A role for glycosphingolipid-enriched microdomains in platelet glycoprotein Ib-mediated platelet activation.

BACKGROUND: Glycoprotein (GP) Ib, a platelet von Willebrand factor (VWF) receptor, plays a crucial role in thrombosis and hemostasis. As recent reports have suggested that GPIb partially locates in a particular region, designated as glycosphingolipid-enriched microdomains (GEMs), we hypothesized that GEMs play a central role in GPIb-mediated platelet activation. METHODS: Platelets were stimulated by VWF/botrocetin to activate platelets through GPIb. GEMs and non-GEMs were isolated by sucrose density gradient ultracentrifugation and the location of signaling molecules characterized. The role of GEMs-mediated signaling in platelet behavior was tested by platelet aggregation and by platelet interaction with immobilized VWF under flow conditions when GEMs were disrupted by methyl-beta-cyclodextrin (MbetaCD). RESULTS: GPIb was partially translocated to GEMs upon VWF/botrocetin stimulation. Immunoprecipitation of GPIb in GEMs and non-GEMs revealed that the tyrosine kinases, Src and Lyn, were associated with GPIb only in GEMs after GPIb-stimulation, and not in non-GEMs. Activation of PLCgamma2 was more intense in GEMs than non-GEMs. Disruption of GEMs by MbetaCD strongly inhibited tyrosine phosphorylation of Syk and PLCgamma2. Functional studies revealed that stable adhesion of platelets to a VWF-coated surface under flow was impaired by GEM disruption by MbetaCD. CONCLUSION: The combined results suggest that GEMs play an important role in GPIb-mediated platelet activation.

The platelet-activating receptor C-type lectin receptor-2 plays an essential role in liver regeneration after partial hepatectomy in mice.

Essentials Regeneration role of C-type lectin receptor-2 (CLEC-2) after 70% hepatectomy (HPx) was investigated. Wild-type or CLEC-2 deleted from platelets of chimeric mice (flKO) underwent HPx. The liver/body weight ratio was significantly lower in the flKO than in the wild-type. CLEC-2 plays an essential role in liver regeneration after HPx. SUMMARY: Background and aim The aim of the present study was to investigate the role of C-type lectin receptor (CLEC)-2 in liver regeneration following partial liver resection in mice. Materials and methods Irradiated chimeric mice transplanted with fetal liver cells from wild-type (WT) mice, CLEC-2-deleted (KO) mice or mice with CLEC-2 deleted specifically from platelets (flKO) were generated. Mice underwent 70% partial hepatectomy (PH). Immunohistochemical staining was performed to investigate the expression of the endogenous ligand for CLEC-2, podoplanin. The accumulation of platelets in the liver was also quantified. The hepatic expression of the IL-6/gp130 and STAT3, Akt and ERK1/2 was also examined. Results The liver/body weight ratio and expression of all cell proliferation markers were significantly lower in the flKO group than in the WT group. The expression of phosphorylated (p) Akt and pERK1/2 was similar in the WT and flKO groups. On the other hand, the expression of pSTAT3 and IL-6 was significantly stronger in the WT group than in the flKO group. The expression of podoplanin was detected in the hepatic sinusoids of both groups. However, the extent to which platelets accumulated in hepatic sinusoids was significantly less in the flKO group than in the WT group. Conclusion CLEC-2 was involved in hepatic regeneration after liver resection and CLEC-2-related liver regeneration was attributed to the interaction between platelets and sinusoidal endothelial cells.

C-type lectin-like receptor 2 promotes hematogenous tumor metastasis and prothrombotic state in tumor-bearing mice.

Essentials The role of C-type lectin-like receptor-2 (CLEC-2) in cancer progression is unclear. CLEC-2-depleted mouse model is generated by using a rat anti-mouse CLEC-2 monoclonal antibody. CLEC-2 depletion inhibits hematogenous tumor metastasis of podoplanin-expressing B16F10 cells. CLEC-2 depletion prolongs cancer survival by suppressing thrombosis and inflammation. SUMMARY: Background C-type lectin-like receptor 2 (CLEC-2) is a platelet activation receptor of sialoglycoprotein podoplanin, which is expressed on the surface of certain types of tumor cells. CLEC-2-podoplanin interactions facilitate hematogenous tumor metastasis. However, direct evidence of the role of CLEC-2 in hematogenous metastasis and cancer progression is lacking. Objective and methods We generated immunological CLEC-2-depleted mice by using anti-mouse CLEC-2 monoclonal antibody 2A2B10 and investigated whether CLEC-2 promoted hematogenous tumor metastasis and tumor growth and exacerbated the prognosis of mice bearing podoplanin-expressing B16F10 melanoma cells. Results Our results showed that hematogenous metastasis was significantly inhibited in CLEC-2-depleted mice. B16F10 cells co-cultured with wild-type platelets, but not with CLEC-2-deficient platelets, showed increased proliferation. However, B16F10 cell proliferation was not inhibited in CLEC-2-depleted mice. Histological analysis showed that thrombus formation in tumor vessels was significantly inhibited and functional vessel density was significantly increased in CLEC-2-depleted mice. These data suggest that CLEC-2 deficiency may inhibit thrombus formation in tumor vessels and increase the density of functional vessels, thus improving oxygen and nutrient supply to tumors, indirectly promoting tumor proliferation. Furthermore, the overall survival of CLEC-2-depleted mice was significantly prolonged, which may be due to the suppression of thrombus formation in the lungs and subsequent inhibition of systemic inflammation and cachexia. Conclusions These data provide a rationale for the targeted inhibition of CLEC-2 as a new strategy for preventing hematogenous tumor metastasis and for inhibiting cancer-related thromboembolism.

Specific biodetection of B16 mouse melanoma in vivo by syngeneic monoclonal antibody.

The specific detection of tumors in vivo using a radiolabeled syngeneic monoclonal antibody made by fusion of P3U1 (BALB/c myeloma cells) and C57BL/6 spleen cells primed with syngeneic B16 melanoma cells was investigated by color imaging, autoradiography, and biodistribution. The radiolabeled antimelanoma antibody specifically accumulated only in the tumor lesions, whereas no radioactivity was observed in normal tissues or organs. The distribution patterns of the radioactive antibody in the tumor lesions depended on the sizes of the tumor. Almost the entire region of the small metastatic tumor in lymph nodes was labeled, whereas the radioactive antibody was irregularly localized mainly in the center of the medium-sized tumor. However, only the peripheral region of the large primary tumor was labeled. The highest uptake of radioactivity (tumor:blood ratio) was observed in the small lymph node metastatic tumor lesions rather than in the large primary tumor. Furthermore, high resolution color imaging of B16 melanoma was also obtained by using 125I-labeled monoclonal antibody. Tumor location was specifically visible without subtraction or enhancement methods 3-5 days after injection of the radiolabeled antibody.

An increase in the in vivo binding of [3H]SCH 23390 induced by MK-801 in the mouse striatum.

The effect of MK-801, a non-competitive NMDA antagonist, on the in vivo binding of [3H]SCH 23390, a dopamine D1 receptor antagonist, was investigated in the mouse brain. The radioactivity following injection of the tracer was measured in the striatum, cerebral cortex and cerebellum. It was found that MK-801 increased the [3H]SCH 23390 binding in the striatum in a dose-dependent manner, but did not influence the binding in the cerebral cortex. The kinetic analysis using the cerebellum as a reference region revealed that the apparent increase in [3H]SCH 23390 binding in the striatum was mainly due to the increase in the input rate of the specific binding component. An in vivo saturation study using varying doses of [3H]SCH 23390 indicated that the maximum binding sites available (Bmax) in the striatum was not altered by MK-801. As the rate constant K3 includes both Bmax and the association rate constant (kon), an increase in the rate constant kon in vivo was the primary factor in the changes in [3H]SCH 23390 binding. The changes in the rate constant kon in vivo seem to be due to a NMDA receptor-mediated process. An in vivo binding method would be applicable for the investigation of the neural interaction between glutamatergic and dopaminergic neurons in intact brain.

The effect of benzodiazepines on the binding of [3H]SCH 23390 in vivo.

The effect of flunitrazepam upon the binding of [3H]SCH 23390 in vivo was investigated. Acute treatment with flunitrazepam decreased the binding of [3H]SCH 23390 in the striatum in a dose-dependent manner. The time course of radioactivity in the striatum, cerebral cortex and cerebellum in controls and flunitrazepam (1 mg/kg)-treated mice was measured after intravenous injection of [3H]SCH 23390. The binding kinetics were calculated, using the cerebellum as a reference region for the estimation of the amount of free ligand in the brain. Flunitrazepam significantly decreased the input rate constant to the receptor compartment and the dissociation rate constant in vivo. An in vivo displacement study, using carrier SCH 23390, also showed significant reduction in the dissociation rate constant of [3H]SCH 23390 in vivo. The drug Ro 15-1788 reversed the effect of flunitrazepam, suggesting that this reduction in binding of [3H]SCH 23390 was mediated by benzodiazepine receptors. To evaluate the relationship between the reduction in binding of [3H]SCH 23390 in vivo and in vivo occupancy of benzodiazepine receptors, in vivo occupancy of benzodiazepine receptors was measured using [3H]Ro 15-1788. A non-linear relationship was found between the reduction in dopamine D1 receptor binding in vivo and the occupancy of benzodiazepine receptors in vivo, indicating that benzodiazepines exerted the maximum change in dopamine receptor binding at a low fractional occupancy of receptors.

Swim stress alters in vivo binding of [3H]N-methylspiperone.

The effect of swim stress on the in vivo binding of [3H]N-methylspiperone in the striatum of the mouse was investigated. Mice were forced to swim for 5 min at 18 degrees C and the time course of radioactivity in the striatum and cerebellum, following intravenous injection of [3H]N-methylspiperone was measured. The ratio of radioactivity in the striatum to that in the cerebellum was plotted as a function of time for the estimation of in vivo binding to dopamine D2 receptors. Immediately after the swim stress, a significant decrease in binding to D2 receptors in vivo was observed. Neither the KD nor Bmax determined by in vitro binding were altered by swim stress. The time course of the changes in binding, within a 24 hr period, following the swim stress was also studied and a rapid reversal of binding, within 1 hr after the swim stress was observed. In vivo binding of [3H]N-methylspiperone in the cerebral cortex, which appeared to involve serotonin receptors, as well as D2 receptors, was not significantly altered by the swim stress. A saturation study of in vivo binding indicated that the decreases in binding to D2 receptors, due to swim stress, were primarily caused by changes in the apparent affinity rather than in the number of binding sites available in vivo. These results support the hypothesis that micro-environmental factors, including the diffusion barrier to the synapse, might be altered by swim stress.

Effects of the GABAergic system on in vivo binding of [3H]N-methylspiperone.

The effects of flunitrazepam, a benzodiazepine receptor agonist and those of NNC-711, a GABA transporter blocker, on the in vivo binding of [3H]N-methylspiperone, a dopamine D2 receptor antagonist, were investigated in mouse brain. Treatment with either flunitrazepam or NNC-711 reduced the specific binding of [3H]N-methylspiperone in the striatum. Flumazenil, a central benzodiazepine receptor antagonist, blocked the effect of flunitrazepam, indicating that the reduction in specific binding in the striatum was mediated via the GABAergic system. The flunitrazepam significantly decreased the specific binding of [3H]N-methylspiperone in the striatum at all time points studied after tracer injection, whereas specific binding in the cerebellum and cerebral cortex was unaltered. This decrease in specific binding in the striatum was found to be due to a reduced input rate constant (k3) of [3H]N-methylspiperone. The maximum number of the binding sites available for dopamine D2 receptors in the striatum was not changed by the flunitrazepam treatment.

Quantitative in vivo analysis of benzodiazepine binding sites in the human brain using positron emission tomography.

Central-type benzodiazepine binding sites were characterized in a single normal human subject, using positron emission tomography (PET) and the radiolabelled benzodiazepine antagonist, carbon-11 labelled flumazenil ([11C] Ro 15-1788). The subject was scanned using tracer alone and tracer plus 4 different concentrations of unlabelled Ro 15-1788, including one concentration of unlabelled Ro 15-1788, chosen to produce maximum displacement of [11C] Ro 15-1788 from specific binding sites. Concentrations of free, unmetabolized [11C] Ro 15-1788 in plasma were estimated using a simple extraction and ultrafiltration method. Radioactivity in the regional exchangeable pool in brain was estimated under non-saturation conditions from the ratio of radioactivity in brain to plasma, under saturation conditions and the kinetics of free ligand in plasma. The specific binding was, then, estimated by the difference between the total radioactivity in brain and exchangeable pool radioactivity. Scatchard analyses were performed to yield Bmax and Kd values under pseudo-equilibrium conditions, which was observed as an increase of specific binding/free with reduction in specific binding. In cerebral cortex and cerebellum, the Bmax values were about 62-73 nmol/l and the Kd values were 3.6-6 nM in the estimation of free ligand in plasma and 12-15 nM in the estimation of exchangeable pool in brain, as free in brain.

Preparation of various Tc-99m dimercaptosuccinate complexes and their evaluation as radiotracers.

The organ distributions of four different Tc-99m dimercaptosuccinate complexes ("Tc-DMS", Complexes 1 to 4) were determined using mice and were evaluated as renal imaging agents. The highest kidney uptake was observed with Complex 2:21.7% of dose, 3 hr after injection. The biologic distributions and gel chromatographic analyses using carrier Tc-99 and Sn-113 indicate that there is little possibility of mixed metal complexes of the type Tc-Sn-DMS: rather they contain only Tc + DMS. The labeling procedure for Tc-99m DMS as a renal agent proceeds in two steps: a rapid formation of Complex 1, and a slower, rate-determining step from Complex 1 to Complex 2. A reproducible lyophilized kit has been prepared. The yield of complex 2 greatly depends on the reconstitution volume of 99mTcO4-; yield averages 89% using 2 ml of 99mTcO4-eluate.

New preparation method for 99mTc-phytate.

A new method is described in which Pt-Sn or Sn-Sn electrodes are used to activate 99mTcO4-. The 99mTc is incorporated into phytic acid by stannous ion released from a tin anode by the corrosive reaction. The most suitable pH for labeling phytic acid by this method was below 5 and the 99mTc-phytate formed could be precipitatedd with Ca+2 at a pH above 3.5. Though 99mTc-phytate is soluble in aqueous solution, it forms an insoluble species with Ca+2 in vivo and is trapped in the reticuloendothelial system. More than 93% of the 99mTc-phytate localized in the liver of mice; here its biologic half-time is about 112 hr. The 99mTc-phytate prepared by this method promises to be useful as a liver-scanning agent. Since our method does not require an applied electric potential, it appears to be one of the most convenient methods for labeling compounds with 99mTc.

Visualization of specific binding sites of benzodiazepine in human brain.

Using 11C-labeled Ro15-1788 and positron emission tomography, studies of benzodiazepine binding sites in the human brain were performed on four normal volunteers. Rapid and high accumulation of 11C activity was observed in the brain after i.v. injection of [11C]Ro15-1788, the maximum of which was within 12 min. Initial distribution of 11C activity in the brain was similar to the distribution of the normal cerebral blood flow. Ten minutes after injection, however, a high uptake of 11C activity was observed in the cerebral cortex and moderate uptake was seen in the cerebellar cortex, the basal ganglia, and the thalamus. The accumulation of 11C activity was low in the brain stem. This distribution of 11C activity was approximately parallel to the known distribution of benzodiazepine receptors. Saturation experiments were performed on four volunteers with oral administration of 0.3-1.8 mg/kg of cold Ro15-1788 prior to injection. Initial distribution of 11C activity following injection peaked within 2 min and then the accumulation of 11C activity decreased rapidly and remarkably throughout the brain. The results indicated that [11C] Ro15-1788 associates and dissociates to specific and nonspecific binding sites rapidly and has a high ratio of specific receptor binding to nonspecific binding in vivo. Carbon-11 Ro15-1788 is a suitable radioligand for the study of benzodiazepine receptors in vivo in humans.

Kinetics of [11C]N,N-dimethylphenylethylamine in mice and humans: potential for measurement of brain MAO-B activity.

Carbon-11-labeled N,N-dimethylphenylethylamine ([11C]DMPEA) was synthesized by the reaction of N-methylphenylethylamine with [11C]methyl iodide. This newly synthesized radiotracer was developed for the purpose of in vivo measurement of monoamine oxidase-B activity in the brain using a metabolic trapping method. Initially, biodistribution was investigated in mice. The rapid and high uptake of 11C radioactivity in the brain was observed following intravenous injection of [11C]DMPEA, the peak of which was reached at 1 min, followed by a decrease at 1-5 min and slowly thereafter. The kinetics of [11C]DMPEA in the human brain were determined using positron emission tomography (PET) and showed that 11C radioactivity increased gradually over 60 min following initial rapid uptake of 11C radioactivity, with basal ganglia and thalamus showing high accumulation.

Changes in apparent rates of receptor binding in the intact brain in relation to the heterogeneity of reaction environments.

Neuroreceptor imaging by PET or SPECT has been widely applied in the field of neurobiology, from basic to clinical investigations, and has the potential to reveal the neurochemical basis of various neurological and psychiatric diseases as well as to provide new knowledge in the field of neuropharmacology. In contrast to the static nature of in vitro systems, neurotransmission systems in the intact brain constitute part of a dynamic and communicating environment. Thus, it is important to develop new functional imaging methods that reflect neural communications and the dynamism of signal transmission in the living brain. In vivo receptor binding can be altered not only by competitive inhibition by endogenous neurotransmitters but by trans-synaptic effects, and investigation of neural interactions by detection of changes in receptor binding therefore presents a potential method for studying this phenomenon. Recently, several PET studies on in vivo neural interactions using the D2 receptor ligand [11C]-raclopride concluded that the phenomenon was mediated by changes in synaptic endogenous dopamine concentrations that compete with [11C]-raclopride binding for neuroreceptor occupancy. However, a growing body of evidence indicates that these changes in in vivo receptor binding cannot be fully explained by competitive inhibition by endogenous ligand, and alternative mechanisms for the interneuronal modulation of receptor binding are addressed. This review highlights some of the discrepancies observed between in vitro and in vivo receptor binding studies with respect to a number of phenomena, including the heterogeneity of the reaction field surrounding receptors. Quantitative receptor binding studies are usually analyzed by using 'static' binding parameters, such as the Bmax, and KD, which are normally determined by in vitro assays. In addition to these parameters, the apparent association and dissociation rate constants (kon, koff) play equally significant roles in receptor binding in the intact brain is expected. The concepts of "diffusion boundary" and "reaction volume" are introduced, and discussions on some of the discrepancies between in vivo and in vitro receptor binding phenomena are presented.

[13N]-beta-phenethylamine ([13N]PEA): a prototype tracer for measurement of MAO-B activity in heart.

[13N]beta-Phenethylamine ([13N]PEA) was evaluated as a radio tracer for the measurement of mouse heart monoamine oxidase (MAO) activity in vivo. After intravenous administration, [13N]PEA was deaminated by MAO-B. 13NH3 formed thereby was taken up by amino acids and trapped in the heart. The relation between the radioactivity trapped in the heart and the enzyme activity was examined. The radioactivity in the heart 15 min after administration was reduced in a dose-dependent manner by pretreatment with a specific MAO-B inhibitor, l-deprenyl, but not with a specific MAO-A inhibitor, clorgyline. A linear correlation existed between the heart radioactivity level and the heart MAO-B activity (0-45%). [13N]1,1-d2-2-Phenethylamine (C6H5-CH2-CD2-13NH2, [13N]d2PEA), a modified tracer with less reactivity towards the enzyme, was tested similarly. This tracer possessed a higher sensitivity than [13N]PEA, and a wider range (0-85%) of MAO-B activity correlated linearly with the trapped radioactivity. These results indicate that [13N]PEA derivatives ([13N]PEA and [13N]d2PEA) can be useful radiotracers for noninvasive measurements of MAO-B activity in the human heart.

An acute effect of triazolam on muscarinic cholinergic receptor binding in the human brain measured by positron emission tomography.

An acute effect of triazolam, a potent benzodiazepine agonist, on cholinergic receptor binding in the human brain was measured by PET (positron emission tomography) using [11C]N-methyl-4-piperidylbenzilate ([11C]NMPB), a potent muscarinic cholinergic receptor antagonist. Two PET scans were performed in each subject: (1) control scan; (2) after oral administration of 0.5 mg triazolam or placebo. The previously discussed amnestic effect of triazolam was measured by immediate and delayed recall of meaningful and meaningless syllables. A compartment model employing the radioactivity in the cerebellum as an input function was used for the quantification of receptor binding. The binding parameter, k3, was decreased after triazolam administration in all measured regions, whereas no change was observed after placebo treatment. The reduction compared to the control study varied from 8.6 +/- 3.7% in the temporal cortex to 16.3 +/- 6.3% in the thalamus. Triazolam administration impaired both immediate and delayed recall of syllables, whereas placebo administration had no effects. Benzodiazepine agonists are reported to decrease the cortical acetylcholine release. The decrease of acetylcholine release in the synaptic cleft might be the explanation for the decreased binding of [11C]NMPB.

D1 dopamine receptor binding in mood disorders measured by positron emission tomography.

D1 dopamine receptor binding in mood disorders was studied by positron emission tomography (PET) using 11C-SCH23390. Ten patients with bipolar mood disorders and 21 normal controls were studied in the drug-free state. The patients were in euthymic (N = 6), depressed (N = 3) and manic (N = 1) states. Regional radioactivity in the brain was followed for 40 min by PET. A two-compartment model was used to obtain the binding potential (k3/k4) for the striatum and frontal cortex. The binding potentials for the frontal cortex for the patients were significantly lower than those for normal controls, whereas those for striatum were not significantly different. These findings suggest that D1 dopamine receptors in the frontal cortex may be in a different state in patients with bipolar mood disorders.

Age-related changes in human D1 dopamine receptors measured by positron emission tomography.

The effects of age on the binding parameters of 11C-SCH23390, the highly selective ligand for central D1 dopamine receptors, at specific binding sites in the brain were studied. Seventeen healthy male volunteers (20-72 years old) participated. Regional radioactivity in the brain was followed for 40 min by positron emission tomography (PET). A high accumulation of radioactivity was observed in the striatum and there was a conspicuous accumulation in the neocortex. A two-compartment model was used to obtain quantitative estimates of rate constants of association (K3) and dissociation (k4). The binding potential (k3/k4) of the dopamine D1 receptors in the striatum and frontal cortex decreased by 35% and 39%, respectively, with age. The value of k3 decreased by 58% in the striatum and 83% in the frontal cortex, whereas the value of k4 decreased by 35% in the striatum and 72% in the frontal cortex with age.

Detection of benzodiazepine receptor occupancy in the human brain by positron emission tomography.

Benzodiazepine receptor occupancy in the brain following oral administration of clonazepam (CZP) with a dose of 30 micrograms/kg in six healthy young men and a further dose of 50 micrograms/kg in one of the subjects was estimated by carbon-11 labeled Ro15-1788 and positron emission tomography (PET). The effects of CZP on the latency of auditory event-related potentials (P300) were also studied. Overall brain 11C uptake was depressed and the % inhibition of 11C uptake in the gray matter of the brain at 30 min after [11C]Ro15-1788 injection was 15.3-23.5% (mean, n = 6) following 30 micrograms/kg CZP when compared with that in the control experiment without any previous treatment. The 11C uptake in the cerebral cortex in the subject who received both doses decreased in a dose-related manner after 30 micrograms/kg and 50 micrograms/kg CZP. The P300 latency was prolonged significantly by 30 micrograms/kg CZP [31.6 +/- 16.3 ms (mean +/- SD, n = 6), P less than 0.05]. The P300 latency in the same subject was prolonged in a dose-related manner by 30 micrograms/kg and 50 micrograms/kg CZP. The technique using [11C]Ro15-1788 and PET permits comparison of the pharmacological effects with the percentage of receptor sites which benzodiazepines occupy in the human brain. P300 also seems to be useful to investigate the pharmacological effects of benzodiazepines.

Serum concentrations of total bile acids in patients with acute Kawasaki syndrome.

OBJECTIVES: To investigate serum concentrations of total bile acids (TBAs) in patients with acute Kawasaki syndrome (KS) and correlate those concentrations with other liver function test results. PATIENTS AND METHODS: Seventy-one Japanese patients with KS (mean age, 1.6 years). Blood samples from 29 patients were obtained during the acute and convalescent phases of KS. RESULTS: Elevated serum TBA concentrations higher than 20 micromol/L were detected in 16 (22.5%) of the patients with acute KS. Serum concentrations of TBAs gradually returned to normal (P=.001, Pearson's correlation coefficient). The mean +/- SD serum concentration of TBAs of patients during the acute phase (37.9 +/- 63.9 micromol/L) was significantly higher than that of patients during the convalescent phase (7.2 +/- 7.5 micromol/L) (P=.02, Pearson's correlation coefficient). The data show a correlation between the serum concentrations of TBAs with those of alanine aminotransferase (P=.03, Pearson's correlation coefficient), total bilirubin (P<.001, Pearson's correlation coefficient), and C-reactive protein (P=.01, Pearson's correlation coefficient) measured during the acute phase. CONCLUSIONS: The cause of the elevated concentration of serum TBAs in patients with acute KS is unclear. Increased bile acid synthesis from cholesterol may occur, or patients with acute KS may suffer from damage to the bile duct cells of their biliary systems caused by cytokine activation. This damage may depress bile acid excretion, thereby increasing the serum concentrations of TBAs.