Mutational analysis of Gbetagamma and phospholipid interaction with G protein-coupled receptor kinase 2.

Agonist-dependent regulation of G protein-coupled receptors is dependent on their phosphorylation by G protein-coupled receptor kinases (GRKs). GRK2 and GRK3 are selectively regulated in vitro by free Gbetagamma subunits and negatively charged membrane phospholipids through their pleckstrin homology (PH) domains. However, the molecular binding determinants and physiological role for these ligands remain unclear. To address these issues, we generated an array of site-directed mutants within the GRK2 PH domain and characterized their interaction with Gbetagamma and phospholipids in vitro. Mutation of several residues in the loop 1 region of the PH domain, including Lys-567, Trp-576, Arg-578, and Arg-579, resulted in a loss of receptor phosphorylation, likely via disruption of phospholipid binding, that was reversed by Gbetagamma. Alternatively, mutation of residues distal to the C-terminal amphipathic alpha-helix, including Lys-663, Lys-665, Lys-667, and Arg-669, resulted in decreased responsiveness to Gbetagamma. Interestingly, mutation of Arg-587 in beta-sheet 3, a region not previously thought to interact with Gbetagamma, resulted in a specific and profound loss of Gbetagamma responsiveness. To further characterize these effects, two mutants (GRK2(K567E/R578E) and GRK2(R587Q)) were expressed in Sf9 cells and purified. Analysis of these mutants revealed that GRK2(K567E/R578E) was refractory to stimulation by negatively charged phospholipids but bound Gbetagamma similar to wild-type GRK2. In contrast, GRK2(R587Q) was stimulated by acidic phospholipids but failed to bind Gbetagamma. In order to examine the role of phospholipid and Gbetagamma interaction in cells, wild-type and mutant GRK2s were expressed with a beta(2)-adrenergic receptor (beta(2)AR) mutant that is responsive to GRK2 phosphorylation (beta(2)AR(Y326A)). In these cells, GRK2(K567E/R578E) and GRK2(R587Q) were largely defective in promoting agonist-dependent phosphorylation and internalization of beta(2)AR(Y326A). Similarly, wild-type GRK2 but not GRK2(K567E/R578E) or GRK2(R587Q) promoted morphinedependent phosphorylation of the mu-opioid receptor in cells. Thus, we have (i) identified several specific GRK2 binding determinants for Gbetagamma and phospholipids, and (ii) demonstrated that Gbetagamma binding is the limiting step for GRK2-dependent receptor phosphorylation in cells.

Muscarinic inhibition of substance P induced ion secretion in piglet jejunum.

We examined the effects of the muscarinic agonist carbachol on ion secretion induced by substance P (SP) in piglet jejunal tissues mounted in Ussing chambers. Tetrodotoxin was present in all solutions to inhibit neural activity. Carbachol added 10 min prior to 0.75 microM SP dose dependently inhibited subsequent SP responses, with 90% inhibition at 10 microM carbachol. Addition of an equipotent dose of SP (7.5 microM) had no effect on subsequent carbachol-induced secretion. Carbachol's inhibition of SP-induced secretion was evident for at least 45 min and was abolished by prior addition of the M3 receptor antagonist 4-diphenylacetoxy-N-methyl-piperidine methiodide (4-DAMP), but remained intact in the presence of the M2 antagonist gallamine or the nicotinic antagonist mecamylamine. Atropine added 10 min after carbachol restored subsequent SP responses toward control levels. Carbachol also reduced secretory responses to histamine and, to a lesser extent, prostaglandin E2 (PGE2). SP-induced secretion was not affected by prior addition of histamine and was reduced by PGE2 only at the highest PGE2 concentration. The results suggest that activation of the epithelial M3 receptor by carbachol inhibits subsequent secretory responses to the calcium-mediated agonists SP and histamine in piglet jejunum. This may reflect muscarinic activation of a negative messenger in epithelial cells that limits Cl- secretion.

Agonist-receptor-arrestin, an alternative ternary complex with high agonist affinity.

The rapid decrease of a response to a persistent stimulus, often termed desensitization, is a widespread biological phenomenon. Signal transduction by numerous G protein-coupled receptors appears to be terminated by a strikingly uniform two-step mechanism, most extensively characterized for the beta2-adrenergic receptor (beta2AR), m2 muscarinic cholinergic receptor (m2 mAChR), and rhodopsin. The model predicts that activated receptor is initially phosphorylated and then tightly binds an arrestin protein that effectively blocks further G protein interaction. Here we report that complexes of beta2AR-arrestin and m2 mAChR-arrestin have a higher affinity for agonists (but not antagonists) than do receptors not complexed with arrestin. The percentage of phosphorylated beta2AR in this high affinity state in the presence of full agonists varied with different arrestins and was enhanced by selective mutations in arrestins. The percentage of high affinity sites also was proportional to the intrinsic activity of an agonist, and the coefficient of proportionality varies for different arrestin proteins. Certain mutant arrestins can form these high affinity complexes with unphosphorylated receptors. Mutations that enhance formation of the agonist-receptor-arrestin complexes should provide useful tools for manipulating both the efficiency of signaling and rate and specificity of receptor internalization.

The beta-adrenergic receptor kinase (GRK2) is regulated by phospholipids.

The beta-adrenergic receptor kinase (beta ARK) is a member of growing family of G protein coupled receptor kinases (GRKs). beta ARK and other members of the GRK family play a role in the mechanism of agonist-specific desensitization by virtue of their ability to phosphorylate G protein-coupled receptors in an agonist-dependent manner. beta ARK activation is known to occur following the interaction of the kinase with the agonist-occupied form of the receptor substrate and heterotrimeric G protein beta gamma subunits. Recently, lipid regulation of GRK2, GRK3, and GRK5 have also been described. Using a mixed micelle assay, GRK2 (beta ARK1) was found to require phospholipid in order to phosphorylate the beta 2-adrenergic receptor. As determined with a nonreceptor peptide substrate of beta ARK, catalytic activity of the kinase increased in the presence of phospholipid without a change in the Km for the peptide. Data obtained with the heterobifunctional cross-linking agent N-3-[125I]iodo-4-azidophenylpropionamido-S-(2-thiopyridyl)-c ysteine ([125I]ACTP) suggests that the activation by phospholipid was associated with a conformational change in the kinase. [125I]ACTP incorporation increased 2-fold in the presence of crude phosphatidylcholine, and this increase in [125I]ACTP labeling is completely blocked by the addition of MgATP. Furthermore, proteolytic mapping was consistent with the modification of a distinct site when GRK2 was labeled in the presence of phospholipid. While an acidic phospholipid specificity was demonstrated using the mixed micelle phosphorylation assay, a notable exception was observed with PIP2. In the presence of PIP2, kinase activity as well as [125I]ACTP labeling was inhibited. These data demonstrate the direct regulation of GRK2 activity by phospholipids and supports the hypothesis that this effect is the result of a conformational change within the kinase.

Arrestin interactions with G protein-coupled receptors. Direct binding studies of wild type and mutant arrestins with rhodopsin, beta 2-adrenergic, and m2 muscarinic cholinergic receptors.

Arrestins play an important role in quenching signal transduction initiated by G protein-coupled receptors. To explore the specificity of arrestin-receptor interaction, we have characterized the ability of various wild-type arrestins to bind to rhodopsin, the beta 2-adrenergic receptor (beta 2AR), and the m2 muscarinic cholinergic receptor (m2 mAChR). Visual arrestin was found to be the most selective arrestin since it discriminated best between the three different receptors tested (highest binding to rhodopsin) as well as between the phosphorylation and activation state of the receptor (> 10-fold higher binding to the phosphorylated light-activated form of rhodopsin compared to any other form of rhodopsin). While beta-arrestin and arrestin 3 were also found to preferentially bind to the phosphorylated activated form of a given receptor, they only modestly discriminated among the three receptors tested. To explore the structural characteristics important in arrestin function, we constructed a series of truncated and chimeric arrestins. Analysis of the binding characteristics of the various mutant arrestins suggests a common molecular mechanism involved in determining receptor binding selectivity. Structural elements that contribute to arrestin binding include: 1) a C-terminal acidic region that serves a regulatory role in controlling arrestin binding selectivity toward the phosphorylated and activated form of a receptor, without directly participating in receptor interaction; 2) a basic N-terminal domain that directly participates in receptor interaction and appears to serve a regulatory role via intramolecular interaction with the C-terminal acidic region; and 3) two centrally localized domains that are directly involved in determining receptor binding specificity and selectivity. A comparative structure-function model of all arrestins and a kinetic model of beta-arrestin and arrestin 3 interaction with receptors are proposed.

Pharmacokinetics of tazobactam M1 metabolite after administration of piperacillin/tazobactam in subjects with renal impairment.

Tazobactam is a new derivative of penicillinic acid sulfone, which functions as an irreversible inhibitor of many beta-lactamases. The disposition of tazobactam M1 metabolite after intravenous (i.v.) infusion of 3 g of piperacillin/0.375 g of tazobactam was evaluated in 26 subjects with various degrees of renal impairment. Participants in the study were 18 subjects with creatinine clearances (ClCR) ranging from 7.4-41.8 mL/min, 4 subjects maintained on continuous ambulatory peritoneal dialysis (CAPD), and 4 subjects undergoing chronic hemodialysis (HD). The pharmacokinetic parameters of piperacillin and tazobactam were evaluated and were similar to previous reports. Tazobactam M1 metabolite maximum plasma concentration increased as renal function declined. The terminal elimination half-life and area under the plasma concentration-time curve of the tazobactam M1 metabolite increased as renal function declined. The mean rate of recovery of the tazobactam M1 metabolite in hemodialysate during a 3- to 4.2-hour HD session 1 hour after the i.v. infusion of piperacillin/tazobactam was 25.3%. However, when HD was performed at 36-48 hours after the i.v. infusion, 57.6% of the tazobactam dose was recovered as M1 metabolite, suggesting further conversion of tazobactam to M1 metabolite. Peritoneal dialysis removed 15.8% (n = 2) of the tazobactam dose as the M1 metabolite. Using a dose of 3 g of piperacillin/0.375 g of tazobactam, the predicted maximum steady-state plasma concentrations of the tazobactam M1 metabolite are 14.6 micrograms/mL, 34.8 micrograms/mL, and 48.8 micrograms/mL for subjects with ClCR 20-40 mL/min (every 6 hour dosing), ClCR < 20 mL/min (every 8 hour dosing), and on CAPD (every 12 hour dosing), respectively.

Expression, purification, and characterization of the G protein-coupled receptor kinase GRK5.

G protein-coupled receptor kinases (GRKs) such as rhodopsin kinase and the beta-adrenergic receptor kinase (beta ARK) play an important role in agonist-specific phosphorylation and desensitization of G protein-coupled receptors. GRK5 is a recently identified member of the GRK family that has greater homology with rhodopsin kinase than with beta ARK. To further characterize the activity of GRK5, it has been overexpressed in Sf9 insect cells and purified by successive chromatography on S-Sepharose and Mono S columns. GRK5 phosphorylates the beta 2-adrenergic receptor (beta 2AR), m2 muscarinic cholinergic receptor, and rhodopsin in an agonist-dependent manner to maximal stoichiometries of approximately 2.5, 1.5, and 1 mol of phosphate/mol of receptor, respectively, with Km values of approximately 0.5 microM for the beta 2AR, approximately 16 microM for rhodopsin, and approximately 24 microM for ATP. Peptide phosphorylation studies suggest that in contrast to beta ARK and rhodopsin kinase, GRK5 preferentially phosphorylates on nonacidic peptides with a Km of approximately 1.5 mM. Heparin and dextran sulfate were found to be potent inhibitors of GRK5 with IC50 values of approximately 1 nM, thereby being at least 150-fold more potent on GRK5 than on beta ARK. GRK5 can also be activated by polycations, with 10 microM polylysine promoting an approximately 2.6-fold activation. Overall, these studies demonstrate that GRK5 has unique properties that distinguish it from other members of the GRK family and that likely play an important role in modulating its mechanism of action.

Isolation and characterization of renal cortical membranes using an aqueous two-phase partition technique.

The aqueous two-phase partition technique is a simple, rapid and inexpensive method for the fractionation of membrane preparations. Aqueous two-phase partitioning separates according to surface properties such as charge and hydrophobicity, making it complementary to established centrifugation techniques, which separate on the basis of density. Although aqueous two-phase partitioning has been successfully applied to animal tissues, there are limited data on the functional properties of the isolated membranes. We have applied the aqueous two-phase partition technique to rat renal brush-border membrane vesicles and sheets. Our aim was to remove organelle contamination while maintaining the functional properties of the membranes. Evidence from marker enzyme analysis and electron microscopy supports the conclusion that renal brush-border membranes are fractionated separate from the mitochondria and endoplasmic reticulum. This separation procedure did not alter the Na(+)-dependent transport of brush-border membrane vesicles. Na(+)-D-glucose symporter and Na(+)-H+ antiporter activity in the fractionated preparation increased to the same extent as did the enrichment of enzyme markers for brush-border membranes.

Thromboxane binding and signal transduction in rat glomerular mesangial cells.

Thromboxane A2 (TxA2) stimulates contraction of glomerular mesangial cells. However, mesangial cell TxA2 receptors have not been previously characterized. We therefore investigated TxA2 binding and TxA2-associated signal transduction pathways in rat glomerular mesangial cells using the specific thromboxane receptor agonist (1S-[1 alpha,2 beta(5Z),3 alpha-(1E,3S)4 alpha])-7-(3-[3-hydroxy-4-(p- iodophenoxy)-1-butenyl]7-oxabicyclo[2.2.1]hept-2-yl)-5-heptenoic acid (IBOP). In these cells, [125I]BOP binding was saturable, displaceable, and of high affinity. Scatchard analysis revealed a single class of binding sites with a dissociation constant (Kd) of 293 pM and a maximal density of binding sites (Bmax) of 33 fmol/mg protein. Specific binding was inhibited by the thromboxane agonist (15S)-hydroxy-11 alpha,9 alpha-(epoxymethano)prosta-5Z,13E-dienoic acid (U-46619) [inhibitor dissociation constant (Ki) = 297 nM] and the TxA2 receptor antagonists SQ 29548 (Ki = 1 nM) and (1R-[1 alpha(Z),2 beta,3 beta,5 alpha])-(+)-7-(5-[(1,1'-biphenyl)- 4-yl-methoxy]-3-hydroxy-2-(1-piperidinyl)cyclopentyl]-4-heptenoic acid (GR 32191) (Ki = 92 nM). Binding was also highly specific for thromboxane because prostaglandin E2 (Ki = 16 microM) and the inactive thromboxane metabolite, TxB2 (Ki = 41 microM), were approximately 1,000-fold less potent at inhibiting binding. IBOP stimulated phosphatidylinositol hydrolysis with an effective concentration of drug that produces 50% of the maximal response of 229 pM, which correlated well with the equilibrium Kd and enhanced phosphorylation of an acidic 80-kDa protein substrate for protein kinase C.(ABSTRACT TRUNCATED AT 250 WORDS)

Characterization of glomerular thromboxane receptors in murine lupus nephritis.

Renal thromboxane (Tx) production is increased in the MRL-lpr murine model of lupus nephritis. To investigate the relationship between increased Tx production and number and affinity of Tx receptors, we measured binding of the Tx receptor antagonist [3H][SQ295481S-1 alpha,2 beta(5Z),3 beta,4 alpha]-7-(3-((2-((phenyl- amino)-carbonyl)hydrozino)methyl)-7-oxabicyclo-(2.2.1)heptan -2-yl)-5-heptenoic acid in glomerular preparations from MRL-lpr mice and both MRL(-)+/+ and LG/J controls. Renal Tx binding was first characterized in normal LG/J mice. In these animals, glomerular binding was specific, saturable and reversible. Scatchard analysis revealed a single class of high-affinity binding sites. We next evaluated Tx production and binding in 12- and 16-week-old MRL-lpr mice and MRL(-)+/+ controls. To assess renal Tx production, excretion of TxB2 was measured in urine. Urinary TxB2 was increased in MRL-lpr mice at 16 weeks of age. This increase in urinary TxB2 was associated with a reduction in density of glomerular Tx binding sites compared to either 12-week-old MRL-lpr mice or MRL(-)+/+ controls. Ligand binding affinity was similar in all groups. To investigate if this alteration in binding was specific for Tx, glomerular binding of [3H]angiotensin II was measured. In MRL-lpr mice, the number and affinity of glomerular angiotensin binding sites were similar at 12 and 16 weeks of age. Thus, in this murine model of lupus nephritis, enhanced renal Tx production is temporally associated with a decrease in glomerular Tx binding sites without a change in receptor affinity.(ABSTRACT TRUNCATED AT 250 WORDS)

Expression and characterization of two beta-adrenergic receptor kinase isoforms using the baculovirus expression system.

The beta-adrenergic receptor kinases, beta ARK1 and beta ARK2, are two recently cloned members of the G protein-coupled receptor kinase family. To further characterize these kinases, bovine beta ARK1 and beta ARK2 have been overexpressed in Sf9 insect cells using the baculovirus expression system. High yields (5-7 mg/L cells) of purified kinase preparations were obtained by sequential chromatography of infected Sf9 cell supernatant fractions on S-Sepharose and Heparin-Sepharose. The expressed kinases were fully active as evidenced by their ability to specifically phosphorylate the agonist-occupied beta 2-adrenergic receptor (beta 2AR) and light-activated rhodopsin. Similar initial rates and maximal stoichiometries of beta 2AR phosphorylation were observed for both beta ARK1 and beta ARK2. Moreover, G protein beta gamma subunits enhanced the initial rates of both beta ARK1 and beta ARK2 mediated beta 2AR phosphorylation by approximately tenfold. In the presence of beta gamma subunits the maximal stoichiometry of beta 2AR phosphorylation was increased from approximately 4 mol phosphate/mol receptor to approximately 10 mol/mol. Detailed kinetic analysis of rhodopsin phosphorylation suggests that both kinases follow a sequential mechanistic pathway and have similar Kms for rhodopsin (approximately 14 microM) and MgATP (60-90 microM). Peptide phosphorylation studies demonstrate that both kinases prefer acidic amino acids amino terminal to a serine. Heparin was found to be the most potent inhibitor for both kinases with IC50s of 1.4 and 1.1 microM for beta ARK1 and beta ARK2, respectively. These studies demonstrate that beta ARK1 and beta ARK2 share very similar kinetic properties and suggest that they may have a similar substrate specificity in vivo.

The receptor kinase family: primary structure of rhodopsin kinase reveals similarities to the beta-adrenergic receptor kinase.

Light-dependent deactivation of rhodopsin as well as homologous desensitization of beta-adrenergic receptors involves receptor phosphorylation that is mediated by the highly specific protein kinases rhodopsin kinase (RK) and beta-adrenergic receptor kinase (beta ARK), respectively. We report here the cloning of a complementary DNA for RK. The deduced amino acid sequence shows a high degree of homology to beta ARK. In a phylogenetic tree constructed by comparing the catalytic domains of several protein kinases, RK and beta ARK are located on a branch close to, but separate from the cyclic nucleotide-dependent protein kinase and protein kinase C subfamilies. From the common structural features we conclude that both RK and beta ARK are members of a newly delineated gene family of guanine nucleotide-binding protein (G protein)-coupled receptor kinases that may function in diverse pathways to regulate the function of such receptors.

Cloning, expression, and chromosomal localization of beta-adrenergic receptor kinase 2. A new member of the receptor kinase family.

The beta-adrenergic receptor kinase (beta ARK) specifically phosphorylates the agonist-occupied form of the beta-adrenergic and related G protein-coupled receptors. Structural features of this enzyme have been elucidated recently by the isolation of a cDNA that encodes bovine beta ARK. Utilizing a catalytic domain fragment of the beta ARK cDNA to screen a bovine brain cDNA library we have isolated a clone encoding a beta ARK-related enzyme which we have termed beta ARK2. Overall, this enzyme has 85% amino acid identity with beta ARK, with the protein kinase catalytic domain having 95% identity. The ability of beta ARK2 to phosphorylate various substrates was studied after expression in COS 7 cells. Although beta ARK2 is essentially equiactive with beta ARK in phosphorylating an acid-rich synthetic model peptide it was only approximately 50% as active when the substrate was the agonist-occupied beta 2-adrenergic receptor and only approximately 20% as active toward light-bleached rhodopsin. As with beta ARK, phosphorylation of the receptor substrates by beta ARK2 was completely stimulus dependent. RNA blot analysis with selected bovine tissues reveals an mRNA of 8 kilobases with a distribution similar to that of beta ARK. More detailed RNA analysis using a ribonuclease protection assay in various rat tissues suggests that the beta ARK2 message is present at much lower levels (typically 10-20%) than the beta ARK message. In the rat the beta ARK2 mRNA is localized predominantly in neuronal tissues although low levels are also observed in various peripheral tissues. The beta ARK2 gene has been localized to a region of mouse chromosome 5 whereas the beta ARK gene is localized on mouse chromosome 19. These data suggest the existence of a "family" of receptor kinases which may serve broadly to regulate receptor function.

Role of acidic amino acids in peptide substrates of the beta-adrenergic receptor kinase and rhodopsin kinase.

The beta-adrenergic receptor kinase (beta-ARK) phosphorylates G protein coupled receptors in an agonist-dependent manner. Since the exact sites of receptor phosphorylation by beta-ARK are poorly defined, the identification of substrate amino acids that are critical to phosphorylation by the kinase are also unknown. In this study, a peptide whose sequence is present in a portion of the third intracellular loop region of the human platelet alpha 2-adrenergic receptor is shown to serve as a substrate for beta-ARK. Removal of the negatively charged amino acids surrounding a cluster of serines in this alpha 2-peptide resulted in a complete loss of phosphorylation by the kinase. A family of peptides was synthesized to further study the role of acidic amino acids in peptide substrates of beta-ARK. By kinetic analyses of the phosphorylation reactions, beta-ARK exhibited a marked preference for negatively charged amino acids localized to the NH2-terminal side of a serine or threonine residue. While there were no significant differences between glutamic and aspartic acid residues, serine-containing peptides were 4-fold better substrates than threonine. Comparing a variety of kinases, only rhodopsin kinase and casein kinase II exhibited significant phosphorylation of the acidic peptides. Unlike beta-ARK, RK preferred acid residues localized to the carboxyl-terminal side of the serine. A feature common to beta-ARK and RK was a much greater Km for peptide substrates as compared to that for intact receptor substrates.

Protein kinase A, cytosolic calcium, and phosphate uptake in human proximal renal cells.

Phosphate uptake by proximal renal cells derived from the human kidney occurs by a saturable process that is approximately 85% dependent on the presence of sodium. Kinetic analysis is consistent with two distinct transport events with Km of 0.08 and 0.63 mM and Vmax of 3.4 and 11.0 nmol.mg-1.3 min-1, respectively. Parathyroid hormone (PTH), isoproterenol, and prostaglandin E2 (PGE2) increased cellular adenosine 3',5'-cyclic monophosphate (cAMP). PTH-stimulated cAMP prevented binding of the photolabel 8-azido[32P]cAMP with a half-maximal effective concentration (EC50) of 1 nM PTH, 30-fold lower than the EC50 for intracellular cAMP accumulation. These data are qualitatively similar to those observed in OK cells. PTH did not inhibit phosphate uptake in human cells, although it activated cAMP-dependent protein kinase and increased cytosolic calcium. Thus phosphate uptake in human proximal renal cells maintained in short-term culture is unresponsive to PTH in spite of increased cytosolic calcium and activation of the cAMP pathway.

Antifibrinolytic therapy in the management of the Kasabach Merritt syndrome.

The Kasabach Merritt syndrome consists of thrombocytopenia, microangiopathic hemolytic anemia, and a localized consumption coagulopathy that develops within the abnormal vascular channels of a hemangioma. In general, these patients demonstrate only mild abnormalities of screening clotting tests, but they can potentially develop life-threatening complications. We present a patient who developed a severe anemia that was refractory to erythrocyte transfusions. Treatment with epsilon-aminocaproic acid to inhibit fibrinolysis and cryoprecipitate to replenish his deficient circulating fibrinogen interrupted the cycle of his systemic coagulopathy and enabled us to transfuse him to a normal hematocrit.

Hemodialysis without anticoagulation. One-year prospective trial in hospitalized patients at risk for bleeding.

This prospective study evaluated a protocol for hemodialysis without anticoagulation in a diverse group of hospitalized patients in unstable condition with relative contraindications to anticoagulation. Of 262 attempts at hemodialysis without anticoagulation in 49 patients, 239 hemodialysis treatments (91 percent) were successfully completed. Approximately 7 percent of the attempts required conversion to a low-dose heparin regimen because of clotting in the extracorporeal dialysis circuit. Fewer than 2 percent of the dialysis treatments resulted in clotting in the extracorporeal circuit sufficient to interrupt hemodialysis. Partial thromboplastin times and activated clotting times did not change during these hemodialysis treatments. Solute clearance, ultrafiltration rate, and decrements in arterial oxygen concentration and platelet count were not different from those in patients who underwent hemodialysis with anticoagulation. There were no episodes of accelerated bleeding associated with this dialysis method. This study indicates that hemodialysis without anticoagulation can be reliable and effective in closely monitored situations.