A GPI-anchored co-receptor for tissue factor pathway inhibitor controls its intracellular trafficking and cell surface expression.

BACKGROUND: Tissue factor pathway inhibitor (TFPI) lacks a membrane attachment signal but it remains associated with the endothelial surface via its association with an, as yet, unidentified glycosyl phosphatidylinositol (GPI)-anchored co-receptor. OBJECTIVES/METHODS: Cellular trafficking of TFPI within aerolysin-resistant ECV304 and EA.hy926 cells, which do not express GPI-anchored proteins on their surface, was compared with their wild-type counterparts. RESULTS AND CONCLUSIONS: Although aerolysin-resistant cells produce normal amounts of TFPI mRNA, TFPI is not expressed on the cell surface and total cellular TFPI is greatly decreased compared with wild-type cells. Additionally, normal, not increased, amounts of TFPI are secreted into conditioned media indicating that TFPI is degraded within the aerolysin-resistant cells. Confocal microscopy and studies using metabolic inhibitors demonstrate that aerolysin-resistant cells produce TFPI and transport it into the Golgi with subsequent degradation in lysosomes. The experimental results provide no evidence that cell surface TFPI originates from secreted TFPI that binds back to a GPI-anchored protein. Instead, the data suggest that TFPI tightly, but reversibly, binds to a GPI anchored co-receptor in the ER/Golgi. The co-receptor then acts as a molecular chaperone for TFPI by trafficking it to the cell surface of wild-type cells or to lysosomes of aerolysin-resistant cells. TFPI that escapes co-receptor binding is secreted through the same pathway in both wild-type and aerolysin-resistant cells. The data provide a framework for understanding how TFPI is expressed on endothelium.

Acetaldehyde disrupts tight junctions and adherens junctions in human colonic mucosa: protection by EGF and L-glutamine.

Acetaldehyde, a toxic metabolite of ethanol oxidation, is suggested to play a role in the increased risk for gastrointestinal cancers in alcoholics. In the present study, the effect of acetaldehyde on tyrosine phosphorylation, immunofluorescence localization, and detergent-insoluble fractions of the tight junction and the adherens junction proteins was determined in the human colonic mucosa. The role of EGF and L-glutamine in prevention of acetaldehyde-induced effects was also evaluated. Acetaldehyde reduced the protein tyrosine phosphatase activity, thereby increasing the tyrosine phosphorylation of occludin, E-cadherin, and beta-catenin. The levels of occludin, zonula occludens-1, E-cadherin, and beta-catenin in detergent-insoluble fractions were reduced by acetaldehyde, while it increased their levels in detergent-soluble fractions. Pretreatment with EGF or L-glutamine prevented acetaldehyde-induced protein tyrosine phosphorylation, redistribution from intercellular junctions, and reduction in the levels of detergent-insoluble fractions of occludin, zonula occludens-1, E-cadherin, and beta-catenin. These results demonstrate that acetaldehyde induces tyrosine phosphorylation and disrupts tight junction and adherens junction in human colonic mucosa, which can be prevented by EGF and glutamine.

The intestine expresses pancreatic triacylglycerol lipase: regulation by dietary lipid.

We identified the enzyme responsible for alkaline lipolysis in mucosa of rat small intestine. RT-PCR was used to amplify a transcript that, by cloning and sequencing, is identical to pancreatic triacylglycerol lipase. In rats fed normal laboratory chow, pancreatic triacylglycerol lipase mRNA was detected in all four quarters of the small intestine, with the first quarter expressing about three times as much of this transcript as was found in the more distal three-quarters combined. Both acutely and chronically administered dietary fat were shown to regulate pancreatic triacylglycerol lipase mRNA expression and lipase activity. The synthesis of pancreatic triacylglycerol lipase protein by the small intestine was demonstrated by in vivo radiolabeling experiments using [(35)S]methionine/cysteine followed by immunoprecipitation with an anti-pancreatic triacylglycerol lipase antibody. Immunohistochemical studies suggest that pancreatic triacylglycerol lipase protein expression is restricted to enterocytes throughout the small intestine. To our knowledge, this is the first report identifying rat small intestinal mucosa as a site of pancreatic triacylglycerol lipase synthesis and the first demonstration of its modulation in the mucosa by dietary fat. We propose that pancreatic triacylglycerol lipase is used by the intestine to hydrolyze the mucosal triacylglycerol that is not transported in chylomicrons.

Effect of increasing lipid loads on the ability of the endoplasmic reticulum to transport lipid to the Golgi.

We have previously shown (Mansbach, C. M. and P. J. Nevin, 1998. J. Lipid Res. 39: 963;-968) that after the development of a mass steady state with respect to triacylglycerol absorption in rats, the introduction of radiolabeled trioleoylglycerol, while maintaining the input rate of trioleoylglycerol constant at 135 micromol/h, was followed by a slow (60 min) achievement of a radiolabel steady state in the intestinal endoplasmic reticulum (ER). We hypothesized that this was due to the large input load and that the time to steady state would be shorter at lower lipid loads. Rats were infused intraduodenally with 22.5, 45, 90, or 135 micromol trioleoylglycerol/h for 6 h to obtain a mass steady state in the intestine. [(3)H]trioleoylglycerol was added to the infusate and the ER and Golgi were isolated from the proximal intestine after 5;-60 min of radiolabel infusion. The time required to reach a radiolabel steady state in the ER lengthened from 10 min at the 22.5 micromol/h infusion rate to 60 min at the 135 micromol/h rate. Similar data were obtained for the Golgi. Incubation of the ER with lipase reduced the ER;-triacylglycerol amount by 43% and increased its specific activity by 73%. The amount of [(3)H]TG-dpm in the ER was not reduced unless taurocholate, 10 mm, and colipase were added. We conclude that as the rate of triacylglycerol infusion is increased, TG movement from the ER to the Golgi progressively lengthens until finally all the triacylglycerol infused cannot be transported. A portion of this triacylglycerol is disposed on the cytoplasmic face of the ER and thus able to be attacked by lipase whereas another fraction is sequestered in the ER lumen and immune to lipase attack unless the ER membrane is solubilized.

Nutrient absorption.

Some key advances occurred last year in understanding mechanisms involved in nutrient absorption. A novel "prechylomicron transport vesicle" was identified; its movement to the Golgi is the rate-limiting step for triacylglycerol absorption. A scavenger receptor (type BI) in the brush border membrane appears to facilitate cholesterol uptake. Several studies define mechanisms for gastrointestinal peptide hormone stimulation of glucose uptake. An oligopeptide transporter, PepT1, is transcriptionally upregulated by certain dietary amino acids and dipeptides. Surprisingly, both insulin and fasting double the maximum velocity for dipeptide uptake (via PepT1), but they act by different mechanisms. Three transporters, SMVT (sodium-dependent multivitamin transporter for biotin and pantothenate), SVCT (for vitamin C), and CaT1 (for Ca uptake from the lumen) have been cloned and are active when expressed in various cells. Additional studies provide insights on Ca absorption and vitamin D action in aging, estrogen deficiency, and adaptation to a low Ca diet. Nramp2, also called DMT1 (divalent metal ion transporter), seems to be a major regulator of transferrin-independent, nonheme iron uptake. Finally, the protein HFE associates with the transferrin receptor and is part of an iron-sensing mechanism that regulates iron absorption. It is defective in hereditary hemochromatosis. HFE and Nramp2 (DMT1) genes are reciprocally regulated.

Nutrient absorption.

Interesting advances occurred recently in nutrient absorption. Kinetics of triacylglycerol appearance in endoplasmic reticulum, Golgi apparatus, and lymph support the hypothesis that endoplasmic reticulum-to-Golgi transport is rate-limiting for lipid absorption. Apolipoprotein B does not appear necessary for initial formation of chylomicron-sized lipid particles in the endoplasmic reticulum, but rather for their movement out of the endoplasmic reticulum and to the Golgi. If peptides are protected from luminal proteolysis by fatty acylation, or if a nonpeptide drug, acyclovir, is esterified with valine to enhance bioavailability, the peptides nevertheless are absorbed by peptide transporters. Experimental conditions needed to use human ileal mucosa for in vitro absorption studies are described. Intestinal mucosa contains leptin receptors, and leptin inhibits galactose absorption, suggesting a new site for leptin's modulation of body mass. The enhancer element for the apoB gene is located much farther from its structural gene in the intestine than in the liver.

Prechylomicron transport vesicle: isolation and partial characterization.

The intestine is able to regulate its output rate of chylomicrons, the major intestinal triacylglycerol (TG) transport vehicle. We have proposed that a vesicle, transporting the developing chylomicron from the endoplasmic reticulum (ER) to the Golgi, is the rate-limiting step in the process of TG transit through the enterocyte [Am. J. Physiol. 273 (Gastrointest. Liver Physiol. 36): G18-G30, 1997]. We wished to isolate and characterize this vesicle. The apical portion of rat intestinal cells were avulsed, and the mucosa was stirred in buffer. The supernatant was centrifuged in two different sucrose gradients, and the top 2.5 ml of the last gradient were collected and concentrated. Electron microscopy showed a 200-nm vesicle. The vesicle contained immunoidentifiable apolipoprotein (apo) B48 and apo A-IV but very little apo A-I, although apo A-I was present in the ER and Golgi. [3H]TG-loaded vesicles delivered [3H]TG to the Golgi but not the ER. Marker enzyme assays also indicate that the isolated fraction is different from the ER and Golgi fractions. We conclude that we have isolated a vesicle that is post-ER but pre-Golgi that vectorially transports TG to the Golgi.

Nutrient absorption.

Interesting advances occurred recently in nutrient absorption. Kinetics of triacylglycerol appearance in endoplasmic reticulum, Golgi apparatus, and lymph support the hypothesis that endoplasmic reticulum-to-Golgi transport is rate-limiting for lipid absorption. Apolipoprotein B does not appear necessary for initial formation of chylomicron-sized lipid particles in the endoplasmic reticulum, but rather for their movement out of the endoplasmic reticulum and to the Golgi. If peptides are protected from luminal proteolysis by fatty acylation, or if a nonpeptide drug, acyclovir, is esterified with valine to enhance bioavailability, the peptides nevertheless are absorbed by peptide transporters. Experimental conditions needed to use human ileal mucosa for in vitro absorption studies are described. Intestinal mucosa contains leptin receptors, and leptin inhibits galactose absorption, suggesting a new site for leptin's modulation of body mass. The enhancer element for the apoB gene is located much farther from its structural gene in the intestine than in the liver.

Intracellular movement of triacylglycerols in the intestine.

The intestine can vary its triacylglycerol output rate depending on differing physiological conditions. The rate-limiting step in the complex process from fatty acid and monoacylglycerol entry to triacylglycerol export is unknown but suggested to be the transport of triacylglycerol from the endoplasmic reticulum to the Golgi. The present studies were carried out to test this hypothesis. The conversion rate of absorbed fatty acid to mucosal triacylglycerol was studied in rats infused intraduodenally with trioleoylglycerol, 135 micromol/h, for 6 h followed by [3H]oleate. In 30 sec, 79% of the mucosal 3H-labeled fatty acid was esterified to [3H]triacylglycerol. The increase in the 3H specific activity of triacylglycerol in the endoplasmic reticulum and Golgi was studied in similarly prepared rats except that the radio-label was [3H]trioleoylglycerol. The endoplasmic reticulum triacylglycerol specific activity was always less than that of the Golgi with a steady state not reached until 60 min of [3H]trioleoylglycerol infusion. The steady state of [3H]triacylglycerol in the lymph was not reached until 70 min of infusion. We conclude that the data are consistent with the rate-limiting step in intestinal triacylglycerol export being the movement of triacylglycerol from the endoplasmic reticulum to the Golgi as the conversion of absorbed fatty acid to triacylglycerol is rapid and the movement of triacylglycerol from the Golgi to the lymph is rapid as well.

Advances in nutrition and gastroenterology: summary of the 1997 A.S.P.E.N. Research Workshop.

BACKGROUND: The 1997 A.S.P.E.N. Research Workshop was held at the annual meeting in San Francisco, on January 26, 1997. The workshop focused on advances in clinical and basic research involving the interface between nutrient and luminal gastroenterology. METHODS: Presentations on the genetic regulation of gastrointestinal development, the molecular biology of small intestinal adaptation, the effect of nutrition support on intestinal mucosal mass, the relationship between nutrition and gastrointestinal motility, nutrient absorption, and gastrointestinal tract substrate metabolism were made by the preeminent leaders in the field. RESULTS: The investigators presented an insightful analysis of each topic by reviewing data from their own laboratories and the published literature. CONCLUSIONS: This workshop underscored the important interactions between nutrition and luminal gastroenterology at the basic science, metabolic/physiologic, and clinical levels. The integration of presentations from the different disciplines provided a unique interaction of information and ideas to advance our understanding of nutrition and gastrointestinal tract.

Determinants of triacylglycerol transport from the endoplasmic reticulum to the Golgi in intestine.

The ability of the intestinal cell to export triacylglycerol (TG) is a physiologically regulatable function. The intracellular site where this occurs is unknown, although available evidence suggests that the step between the endoplasmic reticulum (ER) and the Golgi is the most likely. We studied this process in rat enterocytes that were isolated from the proximal intestine. A novel system was developed in which [3H]TG was transported from ER to the Golgi. This process was time, ATP, temperature, and cytosol dependent. The cytosolic factor(s) was heat and trypsin sensitive. TG transport was directly proportional to the amount of added nonradiolabeled acceptor Golgi. The rate of TG transported to the Golgi was the fastest in cells isolated from rats that had been intraduodenally infused in vivo with glyceryltrioleate (TO) plus phosphatidylcholine and slowest in cells isolated from bile-fistulated rats infused with TO in vivo compared with cells from in vivo TO-infused, bile duct intact rats, mimicking the relative transport rates seen in vivo. TG transport in vitro could not be quenched by adding TG emulsions, chylomicrons, liposomes, or guanosine 5'-O-(3-thiotriphosphate). Cytosol from the liver and kidney supported TG transport, but the Golgi from liver or kidney did not accept TG from intestinal ER. We conclude that an intestinally specific, active transport mechanism transports TG from the ER to the Golgi and that this might be a regulatory step in TG export from the intestinal cell.

Chylomicron remnant uptake by enterocytes is receptor dependent.

During glyceryl trioleate absorption in the rat, mucosal triacylglycerol (TG) fatty acids have been shown to consist of only 71% exogenous oleate. Chylomicron remnants are enriched with endogenous TG fatty acids, compared with their parent chylomicrons, which consist primarily of exogenous TG fatty acids. Because enterocytes have the apolipoprotein B-100/E receptor, this study was directed at determining whether the cells can take up and metabolize chylomicron remnants and, if so, whether this was receptor mediated. Isolated enterocytes were incubated with purified 3H-labeled chylomicron remnants. The remnants were shown to be taken up by the basolateral membrane, not the apical membrane. Remnant uptake was proportional to time and number of enterocytes, and saturation kinetics were observed. Nonradiolabeled remnants, human low-density lipoprotein (LDL), anti-LDL receptor antibody, and receptor-associated protein, an LDL-related receptor inhibitor, were all shown to compete for or reduce 3H-remnant uptake. Remnants taken up by the enterocytes could not be removed on incubation with excess human LDL. Uptake was shown to be greatest in the villus tips of the proximal intestine. These studies suggest that enterocytes take up chylomicron remnants by a receptor-mediated process from their basolateral membranes and that the remnants could provide a source of endogenous TG fatty acids for the enterocytes.

Intestinal triacylglycerol storage pool size changes under differing physiological conditions.

The intestinal mucosal triacylglycerol storage pool consists of triacylglycerol that is predominantly transported from the intestine via the portal vein rather than in chylomicrons (Am. J. Physiol. 1991. 261: G530-G538). Here we examined the size of the storage pool under varying physiological conditions. Four groups of rats were infused intraduodenally for 4 h. Group A was fasted; group B was infused with trioleoylglycerol, 135 mumol/h; group C was infused with trioleoylglycerol, 135 mumol/h plus phosphatidylcholine, 9 mumol/h; and group D was bile-diverted and infused with trioleoylglycerol, 135 mumol/h. The amount of triacylglycerol in the mucosa increased from groups A to D (A > B > C > D) but the storage pool triacylglycerol was least in groups A and C and greatest in groups B and D. The percentage of trioleoylglycerol in mucosal triacylglycerol was greater in groups B and D than in group A and greater in all groups than the percentage of oleate in the total fatty acids. We conclude that the triacylglycerol storage pool size varies inversely with the efficiency of lymphatic lipid output, which is greatest in rats infused with trioleoylglycerol plus phosphatidylcholine (group C) and least in bile-diverted rats infused with trioleoylglycerol (group D).

Role of the intestine in chylomicron remnant clearance.

When 810 mumol of [3H]glyceryl trioleate (TO) were infused intraduodenally over 6 h into rats, 29% of the triacylglycerol (TG) acyl groups in the mucosa were not from the infusate. We tested the hypothesis that chylomicron remnants contribute to the mucosal pool of nondietary TG acyl groups, since the acyl group composition of the chylomicron remnants was 58% oleate, compared with 90% in their parent chylomicrons. Purified 3H-labeled remnants were generated from chylomicrons formed in rats receiving TO intraduodenally, with 95% of the remnant disintegrations per minute (dpm) being in TG. The 3H-remnants were infused intravenously into rats receiving either saline or 135 mumol/h TO intraduodenally. In the saline-infused rats, 32% of the infused 3H dpm were in the proximal and 19% in the distal intestine and 32% were in the liver. In the fat-infused rats, 12% of the infused 3H dpm were in the proximal and 5% were in the distal gut and 29% were in the liver. When [3H]cholesterol-labeled remnants were infused intravenously and saline was infused intraduodenally, the percentage uptake into the mucosa was nearly the same as with the TG label, but comparable uptake by the liver increased. We conclude that the intestine competes with the liver for chylomicron remnant TG and cholesterol.

Effect of brefeldin A on lymphatic triacylglycerol transport in the rat.

Brefeldin A (BFA) has been shown in in vitro studies to either collapse the Golgi into the endoplasmic reticulum (ER) or the peripheral organelles into the trans-Golgi network. Our goal was to determine the effect of BFA on intestinal lipid transport, since the Golgi is thought to play an important role in this process and simultaneously establish the effectiveness of BFA in an in vivo system. We infused rats intraduodenally with glyceryl tri-[3H]oleate at 135 mumol/h for 15 h and included BFA, 750 micrograms/h, during hours 4-7 of infusion. Mass and lipid disintegrations per minute output into the lymph fell to 9% of input rates at 8 h of infusion and returned to steady-state values at 12 h of infusion. Both chylomicron and very low-density lipoprotein output were severely affected by the BFA. Electron microscopy showed that the Golgi was collapsed into the ER. Mucosal triacylglycerol (TG) mass and disintegrations per minute were increased at 7 h of infusion in BFA infused rats vs. controls in the proximal half of the intestine. Lipid absorption, lipase activity, and mucosal TG synthesis were normal in the BFA-treated rats. We conclude that BFA works in vivo and in the intestine collapses the Golgi into the ER. As a consequence, lymphatic TG transport was severely affected.

Alkaline lipase in rat intestinal mucosa: physiological parameters.

The present studies describe an alkaline active lipase in the rat intestinal mucosa. Alkaline lipase activity was determined using a glyceryl tri[14C]oleate emulsion at pH 8.0. Subcellular fractions were prepared from mucosal homogenates by differential centrifugation. Cells from villus to crypt were sequentially released using citrate and EDTA buffer. The enzyme was found to be most active in the proximal quarter of the intestine and in villus tips. Most of the activity was found in the cytosolic fraction. Bile salts stimulated the enzyme activity threefold. The presence of both Ca2+ and taurocholate was essential for optimal activity. Mucosal activity was greatly reduced on intraduodenal glyceryl trioleate infusion. Activity was restored when phosphatidylcholine was added to the glyceryl trioleate infusion. The fact that mucosal lipase has its greatest activity in the villus tips of the proximal intestine and in cytosol suggests that it may play an important role in mucosal glyceryltrioleate metabolism.

Portal transport of long acyl chain lipids: effect of phosphatidylcholine and low infusion rates.

The transport of absorbed long acyl chain lipids in the portal vein of rats has been shown to be 39% when the duodenal input rate is 135 mumol/h glyceryl trioleate (TO) [C. M. Mansbach II, R. F. Dowell, and D. Pritchett, Am. J. Physiol. 255 (Gastrointest. Liver Physiol. 18): G530-G539, 1991]. These calculations were based on a new experimental model in which portal flux is calculated from the knowledge of portal flow and the concentration of the lipids in excess in the portal vein vs. the carotid artery. To test this model, rats were infused for 6 h with a low rate of [3H]TO (27 mumol/h) with or without phosphatidylcholine (9 mumol/h) or with [3H]TO (135 mumol/h) plus phosphatidylcholine (9 mumol/h) or with [3H]TO (135 mumol/h) plus phosphatidylcholine (9 mumol/h). In all three cases, portal flux was expected to be less. Portal transport was 16.5% of the input rate in the low-dose group, 1.4% in the high-dose group given phosphatidylcholine, and 0.5% in the low-dose plus phosphatidylcholine group. There was no net transport of fatty acid in the portal vein in any of the three cases. These data show that portal lipid transport is dependent on the lipid load and that it is greatly reduced at high loads by including phosphatidylcholine in the lipid infusion.

Uptake and metabolism of circulating fatty acids by rat intestine.

The present study was designed to investigate the uptake and metabolism of circulating fatty acids by the intestinal mucosa in rats actively absorbing glyceryl trioleate given intraduodenally to determine the plasma fatty acid contribution to mucosal triacylglycerol. Rats with duodenal, femoral vein, carotid artery, and mesenteric lymph duct cannulas were used. [3H]oleate was constantly infused into the femoral vein while glyceryl trioleate was infused into the duodenum (135 mumol/h). After 5 h of infusion, a mass and radioactive steady state existed in the plasma and mucosa. At 6 h of infusion, the plasma oleate specific activity was sixfold greater than mucosal oleate and 50 times greater than mucosal triacylglycerol oleate; 86% of the mucosal oleate disintegrations/minute were in triacylglycerol. Chylomicron triacylglycerol oleate specific activity was less than that of the mucosa. Furthermore, the percentage of mucosal triacylglycerol acyl groups composed of oleate was greater than the percentage of oleate in mucosal free fatty acids. The data indicate that fatty acids are taken up by the mucosa during active fat absorption and metabolized primarily to triacylglycerols by the mucosa. The triacylglycerols in the mucosa synthesized from circulating fatty acids are selected against as a precursor of chylomicron triacylglycerol. The results support our previous hypothesis suggesting that the mucosa has at least two pools of neutral lipid (J. Lipid Res. 23: 1009-1019, 1982) and that steady-state conditions as performed here yield different results from previous work using bolus tracer injection techniques.

Topical application of WR-2721 to prevent radiation-induced proctosigmoiditis. A phase I/II trial.

Patients undergoing x-ray therapy to the pelvis have intestinal symptoms proportional to the volume treated and the dose delivered. WR-2721, S-2 (3-aminopropylaminoethyl) phosphorothioic acid, is an organic thiophosphate compound that selectively protects normal tissues against radiation effects. A Phase I/II study was done to test the ability of topical application of WR-2721 to protect the mucosa of the rectosigmoid from radiation damage. Thirty-one patients were enrolled in this study, of which, seven were control subjects. Twenty-four patients received WR-2721 daily, in enema form, 45 minutes before treatment. The patients were assigned by groups of three to receive increasing doses of WR-2721 beginning with 100 mg/enema to 450 mg/enema. Rectal mucosal biopsies were obtained within the treated field before, during, and at the end of therapy. The degree of damage to the rectal mucosa was scored on the basis of a 0 to 4 scale (with 0, least damage to 4, most damage) as determined by the percentage of damaged mucosal crypt glands. The patients' symptoms were recorded once a week during the entire course of therapy. The biopsy scores of the control group were slightly higher than those of the treatment groups; however, this difference did not appear to be significant. In the treated groups, there was a slight decrease in the biopsy scores with increasing doses of WR-2721, but this trend was not sustained. There were no differences among any of the groups in the symptoms experienced during the course of therapy. This study showed that WR-2721 could be administered safely in enema form in doses ranging from 100 to 450 mg/enema, but this drug did not protect the rectosigmoid mucosa from radiation damage at the doses administered.