AOAC SPIFAN Review Team Meeting Book-July 22, 2015

BVIT-02 FOR ERP/WORKING GROUP USE ONLY DO NOT DISTRIBUTE

ABBOTT LABORATORIES INTEROFFICE CORRESPONDENCE

Date: 2/13/12

From:

Louis Salvati

Subject: Validation of LC/MS Determination of Total B 6 , B 1

, and B 2

Using an

Enzymatic Preparation.

Background Although internal methodologies are available for the determination of the fortified forms of pyridoxine (PYR), thiamin (THI), and riboflavin (RIB); Abbott Nutrition does not currently have a method for the determining the total content of these three nutrients. While the overwhelming majority of each vitamin in Abbott Nutrition’s products comes from the fortified form, some comes from the natural ingredients used in the products. Natural ingredients such as milk contain two forms or pyridoxine: pyridoxal (PAL) and pyridoxamine (PAM) as well as phosphorylated versions of these two forms. Milk also contains the fortified form of riboflavin with two phosphorylated analogs and the fortified form of thiamin with two phosphorylated analogs. Government regulations in certain countries require the total content to be determined. This method would also be a useful internal research tool. In order to assay the inherent nutrient content that is bound in the sample, some form of matrix digestion is needed; and additionally, since those forms are phosphorylated, a dephosphorylating agent is also necessary. The matrix digestion can be carried out by strong acid and heat, or enzymatically. A potential procedure was proposed by C. Hasselmann 1 which used papain, and alpha-amylase to break down the sample matrix to free bound nutrients. Acid phosphotase is also added to dephosphorylate the nutrients. The separation developed is based loosely on work done by Hong Zhang 2 using an Acquity BEH C18 column (2.1 mm x 50 mm, 1.7 um) and a 20 mM ammonium formate and methanol gradient. The separation is done in less than 7 minutes and is illustrated in Figures 1 and 2. A limited sample group consisting only of milk-based infant formulas was chosen for this validation because milk based ingredients are the dominant source of the inherent vitamins. The goal was to first establish a solid basis using the most affected sample type. Two standard reference materials were analyzed along with a commercial infant formula and a placebo infant formula not fortified with these nutrients. Samples spiked with the phosphorylated forms of each nutrient were tested; however, the spiking levels

8