine in the same samples using the precision-fed cecectomized rooster assay.

Samples from 10 commercial lots of full-fat soybeans that had been collected over a period of a year were analyzed for KOHPS in three different laboratories. The KOHPS results varied widely among the three laboratories, from 43.8% to 85.5%. Even though the values were quite different, the solubility trends were similar for the three laboratories, according to the authors. Samples that were low in solubility in laboratory 1 were also low in solubility in laboratories 2 and 3, although numerically different.

Identical subsamples of the 10 full-fat soybean meals were used to determine the in vivo digestible lysine coefficients in cecectomized roosters, which varied from 81.0% to 88.8%. Correlation coefficients were calculated for the KOHPS from the three laboratories and the digestible lysine values. The correlation coefficient for laboratory 1 was actually negative (R-square was -0.66), while the correlation coefficient was not significant for laboratories 2 and 3.

The authors said these findings indicate that the KOHPS test method is not a good indicator of lysine digestibility (and protein quality) of full-fat soybeans.

Comments

Why do these results indicate that the KOHPS test is not a good indicator of protein quality of full-fat soybeans when the test seems to be a reliable indicator of the protein quality of soybean meal?

It is easy to leap to the conclusion that the difference is the fat content of full-fat soybeans, as soybean meal contains very little fat, while full-fat soybeans contain much higher levels of fat. Hence, the fat must be interfering with the efficiency of the KOHPS test.

However, Whittle and Araba (1992) published a paper on the sources of variability in the protein solubility assay for soybean meal (KOHPS). Four potential sources of variability evaluated were particle size, processing method, duration of extraction and fat content.

In their research, the effect the fat content of the soybean meal had on the KOHPS test was evaluated in two trials. In each trial, the starting material was a sample of raw soybeans.

In trial 1, the raw soybeans were first ground in a Wiley mill to pass a number-60 sieve, then heated in an autoclave at 15 psi for 15 minutes and air dried for two hours. In trial 2, raw soybeans were heated first and then ground. In each trial, six duplicate samples of 10 g each were prepared by extracting the heated and ground soybean meal in 40 mL of petroleum ether for various periods of time.

The petroleum ether extraction step resulted in samples that varied widely in fat content. The post-extraction fat content of the samples was determined using a standard Soxtec method.

In trial 1, the fat content of the samples varied from a low of 0.46% (71.4% protein solubility) to a high of 9.93% (69.0% protein solubility). In trial 2, the fat content varied from a low of 0.17% (80.4% protein solubility) to a high of 13.25% (79.8% protein solubility).

Whittle and Araba concluded that the fat content of the sample did not significantly influence the results of the KOHPS test. They added that they believe the assay (KOHPS) can be applied to full-fat soybeans and stated that full-fat soybeans have been analyzed for protein solubility successfully and repeatedly in their laboratory (unpublished).

Whittle and Araba also suggested that finely ground (number-60 sieve) full-fat soybeans have a tendency to adhere to the sides of the beaker in which the protein extraction step is conducted, so precautions should be exercised in order to achieve homogeneous solubilization.

In their research, Whittle and Araba found that the particle size of the sample, methods of processing the sample and duration of protein extraction with potassium hydroxide all significantly affected the KOHPS values obtained.

It is important to note that KOHPS is not a direct measure of protein quality or biological value; it is an indirect estimate of protein quality. Biological value is the gold standard typically measured by in vivo responses like growth performance or nutrient digestibility.

Whittle and Araba and others have identified a number of factors that significantly affect KOHPS values, but unless these can be associated with changes in biological value, they represent little more than noise in the test that needs to be reduced in order to improve the results.

Research reported by Nelson and Parsons is different from most other research in that they did not start with standardized samples. Ten commercial samples were collected from one or more commercial processing plants over a period of one year. Clearly, there are opportunities for major differences in processing methods, either from plant to plant or from time to time. The analytical results obtained suggest major differences among laboratories using what should be a relatively standard procedure.

These research methods are more representative of commercial reality and serve to validate that the "standard" KOHPS test results are not satisfactory for evaluating the protein quality of commercial sources of full-fat soybean meal. More research is required to identify and quantify the noise in the system.

The Bottom Line

Ruiz and Parsons concluded that protein solubility in potassium hydroxide is not a quality predictor for full-fat soybeans. Anyone using KOHPS as an important part of the quality assurance program for selecting sources of full-fat soybeans for use in the feed may want to re-evaluate this part of the program.

References

Parsons, C.M., K. Hashimoto, K.J. Wedekind and D.H. Baker. 1991. Soybean protein solubility in potassium hydroxide: An in vitro test of in vivo protein quality. J. Anim. Sci. 69:2918-2924.

Ruiz, N., and C. Parsons. 2015. Protein solubility in KOH is not correlated to poultry digestible lysine in full-fat soybeans. Proceedings of the Poultry Science Assn. annual meeting. Abstract 218. Accessed at: www.poultryscience.org/abstracts.asp.

Whittle, E., and M. Araba. 1992. Sources of variability in the protein solubility assay for soybean meal. J. Appl. Poultry Res. 1:221-225.

Volume:88 Issue:01