Progress 01/01/03 to 12/31/03
An earlier accomplishment in this project was the development of a procedure for the bulk purification of active emulsifier components from gum arabic. Commercial gum arabic was mixed with hexadecane to form an emulsion which was then diluted and density-fractionated to obtain the approximately 1% of the gum arabic molecules associated with oil droplets, these being the emulsifiers. The last report outlined our initial attempts to use reversed-phase chromatography to further fractionate this 1% of gum arabic. Several varying attempts at reversed-phase fractionation since then consistently produced two fractions, one that passed through with little affinity for the reversed-phase column and another that bound to the column and eluted with acetonitrile. The two fractions were very similar by both chemical analysis and emulsification assay. Our interpretation of these results is that the molecules have a considerable tendency to self-associate, putting their hydrophobic
sites towards the center of the aggregate where they cannot interact with the reversed-phase column. Our attempts to find solvent additives to completely solubilize the gum arabic molecules have not been successful. Among urea, guanidine hydrochloride, methylsulfoxide, and lithium-3,5-diiodosalicylate (all powerful chaotropic reagents), only urea holds any promise, but even this additive does not completely solubilize the 1% gum arabic fraction. It appears that the other 99% of gum arabic, while not directly associating with oil droplets, assists in emulsification by solubilizing the 1% of the molecules that do associate with oil droplets. In our investigation of gums of Venezuelan origin, we reported last year that the gums from ACACIA TORTUOSA and ACACIA MACRACANTHA have higher lipid content and are more potent emulsifiers than gum arabic from ACACIA SENEGAL. We now report that we have used anion-exchange chromatography to fractionate these gums, as well as the gum from ACACIA
GLOMEROSA. All three of these Venezuelan gums yield fractions that are chemically distinct, for example by protein content, but all of which have significant emulsifying activity. By the criteria of carbohydrate composition (content of arabinosyl, galactosyl, and glucuronosyl residues), the A. TORTUOSA and A. MACRACANTHA gums are somewhat similar to gum arabic. Remarkably, however, the arabinogalactan-protein content of either of these Venezuelan gums is less than 2% that of gum arabic. Our limited analyses of other Venezuela gums, those from SPONDIAS CYTHEREA, ENTEROLOBIUM CYCLOCARPUM, and HYMENEA COURBARRIL, have not revealed them to be exceptional emulsifier candidates. We are also examining plant phytoglycolipids as emulsifier candidates. These molecules are ceramide-containing lipids with oligosaccharide head groups. Purification of these phytoglycolipids by methods that might be commercially practical is challenging, but we have made some progress using counter-current
distribution. Some of the fractions thus obtained appear to be superior emulsifiers when compared with gum arabic on a weight-to-weight basis. Our chemical analyses of these phytoglycolipids are ongoing.
This project involves several approaches to improving our understanding of how emulsifiers work and to broadening the sources of such emulsifiers. The applications of gum arabic, the well-known emulsifier studied here, and the potential applications of the novel emulsifiers are in food and cosmetic products. The project aims towards improving the quality and/or reducing the cost of these products.
- No publications reported this period
Progress 12/01/00 to 11/30/03
Gum arabic, an exudate from ACACIA SENEGAL trees, has a unique combination of excellent emulsifying properties and low solution viscosity. These properties make gum arabic very useful in several industries but especially in the food industry where it is used as a flavor encapsulator and stabilizer of citrus oil emulsion concentrates in soft drinks. The molecular basis of the emulsifying activity of gum arabic is not adequately understood. The hypothesis of this project has been that the emulsifying activity is due to a subset of the molecules in the mixture that comprises this natural product. Many of the molecules in the mixture are arabinogalactan proteins, and the hypothesis holds that a subset of these arabinogalactan-proteins have covalently attached glycosylphosphatidylinositol (GPI) lipid, and it is these lipid-modified arabinogalactan-proteins that are most responsible for the emulsifying activity. Such molecules would spontaneously localize at an oil-water
interface, orientating with the polar arabinogalactan carbohydrate chains in the water and the nonpolar lipid group in the oil. To test this hypothesis, chemical treatments expected to cleave GPI lipid anchors were applied to gum arabic, and the resulting effects on carbohydrate composition and emulsifying activity were examined. Treatment of gum arabic with nitrous acid resulted in diminished emulsifying activity without affecting the principal carbohydrate composition. Treatment with 50% aqueous HF at 0 degrees Celsius resulted in diminished emulsion properties but also detectable effects on glycosyl composition. The approximately 1-3% subfraction of gum arabic components that adsorb at the surface of oil droplets has higher abundance of GPI linker components, much higher relative lipid content, and somewhat higher emulsifying action than the whole gum. These results are generally consistent with the hypothesis and have been written into a manuscript and submitted for publication,
with acceptance now pending on suitable revision of the manscript. Other work aimed to determine if highly effective emulsifiers could be synthesized by chemical attachment of exogenous lipids to the large proportion of gum arabic molecules that do not contain endogenous lipid groups. Best reaction conditions resulted in attachment of about six stearic acid molecules to each gum arabic molecule, and this modified gum arabic had approximately 40% higher emulsifying activity than native gum arabic. In collaboration with Venezuelan scientists, emulsifying activities and structures of gums from several Venezuelan species were analyzed. The gums from ACACIA TORTUOSA and ACACIA MACRACANTHA both had much higher lipid content and were more potent emulsifiers than gum arabic. Plant phytoglycolipids were also examined as emulsifier candidates. These molecules are ceramide-containing lipids with oligosaccharide head groups. Purification of phytoglycolipids by commercially viable methods was
challenging, but progress was made using counter-current distribution. Some of the fractions thus obtained were superior emulsifiers when compared with gum arabic on a weight-to-weight basis.
Gum arabic is a natural product with economic importance because of its use as an emusifying agent and flavor encapsulator in the food and beverage industry. At present, the U.S. is entirely dependent upon import of gum arabic from the politically sensitive areas of Africa where ACACIA SENEGAL trees, the source of gum arabic, are grown. This project involved several approaches to improving our understanding of the molecular basis behind the emulsifying activity of gum arabic, and then using this knowledge to improve gum arabic so the same effect can be obtained with less material, and to identify alternative sources of comparable or better emulsifiers. The results obtained provide new information about lipid components of the gum arabic molecules that greatly influence emulsifying activity. Synthetically increasing the amount of lipid components increases the emulsifying activity so emulsions can be stabilized with less gum arabic. Gums from certain Venezuelan trees
have characteristics that could make them useful as emulsifiers that are competitive with, or even superior to, gum arabic. Phytoglycolipids, which can be extracted from many types of plants and plant parts, also hold promise as effective emulsifiers for the food industry.
- Yadav, M.P. and Nothnagel, E.A. 2004. Highly effective emulsifier subfraction of gum arabic. Chem. Eng. News 82(31):4-TECH.
Progress 01/01/02 to 12/31/02
In the first year of the project, we developed a novel method for the bulk separation of active emulsifier components from gum arabic. This method relied on emulsification itself and involved upward flotation of emulsifier-coated hexadecane droplets in a high density solution. The active emulsifier components were recovered from the floated creamy layer, leaving the non-emulsifier components of gum arabic dissolved in the solution below. Analysis of the emulsifier components showed that glucosamine, inositol, and phosphorous were about two times more abundant in the creamy layer than in total gum arabic, and that fatty acids were eight to twelve times more abundant than in total gum arabic. These results were consistent with our hypothesis that a glycosylphosphatidylinositol lipid is present on some gum arabic components and is important for emulsifying activity. Continuing this line of work in the second year, we applied the flotation procedure to three different gum
arabic samples and analyzed the resulting creamy layer fractions for emulsifying activity. As expected, in each case the creamy layer fraction was a more potent emulsifier than the corresponding original gum. The enhancement of emulsifying activity, however, was only about two-fold based on the amount of gum fraction required to emulsify a standard amount of oil. This enhancement was less than we expected and indicated that further purification of the creamy layer fraction was required. Thus, we applied reversed-phase chromatography to the creamy layer fraction. The initial run of this column was encouraging in that approximately 60% of the creamy layer material passed through quickly, not adhering to the column, and about 30% of the material eluted much later as expected for lipid-containing molecules. The results of our chemical composition and emulsifier activity analyses were only partially as expected, however, and this caused us to suspect anomalous behavior of the creamy layer
molecules on the column. We found that if the creamy layer molecules that passed rapidly through the column were collected and reapplied to the column, much of the material would again pass through quickly, but approximately 20% of the material would adhere and elute later. These and other results caused us to hypothesize that the emulsifier molecules have a strong tendency to self-associate through their hydrophobic domains, thereby preventing their interaction with the reversed phase column. In this model, only the monomers adhere to the column. If this hypothesis is correct, it could have important implications for understanding the emulsifying activity of gum arabic. In another approach to finding and understanding better emulsifiers, we are collaborating with Venezuelan scientists to analyze the emulsifying activities and structures of some gums from Venezuelan species. We have focused thus far on gums from ACACIA TORTUOSA and ACACIA MACRACANTHA and have found that both gums have
much higher (at least 10X) lipid content than gum arabic from ACACIA SENEGAL, and both are more potent emulsifiers than gum arabic, particularly when used at low gum-to-oil ratios.
The results obtained in this period indicate that a subfraction of the molecules in gum arabic have a strong tendency to interact in a manner that limits exposure of their hydrophobic domains. If a method can be devised to reduce this interaction, it is likely that a more efficient emulsifier would result. The studies on Venezuelan gums suggest that these natural products could prove to be superior to gum arabic as emusifiers.
- No publications reported this period
Progress 01/01/01 to 12/31/01
The central hypothesis of this project is that the valuable emulsifying activity of gum arabic is due to a subset of the molecules in the mixture that comprises this natural product. Many of the molecules in the gum arabic mixture are arabinogalactan-proteins, and the hypothesis holds that a subset of these arabinogalactan-proteins have covalently attached glycosylphosphatidylinositol lipid, and it is these lipid-modified arabinogalactan-proteins, in particular, that are responsible for the emulsifying activity. These molecules would likely spontaneously localize at an oil-water interface, orientating such that the polar arabinogalactan carbohydrate chains are in the water and the nonpolar lipid group is in the oil. This capacity to occupy an oil-water interface would enable these lipid-modified arabinogalactan-proteins to coat oil droplets and prevent their fusion, this being the key functional property of effective emulsifiers of dilute oil-in-water mixtures. One
specific aim in the current period was to fractionate gum arabic and assay the resulting fractions for emulsifying activity and for the structural components of glycosylphosphatidylinositol lipid anchors. We developed a very effective fractionation based on preparing a concentrated emulsion of gum arabic with pure hexadecane and then floating this emulsion in a concentrated solution of potassium bromide. The creamy layer is collected from the surface, leaving the non-emulsifier components of gum arabic dissolved in the potassium bromide solution below. The emulsion in the creamy layer is then broken by the addition of isopropanol, and the hexadecane is removed by washes with organic solvents. The mass of gum components recovered in the creamy layer amounted to 0.65-0.75% of the mass of total gum emulsified with the hexadecane, but this subset of gum molecules was enriched with eight to twelve times more covalently-attached fatty acid lipids than the total gum. These results are
consistent with the central hypothesis and show that the flotation procedure is quite effective in purifying the active emulsifier molecules. Another area of emphasis was to determine if highly effective emulsifiers can be synthesized by chemical attachment of exogenous lipids to the large proportion of gum arabic molecules that do not contain endogenous lipid groups. The best reaction condition that we have found so far results in attachment of an average of six stearic acid molecules to each gum arabic molecule. Emulsifier assays indicate that gum arabic modified in this way is approximately 40% more effective than native gum arabic.
These studies are helping us understand how gum arabic acts as an emulsifier in dilute oil-in-water emulsions. This knowledge is being applied to develop methods for chemical modification of gum arabic to further improve its emulsifier activity and will also be useful in identifying promising sources of emulsifiers from other natural sources. These enhanced emulsifiers should be economical for use in the food industry.
- NOTHNAGEL, E.A. and YADAV, M.P. 2001. Lipid components of arabinogalactan proteins and gum arabic. Chem. Eng. News 79(10):85.
- YADAV, M. P. and NOTHNAGEL, E.A. 2001. Carbohydrate and lipid composition of effective emulsifer fractions of gum arabic. Chem. Eng. News 79(10):83.