The functional interplay of low molecular weight thiols in Mycobacterium tuberculosis.

BACKGROUND: Three low molecular weight thiols are synthesized by Mycobacterium tuberculosis (M.tb), namely ergothioneine (ERG), mycothiol (MSH) and gamma-glutamylcysteine (GGC). They are able to counteract reactive oxygen species (ROS) and/or reactive nitrogen species (RNS). In addition, the production of ERG is elevated in the MSH-deficient M.tb mutant, while the production of MSH is elevated in the ERG-deficient mutants. Furthermore, the production of GGC is elevated in the MSH-deficient mutant and the ERG-deficient mutants. The propensity of one thiol to be elevated in the absence of the other prompted further investigations into their interplay in M.tb. METHODS: To achieve that, we generated two M.tb mutants that are unable to produce ERG nor MSH but are able to produce a moderate (ΔegtD-mshA) or significantly high (ΔegtB-mshA) amount of GGC relative to the wild-type strain. In addition, we generated an M.tb mutant that is unable to produce GGC nor MSH but is able to produce a significantly low level of ERG (ΔegtA-mshA) relative to the wild-type strain. The susceptibilities of these mutants to various in vitro and ex vivo stress conditions were investigated and compared. RESULTS: The ΔegtA-mshA mutant was the most susceptible to cellular stress relative to its parent single mutant strains (ΔegtA and ∆mshA) and the other double mutants. In addition, it displayed a growth-defect in vitro, in mouse and human macrophages suggesting; that the complete inhibition of ERG, MSH and GGC biosynthesis is deleterious for the growth of M.tb. CONCLUSIONS: This study indicates that ERG, MSH and GGC are able to compensate for each other to maximize the protection and ensure the fitness of M.tb. This study therefore suggests that the most effective strategy to target thiol biosynthesis for anti-tuberculosis drug development would be the simultaneous inhibition of the biosynthesis of ERG, MSH and GGC.

Activities of E1210 and comparator agents tested by CLSI and EUCAST broth microdilution methods against Fusarium and Scedosporium species identified using molecular methods.

Fusarium (n = 67) and Scedosporium (n = 63) clinical isolates were tested by two reference broth microdilution (BMD) methods against a novel broad-spectrum (active against both yeasts and molds) antifungal, E1210, and comparator agents. E1210 inhibits the inositol acylation step in glycophosphatidylinositol (GPI) biosynthesis, resulting in defects in fungal cell wall biosynthesis. Five species complex organisms/species of Fusarium (4 isolates unspeciated) and 28 Scedosporium apiospermum, 7 Scedosporium aurantiacum, and 28 Scedosporium prolificans species were identified by molecular techniques. Comparator antifungal agents included anidulafungin, caspofungin, itraconazole, posaconazole, voriconazole, and amphotericin B. E1210 was highly active against all of the tested isolates, with minimum effective concentration (MEC)/MIC(90) values (µg/ml) for E1210, anidulafungin, caspofungin, itraconazole, posaconazole, voriconazole, and amphotericin B, respectively, for Fusarium of 0.12, >16, >16, >8, >8, 8, and 4 µg/ml. E1210 was very potent against the Scedosporium spp. tested. The E1210 MEC(90) was 0.12 µg/ml for S. apiospermum, but 1 to >8 µg/ml for other tested agents. Against S. aurantiacum, the MEC(50) for E1210 was 0.06 µg/ml versus 0.5 to >8 µg/ml for the comparators. Against S. prolificans, the MEC(90) for E1210 was only 0.12 µg/ml, compared to >4 µg/ml for amphotericin B and >8 µg/ml for itraconazole, posaconazole, and voriconazole. Both CLSI and EUCAST methods were highly concordant for E1210 and all comparator agents. The essential agreement (EA; ±2 doubling dilutions) was >93% for all comparisons, with the exception of posaconazole and F. oxysporum species complex (SC) (60%), posaconazole and S. aurantiacum (85.7%), and voriconazole and S. aurantiacum (85.7%). In conclusion, E1210 exhibited very potent and broad-spectrum antifungal activity against azole- and amphotericin B-resistant strains of Fusarium spp. and Scedosporium spp. Furthermore, in vitro susceptibility testing of E1210 against isolates of Fusarium and Scedosporium may be accomplished using either of the CLSI or EUCAST BMD methods, each producing very similar results.

Huntington's disease: degradation of mutant huntingtin by autophagy.

Autophagy is a nonspecific bulk degradation pathway for long-lived cytoplasmic proteins, protein complexes, or damaged organelles. This process is also a major degradation pathway for many aggregate-prone, disease-causing proteins associated with neurodegenerative disorders, such as mutant huntingtin in Huntington's disease. In this review, we discuss factors regulating the degradation of mutant huntingtin by autophagy. We also report the growing list of new drugs/pathways that upregulate autophagy to enhance the clearance of this mutant protein, as autophagy upregulation may be a tractable strategy for the treatment of Huntington's disease.

Specificity of mood stabilizer action on neuronal growth cones.

OBJECTIVES: Lithium, valproic acid (VPA) and carbamazepine (CBZ) are commonly used mood stabilizers, but their therapeutic mechanism is unclear. These drugs all cause the same morphological effects on postnatal rat neuronal dorsal root ganglia (DRG) growth cones via an inositol-reversible mechanism. However, due to limitations in earlier analysis, the effects of combining drugs, drug specificity and inositol stereoisomer specificity are unknown. We devised an improved analytical method to address these issues. METHODS: Dorsal root ganglia explants were cultured individually and incubated with combinations of psychotropic drugs and inositol stereoisomers. We recorded axonal growth cone morphology and calculated growth cone area per a modified method described by Williams et al. (Nature 2002; 417: 292-295). Statistically significant changes in area were calculated using non-parametric statistical testing. RESULTS: (i) Lithium and VPA showed an additive effect on growth cone spreading. (ii) Among eight additional psychotropic drugs to those previously tested, only imipramine and chlorpromazine altered DRG growth cone morphology. As this effect was not reversed by myo-inositol, it arises from a different mechanism to the mood stabilizers lithium, VPA and CBZ. (iii) Myo-inositol, but not scyllo- or epi-inositol, causes a significant reversal of the lithium effect on the growth cones spreading, consistent with the inositol depletion hypothesis. CONCLUSIONS: These results show that lithium, VPA and CBZ are unique in causing altered neuronal morphology via myo-inositol depletion.

The mshA gene encoding the glycosyltransferase of mycothiol biosynthesis is essential in Mycobacterium tuberculosis Erdman.

Mycothiol is the major low-molecular-weight thiol found in actinomycetes, including Mycobacterium tuberculosis, and has important antioxidant and detoxification functions. Gene disruption studies have shown that mycothiol is essential for the growth of M. tuberculosis. Because of mycothiol's unique characteristics, inhibitors directed against mycothiol biosynthesis have potential as drugs against M. tuberculosis. Four genes have been identified in mycobacteria that are involved in the biosynthesis of mycothiol. Two genes, mshB and mshD, are not essential for growth of M. tuberculosis. Mutants in these genes produce significant amounts of mycothiol or closely related thiol compounds. A targeted gene disruption in the mshC gene is lethal for M. tuberculosis, indicating that MshC is essential for growth. The remaining gene, mshA, encodes for a glycosyltransferase. In the present study, we attempted to produce a directed knock-out of the mshA gene in M. tuberculosis Erdman but found that this was only possible when a second copy of mshA was first incorporated into the chromosome. Bacteria with only a single copy of mshA that grew after mutagenesis produced normal levels of mycothiol. We therefore conclude that the mshA gene, like the mshC gene, is essential for the growth of M. tuberculosis.

A model of inositol compartmentation in astrocytes based upon efflux kinetics and slow inositol depletion after uptake inhibition.

Intracellular compartmentation of inositol was demonstrated in primary cultures of mouse astrocytes, incubated in isotonic medium, by determination of efflux kinetics after "loading" with [3H]inositol. Three kinetically different compartments were delineated. The largest and most slowly exchanging compartment had a halflife of approximately 9 hr. This slow release leads to retention of a sizeable amount of pre-accumulated inositol in the tissue 24 hr after the onset of uptake inhibition, as confirmed by the observation that the inositol uptake inhibitor fucose caused a larger inhibition of unidirectional inositol uptake than of inositol pool size, measured as accumulated [3H]inositol after 24 hr of combined exposure to the inhibitor and the labeled isotope. Based upon the present observations and literature data, it is suggested that the large, slowly exchanging compartment is largely membrane-associated and participating in signaling via the phosphatidylinositide second messenger system, whereas inositol functioning as an osmolyte is distributed in the cytosol and located in one or both of the compartments showing a faster release.

Inhibition of inositol uptake in astrocytes by antisense oligonucleotides delivered by pH-sensitive liposomes.

An oligonucleotide of 20 bases, complementary to a region of the sodium/myo-inositol cotransporter (SMIT) mRNA, was used to investigate the uptake efficiency and activity of transferred antisense oligonucleotides with regard to substrate uptake. We compared the efficiency of oligonucleotide delivery after application of either free or liposome-encapsulated material. Delivery of liposome-encapsulated material (marker or oligonucleotides) into astrocytoma cells and primary astrocyte cultures was more effective with pH-sensitive liposomes [dioleoylphosphatidylethanolamine (DOPE)/cholesteryl hemisuccinate (CHEMS)] than with non-pH-sensitive liposomes (soy lecithin) or free material in solution. Antisense activity was evaluated by determination of myo-inositol uptake and detection of SMIT transcripts by RT-PCR. Encapsulation of oligonucleotides in pH-sensitive liposomes increased the inhibition of inositol uptake at least 50-fold compared with application of free oligonucleotides in solution.

Nordidemnin potently inhibits inositol uptake in cultured astrocytes and dose-dependently augments lithium's proconvulsant effect in vivo.

It has been suggested that inositol uptake across the cell membrane is of importance for maintenance of the inositol pool involved in lithium's therapeutic effect in bipolar disease and in the lithium-pilocarpine seizure test in freely moving rats (measuring the latency of a normally subconvulsive concentration of pilocarpine to seizure induction in the additional presence of lithium). We have tested this hypothesis by: 1) demonstrating an extremely high potency of nordidemnin as an inhibitor of myo-inositol uptake in primary cultures of mouse astrocytes; and 2) determining the dose-response correlation of a nordidemnin-induced decrease in the latency before appearance of seizures in the lithium-pilocarpine test after intracerebroventricular injection of minute samples (10 microl) of virtually isotonic saline solution.

Epi-inositol and inositol depletion: two new treatment approaches in affective disorder.

Inositol is a simple polyol precursor in a second messenger system important in brain myo-insitol, the natural isomer, which has been found to be therapeutically effective in depression, panic disorder, and obsessive-compulsive disorder in double-blind controlled trials. Recently, epi-inositol, an unnatural stereoisomer of myo-inositol, was found to have effects similar to those of myo-inositol to reverse lithium-pilocarpine seizures. We measured the behavior of rats in an elevated plus maze model of anxiety after chronic treatment of 11 daily intraperitoneal injections of epi-inositol, myo-inositol, or control solution. Epi-inositol reduced anxiety levels of rats compared with controls, and its effect was stronger than that of myo- inositol. Lithium has been hypothesized to alleviate mania by reducing brain inositol levels. Inositol in brain derives from the second messenger cycle, from new synthesis, or from diet via transport across the blood brain barrier. Because the first two are inhibited by lithium, we propose that an inositol-free diet will augment lithium action in mania by enhancing restriction of inositol.

Augmentation of lithium's behavioral effect by inositol uptake inhibitors.

Lithium inhibits the enzyme inositol monophosphatase and thus obstructs the enzymatic degradation of inositol triphosphate (IP3) to inositol in the phosphate-phosphoinositide (PIP) cycle. This inhibition may result in reduced availability of the second messengers IP3 and DAG that are derivates of the PIP cycle, and this action is currently a leading hypothesis regarding lithium's therapeutic and prophylactic effect in affective disorders. Inositol is also available to the cell by uptake from the intercellular matrix, and therefore it is possible that compounds that block the uptake may have lithium-like effects. To test this hypothesis, the present study evaluates the effects of two inositol uptake inhibitors, the carbohydrate L-fucose and the cyclodepsipeptide nordidemnin, in a behavioral model of pilocarpine-induced seizures known to be enhanced by lithium. We tested the possibility that L-fucose produces lithium-like effects, or that L-fucose or nordidemnin augment lithium's behavioral effects. Results indicate that acute ICV treatment with L-fucose did not by itself have a lithium-like effect in the behavioral model, but significantly augmented lithium's effect when combined with lithium treatment. Nordidemnin treatment showed similar effects. The results suggest that when inositol monophosphatase is inhibited by lithium, further restriction of cellular inositol availability may result in an augmentation of lithium's behavioral effects. It is possible that such manipulations may be applicable in the treatment of patients with affective disorders, especially patients who are poor responders to lithium monotherapy.

Effects of inhibition of myo-inositol transport on MDCK cells under hypertonic environment.

To investigate the role of myo-inositol under hypertonic conditions, we examined the effects of inhibition of myo-inositol transport in Madin-Darby canine kidney (MDCK) cells using an analog of myo-inositol, 2-O,C-methylene-myo-inositol (MMI). We first characterized the inhibitory effects of MMI on myo-inositol transport in MDCK cells. The Na+-dependent component of [3H] myo-inositol uptake was inhibited by MMI in a concentration-dependent manner, although MMI did not inhibit the activities of the betaine transporter and system A neutral amino acid transporter. We found decreased affinity for myo-inositol in the presence of MMI, whereas the maximal velocity (Vmax) of the transporter did not change. Thus MMI behaves as a competitive inhibitor of myo-inositol transport with a relatively high inhibition constant (K(i)) value (1.6 mM). Myo-inositol content in hypertonic MDCK cells was markedly reduced in the presence of 5 mM MMI, but MMI itself did not accumulate in these cells. The hypertonic cells began to detach in the presence of MMI 3 days after increasing medium osmolality, whereas MMI did not affect the cells in isotonic medium. We also examined the effects of MMI on colony-forming efficiency of MDCK cells. MMI decreased colony-forming efficiency in a concentration-dependent manner, and addition of myo-inositol returned the efficiency to the value without MMI. Addition of betaine also increased colony-forming efficiency in the presence of MMI. These results indicate that myo-inositol plays an important role in survival and growth under hypertonic environment.

Activators of protein kinase A and of protein kinase C inhibit MDCK cell myo-inositol and betaine uptake.

Amino acid sequences of the myo-inositol and betaine cotransporters that are induced in MDCK cells by hypertonicity include consensus sequences for phosphorylation by protein kinase A and by protein kinase C. To test for the effect of activation of protein kinases A and C on the activity of those cotransporters, MDCK cells were exposed to activators of each kinase and the activity of both cotransporters was assayed. Incubation with 8-bromoadenosine 3':5'-cyclic monophosphate (8Br-cAMP) or 3-isobutyl-1-methylxanthine (IBMX), activators of protein kinase A, and incubation with an active phorbol ester or with an active diacylglycerol, activators of protein kinase C, inhibited the activity of both cotransporters by about 30%. The relative effect of the activation of protein kinase A and of protein kinase C was similar in hypertonic and isotonic cells. The effects of activators of protein kinase A and of protein kinase C were not additive. The two cotransporters behaved differently when protein kinase C activity was down-regulated by prolonged incubation with a higher concentration of phorbol 12-myristate 13-acetate. There was a doubling of activity of the myo-inositol cotransporter and no change in the activity of the betaine cotransporter in hypertonic and isotonic cells. Although the mechanisms of the effects of activation of the two kinases remain to be established, it is clear that the kinases can mediate post-translational regulation of the uptake of compatible osmolytes.

Stimulatory effect of hormonal signaling factors on (Ca(2+)-Mg2+)-ATPase activity in rat liver plasma membranes: cross talk with regucalcin.

The effect of hormonal signaling factors on (Ca(2+)-Mg2+)-ATPase activity in rat liver plasma membranes was investigated. The presence of inositol-glycan (10(-7)-10(-5) M), dibutyryl cAMP (10(-4) and 10(-3) M) or inositol 1,4,5-trisphosphate (IP3; 10(-6) and 10(-5) M) in the enzyme reaction mixture produced a significant increase in (Ca(2+)-Mg2+)-ATPase activity. These effects were completely inhibited by the presence of vanadate (10(-4) M), an inhibitor of the enzyme phosphorylation, and N-ethylmaleimide (5 x 10(-3) M), a SH group modifying reagent. Meanwhile, regucalcin, a Ca(2+)-binding protein isolated from rat liver cytosol, increased the enzyme activity by binding to the SH groups of (Ca(2+)-Mg2+)-ATPase in liver plasma membranes. The presence of regucalcin (0.25 microM) with an effective concentration completely inhibited the effect of inositol-glycan (10(-5) M) to increase (Ca(2+)-Mg2+)-ATPase activity, while the effect of dibutyryl cAMP (10(-3) M) or IP3 (10(-5) M) was not altered. The inositol-glycan effect was not modulated by the presence of dibutyryl cAMP or IP3. Now, the preincubation of the plasma membranes with regucalcin did not modify the effect of inositol-glycan on the enzyme activity, suggesting that regucalcin competes with inositol-glycan for the binding to the plasma membranes. The present results suggest that there may be a cross talk with regucalcin and hormonal signaling factors in the regulation of (Ca(2+)-Mg2+)-ATPase activity in liver plasma membranes.

Effects of high glucose concentrations on human mesangial cell proliferation.

High concentrations of glucose in vitro inhibit cell proliferation and stimulate matrix protein synthesis. These studies sought to characterize the relationship between the effects of glucose on cell proliferation and matrix synthesis and to assess the mechanism(s) responsible for these cellular effects of glucose. The initial experiments showed that high glucose levels stimulate fibronectin (FN) synthesis by human mesangial cells (HMC) but only in those cultures in which cell proliferation was inhibited by glucose. To assess whether this relationship was due to an effect of glucose on the capacity of HMC to respond to cytokines, the responses of HMC to serum or cytokines were measured in the presence of different glucose concentrations. High concentrations of glucose inhibited (3H)thymidine incorporation in response to serum and platelet-derived growth factor. Under the conditions of these experiments, transforming growth factor-beta (TGF-beta) also stimulated thymidine incorporation by HMC, and high glucose concentrations inhibited thymidine incorporation in response to TGF-beta. In contrast, high concentrations of glucose did not inhibit the stimulation of FN synthesis caused by platelet-derived growth factor, serum, or TGF-beta. The antiproliferative effects of high glucose levels were first observed after 48 h of incubation and were reversible after the withdrawal of high glucose from the media. The following evidence suggest that the effects of glucose may be mediated via protein kinase C (PKC): (1) incubation with high glucose concentrations caused an increase in HMC PKC levels; (2) PKC activation with phorbol esters inhibited HMC proliferation; and (3) depletion or inhibition of PKC stimulated HMC proliferation and prevented the antiproliferative effects of glucose. In contrast to these findings, inhibitors of protein glycosylation and myo-inositol supplementation of culture media did not prevent the antiproliferative effects of glucose. In conclusion, high glucose concentrations acutely and reversibly inhibit HMC proliferation, perhaps by a PKC-dependent mechanism. Because PKC can also stimulate FN synthesis, glucose-induced changes in PKC may explain the relationship between the effects of high glucose concentrations on cell proliferation and FN synthesis.

High-dose peripheral inositol raises brain inositol levels and reverses behavioral effects of inositol depletion by lithium.

Lithium (Li) reduces brain inositol levels. Berridge has suggested that this effect is related to Li's mechanism of action. It had previously been shown that pilocarpine causes a limbic seizure syndrome in lithium treated rats, and that these lithium-pilocarpine seizures are reversible by intracerebroventricular inositol administration to rats. We now show that although inositol passes the blood-brain barrier poorly, large doses of intraperitoneal (IP) inositol can also reverse Li-pilocarpine seizures. Using gas chromatography, IP inositol can raise brain inositol levels. Demonstration that inositol enters brain after peripheral administration provides a basis for possible pharmacological intervention in psychiatric disorders at the level of second messengers linked to the phosphatidylinositol cycle.

The uptake of 3H-labelled monodeoxyfluoro-myo-inositols into thymocytes and their incorporation into phospholipid in permeabilized cells.

Monodeoxyfluoro-myo-inositols were applied to electropermeabilized and intact thymocyte preparations to study their metabolism and uptake in order to investigate their suitability as potential inhibitors of phosphoinositide-mediated cellular responses. Only three of the monodeoxyfluoro-myo-inositols were incorporated into the phospholipids of thymocytes: 1D-3-deoxy-3-fluoro-myo-inositol, 5-deoxy-5-fluoro-myo-inositol and 1D-6-deoxy-6-fluoro-myo-inositol, all of which were weaker substrates for phosphatidylinositol synthase than was myo-inositol. The 3-, 5- and 6-fluoro analogues also behaved as competitive inhibitors, with K1 values of 350 +/- 5 microM, 350 +/- 5 microM and 2.9 +/- 2 mM respectively, compared with a Km for myo-inositol of 31 +/- 4 microM. When incubated with electropermeabilized thymocyte preparations, these three analogues of myo-inositol all formed phospholipids with chromatographic properties which corresponded to those of substituted phosphatidylinositol and phosphatidylinositol monophosphate. The uptake of myo-inositol and of the monodeoxyfluoro-myo-inositols into intact thymocytes was studied by a dual-label technique. All the monodeoxyfluoro-myo-inositols were taken up to some extent, but only 2-deoxy-2-fluoro-myo-inositol and 1D-3-deoxy-3-fluoro-myo-inositol were actively concentrated. The monodeoxyfluoro-myo-inositols were also assayed for their ability to inhibit the uptake of myo-inositol into cells. Both 2-deoxy-2-fluoro-myo-inositol and 1D-3-deoxy-3-fluoro-myo-inositol were effective inhibitors of myo-inositol uptake. Furthermore, 1D-1-deoxy-1-fluoro-myo-inositol, which was not taken up actively, was an effective inhibitor of myo-inositol uptake. The three effective inhibitors all showed Ki values of approximately 150 microM, close to the apparent Km for inositol uptake of 180 microM, and the 4-, 5- and 6-fluoro analogues had Ki values in excess of 10 mM.

Insulin mediators are the signal transduction system responsible for insulin's actions on human placental steroidogenesis.

To test the hypothesis that insulin mediators serve as the signal transduction system for insulin's steroidogenic actions in human placental cytotrophoblasts, we examined the effects of two inositolglycan insulin mediators, the insulin pH 2.0 chiro-inositol mediator (IM-pH 2.0) and the insulin pH 1.3 myo-inositol mediator (IM-pH 1.3), on cytotrophoblastic steroidogenesis. When human cytotrophoblasts were incubated in medium supplemented with androstenedione for 24 h, treatment with IM-pH 2.0 or IM-pH 1.3 suppressed aromatase activity by 15% (P less than 0.05) and 49% (P less than 0.05), respectively, compared to insulin, which suppressed aromatase activity by 21% (P less than 0.05). When cytotrophoblasts were incubated in medium supplemented with pregnenolone for 24 h, treatment with IM-pH 2.0 or IM-pH 1.3 stimulated 3 beta-hydroxysteroid dehydrogenase (3 beta HSD) activity by 145% (P less than 0.05) and 168% (P less than 0.05), respectively, compared to insulin, which stimulated 3 beta HSD activity by 63% (P less than 0.05). Suppression of aromatase activity and stimulation of 3 beta HSD activity by inositolglycan mediators were both concentration dependent. Moreover, preincubation of cytotrophoblasts with the antiinositolglycan antibody alpha IGP completely abolished insulin's ability to either inhibit aromatase or stimulate 3 beta HSD activity. These results indicate that insulin mediators mimic insulin's effects on cytotrophoblastic aromatase and 3 beta HSD activities and suggest that inositolglycan mediators are the signal transduction mechanism responsible for insulin's regulation of human placental steroid hormone biosynthesis.

Myo-inositol and betaine transporters regulated by tonicity are basolateral in MDCK cells.

Myo-inositol and glycinebetaine are compatible osmolytes accumulated in the renal medulla and in MDCK cells cultured in hypertonic media. Both osmolytes are taken up by MDCK cells on Na-coupled transporters. The maximal velocity (Vmax) of both cotransporters is increased by culture in hypertonic medium. When hypertonic MDCK cells are shifted to isotonic medium there is a large transient efflux of osmolytes. To determine the polarity of the cotransporters and the transient efflux, we grew MDCK cells on a porous support to assay transport separately at their apical and basolateral surfaces. In hypertonic cells, basolateral uptake of both osmolytes was 1) more than 10-fold apical uptake, 2) greater than 96% Na dependent, 3) 25- (myo-inositol) and 16-fold (glycinebetaine) uptake in isotonic cells, reaching a maximum 24 h after the switch to hypertonic medium. When medium osmolarity was decreased from hypertonic to isotonic, myo-inositol uptake reversed to the isotonic level within 1 day; glycinebetaine uptake decreased more slowly. When medium osmolarity was decreased from hypertonic to isotonic, there was a large transient increase in basolateral efflux of both osmolytes.