Release of osmolytes induced by phagocytosis and hormones in rat liver.

Betaine, taurine, and inositol participate as osmolytes in liver cell volume homeostasis and interfere with cell function. In this study we investigated whether osmolytes are also released from the intact liver independent of osmolarity changes. In the perfused rat liver, phagocytosis of carbon particles led to a four- to fivefold stimulation of taurine efflux into the effluent perfusate above basal release rates. This taurine release was inhibited by 70-80% by the anion exchange inhibitor DIDS or by pretreatment of the rats with gadolinium chloride. Administration of vasopressin, cAMP, extracellular ATP, and glucagon also increased release of betaine and/or taurine, whereas insulin, extracellular UTP, and adenosine were without effect. In isolated liver cells, it was shown that parenchymal cells and sinusoidal endothelial cells, but not Kupffer cells and hepatic stellate cells, release osmolytes upon hormone stimulation. This may be caused by a lack of hormone receptor expression in these cells, because single-cell fluorescence measurements revealed an increase of intracellular calcium concentration in response to vasopressin and glucagon in parenchymal cells and sinusoidal endothelial cells but not in Kupffer cells and hepatic stellate cells. The data show that Kupffer cells release osmolytes during phagocytosis via DIDS-sensitive anion channels. This mechanism may be used to compensate for the increase in cell volume induced by the ingestion of phagocytosable material. The physiological significance of hormone-induced osmolyte release remains to be evaluated.

Expression of glutamine synthetase in macrophages.

We studied the expression of glutamine synthetase in liver macrophages (Kupffer cells, KCs) in situ and in culture. Glutamine synthetase was detectable at the mRNA and protein level in freshly isolated and short-term-cultured rat liver macrophages. Enzyme activity and protein content were about 9% of that in liver parenchymal cells. In contrast, glutamine synthetase mRNA levels in liver macrophages apparently exceeded those in parenchymal liver cells (PCs). By use of confocal laser scanning microscopy and specific macrophage markers, immunoreactive glutamine synthetase was localized to macrophages in normal rat liver and normal human liver in situ. All liver macrophages stained positive for glutamine synthetase. In addition, macrophages in rat pancreas contained immunoreactive glutamine synthetase, whereas glutamine synthetase was not detectable at the mRNA and protein level in blood monocytes and RAW 264.7 mouse macrophages. No significant amounts of glutamine synthetase were found in isolated rat liver sinusoidal endothelial cells (SECs). The data suggest a constitutive expression of glutamine synthetase not only, as previously believed, in perivenous liver parenchymal cells but also in resident liver macrophages.

Involvement of CD95 (Apo-1/Fas) ligand expressed by rat Kupffer cells in hepatic immunoregulation.

BACKGROUND & AIMS: CD95 (Apo-1/Fas) ligand suppresses inflammatory responses in immune-privileged organs. In this study, modulation of the hepatic CD95 receptor/ligand system by interferon gamma and cyclosporin A was investigated. METHODS: CD95 receptor and ligand expression were measured at the messenger RNA level by using quantitative reverse-transcription polymerase chain reaction and immunocytochemistry in primary cultures of rat Kupffer cells, hepatocytes, and T lymphocytes. Soluble CD95 in culture supernatants was detected by enzyme-linked immunosorbent assay and apoptosis by the TUNEL method. RESULTS: Interferon gamma treatment led to an increase in CD95 ligand messenger RNA levels in Kupffer cells followed by an overexpression of the soluble CD95 receptor. Supernatants derived from 24-hour but not from 48-hour interferon gamma-treated Kupffer cells killed lymphocytes by a CD95-dependent mechanism. Cyclosporin A inhibited CD95 ligand expression in Kupffer cells and lymphocyte killing. In liver parenchymal cells, interferon gamma increased messenger RNA levels of the transmembrane CD95 isoform and sensitivity of these cells toward CD95-mediated apoptosis. CONCLUSIONS: The expression pattern of CD95 receptor and ligand in response to interferon gamma points to a coordinated interplay between Kupffer cells, hepatocytes, and T lymphocytes in which Kupffer cells may regulate programmed cell death of T lymphocytes and hepatocytes.

Organic osmolyte transport in quiescent and activated rat hepatic stellate cells (Ito cells).

Activation of hepatic stellate cells (HSCs) results in multiple alterations of cell function, but nothing is known about organic osmolytes in these cells. Organic osmolyte transport and transporter messenger RNA (mRNA) expression was studied in quiescent rat HSCs and after their transformation into alpha1-smooth muscle actin-positive myofibroblastlike cells. Quiescent stellate cells expressed in an osmosensitive manner the mRNA levels of the transporters for taurine (TAUT) and myoinositol (SMIT), whereas that for betaine was not detectable. However, these cells showed osmosensitive uptake not only of taurine and myoinositol but also of betaine. Osmosensitive betaine uptake was mediated by amino acid transport system A. After transformation into myofibroblasts, taurine and myoinositol uptake increased 5.5-fold and 4.5-fold, respectively, together with the respective transporter mRNA levels. Betaine uptake increased twofold because of osmosensitive induction of BGT1 expression. In both quiescent and activated HSCs, hypoosmotic cell swelling induced a rapid and 4, 4'-diisothiocyanatostilbene-2,2'-disulphonic acid-sensitive osmolyte efflux. In quiescent HSCs, hyperosmotic exposure increased the messenger RNA (mRNA) level of cyclooxygenase-2, which was counteracted by taurine but not by betaine or myoinositol. The study identifies taurine, myoinositol, and betaine as osmolytes in HSCs. Transformation of HSCs is accompanied by enhanced osmolyte transport activity and induction of the BGT1 transporter, which may be another activation marker of HSCs.

Activation of rat hepatic stellate cells in culture is associated with increased sensitivity to endothelin 1.

The effect of endothelin (ET) 1 on intracellular Ca2+ transients in cultured rat hepatic stellate cells (HSCs) during transformation was studied by use of single-cell fluorescence. Regardless of the duration of HSC culture, ET-1 caused a BQ-123-sensitive but IRL-1038-insensitive elevation of [Ca2+]i, indicating the involvement of ETA but not ETB receptors. HSCs in early culture ("quiescent HSCs") were mildly responsive to ET-1: the ET-1 concentration required to obtain a [Ca2+]i transient in 50% of the cells (RC50) was 7 nmol/L, and all cells responded to ET-1 concentrations above 40 nmol/L. With culture time, -smooth muscle actin (-SMA) expression increased, as did the ET-1 sensitivity of cells, resulting in a shift of the RC50 value from 7 nmol/L to 13 pmol/L within 8 days. Independent of the duration of culture, ET-1 sensitivity was higher in -SMA-expressing cells. On the other hand, sensitivity of HSCs to produce a [Ca2+]i response to extracellular uridin 5'-triphosphate (UTP) or phenylephrine did not change during the activation process. There was no difference between quiescent and activated HSCs with respect to the sharing of intracellular Ca2+ stores, which could be mobilized by ET-1, UTP, and phenylephrine, respectively. The data suggest three conclusions. (1) A marked increase in ET-1 sensitivity of HSCs during the activation process suggests a potentiation of autocrine/paracrine stimulation. (2) HSCs are susceptible to -adrenergic and purinergic stimulation, but sensitivity to phenylephrine and UTP is not affected during the transformation process. (3) The ET-1-mobilizable Ca2+ store is contained in and is smaller than the Ca2+ pool, which is mobilized by phenylephrine or UTP.

De novo expression of glutamine synthetase during transformation of hepatic stellate cells into myofibroblast-like cells.

The expression of glutamine synthetase (GS) was studied in cultured quiescent hepatic stellate cells (HSC) and during their transformation into myofibroblast-like cells. GS mRNA was detectable in quiescent HSC (1-day culture); however, the enzyme protein was not expressed, as assessed by Western blot analysis, immunocytochemistry and the absence of detectable enzyme activity. Similar findings were obtained after 2 days of culture; in addition, the mRNA levels had dropped by about 70%, but they increased again thereafter during the process of HSC transformation in culture, as indicated by the expression of alpha-smooth-muscle actin. In parallel with the accumulation of alpha-smooth-muscle actin, GS was expressed, as shown by Western blot analysis and immunocytochemistry, and enzyme activity increased from undetectable levels in quiescent cells to 0.13+/-0.01 micromol/h per mg of cell protein within 7-14 days. This value compares with GS activity in liver parenchymal cells of 0.57+/-0.03 micromol/h per mg of cell protein. The findings suggest that activation of HSC results in the de novo expression of GS protein and activity, and this may serve as another marker of HSC transformation.

Activation of mitogen-activated protein kinases and IL-6 release in response to lipopolysaccharides in Kupffer cells is modulated by anisoosmolarity.

BACKGROUND/AIMS: The influence of anisoosmolarity on the activation of the extracellular signal-regulated kinases-1 and -2 and on interleukin-6 release was studied in lipopolysaccharide-stimulated rat liver macrophages. METHODS: Experiments were performed with rat liver macrophages. Activation of the extracellular signal-regulated kinases was determined by kinase shift assay and immune complex kinase assay. Interleukin-6 mRNA was measured by Northern blot analysis and interleukin-6 production by enzyme-linked immunosorbent assay. RESULTS: Lipopolysaccharide-induced activation of the extracellular signal-regulated kinases-1 and -2 was enhanced in hypoosmotic media (205 mosm/l) and diminished by hyperosmotic (405 mosm/l) exposure when compared to normoosmotic (305 mosm/l) conditions. These effects were paralleled by changes in lipopolysaccharide-stimulated interleukin-6 mRNA expression, when determined after 4 h and interleukin-6 release after 18 h. The mitogen-activated protein kinase-kinase inhibitor PD 098059 abolished phosphorylation of the extracellular signal-regulated kinases-1 and -2 in response to lipopolysaccharide, irrespective of the medium osmolarity, and diminished lipopolysaccharide-induced interleukin-6 mRNA expression and interleukin-6 production under normo- and hypoosmotic conditions by about 50%; it also resulted under hyperosmotic conditions in an about 80% inhibition. SB 203580, a specific inhibitor of p38 largely abolished interleukin-6 mRNA expression and interleukin-6 production, irrespective of medium osmolarity, whereas phosphorylation of the extracellular signal-regulated kinases was not affected. CONCLUSIONS: The data indicate a modulation of lipopolysaccharide-induced interleukin-6 production by ambient osmolarity and an involvement of both p38 and the extracellular signal-regulated kinases-1 and -2 in the stimulation of interleukin-6 production by lipopolysaccharide.

Compatible organic osmolytes in rat liver sinusoidal endothelial cells.

Compatible organic osmolytes, such as betaine and taurine are involved in the regulation of Kupffer cell (KC) function, but nothing is known about osmolytes in liver endothelial cells. This was investigated here by studying the effect of aniso-osmotic exposure of rat liver sinusoidal endothelial cells (SEC) on osmolyte transport and the messenger RNA (mRNA) levels for the transport systems for betaine (BGT1), taurine (TAUT), and myo-inositol (SMIT). Compared with normo-osmotic exposure (305 mosmol/L), hyperosmotic exposure (405 mosmol/L) of SEC led to an increase in the mRNA levels for these transport systems and simultaneously to a stimulation of betaine, taurine, and myo-inositol uptake, which led to an increase of cell volume. Conversely, hypo-osmotic exposure decreased osmolyte uptake. When hyperosmotically pre-exposed SEC were loaded with betaine, taurine, or myoinositol, hypo-osmotic stress stimulated the efflux of these osmolytes from the cells. Studies on osmolyte tissue levels revealed that taurine was an important compatible organic osmolyte under normo-osmotic conditions and predominantly released following hypo-osmotic stress. Conversely, following hyperosmotic exposure, the increase in cellular betaine and myo-inositol exceeded that of taurine. In lipopolysaccharide (LPS)-treated SEC, hyperosmotic exposure markedly raised the mRNA levels for cyclo-oxygenase-2 (COX-2), but not for inducible nitric oxide synthase (iNOS). The increase of COX-2 mRNA levels was counteracted by betaine and taurine and, to a lesser extent, by myo-inositol. The findings indicate that SEC use taurine, betaine, and myo-inositol as compatible organic osmolytes.