The whole-cell and perforated patch configurations of the patch-clamp technique were used to characterise the volume-sensitive anion channel in rat pancreatic β-cells. The channel showed high permeability (P ) relative to Cl- to extracellular monovalent organic anions (PSCN/PCll = 1.73, Pacetate/PCll = 0.39, Plactate/PCll = 0.38, Pacetoacetate/PCll = 0.32, Pglutamate/PCll = 0.28) but was less permeable to the divalent anion malate (Pmalate/PCll = 0.14). Channel activity was inhibited by a number of putative anion channel inhibitors, including extracellular ATP (10 mM), 1,9-dideoxyforskolin (100 µM) and 4-OH tamoxifen (10 µM). Inclusion of the catalytic subunit of protein kinase A in the pipette solution did not activate the volume-sensitive anion channel in non-swollen cells. Furthermore, addition of 8-bromoadenosine 3',5'-cyclic monophosphate (8-BrcAMP) or forskolin failed to activate the channel in intact cells under perforated patch conditions. Addition of phorbol 12,13-dibutyrate (200 nM), either before or after cell swelling, also failed to affect channel activation. Our findings do not support the suggestion that the volume-sensitive anion channel in pancreatic β-cells can be activated by protein kinase A. Furthermore, the β-cell channel does not appear to be subject to regulation via protein kinase C. Experimental Physiology (2001) 86.2, 145-150.

A novel isolated Langendorff perfused rabbit heart preparation with intact dual autonomic innervation is described. This preparation allows the study of the effects of direct sympathetic and vagus nerve stimulation on the physiology of the whole heart. These hearts (n = 10) had baseline heart rates of 146 ± 2 beats min-1 which could be increased to 240 ±11 beats min-1 by sympathetic stimulation (15 Hz) and decreased to 74 ± 11 beats min-1 by stimulation of the vagus nerve (right vagus, 7 Hz). This model has the advantage of isolated preparations, with the absence of influence from circulating hormones and haemodynamic reflexes, and also that of in vivo preparations where direct nerve stimulation is possible without the need to use pharmacological agents. Data are presented characterising the preparation with respect to the effects of autonomic nerve stimulation on intrinsic heart rate and atrioventricular conduction at different stimulation frequencies. We show that stimulation of the right and left vagus nerve have differential effects on heart rate and atrioventricular conduction. Experimental Physiology (2001) 86.3, 319-329.

The actions of arachidonic acid (AA) were investigated in guinea-pig isolated ventricular myocytes. Exposure of myocytes to 10 µM AA reduced the amplitude of contractions and calcium transients accompanying action potentials at a frequency of 1 Hz. AA (10 µM) also reduced the amplitude of calcium currents recorded under voltage-clamp conditions. The suppression of contraction by AA was not prevented by either 10 µM trihydroindomethicin (to inhibit cyclo-oxygenase) or 10 µM ETYA (5,8,11,14-eicosatetraynoic acid, to inhibit AA metabolising enzymes), showing that the actions of AA appeared not to be mediated by these metabolites. The reduction of contraction by 10 µM AA was also not prevented by the protein kinase C inhibitor, Ro31-8220 (1 µM), showing that this pathway appeared not to be required for the observed effect. Direct effects of AA may be involved. A further action of 10 µM AA was to suppress spontaneous electrical activity induced by either the β-adrenergic agonist isoprenaline or the Na+ pump inhibitor, ouabain. This effect of AA on spontaneous activity might be associated with the observed reduction of calcium entry through L-type calcium channels, although additional effects of AA on calcium release from the sarcoplasmic reticulum might also be involved. Experimental Physiology (2001) 86.4, 437-449.

Contents Page

Introduction 1

Cardiac sympathetic imbalance and arrhythmia 2

Nitric oxide-cGMP pathway and the cholinergic modulation of cardiac excitability 3

Nitric oxide-cGMP pathway and the sympathetic modulation of cardiac excitability 7

Functional significance of nitric oxide in the autonomic regulation of cardiac excitability 9

Summary 9

References 10

Experimental Physiology (2001) 86.1, 1-12.

In the vagally intact anaesthetised dog, we have investigated the role of nitric oxide (NO) on a normal sinus arrhythmia using an inhibitor of neuronally released NO, 1-(2-trifluoromethylphenyl) imidazole (TRIM). The mean and S.D. of the R-R interval was used to describe mean heart rate and heart rate variability, respectively. TRIM (0.8 mg I.C.) injected into the sinus node artery increased the mean heart rate slightly but reduced heart rate variability 3-fold from a control of 790 ± 124 ms (mean ± S.D.; n = 5) to 666 ± 36 ms (P < 0.01 Student's paired t test, n = 5). These results suggest that neuronally released NO may have a vagal facilitatory role in the maintenance of sinus arrhythmia in the normal heart. Experimental Physiology (2001) 86.5, 539-541.

Voltage-dependent calcium currents in vertebrate (Scyliorhinus canicula), protochordate (Branchiostoma lanceolatum), and invertebrate (Alloteuthis subulata) skeletal and striated muscle were examined under whole-cell voltage clamp. Nifedipine (10 µM) suppressed and cobalt (5 mM) blocked striated/skeletal muscle calcium currents in all of the animals examined, confirming that they are of the L-type class. Calciseptine, a specific blocker of vertebrate cardiac muscle and neuronal L-type calcium currents, was applied (0.2 µM) under whole-cell voltage clamp. Protochordate and invertebrate striated muscle L-type calcium currents were suppressed while up to 4 µM calciseptine had no effect on dogfish skeletal muscle L-type calcium currents. Our results demonstrate the presence of at least two sub-types of L-type calcium current in these different animals, which may be distinguished by their calciseptine sensitivity. We conclude that the invertebrate and protochordate L-type current sub-type that we have examined has properties in common with vertebrate 'cardiac' and 'neuronal' current sub-types, but not the skeletal muscle sub-type of the L-type channel. Experimental Physiology (2001) 86.6, 689-694.

Effects of prostaglandin E2 (PGE2) on exocytosis of mucin were studied in mucous cells isolated from guinea-pig antrum using video-microscopy. Stimulation with PGE2 elicited a sustained increase in the frequency of exocytotic events in a dose-dependent manner, which was under regulation by both Ca2+ and cAMP. Stimulation with a selective prostanoid EP4 receptor agonist (ONO-AEI-329, 10 µM), which activates cAMP signals, elicited a sustained increase in the frequency of exocytotic events (30 % of that evoked by 1 µM PGE2). Stimulation with an EP1 agonist (17-P-T-PGE2, 1 µM), which activates Ca2+ signals, increased the frequency of exocytotic events to a lesser extent (5 % of that evoked by 1 µM PGE2), while addition of an EP1 antagonist (ONO-8713, 10 µM) decreased the frequency of exocytotic events (approximately 40 % of that evoked by 1 µM PGE2). However, addition of the EP1 agonist potentiated the frequency of exocytotic events evoked by the EP4 agonist or forskolin (which elevates cAMP levels) and increased the sensitivity of the exocytotic events to forskolin. These results suggest that the Ca2+ signal activated via the EP1 receptor potentiates the cAMP-regulated exocytotic events activated via the EP4 receptor during PGE2 stimulation, by increasing the sensitivity of the exocytotic response to cAMP. In conclusion, exocytotic events in PGE2-stimulated antral mucous cells were regulated by interactions between EP1 and EP4 receptors. Experimental Physiology (2001) 86.4, 451-460.

Adrenoceptor stimulation enhances repolarising and depolarising membrane currents to different extents in cardiac myocytes. We investigated the opposing effects of the repolarising Ca2+-activated Cl- current (ICl(Ca)) and depolarising L-type Ca2+ current (ICa,L) on the action potential configuration of sheep ventricular myocytes stimulated with noradrenaline. Whole-cell current-clamp recordings revealed that noradrenaline accelerated and prolonged phase-1 repolarisation. We define the minimal potential at the end of phase-1 repolarisation as 'notch level'. Noradrenaline (1 µM) caused the notch level to fall from 14 ± 2.6 to 7.8 ± 2.8 mV (n = 24), but left action potential duration, resting membrane potential or action potential amplitude unaffected. Whole-cell voltage-clamp recordings showed that 1 µM noradrenaline increased both ICa,L and ICl(Ca), but it had no significant effect on the principal K+ currents. Blockage of ICl(Ca) by 0.5 mM 4,4'-diisothiocyanatostilbene-2,2'-disulphonic acid (DIDS) in both the absence and the presence of noradrenaline abolished phase-1 repolarisation. In the presence of noradrenaline, DIDS caused elevation of the plateau phase amplitude and an increase in the action potential duration. In conclusion, elevation of the plateau phase amplitude and action potential prolongation associated with an increased ICa,L upon adrenoceptor stimulation is prevented by an increased ICl(Ca) in sheep ventricular myocytes. Experimental Physiology (2001) 86.2, 151-159.

Using a genetically modified herpes simplex virus encoding green fluorescent protein we sought to establish if this viral modification could be used in transneuronal tracing studies of the sympathetic nervous system. The herpes simplex virus encoding green fluorescent protein was injected into the adrenal medulla of three hamsters and six rats. After a suitable survival period, neurones in the sympathetic intermediolateral cell column of the thoracolumbar spinal cord, rostral ventral medulla and paraventricular nucleus of the hypothalamus were clearly identified by the presence of a green fluorescence in the cytoplasm of the neurones of both species. Thus, herpes simplex virus encoding green fluorescent protein labelled chains of sympathetic neurones in the hamster and rat and therefore has the potential to be used in transneuronal tracing studies of autonomic pathways in these species. Experimental Physiology (2001) 86.6, 695-702.

The dynamics and kinetics of thyroid hormone transport in the isolated rat heart were examined using the modified unidirectional paired tracer dilution method. The uptake of 125I-thyroxine (125I-T4) and 125I-triiodothyronine (125I-T3) from the extracellular space into heart cells was measured relative to the extracellular space marker 3H-mannitol. The thyroid hormone maximal uptake was 54.4 % for 125I-T4 and 52.15 % for 125I-T3. The thyroid hormone net uptake was 25.69 % for 125I-T4 and 25.49 % for 125I-T3. Backflux from the intracellular space was 53.17 % for 125I-T4 and 61.59 % for 125I-T3. In the presence of unlabelled thyroid hormones, 125I-T4 and 125I-T3 maximal uptakes were reduced from 10.1 to 59.74 % and from 34.6 to 65.3 %, respectively, depending on the concentration of the unlabelled hormone, suggesting a saturable mechanism of the thyroid hormone uptake by the heart cells, with Km(T4)= 105.46 µM and the maximal rate of 125I-thyroid hormone flux from the extracellular space to heart cells (Vmax(T4)) = 177.84 nM min-1 for 125I-T4 uptake, and Km(T3) = 80.0 µM and Vmax(T3) = 118.5 nM min-1 for 125I-T3 uptake. Experimental Physiology (2001) 86.1, 13-18.

The present investigation was designed to examine whether calmodulin is involved in the inhibition of the ATP-sensitive K+ (KATP) channel by glucagon-like peptide 1(7-36) amide (GLP-1) in mouse pancreatic β-cells. Membrane potential, single channel and whole-cell currents through the KATP channels, and intracellular free Ca2+ concentration ([Ca2+]i) were measured in single mouse pancreatic β-cells. Whole-cell patch-clamp experiments with amphotericin-perforated patches revealed that membrane conductance at around the resting potential is predominantly supplied by the KATP channels in mouse pancreatic β-cells. The addition of 20 nM GLP-1 in the presence of 5 mM glucose significantly reduced the membrane KATP conductance, accompanied by membrane depolarization and the generation of electrical activity. A calmodulin inhibitor N-(6-aminohexyl)-5-chloro-1-naphthalenesulphonamide (W-7, 20 µM) completely reversed the inhibitory actions of GLP-1 on the membrane KATP conductance and resultant membrane depolarization. Cell-attached patch recordings confirmed the inhibition of the KATP channel activity by 20 nM GLP-1 and its restoration by 20 µM W-7 or 10 µM calmidazolium at the single channel level. Bath application of 20 µM W-7 also consistently abolished the GLP-1-evoked increase in [Ca2+]i in the presence of 5 mM glucose. These results strongly suggest that the mechanisms by which GLP-1 inhibits the KATP channel activity accompanied by the initiation of electrical activity in mouse pancreatic β-cells include a calmodulin-dependent mechanism in addition to the well-documented activation of the cyclic AMP-protein kinase A system. Experimental Physiology (2001) 86.3, 331-339.

The hypotensive and vasorelaxing peptides adrenomedullin (AM) and its gene-related peptide, proadrenomedullin N-terminal 20 peptide (PAMP), were found to be distributed on the surface of the colonic mucosa. AM and PAMP showed dose-dependent antimicrobial activity against E. coli. The results suggest that the novel vasoactive peptides AM and PAMP play an important role in mucosal defence. Experimental Physiology (2001) 86.5, 543-545.

The pathogenesis of chronic fatigue syndrome (CFS) remains unknown. Thresholds and latencies of motor evoked potentials (MEPs) in response to transcranial magnetic stimulation (TMS) are normal but intracortical inhibition has not been investigated. Eleven patients with CFS were compared with 11 control subjects. Each patient completed a questionnaire using visual analogue indices of pain, fatigue, anxiety and depression. Subjects released a button to initiate simple (SRTs) and choice reaction time (CRTs) tasks; for each task, movement times were measured between release of the initiation button and depression of a second button 15 cm away. Subjects held a 10 % maximum voluntary contraction in the thenar muscles of their dominant hand while TMS was applied to the motor cortex; the duration and extent of inhibition of surface electromyographic (EMG) activity were assessed at stimulus strengths above and below the threshold for MEPs. Patients had significantly (P < 0.05) higher mean indices of fatigue than of pain, anxiety or depression. Mean (± S.E.M.) SRTs (but not CRTs) were longer in patients (309 ± 45 ms) than in controls (218 ± 9 ms). Movement times were longer in patients for both SRTs and CRTs. TMS thresholds, expressed as a percentage of the maximum stimulator output, were not significantly (P > 0.05) different in both groups for both MEPs (patients, 34 ± 3 %; controls, 36 ± 3 %) and inhibition of voluntary contraction (patients, 29 ± 2 %; controls, 34 ± 4 %). The duration and extent of inhibition did not differ significantly between groups at any stimulus strength. The pattern of change in duration and extent of inhibition with increasing stimulus intensity was no different in the two groups. The duration and extent of corticospinal inhibition in patients with CFS did not differ from controls, adding further evidence to the notion that the feeling of fatigue and the slowness of movement seen in CFS is not manifest in corticospinal output pathways. Experimental Physiology (2001) 86.5, 547-550.

Oxidants have been suggested to enhance contractile function in unfatigued muscle. In this study we aimed to determine the effect of oxidants on 'chemically skinned' diaphragm muscle fibre bundles. The sarcoplasmic reticulum and contractile proteins were exposed to superoxide anions (O2-) and hydrogen peroxide (H2O2) under controlled conditions. Application of O2- initially increased maximum Ca2+-activated force but subsequently reduced maximum Ca2+-activated force without altering myofilament Ca2+ sensitivity. Unlike myocardium, caffeine-induced Ca2+ release from the sarcoplasmic reticulum was also inhibited by O2- exposure in diaphragm fibre bundles. Application of H2O2 also increased maximum Ca2+-activated force but had additional effects on resting tension (which increased to 25 % of the control maximum Ca2+-activated force). H2O2 was without effect on myofilament Ca2+ sensitivity or caffeine-induced Ca2+ release from the sarcoplasmic reticulum. These data demonstrate that oxidants can potentiate contractile force in the diaphragm through a direct action on the contractile proteins. The potentiation of force is not sustained, however, and under these conditions the detrimental effects of O2- on Ca2+ release from the sarcoplasmic reticulum combined with the effects of oxidants on the contractile proteins will ultimately compromise excitation-contraction coupling in the diaphragm. Experimental Physiology (2001) 86.2, 161-168.

Despite minimal model characterisation Langendorff perfused murine hearts are increasingly employed in cardiovascular research, and particularly in studies of myocardial ischaemia and reperfusion. Reported contractility remains poor and ischaemic recoveries variable. We characterised function in C57/BL6 mouse hearts using a ventricular balloon or apicobasal displacement and assessed responses to 10-30 min global ischaemia. We examined the functional effects of pacing, ventricular balloon design, perfusate filtration, [Ca2+] and temperature. Contractility was high in isovolumically functioning mouse hearts (measured as the change in pressure with time (+dP/dt), 6000-7000 mmHg s-1) and was optimal at a heart rate of ~ 420 beats min-1, with the vasculature sub-maximally dilated, and the cellular energy state high. Post-ischaemic recovery (after 40 min reperfusion) was related to the ischaemic duration: developed pressure recovered by 82 ± 5 %, 73 ± 4 %, 68 ± 3 %, 57 ± 2 % and 41 ± 5 % after 10, 15, 20, 25 and 30 min ischaemia, respectively. Ventricular compliance and elastance were both reduced post-ischaemia. Post-ischaemic recoveries were lower in the apicobasal model (80 ± 4 %, 58 ± 7 %, 40 ± 3 %, 32 ± 7 % and 25 ± 5 %) despite greater reflow and lower metabolic rate (pre-ischaemic myocardial O2 consumption (V˙O2,myo) 127 ± 15 vs. 198 ± 17 µl O2 min-1 g-1), contracture, enzyme and purine efflux. Electrical pacing slowed recovery in both models, small ventricular balloons (unpressurised volumes < 50-60 µl) artificially depressed ventricular function and recovery from ischaemia, and failure to filter the perfusion fluid to < 0.45 µm depressed pre- and post-ischaemic function. With attention to these various experimental factors, the buffer perfused isovolumically contracting mouse heart is shown to be stable and highly energised, and to possess a high level of contractility. The isovolumic model is more reliable in assessing ischaemic responses than the commonly employed apicobasal model. Experimental Physiology (2001) 86.6, 703-716.

The effects of low-intensity, prolonged swimming on functional recovery of the rat heart (Langendorff preparations) from ischaemia-reperfusion (I/R) were investigated. Three groups of rats (120 days old) were used: sedentary rats (S) and rats exercised by a single bout of swimming lasting 5 (E5) or 8 h (E8), respectively. The effect of exercise on the response to I/R was related to an index of oxidative damage such as lipid peroxidation, as well as to the tissue antioxidant capacity and the response of heart tissue to in vitro oxidative stress. The intrinsic performance of E5 Langendorff preparations paced at 220 beats min-1 was also determined. A group of sedentary animals was used for H2O2-treated preparations. The effect of antioxidant treatment on inotropic recovery during reperfusion was studied on preparations from 5 or 8 h swimming vitamin E-treated (EVT5 and EVT8) and 5 or 8 h swimming untreated (EVU5 and EVU8) rats. Hearts from exercised animals displayed a reduced preischaemic inotropism, which in E5 rats was accompanied by an increase in the intrinsic heart rate. The lower intrinsic cardiac inotropism of E5 animals was confirmed in the paced preparations. The reduced contractility found in control hearts after addition of H2O2 to perfusion medium suggested that the low inotropism of E5 and E8 hearts was due to an exercise-induced increase in reactive oxygen species. Inotropic recovery during reperfusion was low in the S hearts, was significantly increased in the E5 hearts, and was again reduced to the S level in the E8 hearts. In the E8 hearts the indexes of cellular damage (LDH release) and oxidative stress increased, and antioxidant capacity decreased, while in E5 hearts there was no evidence of significant changes in such parameters. Performance and reperfusion recovery of hearts from 5 h swimming rats was not affected by vitamin E treatment, while those of hearts from 8 h swimming rats was the highest observed. We suggest that the higher inotropic recovery during reperfusion in the hearts from the E5 rats is related to the negative inotropic effect of exercise. The fall in recovery following the 8 h exercise was instead related to the increased oxidative stress. Experimental Physiology (2001) 86.3, 341-348.

The properties of capacitative Ca2+ influx were studied using the whole-cell patch-clamp technique in crypts isolated from rat distal colon. Store-operated cation influx was evoked by increasing the intracellular buffering capacity for Ca2+ in the pipette solution; contamination by Cl- currents was reduced by the use of NMDG gluconate as the main electrolyte in the pipette solution. The permeability of the non-selective cation conductance stimulated by store depletion had the following sequence for monovalent cations: Cs+ > Na+ [ges ] Li+. The store-operated conductance is permeable to Na+ and Ca2+, but in contrast to Na+, Ca2+ also exerts a (feedback) inhibition on its own influx. Other divalent cations shared this inhibitory action with the sequence: Ca2+ [ges ] Mg2+ [ges ] Ba2+ [ges ] Sr2+. Fura-2 experiments revealed that replacement of extracellular Na+ by NMDG+ induced an increase in the intracellular Ca2+ concentration, which was suppressed by the Na+-Ca2+ exchange inhibitor, dichlorobenzamil, indicating the presence of a Na+-Ca2+ exchanger within the colonic crypt cells. In Ussing chamber experiments dichlorobenzamil induced an increase in short-circuit current (Isc) in the majority of tissues tested indicating that this exchanger acts as a Ca2+-extruding transporter under physiological conditions. When Ca2+-dependent anion secretion was stimulated by the acetylcholine analogue carbachol, dichlorobenzamil no longer evoked an increase in Isc, indicating that after stimulation of the store-operated cation conductance the Na+-Ca2+ exchanger is turned off. Therefore, it is concluded that the influx of Na+ across the non-selective store-operated cation conductance serves to reduce the driving force for Ca2+ extrusion via the Na+-Ca2+ exchanger and thereby maintains the increase in the intracellular Ca2+ concentration during induction of secretion. Experimental Physiology (2001) 86.4, 461-468.

Intense exercise or myocardial ischaemia can significantly increase both the concentration of extracellular potassium ([K+]o) and cardiac sympathetic nerve activity. Since changes in [K+]o modulate membrane currents involved in sino-atrial node pacemaking, in particular the voltage-sensitive hyperpolarization-activated current (If), we investigated whether raised [K+]o (from 4 mM to 8 or 12 mM) could directly affect the heart rate response to cardiac sympathetic nerve stimulation (SNS). In the isolated guinea-pig atrial-right stellate ganglion preparation, raised [K+]o significantly decreased the maximum diastolic potential, amplitude and maximum rate of rise of the upstroke of sino-atrial node pacemaker action potentials in 8 and 12 mM [K+]o (P < 0.05). At 12 mM [K+]o these effects were associated with significant decreases in baseline heart rate (4 mM [K+]o = 187 ± 5 beats min-1 (bpm); 12 mM = 144 ± 11 bpm; P < 0.05) and the heart rate response to SNS (1, 3 and 5 Hz; P < 0.05). A 10 % increase in the baseline heart rate with sympathetic activation (3 Hz) was associated with a significant enhancement of the slope of the pacemaker diastolic depolarization at 4 mM [K+]o (increased by 16 ± 6 %; n = 7; P < 0.05), but not with raised [K+]o. When the If current was blocked with 2 mM caesium (n = 8), 12 mM [K+]o had no effect on baseline heart rate and the heart rate response to 3 Hz SNS. The heart rate response to bath-applied noradrenaline (0.01-100 µM) was significantly attenuated by 12 mM [K+]o (at 4 mM [K+]o, EC50 = -6.31 ± 0.18; at 12 mM [K+]o, EC50 = -5.80 ± 0.10; n = 6, ANOVA, P < 0.05). In conclusion, extreme physiological levels of [K+]o attenuate the positive chronotropic response to cardiac sympathetic activation due to decreased activation of the If current. This is consistent with raised [K+]o protecting the myocardium from potentially adverse effects of excessive noradrenaline. Experimental Physiology (2001) 86.1, 19-25.

Muscle cells in the nematode Ascaris suum undergo bouts of oscillation in the behaving worm, these being correlated with progressing waves of contraction along the body of the worm. The bouts have three time scales: a rough period of 7-20 s, 10-20 short bursts of spikes within a bout, and three to eight spikes per burst. This paper has two aims: showing that there is consistency between measurements of individual currents and measurements on whole muscle cells, and creating a building block to eventually explain the locomotion in this 'simple' system. A realistic model for a single Ascaris suum muscle cell is developed using existing data from experiments which have established the types of ionic currents in the muscle cell and many of their kinetic properties. Numerical simulations are carried out. The model cell reproduces the two shorter time scales present. It has some robustness with regard to parameter changes, but also allows for the different numbers of spikes per burst seen in recordings from muscle cells. The third time scale, the length of a bout, may originate from some system effect, a combination of neural and stretch effects and possibly a secondary effect of the calcium-activated chloride channel. Experimental Physiology (2001) 86.5, 551-559.

In vitro evidence suggests that the 'recycling' of K+ ions through luminal K+ channels in the thick ascending limb of the loop of Henle (TALH) is essential for the normal operation of the luminal Na+-K+-2Cl- co-transporter. In the present study these channels were investigated in vivo by perfusing superficial loops of Henle in anaesthetised rats with and without the K+ channel blocker barium. Using a standard perfusate, intraluminal barium (5 mmol l-1) reduced sodium reabsorption (JNa) from 1887 ± 50 to 1319 ± 53 pmol min-1 (P < 0.001). When the experiment was repeated using a low-Na+ perfusate, designed to inhibit reabsorption in the pars recta (the initial segment of the loop of Henle), a similar reduction in JNa was observed (from 698 ± 47 to 149 ± 23 pmol min-1, P < 0.001), strongly suggesting that the effect of barium is localised to the TALH. The magnitude of the reduction in JNa during blockade of K+ channels confirms the importance of K+ recycling in facilitating Na+ reabsorption in the TALH in vivo. However, the reduction in JNa was not associated with a fall in the K+ concentration of the fluid collected at the early distal tubule. When bumetanide, an inhibitor of the Na+-K+-2Cl- co-transporter, was included in the low-Na+ perfusate, net K+ secretion was observed. Addition of barium to this perfusate reduced, but did not abolish, the secretion, suggesting that bumetanide-induced K+ secretion results partly from paracellular transport. Experimental Physiology (2001) 86.4, 469-474.