Intestinal absorption of dietary maize glucosylceramide in lymphatic duct cannulated rats.

Sphingolipids are ubiquitous in all eukaryotic organisms. Various physiological functions of dietary sphingolipids, such as preventing colon cancer and improving the skin barrier function, have been recently reported. One of the common sphingolipids used as a foodstuff is glucosylceramide from plant sources, which is composed of sphingoid bases distinct from those of mammals. However, the fate of dietary sphingolipids derived from plants is still not understood. In this study, we investigated the absorption of maize glucosylceramide in the rat intestine using a lipid absorption assay of lymph from the thoracic duct. The free and complex forms of trans-4,cis-8-sphingadienine, the predominant sphingoid base of maize glucosylceramide, were found in the lymph after administration of maize glucosylceramide. This plant type of sphingoid base was detected in the ceramide fraction and N-palmitoyl-4,8-sphingadienine (C16:0-d18:2) and N-tricosanoyl-4,8-sphingadienine (C23:0-d18:2) were identified by LC-MS/MS. The cumulative recovery of 4t,8c-sphingadienine in the lymph was very low. These results indicate that dietary glucosylceramide originating from higher plants is slightly absorbed in the intestine and is incorporated into ceramide structures in the intestinal cells. However, it appears that the intact form of sphingoid bases is not reutilized well in the tissues.


Cannulation of the thoracic ducts of rats
This study was conducted in conformity with the policies and procedures detailed in the Animal Experiment Guidelines of Tohoku University. Surgeries and maintenance of rats and all other procedures were as described previously ( 29 ). Male Sprague-Dawley rats (9 weeks old) were obtained from Japan SLC (Hamamatsu, Japan) and were housed in stainless steel wire-mesh cages in a room kept at 23 ± 1°C with a 12 h-light: dark cycle. Each experimental or control group had fi ve rats. After acclimatization for 1 week with MF Standard Rodent Chow (Oriental Yeast, Tokyo, Japan) and distilled water (free access), the rats were anesthetized and a cannula (SV35, Dural Plastics) was inserted into their left thoracic channel to collect lymphatic fl uid, and a catheter (SP-55, Dural Plastics) was inserted into their stomach. After surgery, each rat was placed in a restraining cage in a warm recovery room. A physiological solution containing 139 mmol/l glucose and 85 mmol/l NaCl was infused continuously overnight at a rate of 3 ml/h through the stomach cannula, and the same solution was also provided as drinking water. On the next morning, after collection of lymph for 2 h as a blank control, the rats were infused with 3 ml of an emulsion as a single bolus through the stomach catheter. Test emulsions containing 200 mg triolein, 5 mg maize glucosylceramide or glucosyl-N-palmitoylsphingosine, 50 mg fatty acid-free albumin, and 200 mg sodium taurocholate were prepared by ultrasonication. Control experiments were done using an emulsion without glucosylceramide. After infusion of those emulsions into the rats, by the intestine is metabolized to fatty acids and a small part is resynthesized to complex sphingolipids ( 21,23 ). We previously investigated the digestion and absorption of plant-derived sphingolipids in the rat digestive tract and by Caco-2 human intestinal cells ( 26,27 ). Our study demonstrated that the digestibility of glucosylceramide from maize is similar to that of mammalian origin, but the metabolic fate of plant-derived sphingoid bases within enterocytes differs from that of sphingosine. Sphingoid bases, except for sphingosine, appear to be transported out of cells across the apical membranes of enterocytes by P-glycoprotein after absorption, and consequently the intestinal uptake is poor. However, the exact fate of dietary plant-derived sphingolipids is still not well understood. In this study, we investigated the absorption of dietary plant-derived glucosylceramide in the rat intestine using a lipid absorption assay of lymph from the thoracic duct.
The maize glucosylceramide was kindly donated by Nippon Flour Mills Co. Ltd. (Atsugi, Japan), and its purity was 96% determined by HPLC equipped with an evaporative light-scattering detector, as described previously ( 8 ). The composition of olic 1 M HCl for 18 h at 70°C ( 31 ). The free sphingoid bases liberated from the complex sphingolipids were then subjected to HPLC analysis for quantifi cation of total sphingolipids (total sphingoid base fraction).

HPLC analyses
The OPA derivatives of the free forms of sphingoid bases were analyzed using a reverse-phase HPLC system equipped with a fl uorescence detector ( 32 ). Each dry sample extract was dissolved in methanol and was then mixed with OPA reagent (methanol-OPA reagent, 4:1, v/v), which was freshly prepared daily by dissolving 5 mg OPA and 50 µL 2-mercaptoethanol in 0.1 ml ethanol and adjusting the volume to 10 ml with a borate buffer (pH adjusted to 10.5 with 1 M KOH). After the solution had been incubated for 30 min at room temperature, aliquots were injected into the HPLC system, which consisted of an LC-10AD pump and an RF-10AXL fl uorescence detector (Shimadzu Co., Kyoto, Japan). Sphingoid bases were separated on a 4.6 × 250 mm TSK gel ODS-80Ts (Tosoh, Tokyo, Japan). The column was eluted using a binary gradient consisting of acetonitrile (A) and water (B). The gradient profi le was as follows: 0-15 min, 45-20% B linear; 15-30 min, 20% B; 30-40 min, 20-0% B linear; 40-60 min, 0% B. The fl ow rate was 1.0 ml/min and the column temperature was maintained at 40°C. OPA derivatives were detected with an excitation wavelength of 334 nm and an emission wavelength of 440 nm. The sphingoid bases were quantifi ed from their peak area by comparison with standard curves ( 26,27 ). After the column eluent passed through the fl uorescence detector, it was analyzed on a LCMS-2010EV mass spectrometer (Shimadzu) equipped with an electrospray ionization interface for further identifi cation of peak components.

LC-MS/MS analyses
For the identifi cation of complex sphingolipids consisting of sphingadienine, the alkaline stable fraction of each lymph extract was further analyzed using a Prominence HPLC system coupled to LCMS-IT-TOF equipped with an electrospray ionization interface (Shimadzu) ( 33 ). A TSK gel super ODS column (2.0 × 50 mm, Tosoh) was eluted with methanol-water (95:5, v/v) containing 5 mM ammonium acetate at a fl ow rate of 0.2 ml/min. The MS was the infusion of the glucose:NaCl solution was continued. Lymph was collected in an EDTA-containing tube for analysis during the following intervals after the test-oil infusion: Ϫ 2 to 0, 0-1, 1-2, 2-3, 3-4, 4-5, and 5-6 h. After the lymph fl ow was measured, lymph was stored at Ϫ 30°C until analyzed.

Lipid extraction
Lipids were extracted from each sample of lymph with chloroform-methanol (2:1, v/v). After evaporation of the collected chloroform phase, the residue was saponifi ed with 0.4 M KOH in methanol at 38°C for 2 h to remove glycerolipids ( 30 ). The alkali-stable fraction was recovered with chloroform and then washed with water. The chloroform phase was evaporated to dryness, dissolved in chloroform-methanol (2:1, v/v), and then subjected to HPLC analysis for quantifi cation of free sphingoid bases (free sphingoid base fraction). A portion of the alkali-stable fraction of lymph extract was degraded with aqueous methan-

Statistical analysis
Data are reported as means ± SEM, n = 5. Statistical analyses were made by one-way ANOVA or Student's t -test.

RESULTS
There was no difference in the lymph output among rats infused with each emulsion, confi rming that the surgery and maintenance of rats had been carried out appropriately ( Fig. 2 ). To quantify the absorption of dietary sphingolipids via the lymph, the OPA derivatives of sphingoid bases in lymph extracts were analyzed using an HPLC system with a fl uorescence detector ( Fig. 3 ). After administration of maize glucosylceramide (1-2 h fraction), 4 t ,8 c -sphingadienine, a dominant constituent of maize glucosylceramide, clearly appeared in the alkali stable fraction (free sphingoid base fraction) of the lymph extract. The peak ascribed to 4 t ,8 c -sphingadienine was prominently increased in the lymph extract by hydrolysis with aqueous methanolic 1 M HCl (total sphingoid base fraction). These results demonstrate that the intact form of 4 t ,8 c -sphingadienine is absorbed from the digestive tract into the lymph. A small peak at the same retention time as 4 t ,8 c -sphingadienine that might be due to  fraction in the lymph after administration. The result indicates that 4 t ,8 c -sphingadienine, one of the principal components of maize glucosylceramide, can be absorbed via the lymph duct and is mainly incorporated into lymphatic complex sphingolipids.
To identify complex sphingolipids consisting of 4,8sphingadienine in the lymph after administration of maize glucosylceramide, the alkaline-stable fraction of lymph extract (total sphingoid base fraction) was analyzed to obtain structural information using LCMS-IT-TOF. Two peaks in the chromatogram of the lymph extracts before and after administration with the same retention time as 4 t ,8 c -sphingadienine showed a molecular ion of 4,8-sphingadienine OPA derivative [M+H] + at m/z 474 ( Fig. 4 ).
The lymphatic recovery of 4 t ,8 c -sphingadienine from the dietary maize-derived glucosylceramide is shown in Fig. 5 . The recovery of sphingadienine from the total sphingoid base fraction, which contains hydrolysates of complex sphingolipids such as glucosylceramide and ceramide, is more abundant than that from the free form in lymph 0 Ϫ 6 h after administration was corrected by the amount of sphingosine in the lymph collected from Ϫ 2 h to 0 h individually. The recovery of sphingosine in the lymph of rats infused with emulsion containing triolein only were increased in a time-dependent manner, because chylomicrons contain endogenous sphingomyelin and ceramide ( 34 ). The lymphatic recovery of sphingosine in glucosyl-N-palmitoylsphingosine-treated rats was signifi cantly higher than in rats infused with triolein ( Fig. 8 ). Figure 9 shows the calculated recovery rate of sphingoid bases from dietary glucosylceramides. The recovery of sphingosine after administration of glucosyl-N-palmitoylsphingosine was corrected by the mean value of sphingosine in rats infused with triolein ( Fig. 8 ). The cumulative recovery of total sphingosine were significantly higher than that of 4 t ,8 c -shingadienine at 1 h and at 2 h ( Fig. 9B ). However, there was no signifi cant difference between the groups 3 h after administration. Up to 6 h after the treatment, the cumulative recovery of 4 t ,8 csphingadienine from maize glucosylceramide and sphingosine from glucosyl-N-palmitoylsphingosine was 0.18 ± 0.04% and 0.55 ± 0.22%, respectively.
Focusing on the difference of sphingoid base structure, the absorption of plant-derived sphingoid bases were compared with those of an authentic glucosyl-N-palmitoylsphingosine used as a model for glucosylceramide of mammalian origin. Sphingosine in the free and total sphingoid base fractions of lymph extract before administration ( Ϫ 2 h to 0 h) was 19.8 ± 4.7 and 4,635 ± 595 pmol/ ml, respectively. Therefore, the recovery of sphingosine  25 We have reported that the uptake of sphingosine is signifi cantly higher than the uptake of other sphingoid bases (including 4,8-sphingadienine) in differentiated Caco-2 cells (as a model for intestinal cells) and that P-glycoprotein probably contributes to this selective absorption of sphingosine ( 27 ). P-glycoprotein, encoded by the MDR1 gene, transports a wide variety of hydrophobic compounds, including drugs, natural products, toxicants, and peptides, and contributes to the barrier function of the gut ( 41 ). In this study, the cumulative recovery of 4 t ,8 c -sphingadienine in the lymph is significantly lower than that of sphingosine until 2 h after the administration of glucosylceramide. It is diffi cult to quantify the exact amount of sphingosine due to the diet in the lymph, because chylomicrons secreted from intestinal cells contain endogenous sphingolipids ( 34 ). However, endogenous sphingosine might not affect the recovery of sphingosine from the diet within short time periods after administration ( Fig. 8 ). Therefore, our observations in this study appear to support an important role for P-glycoprotein in the effl ux of 4,8-sphingadienine, but not sphingosine, across the apical membranes of enterocytes after absorption.
In this study, the total recovery of sphingosine in an intact form after administration of glucosyl-N-palmitoylsphingosine was lower than reported in other studies, which used labeled sphingolipids ( 42 ), because most of the dietary sphingosine is converted to palmitic acid in the intestinal mucosa and is incorporated into chylomicron the predominant structure of sphingoid bases in higher plants, can be absorbed into the lymph by intestinal cells after administration of maize glucosylceramide, but the recovery is extremely low. A large portion of the 4,8sphingadienine absorbed by intestinal cells is a constituent of complex sphingolipids, including at least ceramide, while the smaller portion is in the free form in the lymph. The cumulative recovery of 4,8-sphingadienine is relatively lower than that of sphingosine after administration of glucosyl-N-palmitoylsphingosine.
Dietary sphingolipids having sphingosine can be degraded in the intestine (21)(22)(23)(24)(25). Several reports have indicated that alkaline sphingomyelinase is present in the intestinal mucosa (35)(36)(37) and that the glycosylceramidase activity in the intestine is due to lactase-phlorizin hydrolase ( 38,39 ). Neutral ceramidase encoded by Asah2 is also present in the small intestine ( 40 ). We have reported that the hydrolytic activity in the small intestine toward glucosylceramide and ceramide derived from maize is consistent with the fi ndings of such activity toward mammalian-derived sphingolipids ( 26 ). Plant-derived sphingolipids might be hydrolyzed on the surfaces of intestinal villi, because free sphingoid bases and ceramide appear in the small intestinal lumen after ingestion. Plant-derived glucosylceramide would be hydrolyzed to free sphingoid bases in the digestive tract before absorption into the lymph. The low lymphatic recovery of sphingadienine (0.18%) can be attributed to the relatively low capacity for hydrolysis of sphingolipids in the intestine.
in intestinal cells during absorption. Interestingly, Mao et al. ( 45 ) surmised that phosphorylation followed by dephosphorylation is required for the incorporation of exogenous sphingoid bases into complex sphingolipids. Considering our results, some 4,8-sphingadienene might be phosphorylated by sphingosine kinase after uptake by intestinal cells. However, there are no data about the phosphorylation in mammals of sphingoid bases that originated from plants.
It is well known that ceramides in the stratum corneum are required for the epidermal permeability barrier. There have been several reports concerning the improvement of skin barrier function by dietary sphingolipids prepared from higher plants (46)(47)(48)(49). However, the structures of sphingoid bases of ceramide in the skin are quite different from those of plant-derived sphingolipids. In addition, the present study shows that the recovery of dietary sphingolipid prepared from higher plants is extremely low. We are interested in elucidating the mechanism for the effect of dietary plant-derived sphingolipids on skin barrier function.
In conclusion, we demonstrated that dietary glucosylceramide derived from higher plants is only slightly absorbed by the intestine and is incorporated into ceramide structures in intestinal cells. It appears that the intact form of sphingoid bases is hardly reutilized in tissues, because the recovery is extremely low. Our fi ndings provide an important insight into the intestinal absorption of dietary sphingolipids derived from plant sources.
triacylglycerol ( 21,23 ). Phosphorylation of the primary hydroxyl group of sphingoid bases is the penultimate step in the degradation of sphingolipid ( 43 ). The phosphate esters formed are subsequently cleaved by sphingosine phosphate lyase into ethanolamine phosphate and the corresponding aldehyde after which the long-chain aldehyde is further converted to fatty acids ( 44 ). Our results show the recovery of the intact form of 4,8-sphingadienine but not the complete recovery of sphingoid bases, including their catabolites absorbed via the lymph. Clarifi cation of the catabolism of various sphingoid bases is important to understanding the metabolism and biological function of dietary sphingolipids from various foodstuffs.
Although fatty acids of maize-derived glucosylceramide are mostly 2-hydroxy fatty acids ( 16 ), ceramides consisting of 4,8-sphingadienine found in the lymph after administration of maize glucosylceramide have nonhydroxy fatty acids (C16 and C23). This result indicates that the 4,8sphingadienine produced by the digestion of maize glucosylceramide in the intestinal tract is metabolized to ceramide