Magnetic orientation of 1,3,5-triphenyl-6-oxoverdazyl radical crystals.

The magnetic orientation has been studied for paramagnetic organic radical crystals 1,3,5-triphenyl-6-oxoverdazyl and 1,5-di-p-tolyl-3-phenyl-6-oxoverdazyl in magnetic fields of 2-80 kOe at temperatures of 77-343 K. The X-ray diffraction measurement has revealed that the crystals are oriented with the crystallographic c axis perpendicular to the field. The anisotropic diamagnetic susceptibility arising from the benzene rings has been estimated for the crystals along the principal magnetic chi 1, chi 2, and chi 3 axes. (The chi 1 axis is at a small angle to the a axis in the monoclinic ac plane, and the chi 3 axis is along the b axis.) Since the paramagnetic susceptibility originating from the verdazyl ring is isotropic (though a large absolute value), it is shown that the magnetic orientation occurs by the anisotropy of the diamagnetic susceptibility in the crystals. The diamagnetic susceptibility is found to have a relation of chi 2 < chi 1 < chi 3 < 0.

Liver-infiltrating T lymphocytes are attracted selectively by IFN-inducible protein-10.

We have demonstrated that interferon-inducible protein-10 (IP-10) is produced in hepatocytes surrounded by infiltrative mononuclear cells in chronic hepatitis. To clarify the role of IP-10 in hepatitis, we examined the chemoattractive activity of IP-10 on liver-infiltrating lymphocytes in experimental animal models of hepatitis. IP-10 was specifically induced in the livers of mice treated intravenously (i.v.) with Con A, while monocyte chemotactic protein-1 (MCP-1) showed a much lower level of induction and neither RANTES nor macrophage inflammatory protein-1alpha (MIP-1alpha) was detected. The liver-infiltrating lymphocytes in Con A-induced hepatitis were attracted only by IP-10, and not by other chemokines such as RANTES, MCP-1 and MIP-1alpha. The chemoattractive effect of IP-10 was dose-dependent and was neutralized by monoclonal antibodies to IP-10. The specific effect of IP-10 on liver-infiltrating lymphocytes was also seen on those obtained from rat livers with fulminant hepatitis induced by sequential treatment with killed Propionibacterium acnes (P. acnes) and LPS. Peripheral blood lymphocytes were slightly attracted by IP-10 as well as RANTES and MIP-1alpha, while hepatic resident lymphocytes were not. On the other hand, thioglycolate-elicited peritoneal macrophages did not respond to IP-10, although they did show a response to RANTES, MCP-1 and MIP-1alpha. These results indicated that IP-10 is a specific chemoattractant for T lymphocytes in the inflammatory liver tissues and may play a specific role in the development of hepatitis.

Increased levels of serum interleukin-12 in Graves' disease.

We investigated serum total interleukin-12 (IL-12) levels in patients with Graves' disease and Hashimoto's thyroiditis. The serum IL-12 levels in Graves' disease were significantly increased in the hyperthyroid state, and were decreased during treatment with methimazole or propylthiouracil in accordance with the decline of free tri-iodothyronine (T(3)) levels, free thyroxine levels and thyroid-binding inhibiting immunoglobulin (TBII) levels. When T(3) was administered orally to normal subjects, serum IL-12 levels were slightly increased. These results suggest that IL-12 might be increased due to prolonged stimulation with thyroid hormone, and thyroid hormone by itself might be a self-perpetuating factor of Graves' disease via increased IL-12 production.

E-selectin gene expression is induced synergistically with the coexistence of activated classic protein kinase C and signals elicited by interleukin-1beta but not tumor necrosis factor-alpha.

We have examined the effect of protein kinase C (PKC) on the expression of the E-selectin and intercellular adhesion molecule-1 (ICAM-1) mRNAs in human umbilical vein endothelial cells. The lower classic PKC activity on pretreatment with phorbol ester (phorbol 12-myristate 13-acetate (PMA)) for 24 h markedly decreased IL-1beta-induced E-selectin mRNA expression in the presence of fetal calf serum and basic fibroblast growth factor, although the induction of ICAM-1 mRNA expression was only influenced a little by the PKC down-regulation. On the other hand, tumor necrosis factor-alpha (TNFalpha)-induced gene expression of these adhesion molecules was unaffected by such PKC modulation. The intracellular signals generated by interleukin (IL)-1beta and TNFalpha themselves are not mediated through classic PKC activation, because the response to neither stimulant was inhibited by the PKC down-regulation in the absence of fetal calf serum and basic fibroblast growth factor. Simultaneous treatment with IL-1beta and PMA synergistically induced E-selectin gene expression but not when TNFalpha was substituted for IL-1beta. ICAM-1 mRNA expression was only additively induced on the cotreatment. The synergistic effect on E-selectin mRNA induction was independent of de novo protein synthesis and mediated by elevated transcriptional activity. Promoter analysis of E-selectin indicated that the NF-ELAM1/activating transcription factor element is critical for the synergistic effect of the cotreatment with IL-1beta and PMA.

Cloning of the murine interferon-inducible protein 10 (IP-10) receptor and its specific expression in lymphoid organs.

To isolate the interferon-inducible protein 10 (IP-10) receptor gene, we searched for cells that respond to IP-10. Among several human and murine T cell lines, only CTLL2 cells ( a murine cytotoxic T cell line) responded to IP-10 with transient elevation of intracellular Ca2+. The murine IP-10 receptor gene has been cloned from cDNA derived from CTLL2 cells using the reverse transcriptase-polymerase chain reaction protocol with two degenerate primers corresponding to conserved regions of chemokine receptors. The cDNA encoding the murine IP-10 receptor has an open reading frame of 1101 bp corresponding to a protein of 367 amino acids that exhibits 86 % identity with the human IP-10 receptor. It mediates Ca2+ mobilization in response to IP-10, but does not recognize other rodent chemokines, including GRO, RANTES, monocyte chemoattractant protein-1 (MCP-1), and macrophage inflammatory protein-1alpha (MIP-1alpha). Northern blot analysis revealed that murine IP-10 and its receptor mRNA were constitutively expressed in the spleen and thymus from normal mouse, while IP-10 and its receptor mRNA were derived from stromal cells and lymphocytes in both tissues, respectively. In vivo treatment with concanavalin A (Con A) for 12 hrs revealed that splenocytes significantly induce IP-10 receptor mRNA expression and show a good chemotactic response to IP-10. Therefore, it is supposed that IP-10 and its receptor are important for lymphocyte trafficking to lymphoid organs and that the IP-10 receptor on lymphocytes is rapidly inducible on inflammation or in immunological events.

Interleukin-1beta induces tissue- and cell type-specific expression of adhesion molecules in vivo.

We examined the tissue distribution of adhesion molecule gene expression in mice treated intravenously with interleukin (IL)-1 beta. E-selectin mRNA expression was selectively induced in the heart by IL-1 beta, but only slight or no induction was observed in other organs. On the other hand, intercellular adhesion molecule-1 mRNA expression was inducible in all organs examined, although it showed the strongest induction in the lung and the weakest responses in the brain and skin. Vascular cell adhesion molecule-1 mRNA was also inducible in all organs with the exception of the skin, but it was induced most markedly in the lung and the heart. The accessibility of IL-1 beta to the heart was less than that to other organs except the brain. Similar tissue-specific induction of these mRNAs was also seen when tumor necrosis factor (TNF)-alpha or lipopolysaccharide was substituted for IL-1 beta. Analysis of E-selectin mRNA expression in the heart by in situ hybridization indicated that expression was most prominent in microvascular endothelial cells and some other stromal cells, but this transcript was not seen in the lung. Although intercellular adhesion molecule-1 mRNA expression was restricted to the endothelium lining the capillaries and small arteries in the heart, its distribution in the lung covered not only the endothelium but also the cells composing the alveolar septa. In contrast, vascular cell adhesion molecule-1 mRNA expression was most prominent in endothelial cells of larger vessels in both the heart and the lung. Our results demonstrate that expression of adhesion molecules is tissue- and cell type-specific and that endothelial cells differentially express adhesion molecules depending on the size of the blood vessels.

Developmental changes in the NGF content in the brain of young, growing, low-birth-weight rats.

The NGF content in each region of the brain of four-week-old rats was ranked in the decreasing order of cerebral cortex, hippocampus, cerebellum, midbrain/diencephalon, and pons/medulla oblongata, and the NGF concentration, in the decreasing order of hippocampus, cerebral cortex, cerebellum, midbrain/diencephalon, and pons/medulla oblongata in both AFD and SFD groups. The NGF content and concentration in the cerebral cortex were about the same value at each age between those in the AFD and SFD groups. Those in the hippocampus were a little higher in the SFD group than in the AFD group at the ages of three and four weeks, unlike those in the other regions, where the values for the cerebellum, midbrain/diencephalon and pons/medulla oblongata tended to be somewhat higher in the AFD group than in the SFD group. The NGF concentrations in the hippocampus and cerebral cortex increased with growth: the concentration in the hippocampus at four weeks of age was about 4-fold of that at one week in the AFD group and about 5.7-fold of that at one week in the SFD group; and likewise the concentration in the cerebral cortex at four weeks of age was about 5.3-fold in the AFD group and about 7-fold in the SFD group. The NGF concentrations in the cerebellum decreased, and those in midbrain/diencephalon and pons/medulla oblongata hardly changed with growth in either AFD or SFD group. From these results NGF may have stronger implications for the neuronal growth in the hippocampus compared with those in the lower brain regions of the SFD rats.

Expression of IFN-inducible protein-10 in chronic hepatitis.

Chemokines such as IFN-inducible protein-10 (IP-10) and JE/monocyte chemotactic protein-1 (MCP-1) are induced in the murine liver in a tissue-specific manner. We examined whether IP-10 and MCP-1 are pathologically involved in chronic hepatitis. Whereas the serum levels of IP-10 and MCP-1 in patients with chronic persistent hepatitis C were elevated compared with those in normal volunteers, both chemokine levels were further significantly higher in patients with the active form (chronic active hepatitis (CAH)). The elevated IP-10 level was not a general phenomenon of inflammation, because it was not seen in patients with rheumatoid arthritis, whereas MCP-1 levels were elevated to the same extent in both patient groups. Better responsiveness to IFN therapy in CAH was related to lesser grades of necroinflammatory activity and was predicted by the lower IP-10 and higher MCP-1 levels. IP-10 levels in patients cured by IFN therapy decreased to the levels in normal volunteers, while the MCP-1 levels only slightly decreased. Serum levels of both chemokines in patients who were not cured remained unchanged after IFN therapy. In situ hybridization analysis of CAH revealed that IP-10 mRNA was expressed mainly in hepatocytes around intralobular focal and periportal piecemeal necrosis, while some MCP-1 mRNA was expressed in some sinusoidal cells. These results suggested that IP-10 plays a specific role in the intralobular accumulation of mononuclear cells and/or the death of hepatocytes in chronic hepatitis.

Fractionated-dose effect of X-irradiation on the induction of neural tube defects in mice.

This experiment examined whether mouse embryos have a radio-adaptive response to the induction of malformations. Pregnant Slc:ICR mice were exposed to 0.05 Gy X-radiation followed by 0.5 or 1.0 Gy 4 hr later on day 8 of pregnancy. To compare this with the effect of single doses, four other groups were exposed to 0.5, 0.55, 1.0 or 1.05 Gy. All dams were put to death on day 18, and fetuses were examined for malformations. The frequency of neural tube defects induced by the fractionated irradiation into 0.05 Gy and 1.0 Gy was significantly higher than that induced by single doses of 1.0 and 1.05 Gy. Among groups exposed to a dose split into 0.05 and 0.5 Gy, and single doses of 0.5 and 0.55 Gy, there was no significant difference in incidence of the malformation. The present findings suggest that mouse embryos are not adapted to X-irradiation, but become more susceptible following a low-dose exposure.

Graves ophthalmopathy: MR evaluation of 10-Gy versus 24-Gy irradiation combined with systemic corticosteroids.

PURPOSE: To establish the most effective radiation dose for treatment of Graves ophthalmopathy (GO). MATERIALS AND METHODS: A combination of 10 Gy (n = 15) or 24 Gy (n = 16) of radiation and corticosteroids was used to treat 31 patients with GO. Magnetic resonance (MR) images obtained before treatment showed swollen extraocular muscles with prolonged T2 relaxation times in all patients. RESULTS: Before therapy, T2 relaxation time of extraocular muscle was 79.6 msec (95% confidence interval, 76.3, 82.9) in the 24-Gy group and 77.4 msec (95% confidence interval, 74.6, 80.1) in the 10-Gy group (P = .32). After therapy, T2 relaxation time was 62.8 msec (95% confidence interval, 61.2, 64.4) in the 24-Gy group and 68.9 msec (95% confidence interval, 66.8, 71.1) in the 10-Gy group. In the 24-Gy group, there was a significant decrease in T2 relaxation times (P = .001) and clinical response to initial treatment was better. At 1- and 3-month follow-up, the resistance rate was lower in the 24-Gy group. CONCLUSION: In treatment of GO, 24 Gy of radiation is a more effective dose than 10 Gy when combined with systemic corticosteroids.

Thyroglobulin and lactic dehydrogenase isozymes in cystic fluid of thyroid nodules.

Although fine needle aspiration biopsy (FNAB) is most valuable in the diagnosis of thyroid cancer, it is hampered by the fact that no specimen suitable for cytological examination can be collected from all cystic lesions. Often inadequate aspirates, consisting only of fluid or a few foamy cells and lacking the necessary epithelial cells, are all that an aspirationist is able to collect. Therefore an alternative method of determining the benign or malignant characteristics of cyst fluid is of vital importance. In this study we examine thyroglobulin (Tg) concentrations and lactic dehydrogenase (LDH) isozyme patterns of cyst fluid and discuss how these variables help us estimate the probability of malignancy. Fifty-three differentiated cancers (39 papillary and 14 follicular carcinomas) and 72 surgically resected benign thyroid nodules (40 adenomas, 19 colloid goiters, and 13 cysts) were analyzed. Only 28 (53%) of 53 malignant lesions were correctly diagnosed by FNAB. The mean logarithmic value for the Tg concentration (log10 Tg) was significantly lower in malignant cyst fluid than it was in benign nodules (mean +/- SD: 5.8 +/- 1.0 vs. 6.8 +/- 1.0; P < 0.001). The LDH 1 and 2 isozyme percentage was greater in the malignant group than in the benign group (49.1 +/- 12.7% vs. 38.1 +/- 16.9%; P < 0.01). In multiple logistic regression analysis, log10 Tg and the total of LDH 1+2 percentage was significant in estimating the probability of malignant nodules. The results of our study suggest that determining the Tg concentration and the LDH isozyme patterns of cyst fluid could provide new information for the evaluation of cystic thyroid nodules.

[Granulocyte colony-stimulating factor treatment (G-CSF) of antithyroid drug-induced granulocytopenia: granulocyte count measurement after 4 hours of G-CSF injection is useful for the detection of recovery from granulocytopenia].

The primary objective of this study was to ascertain the usefulness of granulocyte count measurement after 4 hours of granulocyte colony-stimulating factor (G-CSF) injections for the detection of recovery from granulocytopenia. Four Graves' patients with antithyroid drug-induced granulocytopenia (granulocyte count between 500 and 1000/mm3) and three Graves' patients with antithyroid drug-induced agranulocytosis (granulocyte count < 500/mm3) each received a daily dose of 75 mu g of G-CSF administered subcutaneously. In all granulocytopenic patients, after 4 hours of G-CSF injection the granulocyte counts increased to 5623, 4050, 8923 and 4647/mm3, and the granulocyte count after 24 hours of G-CSF injection was 3008, 4634, 4854, 4200/mm3. In one of the three agranulocytic patients, the granulocyte count increased from 238/mm3 to 5982/mm3 after 4 hours of G-CSF injection, and the granulocyte count after 24 hours of G-CSF injection was 4800/mm3. Although the granulocyte counts before G-CSF injection of the remaining two agranulocytic patients were 138 and 126/mm3, the granulocyte counts after 4 hours of G-CSF injection were 837 and 59/mm3 and those after 24 hours of G-CSF injection were 817 and 0/mm3. These results indicated that granulocyte count measurement after 4 hours of G-CSF injection was useful for detecting the recovery from granulocytopenia and agranulocytosis.

[Antithyroid drug-induced agranulocytosis: special reference to normal white blood cell count agranulocytosis].

This retrospective study was aimed at establishing the importance of the leukocyte differentiated count and not only routine white blood cell count in patients treated with antithyroid drug. From 1975 to September 1992, 77 patients with antithyroid drug-induced agranulocytosis were examined. In 12 patients (15.6%), the total white blood cell (WBC) count was greater than 3000/mm3. Eight of them showed a downward trend in their leukocyte counts (3000-4000/mm3). Consequently, granulocyte counts were measured. Two of the 12 patients had "symptomatic" agranulocytosis detected after the occurrence of infection. Because antithyroid drug-induced agranulocytosis was strongly suspected, granulocyte counts were checked. In the remaining two patients, the total WBC count was 5700/mm3 and 5900/mm3, respectively. One was hospitalized to receive thyroid surgery. Although she was asymptomatic, agranulocytosis was unexpectedly detected on a routine preoperative examination. The other was diagnosed as agranulocytosis by routine WBC and granulocyte count monitoring since June 1989. Correct diagnosis was based on the leukocyte differentiated counts. We concluded that the leukocyte differentiated count and not only routine white blood count was critically important for the correct diagnosis of antithyroid drug-induced agranulocytosis in patients with Graves' disease.

Effects of changes in gonadal hormones on the amount of aromatase messenger RNA in mouse brain diencephalon.

Aromatase in brain is known to play a crucial role in development and maintenance of androgen dependent sexual behavior of males. Biochemical studies have shown that the aromatase activity in brain is influenced by androgen. In this study, we measured the aromatase mRNA content of mouse brain quantitatively to gain a deeper insight into this phenomenon. We found that in the diencephalic area it is sexually dimorphic, as reported for aromatase activity. The amount of aromatase mRNA in this area in males was 150% higher than that in females and decreased to the level in females after castration. The content of aromatase mRNA in castrated males was elevated by 2-fold by injection of testosterone and restored to the level observed in intact males. Injection of testosterone also affected the level of aromatase mRNA in normal mice of both sexes, though to lesser extent. On the contrary, injection of estrogen decreased the amount of aromatase mRNA in gonadectomized mice of both sexes. These results show that transcriptional control of the aromatase gene is involved in the mechanism of testosterone to affect sexual behavior.

[Studies on onset patterns in antithyroid drug-induced agranulocytosis].

We studied the patterns of the onset of antithyroid drug-induced agranulocytosis. From 1975 to 1990, 19,050 patients with Graves' disease receiving treatment with antithyroid drugs were seen at our clinic. For all patients with Graves' disease treated with an antithyroid drug, a routine white blood cell count was done every 2 weeks until euthyroid state was gained, and a count was done once every moth thereafter. Of these, 70 were found to have agranulocytosis. Agranulocytosis was defined as a granulocyte count of 500/mm3 or less. In only 19 of the 70 was agranulocytosis detected after the occurrence of infection (symptomatic: classical agranulocytosis). The remaining 51 patients were asymptomatic when agranulocytosis was detected during routine white blood cell and granulocyte count monitoring. However, 17 of the 51 patients became symptomatic several days after the withdrawal of antithyroid drug treatment (shifted from asymptomatic to symptomatic agranulocytosis). Thirty-four patients had no symptoms of infection throughout the course of the disease (asymptomatic agranulocytosis). In conclusion, 1) We found three patterns in the onset of antithyroid drug-induced agranulocytosis: classical (symptomatic), a shift from asymptomatic to symptomatic, and asymptomatic agranulocytosis, 2) Unexpectedly, classical (symptomatic) agranulocytosis was seen in only 19 of the 70 patients, 3) We were again remained of the importance of routine white blood cell and granulocyte count monitoring.

Serum sialic acid levels in the diagnosis and follow-up of subacute granulomatous thyroiditis.

To evaluate the usefulness of monitoring serum sialic acid (SA) levels for diagnosis and follow-up of subacute granulomatous thyroiditis (SAT), 43 patients were studied at our clinic. In the acute phase of the disease their SA levels averaged 104.9 +/- 19.7 mg/dl (normal 44-69 mg/dl). In the recovery phase SA levels returned to a range of 60.5 +/- 6.9 mg/dl. However, an increase in SA (87.4 +/- 18.2 mg/dl) was detected at the time of recurrence in 14 patients. In 29 non-recurrent patients, serum SA gradually reduced during the course of therapy and normalized in all patients by the time glucocorticoid therapy was discontinued. Thyroglobulin (Tg) and the erythrocyte sedimentation rate (ESR), however, had normalized in only half the cases even at the time of cessation of therapy (Tg 5/11, ESR 4/8). C-reactive protein (CRP) returned to negative in most patients (19/24) only one week after initiation of the therapy. These results suggested that the monitoring of SA levels can be a useful tool in diagnosis and follow-up of SAT.

Granulocyte colony-stimulating factor treatment of antithyroid drug-induced granulocytopenia.

The primary objective of this study was to ascertain the effectiveness of granulocyte colony-stimulating factor in the treatment of antithyroid drug-induced granulocytopenia of varying degree. Sixteen patients with Graves' disease with antithyroid drug-induced granulocytopenia (granulocyte counts < 1.0 x 10(9)/L) each received a daily dose of 75 micrograms of granulocyte colony-stimulating factor administered subcutaneously. Within 24 hours of the first injection, the granulocyte count increased (0.6 to 12.3 x 10(9)/L) in all 10 patients with mild granulocytopenia (granulocyte counts between 0.5 and 1.0 x 10(9)/L) and all three with moderate granulocytopenia (granulocyte counts < 0.5 x 10(9)/L). The three remaining patients with severe granulocytopenia (agranulocytic), whose granulocyte counts were zero, did not recover from granulocytopenia until the 6th, 7th, and 14th days of treatment with granulocyte colony-stimulating factor. Examination of bone marrow taken at the onset of the disease in all three agranulocytic patients showed a prominent decrease in granulocytic series, while identical examination in six of eight patients with mild to moderate granulocytopenia showed close to normal granulocytic series. There was no elevation of serum granulocyte colony-stimulating factor concentration in four patients with mild granulocytopenia and one with moderate granulocytopenia at the onset of their disease, whereas those of the remaining three patients with severe granulocytopenia (agranulocytic) increased at onset of agranulocytosis. This information led us to conclude that: (1) granulocyte colony-stimulating factor is effective in the treatment of antithyroid drug-induced mild to moderate granulocytopenia and (2) in severe agranulocytic cases, granulocyte colony-stimulating factor is not effective. Accordingly, we were again reminded of the importance of early diagnosis and treatment of antithyroid drug-induced agranulocytosis.

Excitatory amino acid receptors in brains of rats with methylazoxymethanol-induced microencephaly.

We used methylazoxymethanol-acetate (MAM), a potent alkylating agent, to produce microencephaly in offspring by injecting it into pregnant rats on day 15 of gestation. Binding activities of central excitatory amino acid receptors were examined in Triton-treated membranes prepared from brains of adult offspring with MAM-induced microencephaly (MAM rats). MAM rats exhibited approximately 40-50% reductions of the wet weights of the cerebral cortex, hippocampus and striatum compared to those in controls. In the cortex and hippocampus of MAM-rats, total bindings of [3H]glutamate (Glu) (which is sensitive to N-methyl-D-aspartate (NMDA) receptor), and strychnine-insensitive [3H]glycine (Gly) and (+)-5-[3H]methyl-10,11-dihydro-5H-dibenzo[a,d]cyclohepten-5,10-imi ne (MK-801; a noncompetitive antagonist of NMDA receptor), were reduced to approximately 40% of those in controls. Similarly, in both regions of MAM rats, total bindings of [3H]kainate and DL-alpha-amino-3-[3H]hydroxy-5-methylisoxazole-4-propionic acid (an agonist of quisqualate receptors), were reduced to approximately 35-50% of those in controls. However, total bindings of these radioligands in the striatum of MAM rats were more than 65% of those in controls, despite the significant loss of striatum mass. However, specific bindings of radioligands in the striatum of MAM rats were elevated by more than 60% of those in controls, and Scatchard analysis revealed that elevations of [3H]Glu, [3H]Gly and [3H]MK-801 bindings were due to a significant increase in the densities of binding sites, with their affinities remaining unaltered. Spatial recognition ability examined by an 8-armed radial maze task was markedly impaired compared to those in controls. These results suggest that the proliferation of neurons bearing excitatory amino acid receptors (EAA) in the striatum is less affected by MAM treatment on day 15 of gestation than that in the cortex and hippocampus in spite of drastic weight loss in these brain regions. The significant reduction of EAA receptors in the cortex and hippocampus may be involved in the impairment of spatial memory observed in MAM-treated rats.

The ligand binding site of the synaptosomal choline transporter: a provisional model based on inhibition studies.

A topographic model of the ligand binding site of the choline transporter was deduced from inhibition studies with the help of CPK molecular models. It is posited that there are two identical or closely similar hydrophilic anionic sites separated from each other by an hinged, essentially planar but conformationally flexible cationic hydrophobic domain. Subsequently to attachment of external choline to either one of the anionic sites, both sites cooperate in enveloping the ligand by a Venus fly-trap mechanism. This leads to rapid configurational changes by which the closed-liganded form of the transporter opens up to the interior to release the bound choline. Intracellular K+, a ligand for the choline-binding site, is proposed to be counter-transported by a reversal of the above mechanism.