Preclinical in vitro cardiac electrophysiology: a method of predicting arrhythmogenic potential of antihistamines in humans?

The cardiac action potential results from a dynamic balance between inward depolarising Na+ and Ca2+ currents and outward K+ repolarising currents. During a cardiac cycle, the resultant of repolarisation phase from all ventricular cells is represented by the QT interval of the surface ECG. Congenital long QT syndrome (LQTS) is characterised by polymorphic ventricular tachycardia sometimes with twisting QRS morphology (torsade de pointes) which, although usually self-limiting, can result in sudden cardiac death. Acquired LQTS can be induced by a variety of drugs, including some nonsedative histamine H1 receptor antagonists (astemizole, terfenadine). The Committee for Proprietary Medicinal Products of the European Union has recently proposed studying the action potential in in vitro heart preparations as a preclinical test for predicting the propensity of noncardiovascular drugs to induce malignant QT prolongation in humans. The effects of several histamine H1 receptor antagonists on the electrically evoked action potential have been evaluated in rabbit Purkinje fibres. In this preparation, astemizole (0.3 to 10 micromol/L) prolongs the duration of the action potential measured at the level where repolarisation is 90% complete (APD90). This effect is dependent on drug concentration, incubation time, pacing frequency and K+ or Mg2+ concentration. Astemizole also markedly depresses the rate of rise of the action potential (Vmax). Terfenadine showed qualitatively similar, but quantitatively smaller, effects in this model. The histamine H1 receptor antagonists cetirizine, ebastine, carebastine, loratadine and fexofenadine do not significantly affect APD90 at 1 micromol/L, but cetirizine and carebastine prolong it slightly at 10 micromol/L. In conclusion, in rabbit Purkinje fibres, astemizole and terfenadine produce adverse electrophysiological effects at concentrations which may be achieved in the human myocardium in certain clinical situations. APD90 lengthening induced by carebastine and cetirizine is minor and occurs at concentrations that are very unlikely to be encountered clinically, since these drugs, in contrast to astemizole and terfenadine, do not accumulate in the myocardium. Direct extrapolation of preclinical results to humans requires great caution, since malignant QT prolongations by terfenadine and astemizole are extremely rare clinical events. However, since prolongation of the QT interval often precedes the development of torsade de pointes, any significant delay in cardiac repolarisation produced by noncardiovascular drugs in preclinical, and particularly in clinical, studies should, in general, be considered to indicate a potential cardiac risk in humans. Its significance should subsequently be evaluated in appropriate studies in patients with conditions known to predispose to arrhythmias.

The naphtosultam derivative RP 62203 (fananserin) has high affinity for the dopamine D4 receptor.

The dopamine D4 receptor is a potential target for novel antipsychotic drugs. Most available compounds with affinity for the dopamine D4 receptor also bind to dopamine D2 receptors. This report describe the affinity of the 5-HT2A receptor antagonist RP 62203 (fananserin) for the human dopamine D4 receptor. Fananserin displaces [3H]spiperone binding to recombinant human dopamine D4 receptors with a Ki of 2.93 nM. This compares with an affinity (Ki) of 0.37 nM for the rat 5-HT2A receptor and of 726 mM for the rat dopamine D2 receptor. [3H]Fananserin can be used to label the recombinant dopamine D4 receptor expressed in Chinese hamster ovary cells with a KD of 0.725 nM. Fananserin is, thus, the first compound to be reported that distinguishes between dopamine D4 and D2 receptors.

The human galanin receptor: ligand-binding and functional characteristics in the Bowes melanoma cell line.

The human galanin receptor has been characterized pharmacologically from the Bowes melanoma cell line. Using porcine [125I]galanin as the radioligand, a single population of non-interacting high-affinity binding sites (KD = 0.05 +/- 0.01 nM; Bmax = 135 +/- 7 fmol/mg protein) was demonstrated. Human galanin peptide competitively inhibited the specific binding of [125I]galanin (IC50 = 0.35 +/- 0.13 nM) and decreased the forskolin-stimulated cAMP production (EC50 = 0.46 +/- 0.05 nM) with a maximal inhibition of 63 +/- 2% at 10(-7) M. Rat and porcine galanin peptides and the chimeric peptides M15, M35, M32, M40 and C7 also dose-dependently inhibited the forskolin-stimulated cAMP production, while the fragment porcine galanin-(3-29) and [D-Trp2]galanin were found to be inactive. The specific binding of [125I]galanin was decreased in a dose-dependent manner by GTP and the cAMP response was inhibited by the pertussis toxin, suggesting the activation of a G-protein dependent process. The Bowes cell line thus appears to be a relevant tool for the study of human galanin receptor.

Septide: an agonist for the NK1 receptor acting at a site distinct from substance P.

The hexapeptide [pGlu6,Pro9]substance P (SP)6-11, septide, has been shown to be an agonist as potent as SP in eliciting smooth muscle contraction in several in vitro preparations, while being a poor competitor of labeled SP binding. These results, as well as other pharmacological data, have suggested the existence of either a specific septide receptor or a septide site on the neurokinin (NK)1 receptor distinct from that for SP. We have used rat recombinant NK1 receptor expressed in COS-1 cells to address this issue. Both functional (agonist-induced inositol phosphate accumulation) and radioligand binding studies were conducted on transiently transfected cells. SP and septide elicited similar maximal increases (4-6-fold) in inositol phosphate levels in transfected cells, with EC50 values of 0.05 +/- 0.02 nM for SP and 5 +/- 2 nM for septide. No additivity of the maximal responses to the two agonists was observed, and neither agonist evoked any response in sham-transfected cells. RP 67580 was a competitive inhibitor of SP responses, with an inhibition constant (KB) of 13 +/- 2 nM, in agreement with displacement studies of [3H]SP binding to membranes and intact transfected cells (Ki values of 10 +/- 4 nM, and 1.16 +/- 0.06 nM, respectively). In comparison, septide responses were inhibited by RP 67580 in an uncompetitive fashion, with an apparent KB* value of 1.5 +/- 0.2 nM. Septide was a weak competitor of [3H]SP binding, with dissociation constants (Ki) of 2.9 +/- 0.6 microM and 3.7 +/- 0.9 microM for membranes and intact transfected cells, respectively. Similarly, septide at concentrations up to 10 microM did not affect [3H]RP 67580 binding. In conclusion, we have demonstrated that septide is a potent functional agonist of the NK1 receptor but it seems to act at a specific subsite different from that for SP. Although not ruling out the existence of selective septide receptors in some tissues, these results could explain some of the discrepancies with regard to the pharmacological properties of septide. Furthermore, a specific septide site on the NK1 receptor could represent an original pharmacological target.

Substance P-evoked calcium mobilization and ionic current activation in the human astrocytoma cell line U 373 MG: pharmacological characterization.

In the human astrocytoma cell line U 373 MG, application of substance P (SP) leads to a transient increase in cytosolic calcium concentration and to a biphasic current response in voltage-clamped cells. Using these two functional assays we have characterized pharmacologically the SP response in U 373 MG cells. SP and [L-Pro9]SP displayed high potencies in both assays with EC50 values of 2.5 x 10(-9) M and 1 x 10(-9) M on calcium responses and 1 x 10(-9) M and 5 x 10(-9) M on ion current responses, respectively. The high potency of SP and [L-Pro9]SP as well as the low potency of [Lys5,MeLeu9,N-Leu10]neurokinin A(4-10) and the inactivity of senktide demonstrate the NK1-type pharmacology of these responses. Furthermore, the NK1 antagonists (+/-)-CP 96,345, its chloro analogue, (+/-)-cis-3-(2-chlorobenzylamino)-2-benzhydrylquinuclidine, and RP 67580 were potent antagonists of both SP responses. For the calcium mobilization induced by SP (10(-7) M), the IC50 values for the three antagonists were 4 x 10(-10) M, 4 x 10(-9) M, and 9 x 10(-9) M, respectively, whereas on the current response evoked by SP (10(-8) M), the IC50 values were 8 x 10(-9) M, 2.4 x 10(-8) M, and 1.2 x 10(-7) M, respectively. Despite differences in the absolute IC50 values obtained with both techniques, the relative potencies of the three antagonists correlate fairly well.(ABSTRACT TRUNCATED AT 250 WORDS)

Characterization of a human NK1 tachykinin receptor in the astrocytoma cell line U 373 MG.

The human NK1 tachykinin receptor in the astrocytoma cell line U 373 MG was characterized using selective agonists and antagonists described for this receptor in the rat. Specific [3H]substance P binding sites were present on cell homogenates, whereas [3H]neurokinin A or [3H]-senktide binding sites were absent. The binding was saturable and reversible. The binding of [3H]substance P was inhibited by very low concentrations of [L-Pro9]substance P and [Sar9,Met(O2)11]substance P; septide was approximately 1,000-fold less potent. The most potent peptide antagonist was trans-4-hydroxy-1-(1H-indol-3-ylcarbonyl)-L-prolyl-N-methyl-N-(phe nylmethyl)-L- tyrosineamide. The rank order of potency for the nonpeptide antagonists was (S,S)-CP 96,345 > (+/-)-CP 96,345 > (+/-)-2-chlorobenzylquinuclidinone > (R,R)-CP 96,345 > RP 67580 > RP 68651. In [3H]-inositol-labeled cells, substance P stimulated phosphatidylinositol turnover. A good correlation was found when the abilities of NK1 receptor agonists for stimulating inositol phosphate production and for inhibiting [3H]substance P binding were compared. Similarly, the binding and functional assays were well correlated for the antagonists. As a result of its high sensitivity and selectivity, the U 373 MG cell line thus appears an excellent tool for investigating the pharmacology of the human NK1 receptor.

Higher potency of RP 67580, in the mouse and the rat compared with other nonpeptide and peptide tachykinin NK1 antagonists.

1. This study was undertaken to compare the potency and selectivity of the nonpeptide (RP 67580, (+/-)-CP-96,345 and its chloro-derivative [(+/-)-cis-3-(2-chlorobenzylamino)-2-benzhydrylquinuclidine] (CP-C1)) and peptide (GR 71,251 and spantide) neurokinin1 (NK1) antagonists in mouse and rat preparations. 2. Among the NK1 antagonists tested, RP 67580 was the most potent in inhibiting the specific binding of [125I]-Bolton Hunter substance P ([125I]-BHSP) to crude synaptosomes from the rat brain (Ki: 2.9 nM). (+/-)-CP-96,345 was about ten fold less potent (Ki: 31 nM) than RP 67580 while other compounds exhibited even less affinity. 3. All NK1 antagonists inhibit competitively the activation of phospholipase C by [Pro9]substance P ([Pro9]SP) in cultured cortical astrocytes from the newborn mouse, a preparation rich in NK1 receptors but devoid of NK2 and NK3 receptors. pA2 values for the most potent compounds, RP 67580 and (+/-)-CP-96,345, were 8.28 and 7.08 respectively. When used alone, all antagonists showed some agonist activity at 10(-5) M, except spantide which was already effective at 10(-6) M. 4. An excellent correlation was found between the potency of the NK1 antagonists in blocking the stimulation by [Pro9]SP of phosphoinositide breakdown in cortical astrocytes and in inhibiting [125I]-BHSP specific binding to rat brain synaptosomes. 5. As shown on single cells by use of the Indo-1 microfluorometric method, RP 67580 (10(-7) M) prevented reversibly the elevation of cytosolic calcium concentration induced by [Pro9]SP (10(-8) M) in cultured cortical astrocytes. 6. Several experiments indicated that the antagonists were highly selective for NK1 receptors. RP 67580 did not modify the noradrenaline-evoked activation of phospholipase C in cortical astrocytes; when used at 10-5 M all antagonists had no or only little affinity for NK2 or NK3 binding sites and did not block the NKA (10-8 M)-induced activation of phospholipase C in the hamster urinary bladder (a selectiveNK2 test).7. In conclusion, RP 67580 appears to be a potent NK1 antagonist in the mouse and the rat. Results obtained with (+/-)-CP-96,345 confirm the lower potency of this compound in these two species when compared with reported data obtained in the guinea-pig or man.

Cytokeratin expression during AFB1-induced carcinogenesis.

The early stages of the carcinogenic process induced by aflatoxin B1 (AFB1) in rat liver during 24 weeks of feeding and the resulting tumours have been studied with respect to cytokeratin (CK) expression. A previously uncharacterized monoclonal antibody, MRCTU/J1, has been shown to recognize rat CK18 and together with antibodies against human CK8, 18 and 19, has been used to examine the possible lineage of tumour cells and also to identify the altered foci that might be most relevant to tumorigenesis. Results suggested that AFB1-induced transformation in liver may occur in more than one cell type, since tumours with the normal hepatocyte CK pattern and those with bile duct or oval cell CK phenotype were identified. Additionally, hepatocytes with a bile duct CK phenotype appeared during the early stages of carcinogenesis. The in vivo pattern of CK expression also appeared to be maintained in one normal and one hepatoma-derived cell line. Overexpression of CKs (particularly of CK19) was a much more selective marker for altered foci, compared to gamma-glutamyltranspeptidase, and was more consistently expressed at high levels in tumours, suggesting that it might be a more reliable way of identifying those cells involved in the transformation process.