The Journal of Nutrition Vol. 128 No. 11 November 1998, pp. 2001-2008

Digestion and Absorption Rates of [³H]-Oleic Acid and [¹⁴C]-Triolein Do Not Differ in Rats Fed Heated () and (+) Gossypol Cottonseed and Soybean Flours^1,2

Josiane Prost^*, Jacques Belleville^{*, 3}, and Véronique Fustier-Bertrand

^* Unité de Nutrition Cellulaire et Métabolique, Université de Bourgogne, Faculté des Sciences Mirande, 21011 Dijon Cedex-France and  Université de Corse, Faculté des Sciences Grossetti, 20250 Corté-France

	ABSTRACT

Abstract Introduction Methods Results Discussion References

This study was conducted to compare in vivo the acute effects of heated (+) and () gossypol cottonseed flours with those of soybean flour on lipid digestion and absorption in growing rats. Rats were fed by gastric intubation mixed [³H]-oleic acid and [¹⁴C]-triolein with heated flours or without flour (control). Lipid digestion and absorption were determined for 6 h after meal intubation. Both radioactivities recovered in gastrointestinal tract were significantly higher in rats fed (+) gossypol cottonseed flour than in all other groups. The majority of both recovered radioactivities was found in stomach contents, then in stomach wall and finally in intestinal wall. The distribution of both radioactivities at different gastrointestinal levels was similar. In stomach contents and wall, [¹⁴C]-radioactivity was primarily in triacylglycerols, but was also recovered in free fatty acids and diacylglycerols. In intestinal wall (mucosa + tunica) [³H]-radioactivity was found at greatest levels in free fatty acids, then in triacylglycerols and diacylglycerols. Greatest [¹⁴C]-radioactivity was found in triacylglycerols, then in free fatty acids, in diacylglycerols and last in phospholipids in rats fed the three flours. Therefore no quantitative differences in lipid digestion and absorption were observed among the rats fed the three flours. In conclusion, both cottonseed flours slowed lipid digestion and absorption compared with soybean flour and this delay was greater in the rats fed (+) gossypol cottonseed flour than in those fed () gossypol cottonseed flour. However, this inhibiting effect was probably too low to induce physiologically important effects on lipid digestion or absorption.

	INTRODUCTION

Abstract Introduction Methods Results Discussion References

Preduodenal lipases hydrolyze triglycerides to free fatty acids and partial glycerides under acidic conditions (Fink et al. 1984, Hamosh et al. 1975, Hamosh, 1984, Liao et al. 1984). Gargouri et al. (1985 and Gargouri et al. 1986) have shown in vitro that human gastric lipase or rat lingual lipase is inhibited by amphiphilic molecules such as soybean proteins that decrease surface tension at the lipid-water interface. Moreover, in the gastrointestinal tract, the binding of pancreatic lipase, but not colipase, to lipids is impaired by amphiphiles (such as bile salts and proteins) that decrease the hydrophobic/hydrophilic balance of the triacylglycerol-water interface. In the presence of such inhibitors, the colipase cofactor may bring the larger lipase enzyme into close contact with the triacylglycerol-water interface and allow opening of the lid of the lipase molecule, exposing the active site, thus facilitating the enzymatic breakdown of lipids (Van Tilbeurgh et al. 1992). Dietary soybean proteins might decrease the in vivo activity of lipases in the gastrointestinal tract and consequently, hydrolysis and absorption of ingested fats (Gargouri et al. 1984). In a previous report, Khalifa et al. (1992) conducted in vivo and in vitro studies to investigate the effects of raw and heated soybean flour consumption on lipid digestion and absorption, thus showing that lipid digestion is not inhibited by soybean flour consumption but that only lipid absorption is diminished. Soybean flour components that decrease absorption are partially thermostable.

Cottonseed protein is the second most-abundant source of oil-seed protein in the world. The only oil-seed protein to exceed cottonseed protein production is soybean protein (Bodwell and Hopkins 1985).

(+) Gossypol (with gossypol, a toxic pigment for monogastric species; a yellowish phenolic compound isolated from the seeds, stems and roots of cotton plants) and () gossypol (with only gossypol traces) cottonseed flours are consumed by humans. Reyes et al. (1984) showed in vitro that gossypol through interaction with polar/apolar interfaces and modification of the electrostatic potential seems to produce dramatic changes in enzymatic and transport properties of biological membranes. In rats and cynomolgus monkeys, Heywood (1988) showed that gossypol causes body weight loss that was greater than expected from food consumption data, indicating impairment of food utilization. As Shandilaya and Clarkson (1982) have shown that gossypol has a hypolipidemic effect in cynomolgus monkeys when fed at 10 mg/(kg.d), the question raised is whether the presence of gossypol in cottonseed flour can alter lipid digestion and absorption. However, to our knowledge, no studies have been published on lipid digestion and absorption in animals fed () / and (+) gossypol cottonseed flours. Yet cottonseed flour, produced in developing countries, is a protein-rich flour and a potential rehabilitation food for people suffering from protein malnutrition if the flour is proven to have no deleterious effects on lipid digestion and absorption.

In this study, the effects of heated () and (+) gossypol cottonseed flours on lipid digestion and absorption in rats were investigated and compared with those of heated soybean flour, with the prospect of recommending the use of cottonseed flours, like soybean flour, in postweaning and rehabilitation diets.

	MATERIALS AND METHODS

Abstract Introduction Methods Results Discussion References

Animals and diets.  Male Wistar rats (Iffa-Credo, L'Arbresle, France) weighing 280 ± 20 g at the beginning of the experiment were housed individually in wire-bottom cages at constant temperature (25°C) and humidity (65%) with a 0700-1900 light cycle. Tap water and commercial diet (UAR, Epinay sur Orge, France) were freely available from weaning. The diet contains per kg diet: 170 g of protein, 30 g of lipid, 587 g of carbohydrate, 43 g of fiber, 50 g of minerals and 20 g of vitamins. We followed the general guidelines of the Council of European Community (1986) for the care and use of laboratory animals.

Compositions of cottonseed and soybean flours are shown in Table 1. Protein concentrations were estimated by their nitrogen contents, using mineralization followed by coloration with Nessler's reactive, and nitrate ammonium as standard. Total lipids of flours were extracted by the method of Folch et al. (1957). Separation of lipids was performed by thin layer chromatography (TLC)⁵ according to Stahl et al. (1956). Phospholipids were quantified by phosphorus according to the Bartlett method (1958). Dietary fiber was measured using heat-stable amylase in the Neutral Detergent Fiber procedure (Van Soest et al. 1991). Partially delipidated flours were autoclaved at 130°C for 10 min under the pressure of 150 kPa. The remaining oleic acid [18:1 (n-9)] represented 36, 17 and 23%, and linoleic acid [18:2 (n-6)] 36, 54 and 53% in (+) gossypol, () gossypol and soybean flours, respectively.

We chose to use partially delipidated commercial flours even though this choice makes the interpretation of results more difficult. However, these flours are of interest because they could be consumed by humans. (+) Gossypol and () gossypol cottonseed flours contain slightly more protein, lipid and mineral and little less carbohydrate and fiber than soybean flour. The composition of the two cottonseed flours is similar except in their gossypol contents. The () gossypol cottonseed flour was obtained after several extractions of the (+) gossypol flour. Its free gossypol concentration was 0.01 g/kg, therefore below the toxicity level of this compound (0.45 g/kg) (Institut de Recherches sur le Coton et les Fibres Tropicales, IRCT, Anonymous, 1970). The (+) gossypol cottonseed flour contained a high free gossypol concentration, 5 g/kg.

After overnight food deprivation, each type of meal was intubated into the stomachs of three rats between 0800 and 0900 h.

Each intubated meal consisted of mixed pure triolein (Nu-Chek-Prep, Elysian, MN) (217 µmol), oleic acid (Prolabo, Paris, France) (326 µmol), [carboxyl-¹⁴C]-triolein (CEA, Saclay, France) (4.44 GBq · mmol¹) and [9,10³H(N)]-oleic acid (CEA) (370 GBq · mmol¹) with 450 mg of heated () and (+) gossypol cottonseed flours (IRCT, Montpellier, France) or soybean flour (Sigma, St. Louis, MO). After dilution, the final specific radioactivities of substrates were 230 kBq · mmol¹ for triolein and 550 kBq · mmol¹ for oleic acid. Triolein and oleic acid addition to the meal highly enriched the lipid concentration of flours from nearly 30 to 425 g/kg diet. We chose this lipid/flour ratio because meals rich in lipids slow down the gastric emptying and provide sufficient lipid quantities for analysis at the different times tested after meal intubation and at the various gastrointestinal levels studied. A meal without flour was used as the control to check whether the presence of flour in general under the conditions used here affects lipid digestion and absorption.

Before intubation, meals were homogenized by ultrasonication in 1.5 mL of distilled water. The syringe was rinsed with 5 × 1 mL of chloroform to determine the radioactivity loss, and a fresh syringe was used for each meal intubation. After meal intubation, (1, 3 and 6 h) under light anesthesia, rats (3 at each time) were killed by a blow on the neck. Stomach, small and large intestines were removed. Stomach, small intestine and cecum were rinsed by injection of 20 mL of NaCl (150 mmol/L) and sodium taurocholate (2 mmol/L) to obtain intraluminal contents with unabsorbed lipids (Clark et al. 1973). Then, stomach, small and large intestine walls and their contents were separately homogenized in 100 mL of chloroform/methanol (1:1, vol/vol), using an ultraturax homogenizer. Total lipids in each sample were extracted according to Folch et al. (1957). Separation of different lipid classes was performed, in triplicate, by TLC according to Stahl's technique (Stahl et al. 1956).

Determination of radioactivity.  After separation of the lipid fractions by TLC, plates were scraped and gel was introduced into a scintillation vial that contained 1 mL of methanol. Two hours later, 7 mL of scintillation liquid, xylene/Permafluor (1:10, vol/vol) (Packard, Canberra Company, Groningen, ND), was added. Radioactivity was counted in a Packard Tricarb model 1900 TR liquid scintillation analyzer (Packard Instrument Company, Meriden, CT). Quenching was automatically corrected.

Statistical analysis.  Data were tested by two-way analysis of variance (time and diet) followed by Fisher's least significant difference test. Data for the treatment group that did not meet the asumption of equal variance were log-transformed before statistical analysis. Differences were considered significant at P < 0.05. Statistical analysis of the data was carried out using STATISTICA (Version 4.1, Statsoft, Inc. 1994, Paris, France). Values are means ± SEM for n = 3 measured in triplicate.

	RESULTS

Abstract Introduction Methods Results Discussion References

Recovered radioactivities in Tables and Figures were expressed as percentages of intubated radioactivities.

Lipid digestion and absorption were rapid when lipids were intubated with the three flours, since only 15 to 30% was left in the gastrointestinal tract 1 h after meal intubation (Table 2). One hour after intubation, both [³H]- and [¹⁴C]-recovered radioactivities were significantly higher in rats fed the control meal (44 and 39% for [³H] and [¹⁴C], respectively) than in rats fed the () gossypol cottonseed flours (25 and 22% for [³H] and [¹⁴C], respectively) followed by the rats fed the soybean flour (14 and 18% for [³H] and [¹⁴C], respectively). In rats fed (+) gossypol cottonseed flour, the disappearance of both radioactivities was significantly slower than in rats fed the soybean flour (27 vs. 14 and 28 vs. 18% for [³H]- and [¹⁴C]-recovered radioactivities, respectively). In rats fed (+) gossypol [¹⁴C]-radioactivity recovery was significantly less than in rats fed () gossypol cottonseed flour (27 vs. 22% of [¹⁴C]-recovered radioactivity). The difference among rats fed the three flours persisted at 3 h. At 6 h, recovered radioactivities were still present but in low quantities. Both [³H]- and [¹⁴C]-recovered radioactivities were different with the three flours, except for [¹⁴C]-radioactivity of rats fed soybean and () gossypol cottonseed flours. At this time, both recovered radioactivities still exceeded 10% of intubated radioactivities in rats fed the (+) gossypol cottonseed flour.

View this table: [in this window] [in a new window]   Table 2. [3H] and [14C] Radioactivities recovered in gastrointestinal tract of rats fed control meal, soybean, () gossypol and (+) gossypol flours 1, 3 and 6 h after intubation of [3H]-oleic acid and [14C]-triolein1,2,3,4

The majority of both recovered radioactivities (Tables 3 and 4) were found in stomach contents: at 1 h, 10-40% of [³H] and 13-36% of [¹⁴C] intubated radioactivities were recovered; at 3 h, 7.2-23% of [³H] and 10-21% of [¹⁴C] intubated radioactivities were recovered. In stomach wall, the highest recovered intubated radioactivities were obtained at 1 h for [³H] (2-4%) and at 3 h for [¹⁴C] (nearly 2%). In small intestine contents, the highest [³H]- and [¹⁴C]-recovered radioactivities were obtained at 1 h in rats fed soybean flour (0.87 and 1.34%, for [³H]- and [¹⁴C]-radioactivities, respectively), () gossypol cottonseed flour (2.6 and 1.6% for [³H]- and [¹⁴C]-radioactivities, respectively) and (+) gossypol cottonseed flour (0.55% for [³H]-radioactivity), and at 3 h in the rats fed the control diet (2.09 and 2.2% for [³H]- and [¹⁴C]-radioactivities, respectively) and in the rats fed (+) gossypol cottonseed flour (0.2% for [¹⁴C]-radioactivity). In small intestine (mucosa + tunica) the highest [³H]-radioactivity was obtained at 1 h for the four meals, at 1 h with the three flours for [¹⁴C]-radioactivity, but only at 6 h for [¹⁴C]-radioactivity in the rats fed the control meal. In stomach, small intestine and cecum contents [³H]- and [¹⁴C]-recovered radioactivities were significantly different among the rats fed the four meals (P < 0.001) and with time (P < 0.001). In stomach wall, small intestine (mucosa + tunica) and cecum wall, [³H]- and [¹⁴C]-recovered radioactivities were significantly different among the rats fed the four meals (P < 0.001) and with time (P < 0.001). Both radioactivities recovered in cecum were very low over time for the four groups.

View this table: [in this window] [in a new window]   Table 3. [3H]-recovered radioactivity distribution in gastrointestinal tract of rats fed control meal, soybean, () gossypol and (+) gossypol flours 1, 3 and 6 h after intubation of [3H]-oleic acid1,2,3,4

View this table: [in this window] [in a new window]   Table 4. [14C]-Recovered Radioactivity Distribution in Gastrointestinal Tract of Rats Fed Control Meal, Soybean, () Gossypol and (+) Gossypol Flours 1, 3 and 6 h after Intubation of [14C]-Triolein1,2,3,4

In stomach contents, [³H]-radioactivity was almost completely recovered in the free fatty acid fraction, which was expected since intubated [³H]-radioactivity was carried by oleic acid (Fig. 1). After 1 h, recovered [³H]-oleic acid was lower in rats fed the three flours than in controls. Recovered [³H]-oleic acid was greater in rats fed the (+) gossypol cottonseed flour than in rats fed soybean and () gossypol cottonseed flours. Recovered [³H]-oleic acid was significantly different among the rats fed the four meals (P < 0.001) and with time (P < 0.001). [¹⁴C]-Radioactivity recovered in triacylglycerol fraction decreased with time after meal intubation. At 1 and 3 h, the highest recovered [¹⁴C]-triolein in triacylglycerol fraction was observed in the groups fed the control meal (22 and 11% at 1 and 3 h, respectively) and the group fed (+) gossypol cottonseed flour (26 and 15%, at 1 and 3 h, respectively), then in the group fed () gossypol cottonseed flour (15 and 9%, at 1 and 3 h, respectively), finally in the group-fed soybean flour (14 and 10%, at 1 and 3 h, respectively). [¹⁴C]-Radioactivity recovered in triacylglycerols was significantly different among the rats fed the four meals (P < 0.001) and with time (P < 0.001). In the three groups fed the flours, low [¹⁴C]-radioactivity was observed in free fatty acids, and in diacylglycerols these values were higher in the rats fed the control meal. [¹⁴C]-Radioactivity recovered in diacylglycerols and in free fatty acids was significantly different among the rats fed the four meals (P < 0.001) and with time (P < 0.001).

View larger version (29K): [in this window] [in a new window]   Fig 1. [3H]-Radioactivity recovered in free fatty acids and [14C]-radioactivity recovered in triacylglycerols, diacylglycerols and free fatty acids in stomach contents of rats fed control meal, soybean, () gossypol and (+) gossypol cottonseed flours, 1, 3 and 6 h after intubation of 320 µmol oleic acid (550 kBq · µmol1) and 217 µmol triolein (230 kBq · µmol1). Values are expressed as percentages of intubated radioactivity. Values are means ± SEM, n = 3. Means with different letters (a>b>c>d) are significantly different among the four groups at each time, and means with different letters (x>y>z) are significantly different within a group over time at P < 0.05. Data for the treatment groups that did not meet the assumption of equal variance were log-transformed before statistical analysis. Control rats were intubated with lipids without flour. Analysis of variance P values: [3H]-free fatty acids; diet, < 0.001; time, < 0.001; diet × time, < 0.001. [14C]-triacylglycerols; diet, < 0.001; time, < 0.001; diet × time, < 0.001. [14C]-diacylglycerols; diet, < 0.001; time, < 0.001; diet × time, < 0.001. [14C]-free fatty acids; diet, < 0.001; time, < 0.001; diet × time, < 0.001.

[³H]-Radioactivity in stomach wall was mainly recovered in free fatty acids and [¹⁴C]-radioactivity in triacylglycerols (Fig. 2). At 1 and 6 h, both recovered radioactivities were higher in the rats fed the three flours than in controls. At 1 h [³H]-radioactivity recovered in free fatty acids of rats fed (+) gossypol cottonseed flour was significantly lower than in the rats fed the () gossypol cottonseed flour, and this value was higher in the rats fed soybean flour than in the rats fed () gossypol cottonseed flour. At 3 and 6 h, compared with the groups fed () and (+) gossypol cottonseed flours, the lowest values of [³H]-radioactivity were obtained in the group intubated with soybean flour. [³H]-Radioactivity recovered was significantly different among the rats fed the four meals (P < 0.001) and with time (P < 0.001). With time, [¹⁴C]-radioactivity in triacylglycerol fraction decreased in the group fed soybean flour but not in the groups fed () and (+) gossypol cottonseed flours. At 3 h, rats fed () and (+) gossypol cottonseed flours had higher [¹⁴C]-recovered radioactivity in triacylglycerol fractions than in the group fed soybean flour. At 6 h, [¹⁴C]-radioactivity was significantly greater in the group fed (+) gossypol cottonseed flour than in the groups fed soybean and () gossypol cottonseed flours. [¹⁴C]-radioactivity recovered was significantly different among the rats fed the four meals (P < 0.001) and with time (P < 0.001).

View larger version (56K): [in this window] [in a new window]   Fig 2. [3H]-Radioactivity recovered in free fatty acids and [14C]-radioactivity recovered in triacylglycerols in stomach wall of rats fed control meal, soybean, () gossypol and (+) gossypol cottonseed flours, 1, 3 and 6 h after intubation of 320 µmol of oleic acid (550 kBq · µmol1) and 217 µmol of triolein (230 kBq · µmol1). Values are expressed as percentage of intubated radioactivity. Values are means ± SEM, n = 3. Means with different letters (a>b>c>d) are significantly different among the four groups at each time, and means with different letters (x>y>z) are significantly different within a group over time at P < 0.05. Data for the treatment groups that did not meet the assumption of equal variance were log-transformed before statistical analysis. Control rats were intubated with lipids without flour. Analysis of variance P values: [3H]-Free fatty acids; diet, < 0.001; time, < 0.001; diet × time, < 0.001. [14C]-Triacylglycerols; diet, < 0.001; time, < 0.001; diet × time, < 0.001.

Most [³H]-radioactivity distributed into different lipids in the intestinal wall (mucosa + tunica) (Fig. 3) was found, at 1 h in triacylglycerols (0.5-1.3%), then, in a decreasing order in free fatty acids (<0.5%), diacylglycerols (<0.25%), phospholipids (0.12%) and monoacylglycerols (0.05%). [³H]-radioactivity recovered in triacylglycerols, fatty acids, diacylglycerols, phospholipids and monoacylglycerols was significantly different among the rats fed the four meals (P < 0.001) and with time (P < 0.001). [¹⁴C]-Radioactivity was essentially carried by triacylglycerols in all groups and decreased with time after intubation, except in rats fed () and (+) gossypol cottonseed flours. At 1 h, 0.05-0.1% of [¹⁴C]-radioactivity recovered was found in free fatty acids in the four groups. [¹⁴C]-Radioactivity recovered in diacylglycerols and monoacylglycerols was low and variable for the four groups. [¹⁴C]-radioactivity recovered in triacylglycerols, free fatty acids, diacylglycerols, and monoacylglycerols was significantly different among the rats fed the four meals (P < 0.001) and with time (P < 0.001). At 6 h, [¹⁴C]-radioactivity recovered in phospholipids was significantly greater in the group fed () gossypol cottonseed flour than in that fed soybean flour. [¹⁴C]-radioactivity recovered in phospholipids was significantly different among the rats fed the four meals (P < 0.001) and with time (P = 0.045).

View larger version (50K): [in this window] [in a new window]   Fig 3. [3H]- and [14C]-Radioactivities recovered in free fatty acids, monoacylglycerols, diacylglycerols, triacylglycerols and phospholipids of intestinal wall (mucosa + tunica) of rats fed control meal, soybean, () gossypol and (+) gossypol cottonseed flours, 1, 3 and 6 h after intubation of 320 µmol of oleic acid (550 kBq · µmol1) and 217 µmol of triolein (230 kBq · µmol1). Values are expressed as percentage of intubated radioactivity. Values are means ± SEM, n = 3. Means with different letters (a>b>c>d) are significantly different among the four groups at each time, and means with different letters (x>y>z) are significantly different within a group over time at P < 0.05. Data for the treatment groups that did not meet the assumption of equal variance were log-transformed before statistical analysis. Control rats were intubated with lipids without flour. Analysis of variance P values: [3H]-free fatty acids; diet, < 0.001; time, < 0.001; diet × time, < 0.001. [3H]-monoacylglycerols; diet, < 0.001; time, < 0.001; diet × time, < 0.001. [3H]-diacylglycerols; diet, < 0.001; time, < 0.001; diet × time, < 0.001. [3H]-triacylglycerols; diet, < 0.001; time, < 0.001; diet × time, < 0.001. [3H]-phospholipids; diet, < 0.001; time, < 0.001; diet × time, < 0.001. [14C]-free fatty acids; diet, < 0.001; time, < 0.001; diet × time, < 0.001. [14C]-monoacylglycerols; diet, < 0.001; time, < 0.001; diet × time, < 0.001. [14C]-diacylglycerols; diet, < 0.001; time, < 0.001; diet × time, < 0.001. [14C]-triacylglycerols; diet, < 0.001; time, < 0.001; diet × time, < 0.001. [14C]-phospholipids; diet, < 0.001; time, < 0.045; diet × time, < 0.001.

	DISCUSSION

Abstract Introduction Methods Results Discussion References

The aim of this study was to investigate the effects of heated () and (+) gossypol cottonseed flours on [¹⁴C]-triolein digestion and absorption and [³H]-oleic acid absorption in rats.

Soybean flour contained more fiber and carbohydrate, and less protein, lipid and minerals than () and (+) gossypol cottonseed flours. Lipid concentrations were 3.8, 3.6 and 3.2% in (+) gossypol, () gossypol and soybean flours, respectively. These slight differences cannot have affected our data because in each meal 450 mg of flour was enriched in fat with 200 mg of triolein and 100 mg of oleic acid. The (+) gossypol cottonseed flour contained 15.3 g/kg of total gossypol compared with 0.22 in () gossypol cottonseed flour. Lawhon et al. (1976) have determined that protein from () and (+) gossypol cottonseed varieties does not differ substantially in amino acid composition.

In this study conducted at 1 and 3 h after lipid intubation, the rats fed diet without flour had lower lipid digestion and absorption probably because the emulsion was less stable and thus adhered more to the gut wall. Flour components were probably the regulators of gastric emptying. Stability and fineness of lipid emulsion with soybean and cottonseed proteins would be responsible for the higher gastric emptying observed 1 h after intubation (Table 2). Soybean proteins are better emulsifiers than cottonseed proteins. Since the presence of flours in the three meals accelerated lipid digestion and absorption, we will discuss the results obtained in rats fed the three flours.

Bertrand et al. (1995) have shown that in vitro heated () gossypol cottonseed flour affects pancreatic digestive enzymes less than heated soybean flour; however () gossypol cottonseed flour decreased lipase activity more than soybean flour.

Our present data (Table 2) indicate that 1 h after meal intubation with () and (+) gossypol cottonseed flours, [³H]-oleic acid and [¹⁴C]-triolein disappearance from the gastrointestinal tract was slower than in rats fed soybean flour (P < 0.001). However, for a given meal, the proportions of radioactivities recovered did not differ. Nevertheless, triolein absorption requires two steps: digestion followed by actual absorption. Because oleic acid and triolein absorption differed only slightly in rats fed the three flours, these results show high efficiency of gastric and pancreatic lipases on triolein hydrolysis. We also observed that radioactivity appearance in serum and liver of rats fed soybean and cottonseed flours did not differ (results not shown), in agreement with previous data (Khalifa et al. 1992).

In a previous report (Bertrand et al. 1995) we showed in vitro that the addition of both () and (+) gossypol cottonseed and soybean flours could inhibit anchorage between lipase and colipase, and this depleted anchorage impairs pancreatic lipase activity on triolein. The lower lipid digestibility observed in rats fed both cottonseed flours compared with those fed soybean flour could be explained by the protein content of flours that were higher in both cottonseed flours than in soybean flour. This inhibitory effect of cottonseed flours on lipase activity could partly explain the results obtained by Radcliffe (1989), who showed that relative to casein-fed rats, those fed cottonseed protein isolate had significantly lower serum triacylglycerol and total lipid concentrations. Proteins are amphiphilic molecules that decrease the surface tension at the lipid-water interface and impair the binding of lipase to lipids and diminish lipase-colipase-dependent activity in intestine (Gargouri et al. 1985, Gargouri et al. 1986). However at 3 h this inhibitory effect was unobservable with () and (+) gossypol cottonseed flours because the values of [¹⁴C]-radioactivity recovered in free fatty acids (Fig. 3) were higher or did not differ from those of rats fed soybean flour. In rats fed the (+) gossypol cottonseed flour, this effect persisted until 6 h after meal intubation. Rats fed cottonseed flour had lower [³H]- and [¹⁴C]-radioactivities than those fed soybean flour. This might be involved by the dietary fiber composition. The nonstarch polysaccharide fraction of cottonseed flours, particularly pectic substances and hemicelluloses, differs from those of other oilseed polysaccharides, i.e., soybean (Aspinall and Cotrell 1971). Soybean and cottonseed fibers consist of water-insoluble fibers (mainly cellulose and lignin) with low viscosity that shortens transit time. Soybean flour contains more of these insoluble fibers (Table 1) than cottonseed flours, and soybean flour could decrease transit time vs. cottonseed flours. Fibers influencing luminal emulsification of lipids could reduce the diffusion rate of digested products toward the intestinal mucosal surface, which may inhibit absorption (Vahouny 1982).

At 1 h, gastric emptying of [³H]-oleic acid was slower in the rats fed () and (+) gossypol cottonseed flours and differed from that of rats fed soybean flour. Gastric emptying of [¹⁴C]-triolein was slower in the rats fed (+) gossypol cottonseed flour and differed from that of rats fed () gossypol cottonseed and soybean flours. At 3 and 6 h, gastric emptying of both radioactivities in the rats fed () gossypol cottonseed flour did not differ from that of rats fed soybean flour. Soybean protein had a higher emulsifying capacity compared with cottonseed proteins. Heated oilseed proteins had similar or better emulsification properties than raw oilseed proteins, and heated () gossypol cottonseed flour increased water- and oil-holding capacities more than raw () gossypol cottonseed flour (Khalil et al. 1985). Our data showed that a relatively large amount of radioactivity was found in the stomach wall, equivalent to or exceeding that found in the intestinal wall. It is unlikely that substantial lipid absorption took place at the stomach level. Because lipid level in the stomach contents was high, part of the lipids may have remained adsorbed on the wall because the meals were particularly rich in lipids and the lipid emulsion may have been unstable.

Since the cottonseed flours differed essentially only in their gossypol concentrations, gossypol was probably more responsible for the lower lipid digestibility and absorption in the rats fed (+) gossypol cottonseed flour than in the rats fed () gossypol cottonseed flour. Even though Chadha et al. (1988, Chadha et al. 1991) have not studied lipid absorption, they have showed that oral administration of gossypol in rats causes a significant reduction in the uptake of several nutrients in the small intestine segments. Gossypol causes a significant decrease in the intestinal brush border membrane-associated enzymes (sucrase, maltase, lactase and alkaline phosphatase). These authors conclude that since gossypol is a lipophilic compound, it actually binds to the lipid bilayer of the membrane by some presently unknown mechanism and decreases membrane fluidity, possibly by causing steric hindrance. Whatever the molecular mechanism of gossypol action, a solidifying effect on the membrane bilayer is apparent. Reyes et al. (1984) showed that gossypol produces an interfacial change in lipid membranes. Another explanation for the mode of gossypol action is the binding of gossypol to the surface of enzyme, particularly in the area of the active center. The inhibitory effect of gossypol on intestinal lipid absorption could account for the hypolipidemic effect observed in cynomolgus monkeys by Shandilaya and Clarkson (1982).

The majority of both radioactivities was found in stomach contents, then in stomach wall and finally in small intestine wall. Therefore we have presented the relative radioactivity distribution into lipid classes only at these levels. One hour after meal intubation, these values were higher in the rats fed the meal without flour than in the rats fed the flours. Rats fed (+) gossypol cottonseed flour had higher levels of both radioactivities recovered than those fed () gossypol cottonseed flour. Rats fed soybean flour had lower levels of both radioactivities than those fed both () and (+) gossypol cottonseed flours. Nevertheless, after 3 h, these values were very low in the rats fed the three flours (Tables 3 and 4). At 1 and 3 h after meal intubation [³H]- and [¹⁴C]-radioactivities recovered in total gastrointestinal wall of rats fed () gossypol cottonseed flour were higher than in the rats fed soybean flour, both radioactivities in the rats fed (+) gossypol cottonseed flour were lower than in the rats fed soybean flour.

As early as 1 h after intubation, in stomach contents, [³H]-radioactivity was almost completely carried by free fatty acids and [¹⁴C]-radioactivity by triacylglycerol, diacylglycerol and free fatty acid fractions (Fig. 1). Only very low proportions of [³H]-free fatty acid and [¹⁴C]-triacylglycerol radioactivities were recovered in stomach wall (Fig. 2). These results indicate partial hydrolysis of the remaining [¹⁴C]-triolein in stomach and imply the same effect of flours on preduodenal lipase activity.

In intestinal wall (mucosa + tunica), [³H]-radioactivity was recovered primarily in free fatty acids, then in triacylglycerols, and finally in diacylglycerols, monoacylglycerols and phospholipids (Fig. 3). Once absorbed from the lumen, long-chain fatty acids are resynthesized into triacylglycerols or phospholipids, prior to their incorporation into chylomicrons or other lipoproteins. This esterification pathway was slightly but significantly slower in rats fed (+) gossypol cottonseed flour. In intestinal wall, [¹⁴C]-radioactivity was observed in triacylglycerols, but with time after intubation, [¹⁴C]-radioactivity found in phospholipids was enhanced (Fig. 3).

In conclusion, the present investigation of the influence of both cottonseed flours on digestion and absorption of [³H]-oleic acid and [¹⁴C]-triolein, after meal intubation, shows that compared with rats fed soybean flour, rats fed both cottonseed flours had a significantly diminished lipid absorption in gastrointestinal tract, probably because cottonseed flours, compared with soybean flour, contain more amphiphilic protein. Feeding (+) gossypol cottonseed flour amplifies this inhibitory effect, but these inhibitory effects are relatively limited. Though delayed, lipid digestion and absorption can be completed in the presence of both cottonseed flours. Therefore, () and (+) gossypol cottonseed flours, which do not impair strongly lipid absorption and digestion, can be used like soybean flour, in postweaning and rehabilitation diets, if the gossypol concentration in the (+) gossypol cottonseed flour is low ( 0.45 g/kg diet).

	FOOTNOTES

Manuscript received 18 September 1997. Initial reviews completed 11 November 1997. Revision accepted 22 June 1998.

	ACKNOWLEDGMENT

The authors thank Anne Magnet, an ESP linguist at the University of Burgundy, for editing the manuscript.

	LITERATURE CITED

Abstract Introduction Methods Results Discussion References

• Aspinall G. O., Cotrell I. W. Polysaccharides of soybean, 6. Neutral polysaccharides from cotyledon meal. Can. J. Chem. 1971; 49:1019-1025
• Bartlett G. R. Phosphorus assay in column chromatography. J. Biol. Chem. 1958; 234:466-469
• Bertrand V., Prost J., Belleville J. In vitro rat pancreatic digestive enzyme activities with raw and heated glanded and glandless cottonseed flours. Int. J. Food Sci. Nutr. 1995; 46:39-45[Medline]
• Bodwell, C. A. & Hopkins, D. T. (1985) Nutritional characteristics of oilseed proteins in: New Protein Foods (Altschul, AA, and Wilcke, H. L. eds.) Vol. 5. pp. 221-257, New York Academic Press,
• Chadha S., Sanyal S. N., Kanwar U. Effects of gossypol acetic acid on the absorptive and digestive function of rat intestine. Biochem. Int. 1988; 17:1117-1133[Medline]
• Chadha S., Sanyal S. N., Kanwar U. Alteration and reversibility of the effects of gossypol acetic acid on the intestinal uptake of end product nutrients in normal and protein calorie-malnourished male rats. Ann. Nutr. Metabol. 1991; 35:53-60[Medline]
• Clark S. B., Lawergreen B., Martin J. V. Regional intestinal absorptive capacities for triolein: an alternative to markers. Am. J. Physiol. 1973; 225:574-585[Free Full Text]
• Council of European Communities (1986) Council instructions about the protection of living animals used in scientific investigations. Off. J. Eur. Com. J. O. 86/609 CEE, 1358: 1-28.
• Fink C. S., Hamosh P., Hamosh M. Fat digestion in the stomach: stability of lingual lipase in the gastric environment. Pediatr. Res. 1984; 18:248-254[Medline]
• Folch J., Lees M., Sloane-Stanley G. H. Isolation and purification of total lipids from animal tissues. J. Biol. Chem. 1957; 226:497-509[Free Full Text]
• Gargouri Y., Julien R., Pieroni G., Verger R., Sarda L. Studies on the inhibition of pancreatic and microbial lipases by soybean proteins. J. Lipid Res. 1984; 25:1214-1221[Abstract]
• Gargouri Y., Pieroni G., Riviere G., Sugihara A., Sarda L., Verger R. Inhibition of lipases by proteins. J. Biol. Chem. 1985; 260:2268-2273[Abstract/Free Full Text]
• Gargouri Y., Pieroni G., Lowe P. A., Sarda L., Verger R. Human gastric lipase. The effects of amphiphiles. Eur. J. Biochem. 1986; 156:305-310[Medline]
• Hamosh, M. (1984) Lingual lipase in: Lipases (Borgström, B., and Brockman, H. L., eds.) Elsevier. pp. 49-81. Amsterdam, ND.
• Hamosh M., Klaeveman H. L., Wolf R. D., Scow R. O. Pharyngeal lipase and digestion of dietary triglyceride in man. J. Clin. Invest. 1975; 55:908-913
• Heywood R. The toxicology of gossypol acetic acid and () gossypol. Contraception 1988; 37:185-190[Medline]
• Institut de Recherches sur le Coton et les fibres Tropicales. Anonymous Cottonseed to join soybean as human protein source. Chem. Eng. 1970; 15:96-98
• Khalifa F., Bertrand V., Belleville J., Sarda L., Prost J. Short term effect of feeding raw or heated soya flour and casein meals on lipid intestinal digestion and absorption in rat. J. Nutr. Biochem. 1992; 3:224-231
• Khalil M., Ragab M., Hassanien F. R. Some functional properties of oilseed proteins. Nahrung 1985; 29:275-282[Medline]
• Lawhon J. J., Cater M. M., Mattil K. F. Evaluation of the food use potential of sixteen varieties of cottonseed. J. Am. Oil Chem. Soc. 1976; 54:75-80
• Liao T. H., Hamosh P., Hamosh M. Fat digestion by lingual lipase: mechanism of lipolysis in the stomach and upper small intestine. Pediatr. Res. 1984; 18:402-409[Medline]
• Radcliffe J. D. A comparison of the effects of dietary casein and cottonseed isolate on serum hepatic lipids in the rats. Nutr. Reports Int. 1989; 40:821-826
• Reyes J., Allen J., Tanphaichitr N., Bellve A. R., Benos D. J. Molecular mechanism of gossypol action on lipid membranes. J. Biol. Chem. 1984; 259:9607-9615[Abstract/Free Full Text]
• Shandilaya L. N., Clarkson T. B. Hypolipidemic effects of gossypol in cyanomolgus monkeys (Macaca fascicularis). Lipids 1982; 17:285-290[Medline]
• Stahl E., Schroter G., Kraft G., Renz R. Thin layer chromatography (the method affecting factors, and a few examples of application). Pharmazie 1956; 11:633-637[Medline]
• Vahouny G. V. Dietary fiber, lipid metabolism, and atherosclerosis. Fed. Proc. 1982; 41:2801-2806[Medline]
• Van Soest P. J., Robertson J. B., Lewis B. A. Method for dietary fiber, neutral detergent fiber, and nonstarch polysaccharides in relation to animal nutrition. J. Dairy Sci. 1991; 74:3583-3597[Abstract]
• Van Tilbeurgh H., Sarda L., Verger R., Cambillau C. Structure of the pancreatic lipase-colipase complex. Nature 1992; 359:159-162[Medline]

0022-3166/98 $3.00 ©1998 American Society for Nutritional Sciences