Alex J. Baertschi

Relative Contributions of Connexins 40 and 43 to Atrial Impulse Propagation in Synthetic Strands of Neonatal and Fetal Murine Cardiomyocytes

Atrial tissue expresses both connexin 40 (Cx40) and 43 (Cx43) proteins. To assess the relative roles of Cx40 and Cx43 in atrial electrical propagation, we synthesized cultured strands of atrial myocytes derived from mice with genetic deficiency in Cx40 or Cx43 expression and measured propagation velocity (PV) by high-resolution optical mapping of voltage-sensitive dye fluorescence. The amount of Cx40 and/or Cx43 in gap junctions was measured by immunohistochemistry and total or sarcolemmal Cx43 or Cx40 protein by immunoblotting. Progressive genetic reduction in Cx43 expression decreased PV from 34±6 cm/sec in Cx43+/+ to 30±8 cm/sec in Cx43+/− and 19±11 cm/sec in Cx43−/− cultures. Concomitantly, the cell area occupied by Cx40 immunosignal in gap junctions decreased from 2.0±1.6% in Cx43+/+ to 1.7±0.5% in Cx43+/− and 1.0±0.2% in Cx43−/− strands. In contrast, progressive genetic reduction in Cx40 expression increased PV from 30±2 cm/sec in Cx40+/+ to 40±7 cm/sec in Cx40+/− and 45±10 cm/sec in Cx40−/− cultures. Concomitantly, the cell area occupied by Cx43 immunosignal in gap junctions increased from 1.2±0.9% in Cx40+/+ to 2.8±1.4% in Cx40+/− and 3.1±0.6% in Cx40−/− cultures. In accordance with the immunostaining results, immunoblots of the Triton X-100–insoluble fraction revealed an increase of Cx43 in gap junctions in extracts from Cx40-ablated atria, whereas total cellular Cx43 remained unchanged. Our results suggest that the relative abundance of Cx43 and Cx40 is an important determinant of atrial impulse propagation in neonatal hearts, whereby dominance of Cx40 decreases and dominance of Cx43 increases local propagation velocity.

Spinal afferents for peripheral osmoreceptors in the rat

Superfusion of the hepatic portal vein with hypertonic saline solutions increases the electrical activity of the hypothalamo-neurohypophysial tract. Bilateral cervical vagotomy had no effect on this response in all rats studied. Section of the hepatic branch of the vagus abolished the hypothalamic response in only two animals, but injection of xylocaine into the spinal cord at thoracic levels abolished the response in all remaining animals. The results suggest that peripheral osmoreceptors of the portal vein activate the hypothalamo-neurohypophysial system through a spinal afferent pathway.

Action of γ-aminobutyric acid on hypothalamo-neurohypophysial axons

Abstract γ-Aminobutyric acid (GABA) and the related compound, muscimol, were bath-applied to a preparation containing the rat median eminence, pituitary stalk and neurohypophysis, and are shown to reversibly reduce the amplitude, and increase the latency, of the antidromic compound action potential recorded from hypothalamoneurohypophysical axons in response to electrical stimulation of the posterior pituitary lobe. Both actions were antagonized by bicuculline, but were unaffected by strychrine. The data suggest that the peptidergic axons of the hypothalamo-neurohypophysial tract are endowed with receptors for GABA, which is believed to be synthesized and released by neurohypophysial glial cells.

Enkephalins, substance P, bradykinin and angiotensin II: Differential sites of action on the hypothalamo-neurohypophysial system

To test the hypothesis that enkephalins, substance P, bradykinin and angiotensin II could act as neurohumoral modulators of hypothalamic function, these peptides (0.01-20 microgram) were injected into the general circulation of anesthetized rats, and changes in hypothalamo-neurohypophysial activity were determined by continuously monitoring the amplitude of antidromic compound action potentials (CAP) in the hypothalamo-hypophysial tract. A decrease of CAP amplitude was taken to indicate an increase of orthodromic impulse traffic. All peptides elicited a CAP decrease. On a molar basis when injected i.v., the enkephalin analog FK 33-824 was the most effective substance, followed by substance P, Leu-enkephalin and angiotensin II. Enkephalins and substance P injected through the internal carotid artery, were 2-5 times more effective than when injected i.v., whereas bradykinin was most effective when it reached the brain through a vertebral route. Angiotensin II produced the same CAP decrease irrespective of the route of administration and, in contradistinction to the other peptides, its effect was not abolished by stalk section. Tachyphylaxis and reversibility with naloxone was observed only for the enkephalins. The data suggest that sites of action are the hypothalamus for enkephalins and substance P, the neurohypophysis for angiotensin II, and the hindbrain for bradykinin.

Sodium-chloride sensitive receptors located in hepatic portal vein of the rat

The superfusion of the hepatic portal vein of anaesthetized rats with hypertonic solutions of NaCl elicits a significant increase of neural activity measured in the hypothalamo-neurohypophysial tract. Superfusions of the liver or hepatic veins with hypertonic saline solutions had no effect, whereas superfusions of the portal vein also increased the firing rates of supraoptic endocrine neurones and the intramammary pressure. Results suggest that sodium chloride sensitive receptors (perhaps osmoreceptors) are located within the wall of the hepatic portal vein and mediate neurophypophysial hormone release.

Differential sensitivity of atrial and ventricular KATP channels to metabolic inhibition

OBJECTIVEnThe aim is to compare the activation of ATP-sensitive potassium channels (K(ATP) channels) in intact and metabolically impaired atrial and ventricular myocytes.nnnMETHODSnThe K(ATP) channel current is measured by whole cell and gramicidin-perforated patch clamp recordings in 164 cultured neonate rat cardiomyocytes.nnnRESULTSnIn whole cell recordings with 84 micromol/l ADP in pipette, spontaneous activity is significantly higher in atrium than ventricle, and EC(50) for the K(ATP) channel opener diazoxide is 0.13 micromol/l (atrium) versus 3.1 micromol/l (ventricle). With an ATP-regenerating system in pipette, EC(50) for diazoxide is 19.7 micromol/l (atrium) versus 54.9 micromol/l (ventricle). In gramicidin-perforated patch recordings, atrial myocytes respond significantly to 100 nmol/l of the mitochondrial protonophore CCCP, while ventricular myocytes do not. EC(50) for diazoxide is 129 micromol/l (atrium) versus >2500 micromol/l (ventricle) for myocytes exposed to CCCP, and 676 versus >2500 micromol/l, respectively, without CCCP.nnnCONCLUSIONSn(1) K(ATP) channels are significantly more sensitive to metabolic inhibition in atrial than ventricular myocytes. (2) Sensitivity of atrium versus ventricle to the channel opener diazoxide increases from 3:1 to > or = 24:1 with ADP or metabolic inhibition. If extended to intact hearts, the results would predict a higher atrial sensitivity to ischemia, and a high sensitivity of the ischemic atrium to K(ATP) channel openers.

Neural actions of several substance P antagonists in the rat spinal cord.

Four substance P (SP) antagonists were tested on anaesthetized rats, by injecting 8 microgram amounts into the spinal cord (T8-T9), and by observing their effects on the hypothalamo-neurohypophysial responses to a presumably painful stimulus, the superfusion of the hepatic portal vein with 1 microM bradykinin. Only two antagonists, the new D-Pro4,D-Trp7,9,10,Val8-SP4-11 and D-Pro4,D-Trp7,9,10-SP4-11 were capable of inhibiting the responses by 50-60%, the former compound having 3 times less agonistic activity. The results suggest that substitution of the aromatic phenylalanine by the non-polar valine in position 8 may significantly improve the overall characteristics of neurally active SP antagonists.

Angiotensin II and tumour necrosis factor α as mediators of ATP-dependent potassium channel remodelling in post-infarction heart failure

AIMSnAngiotensin II (Ang II) and tumour necrosis factor alpha (TNFalpha) are involved in the progression from compensated hypertrophy to heart failure. Here, we test their role in the remodelling of ATP-dependent potassium channel (K(ATP)) in heart failure, conferring increased metabolic and diazoxide sensitivity.nnnMETHODS AND RESULTSnWe observed increased expression of both angiotensinogen and TNFalpha in the failing rat myocardium, with a regional gradient matching that of the K(ATP) subunit Kir6.1 expression. Both angiotensinogen and TNFalpha expression correlated positively with Kir6.1 and negatively with Kir6.2 expression across the post-infarction myocardium. To further identify a causal relationship, cardiomyocytes isolated from normal rat hearts were exposed in vitro to Ang II or TNFalpha. We observed increased Kir6.1 and SUR subunit and reduced Kir6.2 subunit mRNA expression in cardiomyocytes cultured with Ang II or TNFalpha, similar to what was observed in failing hearts. In patch-clamp experiments, cardiomyocytes cultured with Ang II or TNFalpha exhibited responsiveness to diazoxide, in terms of both K(ATP) current and action potential shortening. This was not observed in untreated cardiomyocytes and resembles the diazoxide sensitivity of failing cardiomyocytes that also overexpress Kir6.1. Ang II exerted its effect through induction of TNFalpha expression, because TNFalpha-neutralizing antibody abolished the effect of Ang II, and in failing hearts, regional expression of angiotensinogen matched TNFalpha expression. Finally, Ang II and TNFalpha regulated K(ATP) subunit expression, possibly through differential expression of Forkhead box transcription factors.nnnCONCLUSIONnThis study identifies Ang II and TNFalpha as mediators of the remodelling of K(ATP) channels in heart failure.

The antidromic compound action potential of the hypothalamo-neurohypophysial tract, a tool for assessing posterior pituitary activity in vivo.

A sensitive technique was developed for measuring continuously the summated neural activity of the rat hypothalamo-neurohypophysial tract. Large compound potentials of 4-9 mV peak amplitude were recorded from the fibre tract in response to electrical stimulation of the neurohypophysis. It was estimated that all fibres in the tract were electrically activated and contributed to the compound potential. It was shown that a decrease of peak amplitude could be used as a measure of changes in endogenous neural activity. The decrease in peak amplitude during an increase of endogenous neural activity was due to fatigue and temporal spread of the compound potential, and also to collision of antidromic and orthodromic action potentials. Decrease of the compound potentials was related to the intensity of vaginal distension, haemorrhage and plasma hyperosmolality, and it correlated with neurohypophysial hormone release. This method may usefully complement hormone assays and single cell analysis.

Peptidyl-Glycine α-Amidating Monooxygenase Targeting and Shaping of Atrial Secretory Vesicles: Inhibition by Mutated N-Terminal ProANP and PBA

ANP (atrial natriuretic peptide) is widely recognized as an important vasorelaxant, diuretic, and cardioprotective hormone. Little is known, however, about how ANP-secretory vesicles form within the atrial myocytes. Secretory vesicles were visualized by fluorescence microscope imaging in live rat atrial myocytes expressing proANP–enhanced green fluorescent protein (EGFP), or N-terminal–mutated fusion proteins thought to suppress the calcium-dependent aggregation of proANP. Results showed the following: (1) aggregates of proANP and coexpressed proANP-EGFP recruited peptidylglycine &agr;-amidating monooxygenase (PAM)-1, an abundant atrial integral vesicle membrane protein; (2) coexpressed N-terminal–mutated (Glu23,24→Gln23,24) and N-terminal–deleted proANP-EGFP inhibited recruitment of PAM-1 by up to 60%; (3) 4-phenyl-3-butenoic acid (PBA) (10 &mgr;mol/L), a pharmacological inhibitor of the lumenal peptidylglycine &agr;-hydroxylating monooxygenase domain of PAM proteins, inhibited recruitment of endogenous PAM-1 and of coexpressed pro-EGFP–PAM-1; (4) PBA had no effect on exocytosis of the potassium inward rectifier KIR2.1; (5) PBA induced a deformation of the secretory vesicles but did not inhibit docking. These findings suggest that recruitment of PAM-1 to secretory vesicles depends on intact N-terminal proANP and on the lumenal domain of PAM-1. Conversely, PAM-1 participates in shaping the proANP-secretory vesicles. The full text of this article is available online at http://circres.ahajournals.org.