Direct measurement of intracellular pH in identified glial cells and neurones of the leech central nervous system

Neutral carrier pH-sensitive double-barrelled microelectrodes were used to investigate intracellular pH (pHi) in leech neuropile glial cells and in Retzius neurones. The mean pHi of the glial cells was 6.87 +/- 0.13 (+/- SD, n = 27) in HEPES-buffered saline (pHo 7.4) and 7.18 +/- 0.19 (n = 13) in solutions buffered with 2% CO2- 11 mM HCO3-. The distribution of H+ ions in both the glia and neurones was found not to be in electrochemical equilibrium. To investigate pHi regulation, the pHi was decreased by exposure to CO2 or by adding and then removing NH4Cl. Acidification by any method was followed by a recovery to normal pHi values within minutes. The pHi recovery from acidification in neuropile glial cells in HEPES-buffered saline and CO2-HCO3- buffered saline was, however, blocked by removing external Na. In HCO3(-)-free solutions the diuretic amiloride (2 mM) reduced the rate of pHi recovery. In the presence of HCO3-, the rate of acid efflux was stimulated; the stilbene 4-acetamido-4'-isothiocyanatostilbene-2,3'-disulfonic acid (SITS; 0.5 mM) slowed pHi recovery. In HEPES buffered and CO2-HCO3- buffered solutions pHi regulation in neurones was inhibited by removing external Na. In HCO3(-)-free solutions amiloride reduced the rate of pHi recovery considerably. In the presence of HCO3-, SITS or amiloride slowed but did not completely block pHi recovery. We conclude that leech glial cells and neurones have two mechanisms of pHi regulation, one being Na+-H+ exchange and the other Na+ and HCO3- dependent.     

Modulation of K+-channels in p-neurones of the leech CNS by phosphorylation

Two types of potassium channels of identified (p-) neurones of the leech (Hirudo medicinalis) were investigated by using the patch-clamp technique. The openstate probability of these channels in cell-attached patches can be reduced by addition of 5-hydroxytryptamine to the bath solution. After excising the patches the application of alkaline phosphatase to the cytosolic face of the patch increases the open probability. The 5-HT_1A-receptor agonist buspirone mimics the effect of 5-HT. Our experiments show that the effect of 5-HT might be due to a channel phosphorylation via a 5-HT_1A-receptor subtype.     

ATP-inhibited and Ca2+-dependent K+ channels in the soma membrane of cultured leech Retzius neurons

The properties of one ATP-inhibited and one Ca²⁺-dependent K⁺ channel were investigated by the patch-clamp technique in the soma membrane of leech Retzius neurons in primary culture. Both channels rectify at negative potentials. The ATP-inhibited K⁺ channel with a mean conductance of 112 pS is reversibly blocked by ATP (K _i = 100 m), TEA (K _i =0.8 mm) and 10 mm Ba²⁺ and irreversibly blocked by 10 nm glibenclamide and 10 m tolbutamide. It is Ca²⁺ and voltage independent. Its open state probability (P _o) decreases significantly when the pH at the cytoplasmic face of inside-out patches is altered from physiological to acid pH values. The Ca²⁺-dependent K⁺ channel with a mean conductance of 114 pS shows a bell-shaped Ca²⁺ dependence of P _o with a maximum at pCa 7–8 at the cytoplasmic face of the membrane. The P _o is voltage independent at the physiologically relevant V range. Ba²⁺ (10 mm) reduces the single channel amplitude by around 25% (ATP, TEA, glibenclamide, tolbutamide, and Ba²⁺ were applied to the cytoplasmic face of the membrane).We conclude that the ATP-dependent K⁺ channel may play a role in maintaining the membrane potential constant—independently from the energy state of the cell. The Ca²⁺-dependent K⁺ channel may play a role in generating the resting membrane potential of leech Retzius neurons as it shows maximum activity at the physiological intracellular Ca²⁺ concentration.This study was supported by the Deutsche Forschungsgemeinschaft (W.-R. Schlue) and by a fellowship of the Konrad-Adenauer-Stiftung (G. Frey). We thank Dr. Draeger (Hoechst AG) for the gift of glibenclamide. The data are part of a future Ph.D. thesis of G. Frey.     

Apparent intracellular Mg2+ buffering in neurons of the leech Hirudo medicinalis

The apparent intracellular Mg2+ buffering, or muffling (sum of processes that damp changes in the free intracellular Mg2+ concentration, [Mg2+](i), e.g., buffering, extrusion, and sequestration), was investigated in Retzius neurons of the leech Hirudo medicinalis by iontophoretic injection of H+, OH-, or Mg2+. Simultaneously, changes in intracellular pH and the intracellular Mg2+, Na+, or K+ concentration were recorded with triple-barreled ion-selective microelectrodes. Cell volume changes were monitored measuring the tetramethylammonium (TMA) concentration in TMA-loaded neurons. Control measurements were carried out in electrolyte droplets (diameter 100-200 microm) placed on a silver wire under paraffin oil. Droplets with or without ATP, the presumed major intracellular Mg2+ buffer, were used to quantify the pH dependence of Mg2+ buffering and to determine the transport index of Mg2+ during iontophoretic injection. The observed pH dependence of [Mg2+](i) corresponded to what would be expected from Mg2+ buffering through ATP. The quantity of Mg2+ muffling, however, was considerably larger than what would be expected if ATP were the sole Mg2+ buffer. From the decrease in Mg2+ muffling in the nominal absence of extracellular Na+ it was estimated that almost 50% of the ATP-independent muffling is due to the action of Na+/Mg2+ antiport.     

Shrinkage-induced activation of Na+/H+ exchange in rat renal mesangial cells

Using thepH-sensitive dye2',7'-bis(2-carboxyethyl)-5(6)-carboxyfluorescein (BCECF),we examined the effect of hyperosmolar solutions, which presumablycaused cell shrinkage, on intracellular pH(pH_i) regulation in mesangialcells (single cells or populations) cultured from the rat kidney. Thecalibration of BCECF is identical in shrunken and unshrunken mesangialcells if the extracellular K⁺concentration ([K⁺])is adjusted to match the predicted intracellular[K⁺]. ForpH_i values between ~6.7 and~7.4, the intrinsic buffering power in shrunken cells (600 mosmol/kgH₂O) is threefold larger than in unshrunken cells (~300mosmol/kgH₂O). In the nominalabsence ofCO₂/HCO₃,exposing cell populations to a HEPES-buffered solution supplementedwith ~300 mM mannitol (600 mosmol/kgH₂O) causes steady-statepH_i to increase by ~0.4. The pH_i increase is due to activationofNa⁺/H⁺exchange because, in single cells, it is blocked in the absence ofexternal Na⁺ or in the presence of50 µM ethylisopropylamiloride (EIPA). Preincubating cells in aCl-free solution for atleast 14 min inhibits the shrinkage-induced pH_i increase by 80%. Wecalculated the pH_i dependence oftheNa⁺/H⁺exchange rate in cell populations under normosmolar and hyperosmolar conditions by summing 1) thepH_i dependence of the totalacid-extrusion rate and 2) thepH_i dependence of theEIPA-insensitive acid-loading rate. Shrinkage alkali shifts thepH_i dependence ofNa⁺/H⁺exchange by ~0.7 pH units.     

NTP,the photoproduct of nifedipine,activates caffeine-sensitive ion channels in leech neurons

Leech P neurons possess caffeine-sensitive ion channels in intracellular Ca(2+) stores and in the plasma membrane. The following results indicate that these channels are also activated by 2,6-dimethyl-4-(2-nitrosophenyl)-3,5-pyridinedicarboxylic acid dimethyl ester (NTP), the photoproduct of the L-type Ca(2+) channel-blocker nifedipine: (1) Just like caffeine, NTP evoked Ca(2+) influx and intracellular Ca(2+) release, as well as the influx of various other divalent cations and that of Na(+). (2) In the presence of high NTP or caffeine concentrations the plasma membrane channels close, suggesting desensitization of the channel-activating mechanism. (3) Depending on the concentration, NTP and caffeine induce cross-desensitization or act additively. (4) NTP was effective in the same neurons as caffeine (P, N, Leydig, 101), and it was ineffective in neurons in which caffeine was also ineffective (AP, T, L, 8, AE). (5) In Retzius neurons, NTP and caffeine evoked intracellular Ca(2+) release but no Ca(2+) influx. Despite these parallels, the effects of NTP and caffeine were not identical, which may be due to differences in the mechanisms of channel activation or desensitization and/or to substance-specific side effects. The caffeine-sensitive ion channels were activated by NTP concentrations > or =10 microM, which is almost three orders of magnitude smaller than the threshold concentration of caffeine.     

Activation and desensitization of the caffeine-sensitive cation channels and calcium stores have no persistent effect on the electrophysiological properties of leech P neurones

In leech P neurones caffeine activates unselective ion channels in the plasma membrane and induces intracellular Ca2+ release (Schoppe, J., Hochstrate, P., Schlue, W.-R., 1997. Caffeine mediates cation influx and intracellular Ca2+ release in leech P neurones. Cell Calcium 22, 385-397). These effects are prominent only upon the first caffeine exposure, while subsequent applications are largely ineffective; i.e. both plasma membrane channels and intracellular Ca2+ release mechanism desensitize irreversibly. In order to examine whether this desensitization is paralleled by irreversible changes in the electrophysiological parameters of the cells, we investigated the action of caffeine on changes in membrane potential and the cytosolic free Ca2+ concentration, which were induced by varying the ionic composition of the extracellular fluid or by application of 5-hydroxytryptamine. Neither the resting values nor any of the experimentally induced shifts in membrane potential or cytosolic Ca2+ concentration were affected by caffeine, which suggests strongly that activation and/or desensitization of the caffeine-sensitive ion channels and Ca2+ stores have no long-lasting effect on the relevant electrochemical gradients, membrane conductances, or transport mechanisms.     

Increased HEV Seroprevalence in Patients with Autoimmune Hepatitis

Background

Hepatitis E virus (HEV) infection takes a clinically silent, self-limited course in the far majority of cases. Chronic hepatitis E has been reported in some cohorts of immunocompromised individuals. The role of HEV infections in patients with autoimmune hepatitis (AIH) is unknown.

Methods

969 individuals were tested for anti-HEV antibodies (MP-diagnostics) including 208 patients with AIH, 537 healthy controls, 114 patients with another autoimmune disease, rheumatoid arthritis (RA), and 109 patients with chronic HCV- or HBV-infection (HBV/HCV). Patients with AIH, RA and HBV/HCV were tested for HEV RNA. HEV-specific proliferative T cell responses were investigated using CFSE staining and in vitro stimulation of PBMC with overlapping HEV peptides.

Results

HEV-antibodies tested more frequently positive in patients with AIH (n = 16; 7.7%) than in healthy controls (n = 11; 2.0%; p = 0.0002), patients with RA (n = 4; 3.5%; p = 0.13) or patients with HBV/HCV infection (n = 2; 2.8%; p = 0.03). HEV-specific T cell responses could be detected in all anti-HEV-positive AIH patients. One AIH patient receiving immunosuppression with cyclosporin and prednisolone and elevated ALT levels had acute hepatitis E but HEV viremia resolved after reducing immunosuppressive medication. None of the RA or HBV/HCV patients tested HEV RNA positive.

Conclusions

Patients with autoimmune hepatitis but not RA or HBV/HCV patients are more likely to test anti-HEV positive. HEV infection should been ruled out before the diagnosis of AIH is made. Testing for HEV RNA is also recommended in AIH patients not responding to immunosuppressive therapy.     

Propriorezeptive Kontrolle der Erregungsmuster eines Schaben-Motoneurons

Bei der Schabe Periplaneta americana L. wurde untersucht, welche Korrelationen zwischen propriorezeptiven Afferenzen und der Efferenz von Bein-Motoneuronen bestehen. Die Afferenzen wurden durch mechanische Reizung einzelner oder mehrerer Sensillen eines Borstenfeldes am Coxa-Trochanter-Gelenk ausgelöst. Die Efferenzen wurden an einem einzelnen, identifizierbaren Motoneuron des Mesothorax-Beinnerv 5 beobachtet. Die Aktivität dieses Motoneurons am unbeeinflußten Präparat wird Grundaktivität genannt. Ihre Frequenz liegt zwischen ≥ 0 und ≤ 15 Spikes/sec (Tab. 1). Nach Abtrennen der Coxa (distal vom Ableitort) oder der Abdominalkonnektive sinkt die Grundaktivität des Motoneurons ab, steigt dann im Verlauf von einigen Minuten wieder an; ihre Frequenz erreicht aber nicht so hohe Werte wie vorher (Abb. 2). Wenn die Unterschlund-Prothoraxkonnektive durchtrennt werden, nimmt die Spike-Aktivität im Beinnerv zu und bleibt über Stunden erhöht. Bei Reizung von Borsten am Coxa-Trochanter-Gelenk treten im Beinnerv 5 sensorische Spikes auf. Bei einmaligem Abbiegen einer einzelnen Borste schnellt ihre Frequenz maximal auf 600—700 Spikes/sec und klingt dann sehr rasch ab. Bei vibratorischer Reizung einzelner Borsten ist die Spikefrequenz im Bereich zwischen 100 und 500 (gelegentlich 700) Hz annähernd gleich der Reizfrequenz (Abb. 3, 4). In dem Beinnerv-Motoneuron steigt kurz (etwa 35—90 msec) nach Beginn eines Vibrationsreizes auf eine einzelne Borste die Spike-Aktivität sprungartig auf eine höhere, konstantbleibende Frequenz (je nach Präparat 24—59 Spikes/sec) an (Abb. 6). Nach Reizende fällt sie sogleich (≤ 500 msec) auf den Wert der Grundaktivität oder tiefer ab. Dasselbe gilt bei gleichzeitiger Reizung mehrerer Borsten (Tab. 2). Die ganglionäre Überleitungszeit der Borstenfaser-Erregungen auf das Beinnerv-Motoneuron beträgt im Minimum 2—3 msec. Die Grundaktivität der Bein-Motoneurone beruht auf Erregungszuflüssen von sensorischen Fasern im Beinnerv, von Interneuronen des Bauchmarks (damit wohl indirekt von Cercal-Sensillen) und wahrscheinlich von anderen, unbekannten Neuronen, die in der Peripherie oder im Zentrum liegen. Die Afferenzen von den Borstenfeldern am Coxa-Trochanter-Gelenk enthalten Angaben über die Bewegungen und wahrscheinlich auch über die jeweilige Stellung dieses Gelenks. Die Erregungen von den Coxa-Trochanter-Borstenfeldern werden im zugehörigen Thoraxganglion mit einer Verzögerungszeit von 1—2 msec monosynaptisch auf das Bein-Motoneuron übertragen.