Kirsty Kliem (University Reading) visits MTT Agrifood Research
Separation of CLA isomers

1. Introduction to the Short Term Scientific Mission (STSM)
Much research into improving the fatty acid profile of bovine milk for human health by changing the dairy cow diet has focussed on reducing milk fat saturated fatty acid composition. This can be achieved by including supplemental oils with high unsaturated fatty acid contents into dairy cow diets. However, due to incomplete rumen biohydrogenation of dietary unsaturated fatty acids, these supplemental oils can also lead to significant increases in trans fatty acid concentrations in milk fat. Many of them are present in milk fat in low concentrations so detection and correct identification is challenging, but essential in order to report a complete picture of the effects of unsaturated oil supplementation on milk fatty acid profile.
Scientists at MTT Agrifood Research have developed methods of sample preparation and analysis that enable accurate identification and quantification of trans monounsaturated and polyunsaturated fatty acids. In particular, the use of silver-ion high performance liquid chromatography (HPLC) to identify conjugated linoleic acid isomers (CLA), and gas chromatography-mass spectrometry (GC-MS) analysis of mono- and polyenoic trans fatty acids are used routinely.
Over a number of years I have developed expertise in milk fatty acid analysis using gas chromatography-flame ionisation detection (GC-FID), partially through collaboration with MTT Agrifood Research when working together on the EU FP6 project “Lipgene”. My PhD thesis was based on outputs from this project, and I am now involved in similar research on reducing the saturated fatty acid content of milk fat by feeding supplemental oilseed preparations. However consumer interest in trans fatty acids is increasing, and I am keen to learn techniques used at MTT to develop my knowledge base so that we can report with confidence a complete trans fatty acid profile.

2. Main Objectives

The main objectives of my STSM were to:
(i) Conduct separation of CLA isomers in milk lipids using HPLC, and gain experience in identification and quantification of the separated isomers. This experience can then be utilised to conduct CLA analysis of milk lipids at the University of Reading.
(ii) Create 4,4-dimethyloxazoline (DMOX) derivatives of milk fatty acid methyl esters in order to separate using GC-mass spectroscopy (GC-MS), so that mass spectra of  three individual chromatographic peaks can be used to aid identification of trans mono- and dienes. In particular, I hoped to gain experience in using mass spectra to identify isomers in the 16:1 and 18:2 chromatogram regions, as these are difficult isomers to obtain standards for and so are extremely hard to identify using GC-FID.

3. Work undertaken

3.1 HPLC and CLA analysis
Prior to the start of the STSM, samples of milk lipid fatty acid methyl esters (FAME) dissolved in hexane were sent to MTT. These 16 samples were prepared from milk from 4 cows involved in a feeding trial with 4 experimental periods (4 x 4 Latin square). The treatment diets were a control diet and three diets including a calcium salt of cis-monounsaturated fatty acids (MUFA) at increasing inclusion levels. The samples sent to MTT were composited triplicates, each of 0.6 ml volume. Prior to HPLC separation, each sample required further concentration by evaporating off some hexane using a gentle nitrogen stream. The HPLC system (Hewlett-Packard, model 1090) was equipped with an autosampler, photodiode array detector, heated column compartment and four silver impregnated silica columns (ChromSpher 5 Lipids, 250 x 4.6 mm; 5 mm particle size, Varian Inc. Ltd., Oxford, UK) coupled in series. The mobile phase was 0.1 % (v/v) acetonitrile in heptane and this needed to be prepared on a daily basis. Acetonitrile is not readily soluble in heptane, so constant agitation is required in situ. Samples were run as a batch together with an external standard (O-5632, Sigma-Aldrich, YA-Kemia Limited, Helsinki, Finland) which contained 12 CLA isomers – 4 with trans/trans configuration, 4 with cis/trans configuration and 4 with cis/cis configuration. The photodiode array detector uses multiple detection wavelengths – 233 nm is commonly used to detect CLA isomers. In addition 268 nm was also used to identify the presence of 18:3 conjugated fatty acids. Each run took approximately 60 minutes, and chromatograms were recorded.
Sample chromatograms were integrated manually. Firstly, due to changes in retention times, each chromatogram required alignment with landmark isomers present in the standard. These were trans-9, trans-11 CLA and cis-9, trans-11 CLA. Once chromatograms were aligned they were saved, and then integrated by comparing peak elution orders with that of the standard and also elution orders from other standards regularly used by MTT. Following integration, results reports were printed that reported proportional (%) peak area for each peak retention time. 4
3.2 GC-MS analysis of DMOX derivatives
Four vials containing milk FAME from the University of Reading were sent to MTT prior to the STSM. Two of these samples were FAME from bulk tank milk from a farm feeding a linseed-based supplement, and two were from individual cow samples from the same calcium salt trial mentioned above. The two pairs of samples were pooled to create two individual samples of greater volume. A subsample was transferred to vials and run together with three standards (GLC463, NuChek-Prep, Elysian, USA and the same standard with additional isomers added by MTT) on the GC-FID (Hewlett-Packard model 6890) using an 100 m CPSil-88 column (Chrompack 7489, Middelburg, The Netherlands), hydrogen as a carrier gas and the same temperature program used by our laboratory at the University of Reading to separate FAME. The resulting chromatograms were printed off in order to compare with the DMOX chromatogram.
To prepare DMOX derivatives, firstly FAME were evaporated to dryness using a gently nitrogen stream. Then appropriate volumes of aminomethylpropanol (approximately 250 mg) were added to each tube. The aminomethylpropanol is solid at room temperature so requires warming before quickly pipetting. Tubes were sealed under nitrogen and incubated in a sand bath at 170°C overnight. Following incubation, DMOX derivatives were then extracted using diethyl ether/hexane (1:1 v/v) and separated using GC-MS (Hewlett-Packard model6890, with selective quadrupole mass detector, model 5973N, Agilent Technologies Inc.) equipped with the same column as above, the same temperature program but helium as a carrier gas.
Both the 16:1 and 18:1/18:2 regions of each chromatogram were printed off and compared with the FAME chromatogram in order to establish which peaks required identity confirmation. Then, mass spectra were obtained for different sections of these “unknown” peaks and compared with published spectra ( Using spectral data, I was able to attempt location of double bonds and even identification of DMOX derivatives. There were still some areas that require future attention, especially where isomers were present in small quantities or co-eluted with others.

4. Outcomes

This STSM resulted in the following outcomes/achievements:
• Basic training in HPLC operation, preparation of DMOX derivatives and separation of derivatives using GC-MS, techniques which I hope to transfer to the University of Reading laboratory.
• Interpretation of HPLC chromatograms and identification of CLA isomers using standards and published literature
• CLA isomer profile of milk lipid samples from the experiment conducted at the University of Reading, which will be incorporated into a scientific paper
• Experience in interpretation of mass spectra which I hope to build on in the future
• Increased knowledge of trans fatty acids present in milk lipid
• Enhancement of relationship between the University of Reading laboratory and MTT, including increasing my contact network with other scientists working in the same field.

5. Proposed publications

• Joint publication based on the results of the calcium salt experiment, which will include the CLA isomer profile measured as part of this STSM (submission to e.g. Animal)
• Joint review article on the range of trans fatty acids present in milk fat. This will involve re-interpretation of some earlier work conducted by myself on the fatty acid profile of milk available at retail, using the experience gathered during this STSM (submission to e.g. Journal of Food Chemistry).

Financial support from the Feed for Health COST Action will be acknowledged in the final submitted manuscripts.

Kirsty Kliem STSM
Design & Programming: Zenlex