AOAC ERP Gluten Assays - Dec 2018

OMAMAN‐47: Quantification of Wheat, Rye, and Barley Gluten in Oat and Oats Products SUMMARY OF COMMENTS  It would be useful to have an estimate of the gluten content for samples 11‐21 based on  outcomes of the gluten in oats working group. The manuscript identified the antibodies used do not detect D‐hordeins, but made no attempt  to create such Ab because it accounts only 5 % of the proteins. There is only one data point for wheat and one data point for rye, as individual contaminants,  in the stated matrices of the collab study. While the number of samples technically meets the  OMA criteria, because this kit is seeing three "analytes", at least one additional sample  representing each of these would have been expected. If the SMPR is being followed, the Collab study should have been able to assess accuracy as  well as precision. If the SMPR conversion factors from Dumas nitrogen to gluten were used to  create the spikes, then the expected gluten values for the spikes prior to processing should be  listed in Table 1.  While several of the matrices were processed in ways that affected the final  gluten concentration, it would not be impossible to estimate the moisture loss from these  processes and estimate a gluten value for each of the spikes, at least in relation to a relatively  unprocessed matrix like the flour. Recovery calculations would not be needed, but it would be  helpful to see if the test results were in line with the expected values. I would also like to see the authors address the unequal extraction or presence of the  prolamin and glutelin fractions in the standard solution (page 6 of Annex B). Does the 2:1 ratio  of prolamins to glutelins represent the actual levels in the flour mix, or is this an issue of  extraction efficiency? Will this ratio change if the day comes when they need to make the  standards from a new starting flour mix, and what affect might a change have on quantitation  of real‐world samples? The cross‐reactivity data tables (page 14‐16 of Annex B) show that some materials which are  assumed to be gluten free produce ODs well above the zero standard (corn starch, tapioca  flour, flax seed, garlic). Did these values calculate out above the stated LoD range from the  Annex B abstract (0.4 ‐ 1.9 ppm) or the LoD provided in the Collab study Discussion (3 ppm)? If  so, will the kit manufacturer make a note in the kit that advises against using any values lower  than the lowest positive standard, especially in those countries that have a threshold of "no  detectable gluten"? There is not enough data provided to comment on the recovery using this method. Some of the data presented in Table 2 needs further justification. See comments in the  attached file. Take care in the writing what mg/kg means. In some sentences it is mg/kg gluten. Perhaps at  times we are also looking at mg/kg gliadin(?). Make sure when it is only mg/kg 'xxxxx???' that  the compound (gluten of gliadin) is given after this. Not always there. Nothing that stands out at the moment

ER 1

ER 2

ER 3 ER 4

ER 5

ER 6

4 OF 10