Following the feeding of a triacylglycerol-rich meal to healthy adult human beings, duodenal contents were aspirated for ex vivo chemical and physical-chemical analyses. The aspirates were collected during established lipid digestion and absorption into a "cocktail" of chemical inhibitors that rapidly inhibited ex vivo lipolysis. Following ultracentrifugation, the lipids separated into a floating oil layer, several interfacial layers, a "clear" or turbid "subphase", and a precipitated "pellet". By chemical and phase analyses, the floating layer was composed of oil-in-water emulsion particles with cores of triacylglycerol (TG), diacylglycerols (DG), and cholesteryl esters (CE) emulsified with a surface coat of partially ionized fatty acids (FA), monoacylglycerols (MG), diacylphosphatidylcholine (PL), and bile salts (BS). The interfacial layers contained similar emulsion particles dispersed among excess emulsifier which adopted a lamellar liquid-crystalline structure. Precipitated pellets were composed principally of emulsifying lipids, with smaller amounts of crystalline calcium soaps and BS. Relative lipid compositions of all but three subphases fell within a two-phase region of the condensed ternary phase diagram (Staggers et al., 1990, companion paper) where saturated mixed micelles composed of BS, FA "acid-soaps", MG, PL, cholesterol (Ch), and traces of DG (and TG) coexisted with unilamellar liquid-crystalline vesicles composed of the same lipids. Attempts to achieve clean separation of vesicles from micelles by repeat ultracentrifugation failed. Compared with the structure and sizes of lipid particles in equilibrated model systems (Staggers et al., 1990), quasielastic light scattering (QLS) analysis revealed that ex vivo micellar sizes (mean hydrodynamic radii, Rh) were similar (less than or equal to 40 A), whereas unilamellar vesicle sizes (Rh = 200-600 A) were appreciably smaller. Two-component QLS analysis of the subphases showed that much larger proportions of lipids were solubilized by micelles than were dispersed as unilamellar vesicles. When followed as functions of time, vesicles frequently dissolved spontaneously into mixed micelles, indicating that, in the nonequilibrium in vivo conditions, the constituent micellar phase was often unsaturated with lipids. These results are consistent with the hypothesis that, during hydrolysis of emulsified DG and TG by luminal lipases, unilamellar vesicles originate in lamellar liquid crystals that form at emulsion-water interfaces in the upper small intestine. In a BS-replete environment, unilamellar vesicles probably represent the primary dispersed product phase of human fat digestion and facilitate the dissolution of lipolytic products into unsaturated mixed micelles.(ABSTRACT TRUNCATED AT 400 WORDS)

We developed equilibrium phase diagrams corresponding to aqueous lipid compositions of upper small intestinal contents during lipid digestion and absorption in adult human beings. Ternary lipid systems were composed of a physiological mixture of bile salts (BS), mixed intestinal lipids (MIL), principally partially ionized fatty (oleic) acid (FA) plus racemic monooleylglycerol (MG), and cholesterol (Ch), all at fixed aqueous-electrolyte concentrations, pH, temperature, and pressure. The condensed phase diagram for typical physiological conditions (1 g/dL total lipids, FA:MG molar ratio of 5:1, pH 6.5, 0.15 M Na+ at 37 degrees C) was similar to that of a dilute model bile [BS/lecithin (PL)/Ch] system [Carey, M. C., & Small, D. M. (1978) J. Clin. Invest. 61, 998-1026]. We identified two one-phase zones composed of mixed micelles and lamellar liquid crystals, respectively, and two two-phase zones, one composed of Ch monohydrate crystals and Ch-saturated micelles and the other of physiologic relevance composed of Ch- and MIL-saturated mixed micelles and unilamellar vesicles. A single large three-phase zone in the system was composed of Ch-saturated micelles, Ch monohydrate crystals, and liquid crystals. Micellar phase boundaries for otherwise typical physiological conditions were expanded by increases in total lipid concentration (0.25-5 g/dL), pH (5.5-7.5), and FA:MG molar ratio (5-20:1), resulting in a reduction of the size of the physiological two-phase zone. Mean particle hydrodynamic radii (Rh), measured by quasielastic light scattering (QLS), demonstrated an abrupt increase from micellar (less than 40 A) to micelle plus vesicle sizes (400-700 A) as this two-phase zone was entered. With relative lipid compositions within this zone, unilamellar vesicles formed spontaneously following coprecipitation, and their sizes changed markedly as functions of time, reaching equilibrium values only after 4 days. Further, vesicle Rh values were influenced appreciably by MIL:mixed bile salt (MBS) ratio, pH, total lipid concentration, and FA:MG ratio, but not by Ch content. In comparison, micellar systems equilibrated rapidly, and their Rh values only slightly influenced by physical-chemical variables of physiological importance. In contrast to the BS-PL-Ch system [Mazer, N. A., & Carey, M. C. (1983) Biochemistry 22, 426-442], no divergence in micellar sizes occurred as the micellar phase boundary was approached. The ionization state of FA at simulated "intestinal" pH values (5.5-7.5) in the micellar and physiologic two-phase zones was principally that of 1:1 sodium hydrogen dioleate, an insoluble swelling "acid soap" compound. By phase separation and analysis, tie-lines for the constituent phase in the two-phase zone demonstrated that the mixed micelles were saturated with MIL and Ch and the coexisting vesicles were saturated with MBS, but not with Ch.(ABSTRACT TRUNCATED AT 400 WORDS)

No abstract