Recent developments in the analysis of food lipids and other lipids

ARTICLE

Gas chromatography (GLC)

In capillary GLC of lipids there has been a long research tradition here in France. Many scientists have shown new applications and improvements in the technique, for example in the separation of positional isomers [1-3], of trans fatty acids [4, 5], of cyclic fatty acids [6, 7], of frying oil artifacts [6], of minor components [8, 9] and contaminants [10], and in many other contexts [3]. In recent years, positional isomer separations and cis-trans isomer separations have become more and more important, e.g. in oil authenticity studies, to investigate sample origin, history and pre-treatment, and partly also to investigate possible health risks [11-14].

Our own work in GLC was fairly straightforward. For example, we needed high resolution capillary GLC when we investigated fish lipids and seed oils with unusual fatty acids and with positional isomers and cis-trans isomers [15, 16].

Fish oils and seed oils: isomer separations and the characteristic 20:1 region

Fish oils are of dietary interest because of their oméga-3 fatty acids [17-19]. Fish oil oméga-3 concentrates, however, sometimes also contain higher levels of an oméga-1 fatty acid, and a vinyl end group can be seen both in ¹H-NMR and in ¹³C-NMR. GLC in these cases may show that 16:4n-1 may be present in a rather high proportion (figure 1). Higher levels (6-10 %) of this fatty acid may indicate that the concentrate was produced from menhaden oil, which is high in 16:4n-1 [20].

In another project, biologists had the idea that differences in lipid metabolism could perhaps be used to fight economically important fish parasites [15]. As usual in phospholipids, the fatty acids from the sn-1-position and those of the sn-2-position of the same phospholipid were quite different. More surprisingly, the lipid composition of the edible fish and its own parasite was usually also quite different [21]. Figure 2 shows the fatty acids obtained from the sn-1 position of phosphatidyl cholin (PC-sn-1) of the fish intestinal tissue, compared with those obtained from the parasite attached to the same tissue.

Figure 2 also shows that a group of peaks in the 20:1 region is very characteristic for the fish parasite, and differs strongly from that of the fish. Important here is the separation of isomers, which may often be of chemotaxonomic value [16]. In particular, the 20:1 region is often highly characteristic, both in marine animals and also in plants. This means that good resolution is required in this area, and one should never say simply "20:1" in a results table without further explanation. "20:1" can be anything, and figure 3 shows this by way of the example of a few seed oils that we have analyzed recently. In each of these examples, a different 20:1 positional isomer is the most prominent member of the group.

Structure-retention relationships, fatty acid families, and evolution

Structure-retention relationships in the GLC of unusual fatty acids have become more and more important, and here the early work of Ackman, Sebedio et al. should be acknowledged [1, 22]. In routine screening for gamma-linolenic and other unusual fatty acids in seed oils of the plant kingdom, we often did not have enough sample material and therefore a more elaborate derivatization and GC-MS investigation was not possible. So we had to rely on GLC retention data on three phases of different selectivity for identification [23].

Ackman et al. [22, 24, 27] have published RRT and ECL tables for Silar 5CP. On our own Silar 5CP column, however, we found that a number of fish oil fatty acid peak elution sequences were reversed, compared with available ECL tables [24]. This may depend on the temperatures used, but it is still advisable to test every column. Moreover, in commercial fish oil fatty acid standards supplied with an authentic test chromatogram, peaks are sometimes mislabelled. We have one authentic test chromatogram, where three minor fish oil fatty acids were wrongly identified by the supplier of the reference mixture.

Fatty acid families are well-known from human and animal fats, and from edible oils. However, in lower marine animals - and this means rather often in seafood - there are other fatty acids that do not fit into the n-x fatty acid family schemes [20, 24, 28-30]. We follow Japanese and American authors who have labelled the fatty acids in their chromatograms both in the "n-x"- way for those fatty acids that fit into families, and in the DELTA-way for those which do not [24]. The latter are often non-methylene-interrupted polyenoic fatty acids [NMIP-fatty acids], and they often contain a DELTA5-double bond [20, 24, 30].

Seed oils can also be rather funny. In edible oils, n-9 fatty acids dominate the C₁₈ to C₂₂ monoenes. However, we had other seed oils, where the n-7 or the n-5 fatty acid family dominated, and other oils where the DELTA5-series or the oméga3-series dominated the monoene fraction (figure 3). There is a huge array of possible fatty acid structures in plant seed oils [31-33] and a few of them have been illustrated recently [34, 35]. DELTA5-fatty acids, such as 20:3DELTA5c,11c,14c seem to be archaic and occur not only in Japanese seafood, but in many rather primitive plants and animals and in gymnosperms [35]. Pinolenic acid (18:3DELTA5c,9c,12c) is typical for conifers only and appears to be the result of another step in the evolution of this branch of the gymnosperms (figure 4). Up to 20 % of DELTA5-NMIP-fatty acids (mostly pinolenic) is present in pineseed oil, a French product which can be found in German food shops on the oil specialities shelf [36]. Little is known about the metabolism and possible physiological effects of this and other positional isomers of linolenic acid. The corresponding trans isomer, columbinic acid, occurs in one subfamily of the plant family Ranunculaceae [37]. This is a rather old plant family where one not only can find gamma-linolenic acid (18:3DELTA6c,9c,12c) [38] but also 9 different DELTA5cis fatty acids and 6 different DELTA5trans fatty acids [35, 37]. These are typical for certain plant genera and may be indicators of phylogenetic evolution. Apparently, the capacity to synthesize DELTA5-fatty acids has been lost during later stages of evolution, which eventually led to those plants which we use to produce our edible seed oils today [35].

Animal and human FA metabolism and the search for gamma-linolenic acid

In human and animal metabolism, DELTA6 double-bonds are introduced into linoleic acid, before chain elongation and further desaturation leads to arachidonic acid and other eicosanoids [19]. However, this process may be impaired or inefficient in older people and under stress. gamma-linolenic acid, which already contains a DELTA6cis double bond, is therefore an important intermediate and is of interest for many dietary and pharmaceutical applications [39, 40].

18:3DELTA5cis-, 18:3DELTA5trans- and 18:3DELTA3trans-fatty acids sometimes interfere with the determination of gamma-linolenic acid. In a screening program of seed oils, the separation of different 18:3 positional isomers is therefore important [23, 41]. GLC conditions can be found where all the naturally occurring unconjugated 18:3 isomers can be separated. This is necessary, because several of these can be expected to occur in seed oils obtained from members of one and the same plant family - for example, the Asteraceae (Compositae) plant family, which is an important supplier of many edible oils [32], or the Ranunculaceae plant family, which is of interest in fatty acid biochemistry and evolution [35, 37].

Analysis of trans-fatty acids

As a government institute, our institute has of course been also involved in the current discussion on trans-fatty acids in olive oils and in partially hydrogenated fats. This has been investigated a lot recently, particularly also by French authors [5, 42, 43], so I can keep this very short.

For reference purposes, a mixture of trans-fatty acids can easily be obtained by NO-isomerisation of linseed oil (figure 5). Fatty acids of this type can be found in refined oils and in heated oils, and their presence can be used to differentiate genuine native oils, or "cold pressed" oils, from oils that had been mixed with refined oils or that were obtained from heated seeds. Oils that were "cold-pressed" from heat-dried seed may already contain trans fatty acids at levels higher than those permitted for virgin olive oil [44].

The separation of trans fatty acids in partially hydrogenated oils and in margarines, where all the positional isomers occur, is more difficult and requires a pre-separation step on a silver nitrate plate or column [13, 45, 46]. If this is not carried out, as in the AOCS method, too high results for the cis monoenes and too low results for the trans monoenes may result. This has been shown by Battaglia, by Wolff and Bayard, by Gertz, Ratnayake, and others [5, 45-48].

By silver TLC plus gas chromatography, one can compare the trans isomer distribution in human and bovine milk samples with those of partially hardened soya bean oil and rapeseed or canola oil. Ratnayake et al. [49-51] could show, that the trans fatty acid pattern in human milk in Canada resembled that of partially hydrogenated vegetable oils found in the diet.

Silver ion HPLC and dynamic impregnation TLC

Adlof et al. [46] showed that the separation of trans from cis fatty acid methyl esters is rather easy by HPLC on a silver-loaded ion exchange column, and a partial separation of the positional isomers can also be achieved. A similar HPLC separation on a silver nitrate-silica column has also been shown fifteen years ago by Battaglia in Switzerland [48]. Ion-exchangers have now replaced silver-loaded silica HPLC columns, and have turned out to be very useful and reliable [6, 52]. However, usually a gradient and a light-scattering detector is needed. Moreover, this approach is useful only when it is used day after day for large numbers of samples.

For those laboratories who need an argentation separation only now and then (or ten or twenty times a year), TLC is still the most useful technique. We had simplified this quite a bit to make it more easy [53-55]. We impregnate plates by running a solution of silver nitrate and phloxin in acetonitrile to the top. Silver and phloxin will also migrate to near the top of the plate, but not quite. The silver front stops about 5 cm from the top of the plate and the phloxin front stops another centimeter or so below the silver front. Below the phloxin front, silver and phloxin are distributed much more evenly than when spraying or dipping a plate.

Fatty acid methyl esters or triglycerides can then be spotted on the dried plate, which is developed in a mixture of hexane and toluene. Our fluorescence indicator, phloxin, does not move with this solvent and acts much like a fluorescent ion exchanger in the Ag⁺ form - and so it facilitates not only the argentation separation as such, but it also provides fluorescence visualization at the same time. The top 5 cm of the plate are free of silver and here the plate can be touched with the bare fingers without any problems [53, 54]. So it is easy to handle the plates, and gloves are not needed. Because phloxin is present already, fluorescent lipid spots are seen under the UV immediately after development and evaporation of the toluene - no spraying is necessary (figure 6).

Fluorescent plates of this type can also be scanned using a scanner [54, 56]. Using our direct impregnation TLC method, one can easily detect symmetrical and asymmetrical isomers of monoenoic triglycerides. Cocoa butter replacers, or lard vs. tallow, can be evaluated in this way. Figure 7 shows direct plate scans with a separation of symmetrical from non-symmetrical monoenoic triglyceride isomers in chemical and enzymatic interesterification mixtures of Chinese Vegetable Tallow with tristearin [54, 56, 57].

A combination of this dynamically impregnated silver-TLC with capillary GLC of the separated FAME zones is a powerful tool for the analysis of more complicated seed oils. This has been demonstrated again recently in analyses of an oil containing DELTA5-fatty acids [55] and of an oil containing gamma-linolenic acid [58].

HPLC of triglycerides with double (RI- and UV-) detection

The reversed-phase HPLC of triglycerides is now a fairly routine practice for most edible fats and vegetable oils [59-61]. It can even be used to separate triglycerides with unusual fatty acids, such as petroselinic, laballenic or gamma-linolenic acid [62, 63] from the normal triglycerides. Usually, only a refractive index (RI) detector is used, but a combination of RI- with short-wavelength UV-detection can be very helpful here [62]. Figure 8 shows that in RP-HPLC of the PN=48 triglycerides of palm oil and Voacanga oil, all the triglyceride peaks which contain one or more linoleic acid residues are enhanced in the short-wavelength UV (see also figure 11 here).

For the determination of LLL [trilinolein] in tests for olive oil adulteration, the EU prescribes an HPLC area-%-method with an RI-detector [64]. However, we believe that a method using a short wavelength UV-detector and an internal standard would be preferable (figure 9). The squared area in figure 9 is then actually all that is needed for the evaluation of the LLL content, and possible adulteration, of an olive oil - in weight-% rather than RI-area-%. The LLL-peak in the UV-chromatogram is much higher - and easier to integrate - than the same peak would be in an RI-chromatogram. Moreover, when an internal standard is used, this analysis could be made much more reliable because it would then be independent of the integration of all the other peaks in the oil (K Aitzetmüller, to be published).

HPLC - gradients and light scattering detection

For the analysis of triglycerides by HPLC, there has always been a search for gradients compatible with detectors, or detectors compatible with gradients [61, 65, 66]. For example, we tried flow gradients with a refractive index detector [61, 66, 67], others tried temperature gradients [68]. On the other hand, light scattering detectors have become more and more popular [52, 69-71]. They can be used with true solvent gradients, but they require individual calibration and are difficult to use in quantitative work [72, 73]. Our old "total artifacts" separation of used frying fats, for example, which worked well with a transport-FID detector where all carbon was converted to methane [66, 74], cannot be carried
out with the laser light scattering detector [69].

HPLC of specialty oils and oxidized oils with short-wavelength UV-detection

HPLC with short-wavelength UV detection can be very useful for highly unsaturated seed oils, for example those containing gamma-linolenic acid, because the higher detector response permits the injection of smaller samples, and thus higher resolution [62]. As already mentioned above, gamma-linolenic acid is important for elderly persons when the activity of the DELTA6-desaturase is reduced. In addition, triglycerides containing alpha-linolenic acid can be separated from those containing the normal gamma-linolenic acid under optimized conditions (figure 10) [62].

Conjugated dienes (conjudienes) and conjutrienes may occasionally also occur in edible oils. Conjudienes may have been formed as products of autoxidation or as products of lipoxygenase reactions. These can be detected in RP-HPLC with a UV-detector set at 235 nm. Conjutrienes are frequently present at high levels in technical oils, but they rarely occur in edible oils. Authentic cherry kernel oils contain lower levels of conjutrienes, and we recently got a sample of imported cherry kernel oil from German customs (figure 11). In RI detection and at short wavelength UV (210 nm), all the triglyceride peaks are seen, but when the UV-detector is set at 272 nm, only the triglycerides containing the conjutriene are visible (Note again here that the peak size ratio of the OLL : 000 peaks is much larger in the UV-210 nm chromatogram than in the the RI chromatogram).

The same can be shown for the estolides in stillingia oil, which is a technical drying oil of South Asian origin and a by-product of the edible Chinese Vegetable Tallow [75]. The estolides contain a conjudiene UV-chromophor (which is further conjugated through to an ester carbonyl), and one can clearly see these estolides in the oil when the HPLC analysis is repeated with the UV detector set at 260 nm [75]. The method can also be used to detect residual kernel oil in Chinese Vegetable Tallow.

Dimeric triglycerides in GPC and HPLC

Gel permeation chromatography (GPC) of used frying oils in THF with RI-detection is now a standard procedure used worldwide [66, 76, 77] (see also figure 3 in [65]). However, calibration may be a problem, particularly when a mono- or dimeric triglyceride peak is composed of both oxidized and non-oxidized lipids at an unknown ratio. For example, polar dimers may have a significantly different refractive index, compared with thermal (non-polar) dimers. Response factors may differ by 10-20% - and this may make area-% calculations rather meaningless. A pre-separation of the used frying fat on a preparative silica column [78, 79], or a calibration with pure triglyceride, may actually even aggravate the situation, because then the polar: non-polar ratio difference between the dimer and monomer peaks will be further increased. Because of this, great caution is required in the interpretation of quantitative data from GPC chromatograms of used frying oils with RI detection.

We had also shown that GPC with RI detection can be combined in certain cases with infrared [IR-] detection at the carbonyl or the hydroxyl frequencies (figure 4 in [65]).

The dimeric triglycerides present in used frying fats are not amenable to reversed phase HPLC separation. Natural dimeric triglycerides, which we found in the edible fruit fat of the Chinese lacquer tree, however, can be separated by RP-HPLC. The dimeric triglyceride fraction there is composed of long-chain alpha,oméga-diacids esterified to two glycerol molecules bearing palmitic and oleic acid residues.

In preparative TLC on silica plates, the dimeric triglyceride fraction can be obtained, and by argentation TLC this fraction can then further be separated into zones of different degree of unsaturation. In contrast to the dimers in used frying fats, these dimeric triglycerides in the fruit fat can be separated using reversed-phase HPLC. Figure 12 shows an example for the HPLC separation of the argentation TLC zones (Jing Li and K. Aitzetmüller, to be published).

¹³C-NMR spectroscopic investigations of triglycerides

¹³C-NMR can be used both for the quick identification of fatty acids in an oil and to determine their position on the glycerol [80-83]. It is particularly useful in the analysis of triglycerides containing DELTA4-, DELTA5- and DELTA6-fatty acids (figure 13) [36, 80, 84]. Here we look at fatty acids not in terms of "fatty acid families"- which is the usual way in medicine, human nutrition and animals - but in groups of DELTA4-, DELTA5- and DELTA6-fatty acids [36]. The ¹³C-NMR signal for the carboxyl carbon atom depends on the position on the glycerol (sn-1/3 or sn-2). Easily seen is this difference when unusual fatty acids are present, as in ¹³C-NMR of fats containing gamma-linolenic, pinolenic or columbinic acid. Whereas gamma-linolenic and columbinic acids are both in the central (sn-2) and the outer (sn-1 and -3) positions on the glycerol, although at different ratios, this is not so with pinolenic acid, which is not found at all in the sn-2 position (figure 13). The bio-availability of gamma-linolenic acid, for example, may depend on whether gamma-linolenic acid occurs in the sn-1,3 or sn-2 position [85] in the fat, such as in seed oils of Oenothera and Borago [86], particularly if one takes into account that sn-1,3-specific pancreatic lipase discriminates against DELTA6-fatty acids such as gamma-linolenic acid [36, 87].

As expected, ¹³C-NMR analyses of lard and tallow, and of samples of human depot fat show the prominent presence of saturated fatty acids in the sn-2 position of the glycerol in lard, the fat from the pig. To a lesser extent, this - and a higher level of linoleic acid - is also seen in human depot fat (Diehl, Herling and Aitzetmüller, unpublished).

The oméga-3 / oméga-6 ratio in pharmaceutical preparations (fish oils and fish oil concentrates) is easily obtained by ¹³C-NMR as the peak area ratio of the two methyl end group signals. Many pharmaceutical and dietary publications link fish consumption and the Eskimo life style with a reduced rate of myocardial infarction. Eskimos, however, eat more seal rather than fish (table). The two oils are very similar in their fatty acid composition and ¹³C-NMR of the methyl end groups clearly shows an almost identical oméga-3/oméga-6 ratio. However, ¹³C-NMR of the C-1 signals shows huge differences between seal and fish. In the seal oil, EPA and DHA are both in sn-1,3 almost exclusively, whereas they are in both positions, although at different ratios, in the fish oil (figure 14) [36].

There have been many projects recently which try to produce eggs with increased oméga-3 levels, by feeding these fatty acids to the hens. A ¹³C-NMR investigation of these eggs shows that oméga-3 enrichment can be achieved but is limited to phospholipids, where these fatty acids occur primarily in the sn-2-position.

HPLC of minor components in fats

Sterols and stigmastadiene

The analysis of sterols and related compounds is very important in the detection of fraud, oil history/processing steps, and oil adulteration [88-90]. It is usually carried out by GLC [88, 91, 92], but some HPLC applications have also been described [89].The determination of stigmastadiene and other steradienes which are formed during oil refining [93] has become very important recently for the analysis of olive oils and other cold-pressed oils [94, 95]. The gas chromatographic method requires training and experienced laboratories. The HPLC-method [96, 97] appears to be more robust and gives better results with those laboratories who analyse olive oils only now and then (figure 15). Ring-test results in Germany were better with the HPLC method.

HPLC of tocopherols

Tocopherol analysis is now best carried out with a diol column [98]. This is an improvement because separations are more reliable than on silica columns, and more independent of traces of water. Moreover, on the diol column the separation of all eight tocopherols and tocotrienols is possible without the use of dioxane or other toxic solvents.

The major remaining problem in tocopherol analysis is the availability of reference standards and internal standards. We had best results in the late seventies with 5,7-dimethyl-tocol, a tocopherol that does not occur in nature
(99, see figure 10 in [64]). This is a very useful internal standard because it has the same chromatographic, fluorescence, and oxidation properties as most tocopherols [99].

Unfortunately, 5,7-dimethyl tocol is no longer available now for normal routine use. Moreover, the available individual tocopherol standards apparently also give rise to all kinds of problems. It would be much better to have a standardized tocopherol- and tocotrienol-containing reference oil mixture (for example a mixture of palm oil and soyabean oil) with a certified tocopherol content. This could be used for the calibration of direct tocopherol analysis using fluorescence detection. In combination with the best internal standard, 5,7-dimethyl-tocol, this would be a vast improvement over present methods.

Vitamin A compounds in margarines

The analysis of margarines for the presence of carotene and vitamin-A products is rather easy in HPLC with diode array detectors or with UV-wavelength switching, as we have shown many years ago [61, 99, 100]. An improvement recently was the use of a nitro column [101]. By wavelength switching, beta-carotene, vitamin-A esters and tocopherols can be analyzed in one run [61, 99-101]. Pure vitamin A propionate can often be used as an internal standard [99, 100]. However, the availability of reference substances for calibration purposes is a problem again, and a standardized reference oil with a certified vitamin-A ester content would be ideal for calibration.

Chlorophyll degradation products

HPLC with fluorescence detection is very useful for the investigation of chlorophyll degradation products in oils. With the advent of so many native and cold pressed oils, it can be important to detect heating of the oil by the formation of pyropheophytins [102, 103]. When olive oils were contaminated with chlorinated solvents, some producers apparently tried to remove these by vacuum distillation and heating. In the HPLC of chlorophyll breakdown products, however, the ratio of pyropheophytin A to pheophytin A can indicate whether or not the oil had been heated above 100°C [102]. Heating or heat conditioning ("toasting") of animal feed which contains whole or shredded rapeseed can also be followed by HPLC of the pyropheophytins [102, 103] (figure 16).

On the other hand, in some cases the presence of other chlorophylls, such as protochlorophylls and their degradation products, can also indicate the nature and history of an oil. Pumpkin seed oils [102] may contain protochlorophylls and their degradation products. The structures of these products are quite different - they elute earlier, because they have more double bonds and often a shorter side-chain [102].

Outlook

There are many other new developments one should mention. Typical examples are chiral separations, supercritical fluid chromatography, electromigration methods, LC-MS and related techniques, and so on. Phospholipids and other polar lipids are so important for membranes and in medical and pharmaceutical applications [72, 73]. Space is far too limited to mention all these.

HPLC of fatty acid derivatives may be the preferred technique for certain labile fatty acids [61, 104]. Grob and others [94, 105] showed very interesting results in coupling LC with GLC. GC-FTIR has been tried for trans-fatty acids [106]. Negative ion tandem mass spectrometry of triglycerides has been used to investigate the triglyceride composition of mother's milk [107]. Direct MS of triglycerides is possible in the field desorption (FD-) MS mode. However, these methods are probably less suitable for the everyday routine.

Coupled MS-techniques such as GC-MS and LC-MS are used mostly in two ways: one is for the confirmation of the structure of a quantitatively determined peak, as in the analysis of dioxins or certain pesticides and other toxic agents. The second, and more important way of using GC-MS is in the identification of unknown peaks in a chromatogram. In GC-MS, a nearly universal separation method and a nearly universal detector is coupled with MS for identification of the structure of unknown separated peaks. HPLC, on the other hand, is at its best when it is especially developed for a known particular structure, or group of structures. This means, to optimize HPLC one already needs to know what structures there are. In HPLC, both the separation column and the detector must be selected, and optimized, for a particular structure, because there is no universal HPLC method and no universal HPLC detector. In HPLC, normally we do not deal with unknown samples containing unknown peaks. It is for this reason that I personally believe that LC-MS will never play the same role as GC-MS in the investigation of unknown samples.

Differential electromigration methods are quickly developing [109], but they are probably less important in the field of lipids. Supercritical fluids are best used for extraction and sample preparation. Supercritical fluid chromatography, SFC, on the other hand, is a technique that can be useful in certain niche applications [108], but I believe it will not be able to replace HPLC, which will remain the more important technique. The reason is that in SFC one usually uses higher temperatures, compared with HPLC. The separation of difficult-to-separate pairs of isomers, however, is improved at low temperatures, or even below room temperature. Separation selectivity is usually lost if one goes to higher temperatures. For this reason, SFC can not really compete with HPLC, except in certain niche applications. The same is true for the high-temperature GLC of triglyceride isomers.

Chiral separations of diastereomeres on silica HPLC columns are also very useful. These have been investigated mostly by Japanese authors and by Christie et al. in Scotland [110-113]. The analysis of the fatty acid position on the glycerol molecule is rather easy as far as the sn-2 position is concerned, but it is still quite difficult to differentiate between the sn-1 and sn-3 position. On the other hand, very few people - mostly in research - are interested in this information. The separation of sn-OPP from sn-PPO, for example, is not possible by HPLC. However, the separation of diastereomeric UV-absorbing derivatives of diglycerides is possible, and fairly easy, even on a regular silica column [114]. This technique may eventually become more wide-spread and it may replace current enzymatic techniques to differentiate between the sn-1- and sn-3-position.

REFERENCES

1. SEBEDIO JL, ACKMAN RG (1982). Calculation of GLC retention data for some accessible C₂₀ isomeric cis-unsaturated fatty acids. J Chromat Sci, 20: 231-4.

2. DUCHATEAU GSMJE, VAN OOSTEN HJ, VASCONCELLOS MA (1996). Analysis of cis- and trans-fatty acid isomers in hydrogenated and refined vegetable oils by capillary gas-liquid chromatography. JAOCS, 73: 275-82.

3. MORDRET FX, COUSTILLE JL (1995). Gas-liquid chromatography: choice and optimization of operating conditions. In: New trends in lipid and lipoprotein analyses, JL Sebedio, EG Perkins, eds. Champaign: AOCS Press, 133-46.

4. CHARDIGNY JM, SEBEDIO JL, GRANDGIRARD A, MARTINE L, BERDEAUX O, VATELE JM (1996). Identification of novel trans isomers of 20:5n-3 in liver lipids of rats fed a heated oil. Lipids, 31: 165-8.

5. WOLFF RL, BAYARD CC (1995). Improvement in the resolution of individual trans-18:1 isomers by capillary gas-liquid chromatography: use of a 100-m CP-Sil 88 column. JAOCS, 72: 1197-201.

6. DOBSON G, CHRISTIE WW, BRECHANY EY, SEBEDIO JL, LE QUERE JL (1995). Silver ion chromatography and gas chromatography - mass spectrometry in the structural analysis of cyclic dienoic acids formed in frying oils. Chemistry and Physics of Lipids, 75: 171-82.

7. SEBEDIO JL, GRANDGIRARD A (1989). Cyclic fatty acids: natural sources, formation during heat treatment, synthesis and biological properties. Progr Lipid Res, 28: 303-36.

8. MORDRET F, COUSTILLE JL, TACONNE L (1984). Dosage rapide du cholestérol dans les margarines. Rev Fr Corps Gras, 31: 503-7.

9. MORDRET F, LAURENT AM (1978). Appli-cation de la chromatographie en phase gazeuse sur colonne capillaire de verre a l'analyse des tocopherols. Rev Fr Corps Gras, 25: 245-50.

10. PREVOT A (1985). Contaminants à l'état de trace dans les corps gras. Rev Fr Corps Gras, 32: 19-28.

11. WOLFF ROBERT L (1995). Content and distribution of trans-18:1 acids in ruminant milk and meat fats. Their importance in European diets and their effect on human milk. JAOCS, 72: 259-72.

12. PFALZGRAF A, STEINHART H (1995). Origin of trans fatty acids and their analysis in food, tissue and blood plasma. European Journal of Medical Research, 1: 86-8.

13. RATNAYAKE WMN, PELLETIER G (1992). Positional and geometrical isomers of linoleic acid in partially hydrogenated oils. JAOCS, 69: 95-105.

14. RATNAYAKE WMN, CHEN ZY, PELLETIER G, WEBER D (1994). Occurrence of 5c,8c,11c, 15t-eicosatetraenoic acid and other unusual polyunsaturated fatty acids in rats fed partially hydrogenated canola oil. Lipids, 29: 707-14.

15. TARASCHEWSKI H, AITZETMÜLLER K, WERNER G, KÜHS K (1995). Lipids of fish parasites and their hosts: fatty acid fingerprints of four species of acanthocephalans and of their host's intestinal tissues. Parasitol Res, 81: 522-6.

16. AITZETMÜLLER K (1993). Capillary GLC fatty acid fingerprints of seed lipids - A tool in plant chemotaxonomy? J High Resol Chromatogr, 16: 488-90.

17. INNIS SN (1989). Sources of omega-3 fatty acids in arctic diets and their effects on red cell and breast milk fatty acids in Canadian Inuit. In: Dietary omega-3 and omega-6 fatty acids, biological effects and nutritional essentiality,
C Galli, AP Simopoulos, eds. New York: Plenum Press, 135-46.

18. EINIG RG, ACKMAN RG (1987). Omega-3 PUFA in marine oil products. JAOCS, 64: 499-502.

19. GALLI C, SIMOPOULOS AP (1989). Dietary omega-3 and omega-6 fatty acids. New York: Plenum Press.

20. ACKMAN RG (1989). Fatty acids. In: Marine biogenic lipids, fats, and oils, RG Ackman ed. Boca Raton: CRC Press, 103-37.

21. AITZETMÜLLER K, TARASCHEWSKI H, FILIPPONI C, WERNER G, WEBER N (1994). Lipids of fish parasites and their hosts: fatty acids of phospholipids of Paratenuisentis ambiguus and its host eel (Anguilla anguilla). Comp Biochem Physiol, 109B: 383-9.

22. ACKMAN RG, HOOPER SN (1973). Additivity of retention data for ethylenic functions in aliphatic fatty acids. II. Polar liquid phases. J Chromatogr, 86: 83-8.

23. AITZETMÜLLER K, WERNER G, TSEVEGSÜREN N (1993). Screening of seed lipids for gamma-linolenic acid: capillary gas-liquid chromatographic separation of 18:3 fatty acids with DELTA-5 and DELTA-6 double bonds. Phytochem Anal, 4: 249-55.

24. ACKMAN RG (1986). WCOT (Capillary) Gas-liquid chromatography. In: Analysis of Oils and Fats, RJ Hamilton, JB Rossell, eds. London: Elsevier Applied Science, 137-206.

25. SCHULTE E (1993). Gas chromatography of acylglycerols and fatty acids with capillary columns. In: CRC handbook of chromatography, analysis of lipids, KD Mukherjee, N Weber, eds. Boca Raton: CRC Press, 139-48.

26. ANTONIOSI FILHO NR, LANÇAS FM (1993). Identification of FAMEs using ECL values and a tridimensional kovats' retention index system. In: Proc 15th Int Symp Capillary Chro-matogr, P Sandra, ed. Heidelberg: Huethig, 1209-18.

27. ACKMAN RG (1984). Straight-chain fatty acids. In: Handbook of chromatography - lipids, G Zweig, J Sherma, eds. Boca Raton: CRC Press, 95-240.

28. ACKMAN RG (1990). Variability of fatty acids and lipids in Seafoods. Omega 3 News, 5: 1-4.

29. PARADIS M, ACKMAN RG (1977). Potential for employing the distribution of anomalous non-methylene-interrupted dienoic fatty acids in several marine invertebrates as part of food web studies. Lipids, 12: 170-6.

30. JOSEPH JD (1989). Distribution and composition of lipids in marine invertebrates. In: Marine biogenic lipids, fats and oils, RG Ackman, ed. Boca Raton: CRC Press, 49-143.

31. BADAMI RC, PATIL KB (1981). Structure and occurrence of unusual fatty acids in minor seed oils. Progr Lipid Res, 19: 119-53.

32. HEGNAUER R (1990). Chemotaxonomie der Pflanzen. Basel: Birkhäuser.

33. POHL P, WAGNER H (1972). Fettsäuren im Pflanzen- und Tierreich (eine Übersicht) II: Trans-ungesättigte, Alkin-, Hydroxy-, Epoxy-, Oxo-, Cyclopropan- und Cyclopropen-Fettsäu-ren. Fette, Seifen, Anstrichm, 74: 541-50.

34. AITZETMÜLLER K (1996). Was können Pflanzen eigentlich alles?. In: Perspektiven nachwachsender Rohstoffe in der Chemie, H. Eierdanz ed. Weinheim: VCH, 209-17.

35. AITZETMÜLLER K (1996). Seed fatty acids, chemotaxonomy and renewable resources. In: Oils-fats-lipids 1995: proceedings of the 21st world congress of the International Society for Fat Research. High Wycombe : PJ Barnes & Associates, 117-20.

36. AITZETMÜLLER K (1995). DELTA4-, DELTA5- und DELTA6- Fettsäuren. In: Ernährungsphysiologische Eigenschaften von Lebensmitteln, E. Schlimme eds. Münster: Landwirtschaftsverlag, 2-13.

37. AITZETMÜLLER K (1995). Fatty acid patterns of Ranunculaceae seed oils: phylogenetic relationships. Plant Syst Evol (Suppl), 9: 229-40.

38. AITZETMÜLLER K, TSEVEGSÜREN N (1994). Occurrence of gamma-linolenic acid in Ranuncula-ceae seed oils. J Plant Physiol, 143: 578-80.

39. HORROBIN DF (1992). Nutritional and medical importance of gamma-linolenic acid. Progr Lipid Res, 31: 163-94.

40. GUNSTONE FD (1992). Gamma linolenic acid - occurrence and physical and chemical properties. Progr Lipid Res, 31: 145-61.

41. TSEVEGSÜREN N, AITZETMÜLLER K (1996). CGC Study of unusual fatty acids in Saussurea spp. and other Compositae seed oils. In: Proc 18th Int Symp Capillary Chromatogr, P Sandra, G Devos, eds. Heidelberg: Huethig, 1080-6.

42. BAYARD CC, WOLFF RL (1996). Analysis of trans-18:1 isomer content and profile in edible refined beef tallow. JAOCS, 73: 531-3.

43. CHARDIGNY JM, SÉBÉLIO JL, GRANDGIRARD ANDRÉ, MARTINE LUCY, BERDEAUX OLIVIER, VATELE JM (1996). Identification of novel trans isomers of 20:5n-3 in liver lipids of rats fed a heated oil. Lipids, 31: 165-8.

44. BRÜHL L (1995). Determination of trans fatty acids in cold pressed oils. European Journal of Medical Research, 1: 89-93.

45. RATNAYAKE WMN, HOLLYWOOD R, O'GRADY E, BEARE-ROGERS JL (1990). Deter-mination of cis and trans-octadecenoic acids in margarines by gas liquid chromatography-infrared spectrophotometry. JAOCS, 67: 804-10.

46. ADLOF RO, COPES LC, EMKEN EA (1995). Analysis of the monoenoic fatty acid distribution in hydrogenated vegetable oils by silver-ion high-performance liquid chromatography. JAOCS, 72: 571-4.

47. GERTZ C (1996). Trans-Fettsäuren in Lebens-mitteln. Lebensmittelchemie, 50: 49-67.

48. BATTAGLIA R, FRÖHLICH D (1980). HPLC-Separation of cis and trans monounsaturated fatty acids. Chromatographia, 13: 428-31.

49. RATNAYAKE WMN, CHEN Z-Y (1996). Trans, n-3, and n-6 fatty acids in Canadian human milk. Lipids, 31: S279-82.

50. CHEN ZY, RATNAYAKE WMN, FORTIER L, ROSS R, CUNNANE SC (1995). Similar distribution of trans fatty acid isomers in partially hydrogenated vegetable oils and adipose tissue of Canadians. Can J Physiol Pharmacol, 73: 718-23.

51. RATNAYAKE WMN, CHEN ZY (1995). Trans fatty acids in Canadian breast milk and diet. In: Development and processing vegetable oils for human nutrition, BE McDonald, R Przybylski, eds. Champaign: AOCS Press, 20-35.

52. BRÜHL L, SCHULTE E, THIER H-P (1993). Fraktionierung der Triglyceride von Mutter-milch durch HPLC an einer Silberionensäule und an RP-18-Material mit dem Lichtstreu-Detektor. Fat Science and Technology, 95: 370-6.

53. AITZETMÜLLER K, GUARALDO GONCALVES LA (1990). Dynamic impregnation of silica stationary phases for the argentation chromatography of lipids. J Chromatogr, 519: 349-58.

54. XIN YAONIAN, AITZETMÛLLER K (1991). The structure analysis of triglycerides of Chinese vegetable tallow with HPLC. China Oils and Fats (Xian), 70 (1): 23-30.

55. TSEVEGSÜREN N, AITZETMÛLLER K (1996). CGC investigation of Cimicifuga spp. seed oils. In: Proc 18th Int Symp Capillary Chromatogr, P Sandra, G Devos, eds. Heidelberg: Huethig, 1087-95.

56. XIN Y, AITZETMÜLLER K (1991). Application of lipozyme in modification of Chinese tallow. China Oils and Fats (Xian), 75 (6): 25-8.

57. XIN Y, AITZETMÜLLER K (1992). The modification of Chinese vegetable tallow by inter-esterification with enzyme. J Chinese Cereals Oils Assoc (Beijing), 7: 55-62.

58. TSEVEGSÜREN N, AITZETMÜLLER K (1996). CGC Study of unusual fatty acids in Saussurea spp. and other Compositae seed oils. In: Proc 18th Int Symp Capillary Chromatogr, P Sandra, G Devos, eds. Heidelberg: Huethig, 1080-6.

59. GOIFFON J-P, REMINIAC C, FURON D (1981). Application de la chromatographie liquide haute performance a l'analyse des triglycerides des corps gras. II - Grandeurs de retention des triglycerides. Rev Fr Corps Gras, 28: 199-207.

60. DEFFENSE E (1984). Application de la chromatographie liquide haute performance a l'analyse des triglycerides des graisses végétales et animales et de leurs fractions obtenues par cristallisation fractionnée. Rev Fr Corps Gras, 31: 123-9.

61. AITZETMÜLLER K (1986). Application of HPLC to the separation of lipids. In: Practice of high performance liquid chromatography, edited by H Engelhardt, ed. Berlin: Springer, 287-321.

62. AITZETMÜLLER K, GRÖNHEIM M (1992). Separation of highly unsaturated triacylglycerols by reversed-phase HPLC with short wavelength UV detection. J High Resol Chromatogr, 15: 219-26.

63. SEHER A, FIEBIG H-J (1983). Stereospezifische Analyse von Triacylglycerinen mit selten vorkommenden Fettsäuren (Stereospecific analysis of triacylglycerides with rarely occurring fatty acids). Fette, Seifen, Anstrichm, 85: 333-8.

64. AITZETMÜLLER K (1991). Anwendungsmögli-chkeiten der HPLC in der modernen Fettana-lytik. Fat Sci Technol, 93: 501-10.

65. AITZETMÜLLER K (1982). Recent progress in the high performance liquid chromatography of lipids. Prog Lipid Res, 21: 171-93.

66. AITZETMÜLLER K (1988). HPLC in the fats and oils laboratory. (Hilditch memorial lecture, 1987). Chem and Ind (London), 1988: 452-70.

67. AITZETMÜLLER K (1990). Flow gradients in the HPLC analysis of triacylglycerols in fats and oils. J High Resol Chromatogr, 13: 375-8.

68. FREDE E (1986). Improved HPLC of trigly-cerides by special tempering procedures. Chromatographia, 21: 29-36/189-92.

69. AITZETMÜLLER K, GRÖNHEIM M (1993). Gradient elution HPLC of fats and oils with laser light scattering detection. Fat Sci Technol, 95: 164-8.

70. PERRIN JL, PREVOT A, TRAITLER H, BRACCO U (1987). Analysis of triglyceride species of blackcurrant seed oil by HPLC via a laser light scattering detector. Rev Fr Corps Gras, 34: 221-3.

71. PERRIN JL, PREVOT A (1986). Use of a laser-light-scattering detector in the HPLC analysis of fats. II. Analysis of the triglycerides in oils and fats. Rev Fr Corps Gras, 33: 437-45.

72. ABIDI SL, MOUNTS TL, FINN T (1996). A preferred solvent system for high-performance liquid chromatographic analysis of soybean phospholipids with evaporative light-scattering detection. JAOCS, 73: 535-6.

73. PICCHIONI GA, WATADA AE, WHITAKER BD (1996). Quantitative HPLC analysis of plant phospholipids and glycolipids using light-scattering detection. Lipids, 31: 217-21.

74. AITZETMÜLLER K (1973). Frontal elution liquid chromatography of a total artefacts peak in frying oils. J Chromatogr, 79: 329-34.

75. AITZETMÜLLER K, XIN Y, WERNER G, GRÖNHEIM M (1992). High-performance liquid chromatographic investigations of stillingia oil. J Chromatogr, 603: 165-73.

76. PERRIN J-L, REDERO F, PREVOT A (1984). Dosage rapide des polymeres de triglycerides par chromatographie d'exclusion. Rev Fr Corps Gras, 31: 131-3.

77. UNBEHEND M, SCHARMANN H, STRAUSS H-J, BILLEK G (1973). Anwendung der Gelpermeationschromatographie auf die Untersuchung thermisch-oxidativ belasteter Fette. Fette Seifen Anstrichm, 75: 689-96.

78. MARQUEZ-RUIZ G, PEREZ-CAMINO MC, DOBARGANES MC (1990). Combination of adsorption and size exclusion chromatography for the determination of fatty acid monomers, dimers and polymers. J Chromatogr, 514: 37-44.

79. DOBARGANES MC, PEREZ-CAMINO MC, MARQUEZ-RUIZ G (1988). High performance size exclusion chromatography of polar compounds in heated and non-heated fats. Fat Sci Technol, 90: 308-11.

80. DIEHL BWK, OCKELS W (1995). Ermittlung der Fettsäureverteilung durch ¹³C-NMR-Spektroskopie. Fat Sci Technol, 97: 115-8.

81. GUNSTONE FD (1993). High resolution ¹³C-NMR spectroscopy of lipids. In: Advances in lipid methodology - Two, WW Christie ed. Dundee: The Oily Press, 1-68.

82. NG SOON (1985). Analysis of positional distribution of fatty acids in palm oil by ¹³C NMR spectroscopy. Lipids, 20: 778-82.

83. WOLLENBERG KF (1990). Quantitative high resolution ¹³C nuclear magnetic resonance of the olefinic and carbonyl carbons of edible vegetable oils. JAOCS, 67: 487-94.

84. AITZETMÜLLER K, DIEHL B, HERLING H, OCKELS W (1994). Investigation of seed oils containing DELTA5- and DELTA6- fatty acids using ¹³C-NMR spectroscopy. Poster presented at: 2nd EUROLIPID and 59th DGF Congress, Münster, Sept 94.

85. ACKMAN RG (1988). Some possible effects on lipid biochemistry of differences in the distribution on glycerol of long-chain n-3 fatty acids in the fats of marine fish and marine mammals. Atherosclerosis, 70: 171-3.

86. MUDERHWA JM, DHUIQUE-MAYER C, PINA M, GALZY P, GRIGNAC P, GRAILLE J (1987). Répartition interne/externe des acides gras des triglycérides de quelques huiles gamma linoléniques (Internal and external distribution of triglyceride fatty acids in a few gamma-linolenic oils). Oléagineux, 42: 207-11.

87. MUKHERJEE KUMAR D, KIEWITT IRMGARD, HILLS MATTHEW J (1993). Substrate specificities of lipases in view of kinetic resolution of unsaturated fatty acids. Applied Microbiology and Biotechnology, 40: 489-93.

88. MORDRET F, COUSTILLE J-L, TACONNE L (1984). Dosage rapide du cholestérol dans les margarines. Rev Fr Corps Gras, 31: 503-7.

89. MORDRET F, AJANA H, GAUCHET C (1985). Fractionnement de l'insaponifiable par CLHP. Isolement de la fraction stérolique. Revue Fr Corps Gras , 32: 305-10.

90. SCHULTE ERHARD, WEBER NIKOLAUS (1987). Analysis of disteryl ethers. Lipids, 22: 49-1049-52.

91. HOMBERG E, BIELEFELD B (1987). Verglei-chende Steringehaltsbestimmungen in Fetten und Ölen Teil I: Gaschromatogra-phische Methoden. Fat Sci Technol, 89: 255-8.

92. ARENS M, FIEBIG H-J, HOMBERG E (1990). Sterine (Isolierung und GC-Untersuchung) - Gemeinschaftsarbeiten der DGF. 114. Mitt: Deutsche Einheitsmethoden zur Untersuchung von Fetten, Fettprodukten, Tensiden und verwandten Stoffen; 86. Mitt.: Analyse von Fett-begleitstoffen VII. Fat Sci Technol, 92: 189-92.

93. HOMBERG E (1975). Veränderung der Sterine durch industrielle Verarbeitungspro-zesse von Fetten und Ölen II: Veränderung- sprodukte bei der Behandlung von Cholesterin mit Bleicher-den. Fette Seifen Anstrichmittel, 77: 8-11.

94. GALLINA TOSCHI T, BENDINI A, LERCKER G (1996). Evaluation of 3,5-stigmastadiene content of edible oils: comparison between the traditional CGC method and the on-line HPLC-CGC analysis. Proc 18th Int Symp Capillary Chromatogr, 3: 2236-43.

95. BRÜHL L, FIEBIG H-J (1995). Qualitätsmerkmale Kaltgepresster Speiseöle. Fat Sci Technol, 97: 203-8.

96. LANZON A, CERT A, ALBI T (1989). Deteccion de la presencia de aceite de oliva refinado en el aceite de oliva virgen. Grasas y Aceites, 40: 385-8.

97. SCHULTE E (1994). Determination of edible fat refining by HPLC of delta-3,5-steradienes. Fat Sci Technol, 96: 125-8.

98. BALZ M, SCHULTE E, THIER H-P (1992). Trennung von Tocopherolen und Tocotrieno-len durch HPLC. Fat Sci Technol, 94: 209-13.

99. AITZETMÜLLER K, ARZBERGER E (1981). HPLC of fat soluble vitamins. (Poster presentation). Fifth international symposium on column liquid chromatography; Avignon, June 11-5.

100. AITZETMÜLLER K (1994). HPLC von vitamin A-Estern in Margarine. Fat Sci Technol, 96: 279-85.

101. BALZ M, SCHULTE E, THIER H-P (1995). Simultaneous determination of retinol esters and tocochromanols in foods using nitro-column HPLC. Fat Sci Technol, 97: 445-8.

102. AITZETMÜLLER K (1989). Analysis of chlorophyll degradation products in fats and oilseeds. Actes du congrès international «Chevreul» pour l'étude des corps gras (Proceedings of the Chevreul International Congress), Angers, 06-10 juin 1989, 177- 84.

103. AITZETMÜLLER K, KÔNKER S (1991). Möglichkeiten zum Nachweis einer Hitzeein-wirkung auf Rapssaat in Zusammen-hang mit Glucosinolat-Bestimmungs-Metho-den. Fat Sci Technol, 93: 256-64.

104. SEBEDIO J-L (1993). High performance liquid chromatography of fatty acids. In: CRC handbook of chromatography, analysis of lipids, KD Mukherjee, N Weber, eds. Boca Raton: CRC Press, 57-70.

105. GROB K, LANFRANCHI M, MARIANI C (1990). Evaluation of olive oil through the fatty alcohols, the sterols and their esters by coupled LC-GC. JAOCS, 67: 626-34.

106. WAHL HG, HABEL S-Y, SCHMIEDER N, LIEBICH HM (1994). Identification of cis/trans Isomers of methyl ester and oxazoline derivatives of unsaturated fatty acids using GC-FTIR-MS. J High Resol Chromatogr, 17: 543-8.

107. KALLIO H, RUA P (1994). Distribution of the major fatty acids of human milk between sn-2 and sn-1,3 positions of triacylglycerols. JAOCS, 71: 985-92.

108. MANNINEN PEKKA, LAAKSO PƒIVI, KALLIO HEIKKI (1995). Method for characterization of triacylglycerols and fat-soluble vitamins in edible oils and fats by supercritical fluid chromatography. JAOCS, 72: 1001-8.

109. VERLEYSEN K, SCÜCS R, SANDRA P (1996). MEKC versus LC for the analysis of phospholipids in lecithins. In: Proc. 18th Int Symp Capillary Chromatogr, P Sandra, G Devos, eds, Heidelberg: Huethig, 2155-63.

110. LAAKSO PAIVI, CHRISTIE WILLIAM W (1990). Chromatographic resolution of chiral diacylglycerol derivatives: potential in the stereo-specific analysis of triacyl-sn-glycerols. Lipids, 25: 349-53.

111. CHRISTIE WW, NIKOLOVA-DAMYANOVA B, LAAKSO P, HERSLOF B (1991). Stereospecific analysis of triacyl-sn-glycerols via resolution of diastereomeric diacylglycerol derivatives by high-performance liquid chromatography on Silica. JAOCS, 68: 695-701.

112. ANDO Y, NISHIMURA K, AOYANAGI N, TAKAGI T (1992). Stereospecific analysis of fish oil triacyl-sn-glycerols. JAOCS, 69: 417-24.

113. TAKAGI T, ANDO Y (1991). Stereospecific analysis of triacyl-sn-glycerols by chiral high-performance liquid chromatography. Lipids, 26: 542-7.

114. CHRISTIE WW (1992). The chromatographic resolution of chiral lipids. In: Advances in lipid methodology - One, WW Christie, ed. Dundee: The Oily Press, 121-48.