The Procter & Gamble Company, Winton Hill Technical Center, Cincinnati, OH 45224
Olestra is a mixture of hexa-, hepta- and octaesters of sucrose formed from long-chain fatty acids from edible oils. Olestra is not hydrolyzed by gastric or pancreatic enzymes (Mattson and Volpenhein 1972
) and consequently is not absorbed from the gut (Miller et al. 1995). Olestra is approved as a replacement for fat used in the preparation of savory snacks (Federal Register 1996). Because olestra is lipophilic and is not absorbed, it can affect the absorption of lipophilic nutrients by making them less available to the micelle-mediated process by which they are absorbed (Jandacek 1982). This interference occurs because nutrients which are sufficiently lipophilic partition between the nonabsorbed olestra phase and the intestinal micelles in the gastrointestinal (GI)4 tract. That portion that partitions into the olestra then passes through the GI tract with the nonabsorbed olestra, thus reducing the total amount absorbed.
Studies in animals and humans consuming olestra have shown that olestra can interfere with the absorption of lipophilic nutrients. Human studies showed that olestra reduced serum concentrations of carotenoids, 
-tocopherol, and 25-hydroxyergocalciferol [25(OH)D2], the metabolite of dietary vitamin D, but did not affect vitamin K status (Jones et al. 1991a and 1991b, Koonsvitsky et al. 1997, Schlagheck et al. 1997a) or the absorption of retinyl palmitate (Daher et al. 1997). Decreases in serum cholesterol concentration (Crouse and Grundy 1979) and increases in fecal cholesterol excretion (Jandacek et al. 1990) have been measured in humans eating olestra. Studies in the pig showed that olestra reduced tissue and circulating levels of vitamins A, D and E but had no effect on the status of vitamin K or the status of water-soluble nutrients (Cooper et al. 1997b-d). Studies in humans (Schlagheck et al. 1997a) and pigs (Cooper et al. 1997a) have shown that the effects of olestra on the status of vitamins A, D and E can be offset by adding extra amounts of these vitamins to olestra or olestra-containing foods.
Eating olestra mixed homogeneously with other dietary components, at every meal, as was done in the pig studies, is not representative of the manner in which humans will consume olestra. Humans are likely to eat olestra and fat-soluble vitamins in different foods or at different times throughout the day. Menu census data show that savory snacks, the initial intended use of olestra, are not eaten every day or at every meal (Webb et al. 1997). At the 90th percentile, snacks are eaten 9-12 times in a 14-d period, depending on age. Also, at the 90th percentile intake (all ages), snacks are eaten with meals only 18% of the time. Therefore, to make a reasonable extrapolation of measurements made in the pig studies to the human situation, it is necessary to know the extent to which dietary context (form of food, eating times) influences the effect of olestra on fat-soluble vitamin status.
|
Table 1.
Description of feeding patterns for pigs fed olestra mixed in the diet or in potato chips1
[View Table]
|
On the basis of the partitioning mechanism, one would expect that homogeneously mixing olestra with the dietary components would produce the maximum effect on the absorption of fat-soluble nutrients because this situation provides maximum opportunity for interaction between olestra and the nutrients. Separating olestra and the nutrients either physically by feeding them in different foods, or temporally by feeding olestra at times other than when the nutrients are fed, would be expected to reduce the effects of olestra. Therefore, the purpose of this study was to determine whether, and to what extent, separating olestra and vitamins physically or temporally affects the influence of olestra on the status of vitamins A, D and E.
MATERIALS AND METHODS 
The study protocol was approved by the Institutional Animal Care and Use Committee at Hazleton-Wisconsin (Madison, WI) where the study was conducted. All procedures were performed in compliance with the Guide for Care and Use of Agricultural Animals in Agricultural Research and Teaching (Consortium 1988). The study was conducted in accordance with the Food and Drug Administration Good Laboratory Practice Regulations for Nonclinical Laboratory Studies.
Animals and housing.
Six groups of 10 pigs each (five castrated males, five females) were used in the study. The pigs were a cross-bred domestic strain, one-half Duroc, one-quarter Landrace, and one-quarter Large White, acquired from the University of Wisconsin-Madison Swine Unit (Madison, WI).
The pigs were received at 5-6 wk of age and were acclimated to the facility for 13 d before treatment was started. The pigs were weaned at about 3 wk of age. Between the time they were weaned and the beginning of the acclimation period they were fed a standard corn-soy-based swine starter diet formulated by the University of Wisconsin. During the 13-d acclimation period, the pigs were housed in groups of 1-3 per pen and were given free access to water and purified basal diet (described below).
During the treatment period, the pigs were housed individually in pens in a sunlight-free barn with a 12-h incandescent light:dark cycle, controlled temperature (above 18°C) and ambient humidity. The pigs were not exposed to ultraviolet light so that 25-hydroxycholecalciferol [25(OH)D3] resulting from UV-induced synthesis of vitamin D3 would not interfere with detection of olestra-related effects on dietary vitamin D2 .
Treatment groups.
At the end of the acclimation period, 10 pigs (five males and five females) were randomly selected and killed as described below to provide base-line data on tissue and circulating vitamin concentrations. The remaining 50 pigs were randomized by weight and assigned to one of five treatment groups (Table 1). The control group (Group 1) was fed purified diet with no olestra. The other four groups were fed purified diet containing 2.2% (wt/wt) olestra either mixed in the diet or in potato chips. This dietary level of olestra was chosen to provide a daily olestra intake, on a gram/day basis, that would be in excess of the 90th percentile chronic human intake in the U.S. from savory snacks, 6.9 g/d, olestra's intended use (Webb et al. 1997). A previous pig study (Cooper et al. 1997c) showed that pigs fed 2.2% olestra mixed in the diet ate, on average, 10.3 g olestra/d during wk 1 of the study and 22.5 g/d during wk 4. This dietary level of olestra produced readily measured reductions in serum vitamin E and serum 25-hydroxyergocalciferol [25(OH)D2] concentrations in 4 wk.
|
Table 2.
Composition of the diets fed to the pigs
[View Table]
|
Group 2 pigs were fed a purified diet in which olestra was mixed thoroughly when the diet was prepared. Group 3 was fed the same amount of olestra in crushed potato chips in equal weighed portions at each daily feeding (0730, 1200 and 1630 h). Group 4 was fed the same amount of olestra in crushed potato chips but only with the feeding at 1200 h; Group 5 received the same amount of olestra in crushed potato chips but was fed about midway between the feedings at 1200 and 1630 h. To ensure complete consumption of the olestra potato chips, the pigs in Groups 3 and 4 were offered the potato chips about 20 min before they were offered their feed.
Any uneaten feed or potato chips from the earlier feedings were added to the final feeding of each day. After the final feeding, any uneaten feed or chips and feed or chips spilled around the feed bowl were collected and weighed. The consumption of feed and chips was corrected for this spillage and uneaten feed and chips.
The pigs were fed sufficient diet daily to provide 95% of the digestible energy for swine recommended by the National Research Council (NRC 1988). The daily allotment of feed was calculated from the pig's weight at the beginning of each week and the projected weight gain over the week, using growth curves for these cross-bred pigs given free access to a standard diet for swine (Martin and Crenshaw 1989).
The quantity of olestra potato chips required to provide 22 g olestra per kg of feed (diet plus potato chips) was determined by calculating feeding tables for each pig based on the amount of diet fed daily and the olestra and nonolestra content of the chips. The quantity of chips fed to each pig was adjusted weekly to ensure that the olestra dose remained constant at 2.2% of the total diet, including the potato chips.
Diets.
The diets were custom prepared by ICN Biomedicals (Cleveland, OH) so that all groups received the same concentration of digestible energy, fat, protein and carbohydrate (Table 2). The digestible energy content of the diet was calculated by summing the digestible energy contributed, per kilogram of diet, by each dietary ingredient. The diets provided 1 time the NRC daily requirements of micronutrients for 5- to 10-kg swine (NRC 1988). The basal diet fed to the control group was a purified diet providing 30% of energy from fat and containing 140 g fat per kg of diet. This diet was the same as that used in other pig studies with olestra (Cooper et al. 1997a, 1997c and 1997d) and produces growth equivalent to that produced by a standard corn-soy-based diet for swine (Cooper et al. 1997d). For the other groups, the ingredients of the basal diet were adjusted to compensate for the addition of olestra and for the protein, carbohydrate and fiber in the potato chips.
The composition of the diet and the vitamin-mineral premix and the concentration and stability of olestra in the diets were confirmed by analyses as described in Cooper et al. (1997d).
Olestra.
The olestra (Olean, Proter & Gamble, Cincinnati, OH) with fatty acids isolated from cottonseed oil (Rizzi and Taylor 1978) and was the same as that used in other pig studies (Cooper et al. 1997a, 1997c and 1997d). It consisted of 71% octaesters, 28% heptaesters, and 1% hexa- and lower esters. The composition of the fatty acid chains was 36% oleic, 34% linoleic, 18% palmitic, and 2% others. After being used to fry the potato chips, the olestra contained 7.9% polymeric species, a level similar to that found in vegetable oils used to fry potato chips (Henry et al. 1992).
Olestra potato chips.
The olestra potato chips were prepared by batch frying in 100% olestra. After frying, the chips were crushed, packed into a single barrel, and stored at room temperature until they were offered to the pigs. The olestra content of the chips was determined gravimetrically by extracting 200-g samples of crushed chips with hot ethylene dichloride. The stability of olestra in the chips was determined by analyzing a sample of crushed chips before and after the 4-wk treatment period.
Examination and observation of animals.
During the acclimation period, physical examination and clinical chemistry and hematology analyses were performed, and body weights were measured. At the start of the treatment period, food and water were withheld overnight and blood was collected from the superior cranial vena cava of pigs in all groups before the morning feeding. The blood samples were stored at 
20 ± 10°C before being analyzed for concentrations of serum vitamin A (retinol), vitamin E (-tocopherol), 25(OH)D2 , and 25(OH)D3.
At the start of the treatment period, 10 pigs were randomly selected, stunned with a captive bolt, and killed by exsanguination after an overnight fast. The livers were removed through an incision in the cranial abdomen, and the left lateral lobe was isolated and perfused with normal saline. The lobe then was sliced, blotted dry, flash frozen with liquid nitrogen, and stored at 70 ± 10°C until analyzed. Liver tissue was analyzed for vitamin A (total retinol and retinyl esters) and vitamin E (-tocopherol).
During the treatment phase, the animals were observed daily for signs of nutritional deficiency, morbidity, mortality and intolerance to olestra. Consumption of feed and potato chips was determined daily. Body weights were measured weekly. Cumulative weight gain, and ratios of cumulative weight gain to wk 0 weight and wk 0 body surface area were calculated. Body surface area was estimated as (wk 0 body weight)0.75 (Kleiber 1975). Weekly and cumulative digestible feed efficiencies were calculated as (body weight gain/digestible energy intake). Digestible energy intake was calculated from the kilograms of feed eaten and the digestible energy of the diet. At the end of the study, blood was collected, the pigs were killed and liver samples obtained as described above.
Analytical methods.
The concentrations of vitamin A (total retinol and retinyl esters) and vitamin E (-tocopherol) in liver and serum and the concentrations of 25(OH)D2 and 25(OH)D3 in serum were measured as described in Cooper et al. (1997c).
Statistical methods.
The effects of olestra were assessed by two-way ANOVA using gender and group as factors (Winer 1971). If no significant differences were detected between the males and females regarding the nature of the response to olestra, the data were combined and intergroup comparisons were made on the combined data set. Pairwise comparisons were based on the protected least significant difference (LSD) test. Among multiple-comparison procedures, the protected LSD test has the highest probability of detecting group mean differences if in fact they exist (Carmer and Swanson 1973, Welsch 1977). The data were examined for homogeneity of variance by using Bartlett's test. The serum vitamin A, vitamin E and 25(OH)D2 data did not pass Bartlett's test; therefore the two-way ANOVA was performed on rank-transformed data (Conover and Iman 1981).
All statistical tests were performed at the two-tailed 0.05 significance level, using PC-SAS software version 6.03 (SAS Institute, Cary, NC).
RESULTS
There were no indications of nutritional deficiency in the pigs, and no unscheduled deaths occurred during the study. There were no antemortem findings indicative of an olestra effect on the general health or nutritional status of the pig. There were no signs of intolerance to the diet.
The concentration of olestra in the diet and the potato chips was within 9% of target values. Analysis of samples of the olestra-containing diet and the potato chips confirmed that olestra was distributed homogeneously within both matrices and that no substantial degradation of olestra had occurred over the course of the study (data not shown). All groups ate essentially the same amount of olestra; the cumulative amounts consumed over the 4-wk period are given in Table 3. During wk 4, the average daily intake of olestra per pig, by Groups 2, 3, 4 and 5, was 27.1, 27.1, 25.7, and 27.1 g, respectively. This intake exceeds the estimated mean chronic intake of olestra from savory snacks, 3.1 g/d, by the total population of U.S. snack eaters by almost seven times (Webb et al. 1997).
|
Table 3.
Cumulative olestra consumption, energy consumption, body weight gain and digestible feed efficiency
for pigs fed olestra mixed in the diet or in potato chips for 4 wk1
[View Table]
|
|
Table 4.
Liver and serum vitamin A concentrations for pigs fed olestra mixed in the diet or in potato chips1
[View Table]
|
There were no statistically significant differences among the diets with respect to the concentrations of vitamins A, D2 and E (data not shown). Analyses showed that the concentration of retinol in the diets averaged about 62% of target, the concentration of -tocopherol averaged about 96% of target, and the concentration of ergocalciferol averaged about 79% of target.
There were no significant differences among the control and olestra-fed groups with respect to mean energy consumption (Table 3). This finding was expected because the diets were adjusted to provide equal amounts of total energy per kilogram of body weight to pigs in all groups.
All groups grew at essentially the same rate (Table 3). The pigs grew from an average weight of about 16 kg at the start of the study to about 35 kg at the end. The growth was similar to that observed for pigs fed a standard diet for swine and to that observed in studies in which pigs were fed a similar diet containing 1.1-7.7% olestra for 12 wk (Cooper et al. 1997a and 1997c). There were no significant differences between males and females with respect to their rate of growth; therefore, the data for males and for females were combined to assess intergroup differences. The mean body weights of the groups did not differ significantly at the start of the 4-wk treatment period; the group mean values ranged from 15.3 to 16.2 kg. Mean cumulative weight gain during the 4-wk treatment period, expressed in kilograms, did not differ among the groups (Table 3). The groups also showed no differences in mean cumulative weight gain expressed in relation to wk-0 body weight or in relation to body surface area (data not shown). Digestible feed efficiency did not differ among the groups (Table 3).
The mean (± SD) liver vitamin A concentration in the control group was 23.4 ± 4.6 nmol/g at the end of the study. This compared to a value of 56.2 ± 15.7 nmol/g measured in the group killed at base line. At wk 4, the mean concentration of vitamin A in the liver for the pigs fed 2.2% olestra mixed in the diet was 56% of the concentration in the control group (Table 4). The mean concentration measured for this group was significantly less than the mean concentrations for all other groups. The liver vitamin A concentration measured for the groups fed olestra in potato chips with all three feedings, or at the noon feeding only, was lower than the concentration in the control group by about 15%; the difference was not statistically significant. The mean liver vitamin A concentration for the group fed olestra in potato chips between the noon feeding and the evening feeding did not differ significantly from the control group.
The mean serum vitamin A concentration for the group fed 2.2% olestra mixed in the diet was 73% of the concentration for the control group (Table 4) and was significantly less than the mean values for all other groups. Mean vitamin A serum concentrations for the groups fed olestra in potato chips were not significantly different from the control value or from each other.
The mean (± SD) concentration of vitamin E in the liver of the control group was 10.9 ± 2.48 nmol/g at the end of the study compared with 6.5 ± 1.7 nmol/g measured in the group of pigs killed at base line. The concentrations of vitamin E in the liver of all groups fed olestra were reduced significantly in relation to the control value (Table 5). The mean concentration for the group fed olestra mixed in the diet was about 40% of control and was significantly less than the values measured for the three groups fed olestra in potato chips. The liver vitamin E concentration for the group fed olestra in potato chips at each daily feeding was 64% of the control value. The liver vitamin E concentrations for all groups fed olestra in potato chips did not differ from each other.
|
Table 5.
Liver and serum vitamin E concentrations for pigs fed olestra mixed in the diet or in potato chips1
[View Table]
|
At wk 4, serum vitamin E concentrations for all olestra-fed groups were reduced significantly relative to the mean concentration for the control group (Table 5). The value for the group fed olestra mixed in the diet was about 45% of control and was significantly less than the values for all groups fed olestra in potato chips. The value concentration for the group fed olestra in potato chips at each feeding was about 81% of the control value. The mean concentrations for the groups fed olestra in potato chips at the noon feeding only or between the noon feeding and evening feedings were about 90% of the control value. There were no significant differences among the serum vitamin E concentrations for any of the groups fed olestra in potato chips.
Serum vitamin E concentrations normalized with respect to serum total lipids showed the same trends as the nonnormalized values. At wk 4, group mean (± SD) serum total lipid concentrations ranged from 103 ± 6 to 119 ± 13 mg/dL. There were no significant differences among the groups.
At wk 4, the mean serum 25(OH)D2 concentration for the group of pigs fed olestra mixed in the diet was about 78% of the mean control value (Table 6). This difference was significant. The mean concentration for this group was significantly less than the mean concentrations for the groups fed olestra in chips at all feedings or at the noon feeding only. The mean serum 25(OH)D2 concentrations for the three groups fed olestra in potato chips did not differ significantly from the mean concentration for the control group or from each other.
|
Table 6.
Serum 25-hydroxyergocalciferol [25(OH)D2] concentrations for pigs fed olestra mixed in the diet or in potato chips1
[View Table]
|
Serum concentrations of 25(OH)D3 were low in all pigs because the animals were not exposed to sunlight or UV light during the study. At wk 4, only 20 of the 50 pigs had a serum 25(OH)D3 concentration that was above the detection limit of the analytical method, about 4.8 nmol/L. For this reason, no attempt was made to draw any conclusions about possible effects of different feeding regimens on serum 25(OH)D3 concentration. However, because 25(OH)D3 arises from endogenously synthesized vitamin D3 , the serum concentration of this metabolite would not be expected to be affected by olestra.
DISCUSSION
Pigs fed 2.2% olestra, either mixed in a diet containing one time the NRC requirements of micronutrients or in potato chips added to the diet, grew at a rate comparable to that of pigs fed purified diet providing 30% of energy as fat and one time the NRC requirements of micronutrients for 5-10 kg swine. The growth of the pigs was similar to that observed for pigs fed standard corn-soy-based swine diet (Cooper et al. 1997d, Martin and Crenshaw 1989). The pigs ate the potato chips readily. Over the course of the study, all groups ate essentially the same amount of olestra; therefore, comparisons of olestra effects among the groups are appropriate.
Fig. 1.
Reduction in concentrations of vitamins A and E in liver and serum concentration of 25-hydroxyergocalciferol [25(OH)D2] in serum when pigs were fed 2.2% olestra for 4 wk, either mixed in purified diet (Group 2, solid bars) or in potato chips fed 20 min before each of the three daily feeding (Group 3, hatched bars) of the diet. Data plotted as percentage reduction from control mean ± SD. Each group consisted of five male and five female pigs.
[View Larger Version of this Image (12K GIF file)]
The tissue and circulating concentrations of vitamins A and E in the control animals showed that the status of these vitamins was not excessive, an indication that the olestra effects were not diminished because of excessive stores. The mean liver vitamin A concentration in the control group at the end of the study, 23.4 nmol/g liver, was about one half the value of 52 nmol/g liver associated with increased risk of subclinical vitamin A deficiency in the pig (Nelson et al. 1962). Concentration of vitamin E in the liver at the start of the study, 6.5 nmol/g, was about twice the level of 3 nmol/g associated with signs of vitamin E deficiency in the pig (Jensen et al. 1988), and increased by less than twofold in the pigs in the control group over the course of the study. Serum 25(OH)D2 concentration in the control group was in the range that others have reported for pigs fed diets containing adequate levels of vitamin D (Engstrom and Littledike 1986, Foley et al. 1990), and was comparable to other values measured in pigs fed this same diet (Cooper et al. 1997c).
The 68% increase in liver concentration of vitamin E for the pigs in the control group, relative to the group killed at base line, indicates that the dietary level of vitamin E was sufficient to meet the requirements of the pigs and to allow some growth of stores. The fact that the control pigs had a liver concentration of vitamin A which was about 42% of the value measured in the group killed at base line was, at least in part, a result of the fact that the basal diet provided only 63% of the target amount of 1 NRC of vitamin A. The diets fed to the other groups provided the same levels of the vitamins as the diet given to the control group; therefore, intergroup comparisons of vitamin status are valid.
Feeding olestra in potato chips rather than mixed in the diet reduced the influence of olestra on the status of all three fat-soluble vitamins. This observation is consistent with the partitioning mechanism proposed to explain how olestra affects the absorption of lipophilic substances (Jandacek 1982). When olestra is mixed in the diet, rather than being in a food matrix separate from the matrix containing the fat-soluble vitamins, there is greater opportunity for physical contact between the olestra and the fat-soluble vitamins. Also, when olestra is mixed in the diet when the diet is prepared, the olestra and the vitamins also can interact for a longer period. According to the partitioning mechanism, both of these factors should increase the olestra effect on the absorption of fat-soluble vitamins by providing greater opportunity for the vitamin to partition into the olestra.
When olestra was physically separated from the vitamins by feeding it in potato chips, the change in the effect of olestra, relative to the effect produced by feeding it mixed in the diet, differed for each vitamin. For all three vitamins, however, the reduction in status was greatest when olestra was fed mixed in the diet than when fed in potato chips. Figure 1 compares the decreases in liver or serum concentrations of vitamins A, D and E produced by feeding 2.2% olestra mixed in the diet and in potato chips fed at each daily feeding (Group 3). The decreases in liver and serum vitamin A concentrations produced by feeding olestra mixed in the diet were 2.8 and 4.5 times greater, respectively, than the decreases produced by feeding olestra in potato chips with the diet. The data suggest that the olestra effect was diminished even further when the olestra-containing potato chips were fed between the noon and the evening feedings.
The decreases in liver and serum concentrations of vitamin E were 1.7 and 2.5 times greater, respectively, when the olestra was fed mixed in the diet than when fed in potato chips at each daily feeding The serum vitamin E data suggest that the olestra effect was reduced further, relative to the effect produced by feeding olestra mixed in the diet, when all of the daily allotment of olestra-containing potato chips was fed at one feeding or between feedings.
The reduction in serum 25(OH)D2 concentration was 22% when the olestra was fed mixed in the diet. When olestra was fed in potato chips in any of the three patterns, it did not affect serum 25(OH)D2 concentration.
The effect of olestra on the status of the fat-soluble vitamins was greatly reduced when the olestra was physically separated from the vitamins by feeding the olestra in potato chips and the vitamins in purified diet. Because the olestra effects on the status of the fat-soluble vitamins were small or nonexistent when olestra was fed in potato chips with the three daily feedings, it was not possible to draw any definitive conclusions about the relative effects produced under the less extreme conditions of feeding olestra and vitamins at different times during the day. At most, the data suggested that the olestra effect on the absorption of the vitamins was lessened by separating the consumption of olestra and of fat-soluble vitamins in time.
These results demonstrate that the magnitude of the olestra effect on the absorption of fat-soluble nutrients depends to a significant extent on how the two are eaten. When they are present in the same food matrix before and during ingestion, the effect will be maximal. Separating the two physically will lessen the effect significantly. Separating the times of ingestion probably would cause a further decrease in the influence of olestra on the status of the vitamins. Because of this, the results of studies in which olestra was fed mixed in the diet, as with the pig (Cooper et al. 1997a-d), or was eaten at every meal, as in controlled human studies (Schlagheck et al. 1997a and 1997b), are exaggerated in relation to real-life conditions, in which olestra is consumed with only about 8% of meals by the average snack eater. Based on the liver concentrations of vitamin A and the liver and serum concentrations of vitamin E measured in this study, the exaggeration may be as great as two- to fivefold.
ACKNOWLEDGMENTS
The authors thank A. L. Kiorpes of Hazleton Wisconsin, Inc., Madison, WI, for supervising the conduct of the study, and K. W. Miller and M. G. Royer for assistance in preparing the manuscript.
![[Feedback]](/icons/banner/feedbackACT.gif){kind=icon}
![[For Subscribers]](/icons/banner/subscriberACT.gif){kind=icon}
![[Archive]](/icons/banner/archiveACT.gif){kind=icon}
![[Search]](/icons/banner/searchACT.gif){kind=icon}
![[Contents]](/icons/banner/tocACT.gif){kind=icon}
ABSTRACT
