Go to contents (site navigation)


Realised by ALMS™
developer of the AIDS-HIV Reference project
Abstract No.: MoP-035
Session: Food and Nutrition
Presentation date: Mon, Aug 28, 2006
Presentation time: 09:50 – 11:20

Authenticity and Identity of Skimmed-milk Powder

Dion Luykx1, Jan Cordewener2, Pasquale Ferranti3, Suzanne van Steenis1, Rob Frankhuizen1, Monique Bremer1, Dick Hooijerink1, Twan America2

1 RIKILT - Institute of Food Safety, Wageningen, Netherlands
2 PRI - Plant Research International, Wageningen, Netherlands
3 University of Napels, Portici, Italy

Correspondence address: Dion Luykx, RIKILT - Institute of Food Safety, Authenticity and identity, Bornsesteeg 45, Wageningen, 6708 PD Netherlands.

Keywords: Digests, Tryptic; Mass Spectrometry, Liquid Chromatography; MS/MS; Protein Identification.

Novel aspect: Applying (untargeted comparative) LC-MS for authenticity and identity of feed.


The EU subsidises the use of skimmed-milk powder in compound feeding stuffs (EC regulation 2799/1999). In this way European dairy surpluses can be reduced. A precondition for receiving any subsidy is that the processed milk powder meets certain specifications. RIKILT checks these specifications on assignment from the General Inspection Service of the Dutch Ministry of Agriculture, Nature and Food Quality. There are indications now for falsified skimmed-milk powder content by adding (hydrolysed) vegetable proteins. These proteins are not allowed and cannot (always) be detected by standard analysis methods. In order to detect the presence of the adulterations (min. 1%) a sensitive protein identification method was developed based on LC-MS technology. A comparison was made between targeted and non-targeted LC-MS approaches. In the targeted LC-MS approach specific soy, pea or wheat peptides were selected from the corresponding LC-MS peptide patterns of various digested soy, pea and wheat protein isolates digested by trypsin, respectively. Subsequently, skimmed milk powder samples in absence and presence of these vegetable proteins (5%) were digested and analysed for the presence of the selected vegetable peptides in the LC-MS peptide pattern. The second LC-MS approach concerned untargeted data dependent LC-MS/MS. In this approach MS/MS spectra were collected from the peptides retrieved from trypsin-digested skimmed-milk powder with vegetable proteins. After multiple injections peptide sequences were identified that originated from the respective plant proteins. The final LC-MS approach concerned untargeted comparative LC-MS. In a two-step procedure, all quantitative differences between LC-MS traces from tryptic digests from adulterated and non-adulterated milk powder were first selected and subsequently targeted for selective LC-MS/MS analysis in a second injection. Here, unique peptides (not corresponding to skimmed-milk powder) were identified. A pre-extraction (via a specific buffer) or pre-separation (via HPLC) of the vegetable proteins from the milk proteins simplified the detection or identification of the vegetable proteins by LC-MS. Although preliminary results show that several LC-MS approaches were suitable for screening and/or identifying vegetable proteins in skimmed-milk powder, untargeted comparative LC-MS looks the most promising approach. In this way the authenticity and identity of skimmed-milk powder can be assessed.