Protein Identification and Quantification in Mixtures of highly-modified Proteins

The identification and quantification of proteins in highly-modified mixtures using
proteomics has been performed. Two research projects have been undertaken which fulfil this
aim.
The first comprised the development of quantitative methods to detect trace amounts of
hazelnut and soy in complex mixtures.
A method for the detection and absolute quantification of Cor a 9, a major hazelnut (Corylus
avellana) allergen was developed based on mass spectrometry. One hundred and ten hazelnut
proteins were detected, five of which were allergens. The peptide chosen for quantification
was from Cor a 9 (11S globulin-like protein). Two specific fragmentation reactions were
chosen in multiplexed Selected Reaction Monitoring (SRM). Forty three hazelnut food
processing imitation samples, varying a range of factors, such as the temperature and
incubation time were analysed. A calibration curve was made for cookies. The developed
method was for home-made cookies, shop-bought cookies and chocolate. The quantities of
Cor a 9 in each sample were determined from the quantification of the target peptide using
isotopic dilution with a heavy isotopically labelled peptide. A second peptide with two
transitions was also proven to be a possible alternative as a detection method for hazelnut.
The presence of soybean allergens in processed food can be detected using the mass
spectrometric identification of a soybean peptide which is resistant to the heating and
chemical reactions associated with food processing. The proteomic analysis of soybeans
allowed the identification of 11 allergens. A method using a peptide (VFDGELQEGR) from
glycinin G1 (Gly m 6.0101) was developed for the detection and semi-quantification of the
allergen in food samples. Two specific fragmentation pathways were chosen in selected
reaction monitoring for unambiguous identification of glycinin G1 and were: 575.3 903.4
Da and 575.3 788.4 Da. Sixteen imitation samples of processed food spiked with soybean
were analyzed, where factors such as temperature and incubation time were varied, and the
chosen transitions were detected. The developed method was specific for home-made cookies
and a shop-bought biscuit. Semi-quantification from both cooked and uncooked cookies was
demonstrated.
The second comprised the identification and quantification of conotoxins in the venom of
Conus textile by the use of isotope coded affinity tagging (ICAT) and label-free
quantification. The extreme variety and complexity of the conotoxins has been insufficiently
documented and this research demonstrates the varied nature of conotoxins found in different
parts of the venom duct and their patterns of expression. Fifteen conotoxins, several with
different post-translational modifications (PTMs), were identified and quantified. Distinctive
patterns emerged, with the largest group of conotoxins increasing, then peaking in the central6
proximal part, before decreasing; whilst the second largest group peaked in the distal region,
generally displaying nothing in the first parts. A new conotoxin, PCCSKLHDNSCCGL*, was
sequenced.