Does the lipase source or oil composition influence enzymatic interesterification!

International News on Fats, Oils and Related Materials : INFORM, Oct 2009 by Cowan, David

Editor's note: The following article is adapted from the author's presentation delivered during the General Processing session at the 1 00th AOCS Annual Meeting & Expo, held May 3-6, 2009, in Orlando, Florida, USA.

Enzymatic interesterification (EIE) is no longer an unknown technology but one that is widely applied. As experience in the field has grown, researchers have learned that not all oil types and lipases interact in the same manner. The following article highlights some of the recent research in this area, aimed at better understanding these interactions and their practical consequences.

OIL QUALITY CRITERIA

One of the major areas of research relating to EIE is the elucidation of the influence of the various oil quality factors on enzyme performance. The understanding of these is critical in achieving an acceptable economy, and product quality, for EIE so that it can compete with other fat- modification technologies. The quality specifications (see Table 1) are very similar to those of oil blends for chemical interesterification (CIE) but not always for the same reasons.

For example, a low FFA (free fatty acid) content is recommended, not because the enzymatic process is sensitive to FFA, but more as an indication of whether the oil has been properly refined and stored. Oxidation product levels are important as these components can interact with the enzyme protein and damage its activity and working life. High levels of moisture are not normally regarded as an issue for enzyme working life, but they do produce loss of yield through hydrolysis of the fats.

The most critical parameter for EIE, which is different from oil blends for CIE, is the presence of inorganic acid residues, derived from phosphoric acid degumming and/or from acid- activated bleaching earths. This can result in small amounts of strong acid becoming dissolved in water entrained in the oil blend. Passage down the immobilized enzyme column results in the uptake of this water by the granules; the internal ? H then drops, reducing the activity of the enzyme and ultimately its working life. The discovery of this factor and the development of a simple analytical technique to demonstrate the presence of this non- fatty acid acidity have made a major contribution to understanding and improving enzyme performance.

ARE ALL OILS AND ENZYMES EQUAL?

When any new process is developed, the level of understanding of all the parameters influencing it progresses as more research is conducted. As research into EIE advanced, it became clear that not all oil blends react in exactly the same manner. Looking for parallels with the closest technology to EIE (i.e., CIE) revealed that the degree of saturation of the fats influences the completeness of the randomization obtained in CIE. In investigating further, we considered this fact as well as the lipase type and the fatty acid chain length.

Laurie fats in particular take longer to achieve the maximal degree of interesterification by EIE (as measured by change in solid fat content). Using a batch reaction setup and 4% Thermomyces lanuginosus lipase, palm olein reached its maximum much more quickly than a blend of 85% palm kernel oil and 15% palm stearine (Fig. 1).

Substituting the T. lanuginosus lipase with one produced in Candida antarctica gave similar results with the lauric oilcontaining blend. Full conversion was reached more slowly than with the nonlauric blend.

The lauric fats such as palm kernel and coconut oil contain much higher levels of C12 fatty acids than palm stearine or soybean oil, and, as lipases are normally characterized by their preferences for chain length, this could be a possible explanation for the observed differences. Two lipase types having preferences for either longer- or shorter-chain fatty acids were compared using three different oil blends. The blends (50% palm olein/50% coconut oil, 80% palm stearine/20% coconut oil, or 50% palm stearine/50% coconut oil) were chosen to provide a range of fatty acid compositions. As these were commercially derived oil blends, produced using phosphoric acid degumming and activated bleaching earth, residual mineral acid contents were high and this parameter was expected to affect productivity (kg oil converted/kg enzyme).

Lipase from C antarctica B, which is known to be more acid tolerant, exhibited a higher productivity when used for hydrolysis than that from T. lanuginosus, which has an alkaline pH optimum. This suggests that, for oils with high residual inorganic acid contents, an acidophilic lipase could be a useful candidate for EIE. When the degree of saturation of the fats was considered, it was not possible to observe any major differences; the level only varied from 70.6 to 79.1% in the fat blends used. However, some lipases have different rates of reaction when highly saturated fats are used.

The rate of reaction using these three blends and the two lipase types showed some differences between both blend and enzyme. In both cases reaction rates were slower on the blends containing 50% lauric fats than on the blend with only 20% inclusion. This was more noticeable for the T. lanuginosus lipase than for the C. antarctica B enzyme. Overall, the rate of reaction for the C. antarctica B lipase was lower than that of the standard T. lanuginosus enzyme, but it was less sensitive to the fatty acid composition.