Alteration of endothelial proteoglycan and heparanase gene expression by high glucose, insulin and heparin.

Heparan sulfate proteoglycans (HSPGs) contain a core protein with glycosaminoglycans attached. Reduced glycosaminoglycan, in endothelial HSPGs syndecan and perlecan, is associated with diabetic cardiovascular complications but changes in core protein remain controversial. Since heparanase degrades heparan sulfate, we wished to determine if changes in endothelial heparanase mRNA, by high glucose (HG), correlate with changes in syndecan and perlecan core proteins, and to observe effects of heparin or insulin. RNA was isolated from cultured human aortic endothelial cells treated with HG (30mM), insulin (0.01 units/mL), heparin (0.5µg/mL), HG plus heparin and/or insulin for 24h. Real time PCR revealed that HG alone significantly increased heparanase, decreased syndecan with no effect on perlecan mRNA. Heparin or insulin significantly prevented the increase in heparanase but decreased perlecan mRNA while heparin, but not insulin, prevented the decrease in syndecan mRNA in HG treated cells. HG plus heparin and insulin increased heparanase and syndecan mRNA compared to all other treatments and decreased perlecan mRNA compared to control and HG alone. Heparin may protect endothelium from HG injury by reducing heparanase and increasing syndecan while insulin inhibits heparanase expression. Effects with insulin plus heparin suggest interference in transcriptional regulation of heparanase and syndecan genes.

Renal protection in diabetes: is it affected by glucose control or inhibition of the renin-angiotensin pathway?

BACKGROUND: Recent reports indicate increased risk of renal failure with long-term use of angiotensin-converting enzyme inhibitors (ACEI) in diabetes. End-stage renal disease (ESRD) in diabetes has increased despite ACEI and angiotensin receptor blocker (ARB) use. This study questions renal protection by ACEI or ARB. Our hypothesis is that uncontrolled hyperglycemia is central to diabetic ESRD while tight glucose control is renoprotective. Cultured endothelial cells show morphological damage that increases with duration of exposure to high glucose and is prevented by insulin and more so by a combination of insulin and heparin. METHODS: Findings from individual patients are compared to clinical trial results wherein ACEI or ARB was emphasized as the prime therapy to prevent progression of diabetic nephropathy to ESRD. Serum creatinine (Scr) changes were the main indicator of renoprotective effects in clinical trials. Creatinine clearance (Ccl), an important marker of glomerular filtration rate, was seldom reported. RESULTS: Our observations show that ACEI-treated patients develop progressive renal failure, whereas renal function remains stable with optimum glucose control. Clinical trials showed that reduction of proteinuria, with ACEI, reduces the risk of ESRD. Our studies show that reduction of proteinuria with use of ACEI or ARB parallels a reduction in Ccl, suggesting that a change in proteinuria is related to Ccl changes. Scr changes are small, giving a deceptive view of renal protection. CONCLUSIONS: Our observations find no evidence of renal protection with ACEI or ARB use in diabetes. Laboratory studies and clinical observations suggest that adequate glucose control is the key to renal protection in diabetes.

Enhanced antithrombotic effects of unfractionated heparin in rats after repeated oral doses and its relationship to endothelial heparin concentration.

BACKGROUND AND PURPOSE: An oral, single dose of 7.5 mg kg(-1) of unfractionated heparin (UFH) reduces thrombosis by 50% in a rat model of venous thrombosis. As long-term use is required clinically, our objectives were to study the antithrombotic effects following repeated oral UFH administration. EXPERIMENTAL APPROACH: Bovine lung UFH was administered by oral gavage to rats in 3 doses of 7.5 mg kg(-1) each 12, 24, 48, and 72 h apart; and in 3 or 15 doses of 1 mg kg(-1) every 48 h. The last dose was given immediately after thrombus initiation where 10% formalin in methanol was applied to the jugular vein. The vessel was examined for thrombosis 4 h later. Amounts of heparin in tissue and endothelium, and plasma anticoagulant activity were measured. KEY RESULTS: When 3 x 7.5 mg kg(-1) heparin was given, thrombotic incidence was most reduced at 48 h dose-intervals and was significantly less than single dose treatment. There was a negative correlation between endothelial heparin content and thrombotic incidence, but not anticoagulant activity. When 3 doses of 1 mg kg(-1) every 48 h were given, thrombotic incidence was similar to single dose treatment. When 15 doses were given, total thrombotic incidence was less than for 3 doses and was similar to that after s.c. administration. CONCLUSIONS AND IMPLICATIONS: Antithrombotic activity increased with repeated doses of oral UFH, with antithrombotic effects similar to s.c. administration. Antithrombotic activity was related to heparin on endothelium.

Electron microscopic analysis of glucose-induced endothelial damage in primary culture: possible mechanism and prevention.

We previously reported that high glucose treated cultured endothelial cells (ECs) showed intercellular gaps by transmission electron microscopy (TEM). These gaps were abrogated with insulin and/or heparin treatment. Our aims were to assess the severity of injury in ECs treated with high glucose for variable duration, and to further study the protective effects of insulin and/or heparin. Cells were also treated with L-buthionine sulfoximine (BSO), a glutathione inhibitor, to help understand the mechanism of high glucose injury. Primary porcine ECs were treated with high glucose (30 mM) for 2, 6 or 10 days; and glucose plus insulin (1 U/ml), glucose plus heparin (5 microg/ml), glucose plus insulin plus heparin for 6 days. ECs were treated with BSO (0.001-0.05 mM) for 2 days. Pellets from trypsinized cells were processed for TEM. High glucose treatment revealed apoptosis or necrosis showing variable cell size, abnormal nuclei, condensation of nuclear chromatin, few mitochondria, cell membrane disruption and needle-shaped structures. Changes increased with duration of exposure. In high glucose plus heparin or insulin treated cultures at least one-half of the cells appeared normal. Most ECs were intact when treated with high glucose plus insulin plus heparin. BSO treatment showed dose-dependent changes with low doses showing apoptosis whereas higher doses revealed necrosis similar to high glucose treatment for 6 or 10 days. High glucose-induced EC injury increased with duration of exposure. These data demonstrate that high glucose injury resembles that of BSO treatment, suggesting that glutathione depletion may be involved in EC injury. Insulin and/or heparin protect against high glucose-induced injury.

Vitamin E protects porcine but not bovine cultured aortic endothelial cells from oxygen-derived free radical injury due to hydrogen peroxide.

The antioxidative effects of vitamin E (VE) are well known and have been demonstrated in in vitro studies. Since we previously observed that dextran sulfate was markedly more protective of porcine versus bovine aortic endothelial cells when damaged by hydrogen peroxide (H2O2), our objectives were to determine if a similar species difference could be observed with VE. The effects of VE or Trolox (a more water-soluble VE) against oxygen-derived free radical (OFR) injury produced by H2O2 was studied in porcine aortic endothelium (PAE) vs. bovine aortic endothelium (BAE) and bovine brain microvessel endothelium (BBME). VE or Trolox was added to culture medium for at least 24 h prior or immediately prior to H2O2 addition. In PAE, pretreatment with VE dissolved in either ethanol (VE-EtOH) or Tween 20 (VE-Tween 20), or Trolox dissolved in DMSO (Trolox-DMSO) was protective, shown by increased percent viable cells and reduced lactate dehydrogenase (LDH) release. EtOH, Tween 20 or DMSO alone was protective in PAE although DMSO or Tween 20 alone was less effective than when added with VE. VE-Tween 20 or Trolox-DMSO protected PAE when added just prior to H2O2 injury, but protection was significantly less than with pretreatment. DMSO immediately prior to H2O2 injury had no protective effect. Tween 20 immediately prior resulted in complete cell death. In BAE and BBME, pretreatment with VE-EtOH, EtOH, Trolox-DMSO, or DMSO alone had little or no protective effect. Pretreatment with VE-Tween 20 or Tween-20 alone was protective of BAE with Tween 20 being more effective than VE-Tween 20 suggesting that Tween 20 was the protective agent. These studies show that the protective effects of VE and Trolox as well as DMSO, EtOH, and Tween-20 are species dependent.

Antithrombotic efficacy in a rat model of the low molecular weight heparin, reviparin sodium, administered by the oral route.

Previous studies in rats show that unfractionated heparin and the low molecular weight heparin logiparin have a dose-dependent antithrombotic effect and are found in endothelium and plasma when administered orally. Objectives of the present study were to determine if similar evidence of absorption could be observed with oral reviparin sodium. Thrombosis incidence was determined 4 h after application of 10% formalin in methanol to the exposed jugular vein. A dose-dependent antithrombotic effect was observed when 0.01 to 7.5 mg/kg (20 rats/group) was administered by stomach tube immediately following thrombus initiation. Thrombotic incidence was also significantly reduced when 0.025 mg/kg was given 4 and 2 h prior to, immediately after, and 2 and 3 h following thrombus initiation. Reviparin was recovered from endothelium and plasma in trace amounts at all doses. At 0.025 mg/kg, peak aortic endothelial reviparin concentrations were found at 1 and 2 h and peak plasma anti-Xa activity was detected at 2 h. Trace amounts of plasma TFPI were found only at 8 h after administration. Dose-dependent antithrombotic activity and recovery from endothelium and plasma support the hypothesis that orally administered reviparin sodium is absorbed.

Marked variation in free radical injury between bovine and porcine endothelial cells cultured in different media.

Previous studies produced models of oxygen-derived free radical (OFR) injury, using H(2)O(2) or xanthine/xanthine oxidase (X/XO), in cultured porcine aortic endothelium (PAE) and rat coronary endothelium. H(2)O(2) at 0.1 mM resulted in 50% viability in both cell types. To determine if comparable H(2)O(2) or X/XO concentrations have the same injurious effect on endothelium from other sources, models of OFR injury were developed for bovine aortic endothelium (BAE) and bovine brain microvessel endothelium (BBME). Varying concentrations of H(2)O(2) (0.01 to 6 mM) or X/XO (10 microM/0.1 to 0.3 U/mL) were added to medium 24 h prior to evaluating cell damage. Injury was assessed using the Trypan blue exclusion test (% viability) and by measuring the release of lactate dehydrogenase into medium. H(2)O(2) concentrations required to produce 50% viability were >6 mM in BAE and BBME versus 1 mM in PAE when cells were grown in Dulbecco's modified Eagle's medium (DMEM). Similarly, BAE and BBME were less sensitive than PAE to damage by X/XO. Cells from both species were more sensitive to H(2)O(2) or X/XO injury when grown in Medium 199 (M199) versus DMEM. The most profound difference was observed with PAE where 50% viability was obtained with 0.12 versus 1.05 mM H(2)O(2) in M199 versus DMEM. These results indicate that bovine endothelial cells from aorta and brain are more resistant to free radical injury than PAE. The presence or absence of key media components (iron, pyruvate, cysteine, histidine) likely influences the extent of OFR injury.

Tissue distribution and antithrombotic activity of unlabeled or 14C-labeled porcine intestinal mucosal heparin following administration to rats by the oral route.

Distribution and antithrombotic activity of orally administered unfractionated porcine heparin were studied. [14C]Heparin was prepared by de-N-acetylation of porcine mucosal heparin followed by re-N-acetylation, using [14C]acetic anhydride. [14C]Heparin and (or) cold heparin (60 mg/kg) were administered by stomach tube to male Wistar rats. Blood, all levels of gut and gut contents, liver, lung, spleen, kidney, and aortic and vena caval endothelium were collected under deep anesthesia at 3, 6, 15, 30, and 60 min and 4 and 24 h (6 rats/group) after administration. Urine and feces were collected at 24 h, using metabolic cages. In three additional rats, drugs were administered in gelatin capsules. Tissues listed above and tongue, esophagus, trachea, brain, heart, thymus, bile ducts, vena caval and aortic walls, ureters, bladder, samples of muscle, skin, hair, and bone marrow were collected at 24 h. Radioactivity and chemical heparin, measured by agarose gel electrophoresis, were observed in all tissues examined as well as gut washes, plasma, urine, and feces. Radiolabel recovered was confirmed to be heparin by autoradiograms of gradient polyacrylamide electrophoretic gels. [14C]Heparin and chemical heparin in gut tissue suggest a transit time of 4 h. Porcine or bovine heparin (7.5 mg/kg), administered by stomach tube, decreased the incidence of thrombosis induced by applying 10% formalin in 65% methanol to the exposed jugular vein of rats. Heparin isolation from non-gut tissue, endothelium, urine, and plasma and the observed antithrombotic effect are consistent with oral bioavailability.

Antithrombotic activity of orally administered low molecular weight heparin (Logiparin) in a rat model.

Previous studies in rats demonstrated that orally administered, unfractionated bovine lung heparin is absorbed and has a dose-dependent antithrombotic effect. The objective of this study was to determine if an oral low molecular weight heparin had a similar antithrombotic effect in the same model. Thrombosis was induced in rats by application of 10% formalin in 65% methanol to the exposed jugular vein. Immediately following, saline, unfractionated heparin (3.3-60 mg/kg) or the low molecular weight heparin, Logiparin (0.025-15 mg/kg; 20-30 rats per group) was placed in the stomach and 4 h later the jugular vein was inspected for a thrombus. Compared to saline, oral Logiparin reduced the incidence of thrombosis at all doses with a dose-dependent effect suggested. A significant increase was observed in the activated partial thromboplastin time and in plasma heparin concentrations, determined by Accuclot Heptest and anti-factor Xa chromogenic assay for rats given oral Logiparin versus saline. A dose-dependent increase in plasma heparin concentration was observed when estimated by the anti-Xa chromogenic assay. Heparin was recovered in 9% of aortic endothelial samples when > or = 0.8 mg/kg Logiparin was administered. A 50% reduction in thrombosis was observed at 0.1 mg/kg for oral Logiparin versus 7.5 mg/kg for unfractionated bovine lung heparin indicating that oral Logiparin is an effective antithrombotic agent at doses lower than unfractionated heparin. Orally administered low molecular weight heparin may be useful for the prevention and treatment of thrombosis.

Orally administered dextran sulfate is absorbed in HIV-positive individuals.

Preliminary in vivo studies suggested that oral dextran sulfate was poorly absorbed, but investigations were limited by inadequate methods for measuring the drug in the body. To determine absorption in HIV-positive subjects, hydrogenated dextran sulfate, average molecular weight 8000 (Usherdex 8), was orally administered in a short-term (single dose, 4 g/day for 5 days, 7 subjects) and in a long-term study (1 g, 4 times per day for 29 to 335 days, 8 subjects), which was a continuation of the short-term study with the inclusion of an additional subject. When an agarose gel electrophoresis technique with toluidine blue staining was used, the drug was recovered from plasma (67%, peak 2.2 microg/mL) and circulating peripheral blood lymphocyte (PBL) samples (50%, peak 333 microg/L blood) obtained at 5 and 15 minutes and 1, 3, 6, and 24 hours after the first day's dose and from plasma (56%) and PBL samples (38%) obtained 5 minutes after administration on 4 subsequent days in the short-term study. In the long-term study, the drug was found in plasma (67%, peak 2.4 microg/mL) and PBL samples (25%, peak 126 microg/L blood) obtained at monthly visits within 4 hours of the last dose. The drug was found in all urine samples from all subjects in both studies (short-term study, 24-hour samples up to 4 days after the final dose; long-term study, monthly samples within 4 hours of the last dose). In the long-term study, bone marrow preparations from 3 subjects showed metachromatic inclusions present in reticular cells when the cells were stained with toluidine blue, indicating the presence of sulfated polyanions. A significant rise in activated partial thromboplastin time and a drop in platelet count (P < .025) were demonstrated, with thrombocytopenia developing in 3 patients. Mild-to-moderate gastrointestinal disturbances were experienced by 6 subjects in the short-term study and by all subjects in the long-term study. One subject experienced mild central nervous system symptoms in the short-term study. These results indicate that dextran sulfate is absorbed after oral administration; therefore, further studies on its efficacy, particularly in the early stages of the disease, along with additional observations on its toxicity, are warranted.

Antithrombotic activity of oral unfractionated heparin.

Although heparin is believed to be poorly absorbed orally, we recently demonstrated that oral heparin rapidly enters the circulation, with most of the drug being taken up by endothelium. To determine the effective antithrombotic dose of oral heparin, we induced thrombosis by applying 10% formalin in 65% methanol to exposed rat jugular vein. Saline or heparin, at doses ranging from 3.25 to 60 mg/kg, was immediately placed in the stomach; 4 h later, the vein was inspected for a thrombus. A dose-dependent decrease in thrombosis was observed with oral heparin. Although there was little change in anticoagulant activity as measured by the activated partial thromboplastin time (APTT) of plasma samples taken 4 h after administration, a significant dose effect was demonstrated by regression analysis. Heparin could be demonstrated chemically in 52% of plasma samples and in 38% of aortic or vena caval endothelial samples. A significant dose effect was observed in aortic endothelial heparin concentrations, with amounts 1,000-fold that determined in plasma. These results indicate that oral heparin exhibits antithrombotic activity in a dose-dependent manner, with low levels in plasma.

Dextran sulphates protect porcine arterial endothelial cells from free radical injury.

1. The ability of dextran sulphate to protect cultured porcine arterial endothelial cells injured by addition of xanthine and xanthine oxidase (X/XO) or hydrogen peroxide to cell medium was examined using a variety of drug preparations. Cell damage was assessed by determining cell viability (by trypan blue exclusion) and release of lactate dehydrogenase into the medium. 2. Dextran sulphates of average molecular weight (M(r)) 5000, 8000 (hydrogenated or unhydrogenated) at 0.05, 0.5, 5 and 50 micrograms ml-1 medium, added 24 h prior to X/XO, protected cells, whereas dextran sulphate M(r) 500,000 was protective only at 0.5 microgram ml-1. 3. None of the dextran sulphates used showed any toxic effect on cells in concentrations up to 500 micrograms ml-1 medium. 4. When the duration of pretreatment with dextran sulphate M(r) 8000 was varied, 6 h was required for a protective effect on cells damaged by X/XO, which was enhanced with durations of 16 and 24 h. 5. Dextran sulphates had a similar protective effect on cells damaged by hydrogen peroxide. 6. This study suggest that dextran sulphates may prevent conditions resulting from free radical injury.

The heparin target organ--the endothelium. Studies in a rat model.

Plasma levels of the antithrombotic drug heparin, as estimated by coagulation tests, are a poor indicator of antithrombotic effectiveness. The interaction of heparin with endothelium is a poorly studied but important factor in the clinical activity of heparin. This study describes the interaction of heparin with endothelium, following intragastric administration. The concentrations of heparin in endothelium and plasma were determined by gel electrophoresis following administration of heparin to rats by various routes. Heparin concentrations in endothelium versus plasma were approximately 100 times greater following intravenous or ex vivo administration and more than 1000 times greater when administered by intrapulmonary, subcutaneous, intraperitoneal and intragastric routes indicating that the route of administration affects the distribution of the drug. At 2.4 and 6 min after intravenous administration, 88 and 51% respectively of the administered dose was found associated with endothelium. Heparin was rapidly absorbed following intragastric administration and could be detected associated with endothelium at 2.4 min. At 6 min less than 1% of the administered dose was found in plasma, and 45% was associated with endothelium. These results show that endothelium is the main site of heparin distribution. Heparins could also be detected in cellular and pericellular fractions of cultured porcine aortic endothelial cells when 125I-heparin was added to medium. Bound radioactivity was released to medium from both cellular and pericellular fractions suggesting that heparin taken up by endothelium can be released. Intragastric administration of heparin and dextran sulphates significantly prevented thrombus formation in a rat model of thrombosis without significant changes in activated partial thromboplastin times.(ABSTRACT TRUNCATED AT 250 WORDS)

Evidence from endothelium of gastric absorption of heparin and of dextran sulfates 8000.

Heparin, hydrogenated dextran sulfate 8000 (Usherdex 8), and dextran sulfate 8000 were administered to rats, and the total drug was separated and determined in endothelium and plasma. A large amount of each drug was recovered from endothelium 2.4 and 6 minutes after intravenous injection. This accounted for the drug missing from plasma. The drugs in water were placed in the stomach by catheter. All three drugs were recovered from the endothelium and identified unchanged by electrophoresis and specific staining. The amounts that were recovered at 2.4 and 6 minutes were equivalent to most of the drug administered. Thus heparin, Usherdex 8, and dextran sulfate 8000 enter the body immediately on oral administration. At longer time intervals after intravenous and oral administration, much of each drug was not demonstrable in endothelium by the method used. Some drug could be detected in endothelium after 4 hours. After oral administration, plasma levels of each drug were rarely more than 0.5% of the dose. Formalin-alcohol was applied to the jugular veins of anesthetized rats to produce a thrombus, (see Blake et al. J Clin Path 1959;12:118-22) and the drugs were immediately introduced into the stomach. Four hours later the injured veins were inspected for thrombi. Incidence of thrombotic plug was 80% in rats that received saline solution, 4% with Usherdex 8, 0% with dextran sulfate 8000, and 0% with heparin. Usherdex 8, dextran sulfate 8000, and heparin demonstrate low, moderate, and high in vitro anticoagulant activity, respectively.(ABSTRACT TRUNCATED AT 250 WORDS)

Heparin protects cultured arterial endothelial cells from damage by toxic oxygen metabolites.

Toxic oxygen metabolites can damage endothelial cells and may play an important role in the initiation and progression of atherosclerotic lesions. Since the antithrombotic drug heparin, interacts with endothelium, we wished to determine if heparin would protect endothelial cells from free radical injury. Endothelial cell injury was produced by the addition of xanthine and xanthine oxidase to cultured cells and assessed by changes in cell viability and release of lactate dehydrogenase (LDH) to the media. Pretreatment with heparin 24 h prior to addition of xanthine and xanthine oxidase significantly decreased cell damage. We suggest that heparin (and related compounds) can protect endothelium from free radical damage, and is therefore prophylactic for ischemic and inflammatory injury, and the development and progression of atheroma.

Determination of absolute amounts of heparin and of dextran sulfate in plasma in microgram quantities.

Heparin and dextran sulfates 8000 are separated from citrated plasma by absorption on epichlorohydrin triethanolamine cellulose columns followed by elution with 1.1 and 1.4 mol/L NaCl in 0.05 mol/L glycine-HCl buffer. The eluate is desalted with Sephadex G25-40, dried, and dissolved in water. A 1 microliters sample is applied to an agarose gel slide. After electrophoresis, the slide is fixed and stained with toluidine blue. The sulfated polysaccharide band(s) is identified by relative electrophoretic migration. The total amount of drug is estimated by matching its optical density with that of a band on one of a set of slides with graded amounts of heparin or dextran sulfate. The reaction with toluidine blue measures the total polyelectrolyte, not just the small proportion of the drug with anticoagulant activity. Pooled normal plasma showed a trace of chondroitin and no heparin. Recovery of heparin and hydrogenated dextran sulfate that was added to pooled normal plasma was complete (lowest concentration tested was 10 micrograms/ml); however, recovery for unhydrogenated dextran sulfate declined consistently by 9 micrograms/ml for concentrations below 50 micrograms/ml, setting a limit for its recovery. Plasma samples taken from patients for coagulation tests were examined by this procedure, and in so doing, steps were ascertained to improve the procedure for routine use. Results were compared with values for prothrombin time and activated partial thromboplastin times obtained on the same samples by the clinical laboratory. Because the procedure provides an independent parameter for measurement in patients who have received heparin therapy, insight into different patient responses to the drug is therefore possible. With minor modifications, the procedure can be used for heparans, dermatans, and chondroitins, because it allows identification and microscale quantitation on the basis of charge, molecular weight, and carbohydrate structure.

The internalization and release of heparins by cultured endothelial cells: the process is cell source, heparin source, time and concentration dependent.

Endothelial cells in vivo and in vitro take up heparin following administration. In in vitro systems, cellular and pericellular extracts of monolayers exposed to heparin, demonstrate internalization as well as cell surface attachment. To further investigate the characteristics of this exogenous heparin partitioning, we have exposed two types of endothelial cells to media containing cold and 125I labelled bovine lung, porcine mucosal or CY222 heparin. Heparin uptake, in pericellular and intracellular fractions of porcine cells, increased with concentration at 24 hrs exposure for all heparins. Only bovine heparin was preferentialy accumulated intracellularly. Cultured murine LE-II endothelial cells showed a greater accumulation in the intracellular fraction than porcine cells when exposed to porcine heparin. When bovine or CY222 heparin was administered for varying times, total heparin uptake increased with time. Heparin in the pericellular fraction was greater than intracellular at exposure durations from 5 mins to 6 hrs. At 16 hrs intracellular heparin markedly increased and far exceeded pericellular. The possibility of release of previously internalized heparin was studied by washing cultures previously exposed to heparin with heparin-free media. Heparin could be recovered up to 96 hours post exposure. A concurrent decrease in pericellular and cellular heparin was observed. These results show a differing distribution of heparin across endothelial cell membranes that is heparin source, concentration and time dependent. A previously unsuspected gradient-time mechanism is suggested. The degree of internalization differs for different endothelial cell sources. The protracted retention of heparin intracellularly by endothelial cells and subsequent release may be of therapeutic and physiological consequence.

The close relationship of heparin and the vessel wall.

For over 100 years heparin has attracted interest because of its anticoagulant powers. Commercial heparin has now been shown to be a mixture of over 100 different closely related sulfated polysaccharides of which only 10% activate antithrombin-III. Fifty years ago the original research teams in Toronto and Stockholm in demonstrating the clinical uses of heparin observed that antithrombotic activity did not correspond to levels of anticoagulation. It has been shown that: (a) Heparin accumulates rapidly and specifically in the endothelium against a concentration gradient of hundreds- to thousands-fold. (b) Experimental thrombosis, however produced, is accompanied by a marked decrease in the electronegative charge of the vessel wall and the charge is restored in all cases by heparin. (c) The normal electronegative charge is due to glycosaminoglycans. Heparin possesses the strongest electronegative charge of these substances and is present in the vessel wall as a component of a larger heparitin (sulfate) proteoglycan molecule. (d) Maintenance of the normal electronegative charge depends on adequate supply of oxygen (adequate blood flow). (e) Commercial heparin releases enzymes from the endothelium, lipoprotein lipase and histaminase (D.A.O.). Lipoprotein lipase changes the composition of plasma lipids and lipoproteins and histaminase provides a check for fat absorption. The release of these enzymes decrease and prevent atherosclerotic changes. (f) After administration of commercial heparin, heparin isolated from the plasma has higher antithrombin activity than that injected. The heparin taken up by the endothelium is returned with greater activity. The anticoagulant effect of administered heparin does not produce hemorrhage since this requires simultaneous occurrence of defects in the vascular factor of hemostasis (the result of stress or pituitary-adrenal imbalance) or platelet defect. Thus, clinical effectiveness of heparin is an expression of its close relationship to the vessel wall.

The intracellular uptake and protracted release of exogenous heparins by cultured endothelial cells.

Heparins from bovine or porcine sources were fed in media for 48 hrs to cultured porcine aortic and human umbilical vein endothelial cells. Heparin was found in pericellular and cellular fractions after extraction by chemical methods and 125I radiolabelled heparins were recovered when radiolabelled heparin was included in the feed. Even after washing and media changes heparin was detected in media and cell fractions up to 6 days post feeding. Metachromatic vacuoles within cells were demonstrated histologically up to 7 days post feeding after staining with toluidine blue. This is the first report of protracted internalization of exogenous heparin by cultured endothelial cells with concurrent prolonged release of the heparin to the media. This clearly demonstrates that the endothelium plays an important role in the distribution and metabolism of heparin.