Ultra low doses: Biological and clinical applications

ULTRA LOW DOSES

Ultra Low Doses Edited by C. Doutremepuich Université Bordeaux France

Taylor & Francis London · Washington, DC 1991

UK Taylor & Francis Ltd., 4 John St. London WC1N 2ET USA Taylor & Francis Inc., 1900 Frost Road, Suite 101, Bristol, PA 19007 This edition published in the Taylor & Francis e-Library, 2005. “To purchase your own copy of this or any of Taylor & Francis or Routledge’s collection of thousands of eBooks please go to www.eBookstore.tandf.co.uk.” Copyright © Taylor & Francis Ltd., 1991 All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, electronic, electrostatic, magnetic tape, mechanical, photocopying, recording or otherwise, without the prior permission of the copyright owners. British Library Cataloguing in Publication Data Available on request Library of Congress Cataloging-in-Publication Data Available on request ISBN 0-203-48143-7 Master e-book ISBN

ISBN 0-203-78967-9 (Adobe eReader Format)

Contents

Preface

I EXPERIMENTAL PHARMACOLOGY First experimental arguments in favor of the effect of very weak doses of copper on digestive motricity in mice and rabbits Santini, R.Tessier, M.Belon, P.Pacheco, H.

2

The effects of some regulatory peptides in femtomolar and lower concentrations on the contraction of lymphatic vessels (LVs) Ashmarin, I.P.Levekova, L.Ts.Sanzhieva.

9

Modulation of experimental rat liver carcinogenesis by ultra low doses of the carcinogens De Gerlache, J.Lans, M.

14

Overcoming tumor cell drug resistance by low doses of recombinant tumor necrosis factor and drug Bonavida, B.Safrit, J.Tsuchitani, T.Zighelboim, J.

22

Effect of acetylsalicylic acid at ultra low dose on the interaction platelets/ vessel wall Lalanne, M.C.De Seze, O.Doutremepuich, C.

36

II BIOPHYSICS A quantum chemical study of some model anti-inflammatory compounds: the preferred conformations and their electrostatic similarities Bhattacharjee, Apurka, K.

44

Influence of several physical factors on the activity of ultra low doses Cazin, J.C.Cazin, M.Chaoui, A.Belon, P.

55

v

III BIOCHEMISTRY—TOXICOLOGY Uranyl nitrate induced corpuscular derangement in rat, an early indication of renel dysfunctioning Gojer, M.Sawant, V.

66

Degranulation of mesenteric mast cells as ‘spot test’ in toxicology Rathinam, K.Mohanan, P.V.Lizzy Michael

72

IV CELL BIOLOGY Simultaneous measurement of oxidative metabolism and adhesion of human neutrophils and evaluation of multiple doses of agonists and inhibitors Bellavite, P.Chirumbolo, S.Signorini, A.Bianchi, I.Dri, P.

76

Non-stimulatory concentrations of concanavalin A can modulate subsequent stimulation by the same mitogen Eskinazi, D.P.Molinaro, G.A.

95

Biological activity of ultra low doses: I/Effect of ultra low doses of histamine on human basophil degranulation triggered by D. pteronyssinus extract Sainte-Laudy, J.Sambucy, J.L.Belon, P.

101

Biological activity of ultra doses: II/Effect of ultra low doses of histamine on human basophil degranulation triggered by anti-IgE Sainte-Laudy, J.Belon, P.

112

V CLINICAL PHARMACOLOGY A study of the effectiveness of ultra low doses of copper in the treatment of hemodialysis-related muscle cramps Hariveau, E.Nolen, P.Holtzscherer, A.

118

Action of aspirin after ingestion at ultra low doses in healthy volunteers Lalanne, M.C.De Seze, O.Le Roy, D.Doutremepuich, C.Boiron, J.

122

Index

129

Preface

Research on Ultra Low Doses (ULD) is a multidisciplinary concept, situated at the crossroads of medical, human, biological and physical sciences. Determination of the mechanism of action is at present being widely developed, with numerous physico-chemical findings. Many questions about ULD efficacy now have a proven answer. However, other questions have arisen for which there are not enough experimental findings to provide definite conclusions. It is therefore of interest to convene an assembly of all people working in this field, to compare their results and conclusions. Fifteen countries were represented in Bordeaux in September 1990, for the first ‘International Congress on Ultra Low Doses’. The proceedings of these communications are summarized in this issue. The scientific committee was professors Aran (INSERM, France), Bonavida (USA), Dufourcq (CNRS, France), Roberfroid (Belgium) and Turner (GB). The editors wish to thank the authors for their contribution. They are also indebted to M.Claire Lalanne, for supporting the copy preparation. They hope that these proceedings may serve as a source of material for all investigators, researchers and students, who will be attracted by this highly interesting area. The kind cooperation and help from the publisher, Taylor & Francis, is hereby acknowledged. Professor C.Doutremepuich

I EXPERIMENTAL PHARMACOLOGY

FIRST EXPERIMENTAL ARGUMENTS IN FAVOR OF THE EFFECT OF VERY WEAK DOSES OF COPPER ON DIGESTIVE MOTRICITY IN MICE AND RABBITS Santini R.*, Tessier M.**, Belon P.*** and Pacheco H.* * INSA 20 Avenue Albert Einstein, 69621 Villeurbanne, France ** 136 Avenue Thiers, 69006 Lyon, France *** Institut BOIRON, 69110 Sainte Foy-Lès-Lyon, France INTRODUCTION During a general program of evaluation of homeopathic treatments in experimental pathology, it was pointed out that preliminary treatment with Cuprum 4 CH inhibited the facilitating effects of Neostigmine, a parasympathomimetic substance, on intestinal transit in mice and rabbits. Cuprum 4 CH (fourth centesimal dilution) is an homeopathic pharmaceutical preparation composed of a metal base of pure copper, made according to the Hahnemannian method of preparation described in the French pharmacopoeia, and corresponding to a molar solution of about 10−10. FIRST STUDY DESIGN—INTESTINAL TRANSIT IN MICE METHODS AND MATERIALS Intestinal transit was studied in mice using a technique derived from that of LOEWE. The transit marker used was phenol-sulfonephtalein or P.S.P., administered to healthy mice “per os” at T=0, in volumes of 0.3 ml. At T=+10 min. the mice were decapitated and the P.S.P. migratory front was observed by applying 20% sodium (NaOH) to the intestin (from the the presence of P.S.P. The percentage of P.S.P. migration was calculated for each mouse by measuring the length of the small intestin and by measuring the marker’s intestinal movement from the pylorus. Two experiments were carried out: In the first experiment 36 OF1 adult male mice, who had not been fed for 24 hours, were randomly separated into 3 groups:

FIRST EXPERIMENTAL ARGUMENTS IN FAVOR 3

Group 1: Intestinal transit comparison or reference group which was given a placebo treatment of distilled water: 0.3 ml., intraperitoneally (I.P.), at T=−24 hrs. and T=−5 hrs., 0.1 ml. was given at T=−10 min. Group 2: Neostigmine group, given 0.3 ml. of distilled water, I.P. at T=−24 hrs. and T=−5 hrs.; and 50 µg/kg of Neostigmine I.P. at T=−10 min. Group 3: Neostigmine+Cuprum group, given 0.3 ml. I.P. of Cuprum 4 CH at T=−24 hrs. and T=−5 hrs, and 50 µg/kg I.P. at T =−10 min. In the second experiment 48 O.F.A. adult female mice, who had not been fed for 24 hours, were randomly separated into 4 groups: 3 groups which were identical to the first experiment; and a fourth group: the Cuprum group, given 0.3 ml. I.P. of Cuprum 4 CH at T=−24 hrs. and T=−5 hrs.; and 0.1 ml. of distilled water at T=−10 min. Statistical analysis was carried out on migration percentages for each mouse of each group, using the Kruskall-Wallis non-parametric rank test. RESULTS The results are indicated in the following tables: Table 1. Intestinal transit in mice. First experiment. Results. Group

Mean migration percentages (M±sm)

Group 1=reference 40.66±4.67 Group 2=Neostigmine 74.91±6.52 51.76±7.35 Group 3=Cuprum+Neostigmine Statistics on 48 values (Kruskall-Wallis) Group 2 different from Group 1 (p<0.001) Group 3 identical to Group 1 (difference not significant) Group 3 different from Group 2 (p<0.02) Table 2. Intestinal transit in mice. Second experiment. Results. Group

Mean migration percentages (M±sm)

Group 1=reference Group 2=Neostigmine Group 3=Cuprum+Neostigmine Group 4=Cuprum

46.27±3.60 85.88±3.75 66.35±7.54 44.50±4.65

Statistics on 48 values (Kruskall-Wallis) Group 2 different from Group 1 (p<0.001) Group 3 different from Group 1 (p<0.02) Group 3 different from Group 2 (p<0.02) Group 4 identical to Group 1 (difference not significant)

4 ULTRA LOW DOSES

Figure 1.

DISCUSSION In both experiments intestinal transit was significantly accelerated (p<0.01) only by Neostigmine. In both experiments, preliminary treatment with Cuprum 4 CH significantly reduced (p<0.02) the facilitating action of Neostigmine, even cancelling it in the first experiment. Isolated treatment with Cuprum 4 CH had no effect on the intestinal transit of mice. SECOND STUDY DESIGN INTESTINAL ELECTROMYOGRAPHIC RECORDINGS IN VIGIL RABBITS METHODS AND MATERIALS Intestinal transit was studied through electroenterographic recording by implanting, in the duodenum of rabbits, chronic digestive electrodes exteriorized on their backs (SANTINI, R.). This proven technique made it possible to carry out repeated experiments and recordings without disturbing the animals. It also made it possible to work on only one animal at a time, since each rabbit served as its own reference animal. The electroenteromyogram displayed two types of electrical activity (Figure 1): – slow electrical activity (“basic electrical rythm”) composed of slow waves of constant frequency for a given species, – rapid electrical activity composed of bursts of potentials (spikes).

FIRST EXPERIMENTAL ARGUMENTS IN FAVOR 5

The injection of Neostigmine did not modify rapid electrical activity. Modifications of the response to Neostigmine under homeopathic treatment with Cuprum 4 CH were studied in two ways: – by analyzing the continuous integral lines of the bursts of potential (LATOUR, A.). – by the statistical analysis of the motricity index (I.M.) values before and after injection of Neostigmine. The motricity index represents the ratio of the number of bursts of spikes over the number of slow waves per minute. The statistical analysis was carried out using the Kruskall-Wallis non-parametric rank test. The study design called for the use of a 3.5 kg. male adult rabbit of the species “Fauve de Bourgogne”. The protocol order was as follows: – at D=+1, under general anesthesia (intravenous pentobarbital at 30 mg. per kilogram), the recording electrodes were implanted at the level of the proximal duodenum (3–4 cm. from the pylorus), – at D=+8, first recording (A), WHITOUT homeopathic treatment. Composed of: one hour of preliminary recording, the injection of Neostigmine (50 µg per kilo, intramuscular), and a new hour-long recording. The experiment was carried out at a set time, between 2:30 p.m. and 5:30 p.m. – at D=+9, start of the homeopathic treatment with a twice-daily injection (9: 00 a.m. and 2:00 p.m.), intraperitoneally, of Cuprum 4 CH, 1 ml. Treatment was continued up to and including D=+15. – at D=+12 and D=+15, second (B) and third recording (C) UNDER homeopathic treatment. – at D=+18, fourth recording (D) WITHOUT homeopathic treatment. In this particular experiment, at D=+21, a new comparison recording was made, not part of the study design. RESULTS – The results are demonstrated by the four integral lines: A, B, C, and D (Figure 2). Line C shows the lack of response to Neostigmine after 7 days of treatment with Cuprum 4 CH. – The results are quantified by the values of the motricity index in Table 3. Neostigmine acted highly significantly (Kruskall-Wallis, p<0.001) in the A and B recordings, and significantly (p<0.05) in the D recording. Neostigmine had no effect in the C recording (p<0.001) after 7 days of treatment with Cuprum 4 CH. At D=+21, once again, the recording (Figure 3) showed highly significant action with Neostigmine (p<0.001).

6 ULTRA LOW DOSES

Figure 2.

FIRST EXPERIMENTAL ARGUMENTS IN FAVOR 7

Table 3. Average value of duodenal motricity indexes in rabbits before and after injection of Neostigmine.

Figure 3.

CONCLUSIONS The results of the preliminary tests demonstrate that a preceding homeopathic treatment by Cuprum 4 CH inhibits the facilitating action of Neostigmine on intestinal transit, both in mice (p<0.02) and rabbits (p<0.001 after 7 days of treatment). These trials should be carried out again in blind. The results give hope for progress in understanding the means of action of certain homeopathic medicines, since the mechanism of action of Neostigmine by inhibition of cholinesterase is well understood. A study design which includes assays of this enzyme is now being developped.

8 ULTRA LOW DOSES

REFERENCES Latour, A., 1973, Un dispositif simple d’analyse quantitative d’électromyogramme intestinal chronique . Annales de la recherche vétérinaire, 4, 347–353. Loewe, S., 1939, “Bioassay of Laxatives on Monkeys (Rhesus) and on Lower Mammalians”. Journal of the American Pharmacology Association, 28, 427–442. Santini, R., Brard, E. and Thouvenot, J., Activité électrique de l’intestin du chien éveillé. Exploration pharmacologique simultanée par electrodes chroniques et électro intraperitoneographie. Pathologie Bioloqie, 26, 151–219–224. Santini, R., Tessier, M., Belon, P. and Pacheco, H., Incidence d’un traitement homéopathique par Cuprum 4 sur le transit intestinal de la souris: étude préliminaire. Comptes-rendus de la Société de Bioloqie de Paris, (In Press).

THE EFFECTS OF SOME REGULATORY PEPTIDES IN FEMTOMOLAR AND LOWER CONCENTRATIONS OF LYMPHATIC VESSELS (LVs) I.P.Ashmarin, T.V.Levekova, L.Ts.Sanzhieva Biological faculty, Moscow State University, Moskow, 119899, USSR The influence of regulatory peptides (RPs) on the frequency of contraction of micro lymphatic vessels in rat mesentery was recorded photometrically. The RPs were introduced intravenously or superficialy. Were found 3 peptides from 20 RPs investigated-thyroliberin (TRH), defensin and tufts in modulated the contractibility in dilution up to 10–12—10–16 mol 1 −1. Other RPs and bioamins were much less effective (103–104 times or more). Possible mechanisms of such high susceptibility of LVs to RPs are under discussion. The LVs contractility regulation is a poorly explored area. We supposed that the role of humoral regulation especially for mesenteric LVs was more important than for blood vessels (Orlov R.S. et al., 1983). Consequently the LVs susceptibility to low concentrations of regulatory compounds could be higher in comparison with blood vessels susceptibility (Levekova T.V. et al., 1989, 1990). But LVs are very difficult objects for investigation. They have very vulnerable thin walls. So it’s very difficult to prepare an viable isolated lymphatic vessel. We made our experiments on the mesentery of an alive rat. The mesentery was placed on the table of microscope equipped with an optical system for registration of the movements of the vessel walls (Chernukh A.M. et al., 1975). The regulatory compounds (peptides, bioamines) were introduced by two ways: 1) superficially on the mesentery—into the water immersion fluid (the Ringer solution) between the lens and mesentery; 2) intravenously into the tail. Having examined about 20 regulatory peptides and 5 other compounds we found three peptides which modulate the contractility in very low concentrations. Thus thyroliberin (TRH) and defensin in concentrations about 10−15−10−16 M stimulated the contractility, but tuftsin in 10−12 M suppressed the contractility (fig. 1, 2, table 1).

10 ULTRA LOW DOSES

Figure 1. The effect of surface application of 10−14 mol 1−1 thyrotrophin-releasing hormone (TRH) on the contraction of an actively contracting rat lymphatic vessel. (a) Control; arrow indicates application of physiological saline. (b) 1–4 min after TRH administration, at time indicated by the arrow, (c) 5–9 min after TRH administration, (d) 41–45 min after TRH administration.

First of all we checked the simplest possible sources of methodological mistakes: the regime of pipettes and vessels changing, the accuracy of dilution procedure (by means of radioactive analogs—up to dilution 10−14) etc. Secondly we compared the effectiveness of a different peptide and nonpeptide regulators listed in the table 1. All other tested substances were much less effective. Even classical stimulators of contraction—noradrenaline and dopamin—acted in concentrations about 10−8−10−9 M (which are close to concentrations stimulating blood vessels). It’s obvious now that TRH, defensin and tuftsin are the most powerful ever known LVs regulators. But the intriguing question about the mechanism of too low effective concentrations and doses remains unsolved. What explanation can we give to the first paradox mentioned above? We know that protein to protein interaction can be characterized by higher KD up to 10−17 −10−19. Perhaps, before binding the receptor there is the stage of binding the peptide by special protein-mediators. Such binding may be accompanied by so deep change of its conformation that the whole complex acquires more affinity to the receptor than the regulatory peptide itself. But it’s only a supposition. There is no experimental support to them despite a lot of modern literature data about the existence of different protein-carriers for regulatory peptides (similar to well known neurophisins). The second paradox mentioned above has more trustworthy explanation. The usual determinations of endogenous peptides concentrations by means of RIAmethod includes the stage of proteins denaturation by ethanol or methanol, which

REGULATORY PEPTIDES IN LYMPHATIC VESSELS 11

Figure 2. The effect of surface application of 10−11−107 mol 1−1; arrow indicates application of tuftsin or physiological saline. Similar results were obtained in experiments with intravenous injections of these peptides. Such low effective concentrations of this three peptides were a shock for us. We realized the existence of two paradoxes: 1), the KD for known peptides-receptors interactions were no less than 10−12 M; 2) the effective concentrations of these three peptides are 103−104 times less than the concentrations of corresponding endogenous peptides in body fluids (blood plasma for example).

may lead to dissociation of protein-peptide complexes. Thus we do not know the real distribution of regulatory peptides among different protein-carriers in body fluids. The real concentration of free peptides is also unknown. Perhaps, practically there are no free regulatory peptides in body fluids at all. If it is so the effectiveness of some exogenous regulators in very low doses becomes more understandable. Now, we began the new cycle of investigations trying to display a mechanism of these strong RPS action. For example we immunized the rats against TRH by means of covalent conjugates of TRH with bovin serum albumine (as antigencarrier). The effective doses of TRH for such immunized animals rises in about one hundred—thousands times. It proves that TRH is the real physiological (not artificial) regulator of LVs contraction. In medical area we began now to utilize the phenomenon of very high activity of TRH in LV regulation. Preliminary clinical experience of the TRH injections in doses about 10−20 M demonstrated their usefulness in treatment of some traumatic diseases and some forms of pancreatitis connected with lymphatic flow suppression.

12 ULTRA LOW DOSES

Table 1. The action of bioregulators on lymphatic vessels contractility.

REFERENCES Chernukh A.M., Aleksandrov P.N. and Alekseev O.V., 1975, Microcirculation, (Moscow). Khugaeva V.K., and Suchkov V.V., 1980, Influence of enkephalin on microcirculation. Bulletin du centre de cardiologie AMH USSR , 3, 1, 92–96. Lelekova T.V., Sanzhieva L.Ts. and Ashmarin I.P., 1990, Thyrotrophin-releasing hormone- a powerful stimulator of lymphatic vessel contraction in rat mesentery. Biomedical Science, 1, 99. Lelekova T.V., Romannoyckyi P.Ja., Aleksandrov P.N. and Ashmarin I.P., 1989, The effect of femto -and picomolarities concentration of Thyrotrophin-releasing hormone and tuftsin on contraction of the rat’s mesentery lymphatic vessels. Bulletin of experimental biology and Medicine, 7, 8–10.

REGULATORY PEPTIDES IN LYMPHATIC VESSELS 13

Orlov R.S., Borisov A.V. and Borisova R.P., 1983, Lymphatics (Leningrad): Nauka). Zverev M.D., 1984, The effect of enkephalins on contraction of the mesentery lymphatic vessels. In: Lymphatics (Leningrad), pp. 72–75.

MODULATIONS OF EXPERIMENTAL RAT LIVER CARCINOGENESIS BY ULTRA LOW DOSES OF THE CARCINOGENS J.De Gerlache and M.Lans Université Catholique de Louvain, Ecole de Pharmacie, Unite de Biochimie Toxicologique et Cancérologique UCL 7369, B– 1200 Bruxelles, Belgique

INTRODUCTION Carcinogenesis is a long lasting multistep process. Injection of chemical carcinogen is the most common experimental mean to initiate carcinogenesis. If initiation is weak, it is a necessary but still insufficient event; even though benign tumors may appear in the treated animals, no malignant tumors will develop within the life time of the animals. Beyond initiation various treatments do exist which speed up or increase the intensity of the carcinogenic process so that such an insufficient initiating treatment becomes full carcinogenic and malignant tumors appear within months after initiation. Such treatments which accelerate a carcinogenic process have been called promoters or more generally “positive modulators” [1]. If such treatments do exist, the hypothesis has to be accepted that experimental means might be found which slow down or reduce the intensity of carcinogenesis. In such conditions malignant tumors might no longer develop within the animal’s life span. Such a phenomen has been called “negative modulation” of carcinogenesis [1], it is the objective of cancer chemoprevention. The major targets for experimental carcinogenesis in rats, that have been used to study chemical carcinogenesis and its modulation, are the liver, the mammary glands and the colon. According to the expertise of the authors in the field of cancer modulation in hepatic carcinogenesis, metabolic and/or systemic unbalances might be determinant in this process [1]. During the last few years, both experimental and clinical results have been published which demonstrate that highly diluted drugs, prepared according to the so-called “homeopathic recipe”, might have biologically and therapeutically significant reproducible effects [2–6]. The aim of this work was to test the effects of such highly diluted drugs on chemically-induced rat liver carcinogenesis.

MODULATIONS OF EXPERIMENTAL RAT LIVER CARCINOGENESIS 15

Based on the results of preliminary experiments, a suitable protocol was designed which was adapted from that of Peraino et al. [7]. It was applied to a large number of rats sacrified sequentially during the carcinogenic process. MATERIALS AND METHODS Chemicals 2—Acetylaminofluorene (2AAF) was from Janssen Chemicals (Beerse, Belgium) and phenobarbital (PB) (Sodium salt) form Ludeco (Bruxelles, Belgique). All other chemicals were of analytical grade from either Sigma (St Louis, Ma, USA), Merck (Darmstadt, FRG) or Miles (USA). Animals and diets Male Wistar rats ICO: Wi (IOPS AF/HAN) were used. They were 22 days old at the beginning of the experiment and housed in standard conditions with 12th light/ 12th dark cycle. They were given free access to water and diet. The diet was a standard pelleted diet (AO4 from UAR, Villemoisson-sur-Orge, France). 2AAF and PB were incorpored in the diet by UAR at a concentration of 0.03% and 0. 05% respectively. Preparation of test dilutions Highly diluted solutions of PB and 2AAF were prepared by the Laboratories Boiron (Ste Foy-Lès-Lyon, France), according to the GMP. Nine successive centesimal dilutions with tridistillated water of a stock solution (1 g/l) give the final so-called 9CH dilution. Fractions of 30 ml were then vortexed mechanically before further dilution in water (20%) and final solutions were pooled and stored in glass bottles delivery to treated animals. The final concentration of PB and 2AAF in drinking water thus +/−2.10−19 M. Experimental design The experimental design is schematized in fig. 1. Rat liver carcinogenesis was induced according to a protocol slightly adapted from that described by Peraino et al. [7]. During 21 days, young wealing rats 22 days old received the carcinogenic agent (2AAF in their diet) as inhibitor. After 10 days of recovery from this intoxication period, the animals received the promoting treatment with PB (0.05% in their diet) for 12 months. At the beginning of PB treatment, animals were assigned to 3 experimental groups. The animals of first group received, in their drinking water, the solution PB 9CH in the proportion of 20%. The animals of the second group received, in the same conditions, the solution

16 ULTRA LOW DOSES

2AAF 9CH while the animals of the reference group received the solvent used to prepare the experimental dilutions and treated according to the described procedure. Between the 6th and the 19th month, groups of animals were sacrified for analysis. Moribund animals or animals presenting large and inflammed tumors of the Zymball gland were systematically sacrified for ethical reasons. Histological analysis At sacrifice, a general autopsy of each animal was performed and the liver weighed. Macroscopically visible lesions were carefully checked and longitudinal slices of approximatively 5 mm thick were cut from each lobe and from each tumor. Tissue sections were fixed in neutral formol, embedded in paraffin and stained with hematoxilin-eosin. Histological analysis was made in blind and liver lesions were classified according tot he criteria usually accepted and previously discussed at a workshop reported by Squire and Levitt [8]. Statistical analysis Statistical analysis was performed by Drs Cherruault and Guillez (Medimat, Institut Biomedical des Cordeliers, Université de Paris VI). The experimental data were examined according to the test of Bartlett described by A.Rosengard. RESULTS A preliminary experiment using small number of rats (+/−40/group) had shown that the 9CH solutions of PB or 2AAF had no effect on the percentage of rats with macroscopic liver lesions (total or malignant) but that they significantly increased the percentage of animals surviving with such lesions. In both PB 9CH and 2AAF 9CH treated groups, 100% of rats with lesions survived as compared to +/−60% in the control group. A second short-term preliminary experiment using a slightly different protocol, confirmed that neither the percentage not the size of early morphologically-altered lesions were affected by the administration of PB 9CH and 2AAF 9CH. Based on these data a large scale experiment was designed to examine, for the long term, the influence of a treatment with solution of PB 9CH or 2AAF 9CH on malignant tumor development and rat survival. It must be emphasized here that PB was given to rats at a dietary dose (+/−20 mg/rat/day) by far exceeding the amount they received in their drinking water (+/−10−19 mol/day). The overall evolution of the different experimental group was first examined. The spontaneous mortality during the whole experiment was about 20% in all groups. Most animals found dead were not included in the analysis because the carefull macroscopic analysis of their liver was often not possible. Tumors of the

MODULATIONS OF EXPERIMENTAL RAT LIVER CARCINOGENESIS 17

Zymball gland, a lesion which shows up very frequently in rats submitted to such a protocol, were observed in about 20% of the animals. The appearance of these tumors was observed from the 6th month upward and all along the experiment. The often rapid evolution of this pathology, associated with an acute inflammatory process generally led to the sacrifice of the animal. No correlation between the presence of a tumor of the Zymball gland and the presence of macroscopically detectable lesions in the liver was observed. Also no difference in the incidence of this pathology among the experimental groups observed. In the liver, small foci of cholangiofibrosis appeared progressively from the 6th month of treatment upward. Hepatocellular neoplastic lesions were also observed but, before the 9th month, those were generally detectable only microscopically. The cells in these lesions were of the types observed in rat liver carcinogenesis: clear cells, mixed acidophilic cells and foci. Neoplastic modules were rare as usual with this protocol [7]. From the 9th month upward, microscopic carcinomas, similar to those described by Pitot et al. [9] with a similar protocol, were observed. Lateron, macroscopically detectable tumors developped. At histological examination, these tumors were generally classified as hepatocellular carcinoma of type A, well differentiated. Area of cholangiofibrosis with kystic lesions were still present at this stage. A distinction was made between the animals in which macroscopic lesions of the liver were observed and the animals in which the lesions were histologically characterised as malignant. All along the experiment (Table 1), there was a difference in the number of animals presenting macroscopically detectable liver lesions the reference group and the PB 9CH group. Up to 11 months, 47% of animals examined in the reference group and 45% in the 2AAF dilution group presented macroscopic lesions while only 32% of the animals examined in the PB 9CH group presented such lesions. Between the 13th and the 19th month, macroscopically detectable lesions of the liver were observed in 82% of the animals examined in the reference group and in 60% of the PB 9CH dilution group. These differences are illustrated in Fig. 2. They were analysed statistically by applying the test of Bartlett. All samples of the PB 9CH group were significantly different from those of the reference group (p<0.05), not only on the whole but also nearly at each time point where the test was possible. Microscopically, a difference in the number of rats bearing malignant liver lesions was observed between the reference group, the group treated with PB 9CH and the group treated with 2AAF 9CH (Table 1). This difference was observed all along the experiment as illustrated in Fig. 3. All the samples were different from each other on the whole and at each point where the comparison was possible (p<0.05).

18 ULTRA LOW DOSES

Table 1. Number of rats with macroscopic and histologically detected-malignant lesions and evolution of the cumulative incidence of cancers in the 3 groups.

DISCUSSION The objective of the present experiment was essentially to determine, in a welldefined protocol, if the potential effect of highly diluted solutions of a pharmacological agent could be evaluated. Clearly, this experimental approach was undertaken independently from any underlying mechanistic or therapeutical consideration. The observation concerning the spontaneous mortality and the incidence of tumors of the Zymball gland indicates that the three experimental groups evolved parallely. The tumors of the Zymball gland were relatively frequent (about 20% of the animals) and this pathology is attributed to a known organotropism of 2AAF for this gland. These tumors developped from the 6th month in animal of all groups and thereafter all along the experiment. In the liver, the putative premalignant neoplastic lesions observed before the 9th month of treatment were generally small foci of cholangial origin and microscopic hepatocellular foci of the various types classically observed in experimental hepatocarcinogenesis: foci of clear or acidophilic cells and mixed foci. The first carcinomas observed were hepatocellular basophilic microcarcinomas similar to those described by Pitot [9] with comparable protocol. Later, tumors of a macroscopically detectable size were also observed. Histologically these tumors were hepatocellular carcinoma of type A (well differentiated in the classification reported by Squire and Levitt [8]. The number of animals in each experimental group was sufficiently high so that the differences observed between the reference group and the group treated with dilutions of PB or of 2AAF were statistically very significative. The most significative difference was observed in the number of rats developping

MODULATIONS OF EXPERIMENTAL RAT LIVER CARCINOGENESIS 19

Fig. 2.

macroscopically detectable lesions. There was indeed less possibility of error on this parameter than on the histological examination of a necessarily limited number of samples from each liver. Nevertheless, up to the 19th month, the number of animals presenting malignant tumors in the liver, including the microcarcinomas, was systematically lower both in the PB dilution group and in the 2AAF dilution group than in the reference group. Obviously, these treatments did not cure the rats nor avoided the progression to malignancy. The number of rats presenting liver alterations in the PB dilution group and in the reference group evolved parallely and the statistical analysis strongly suggested that, in the treated group, the carcinogenic process was delayed increasing latency for its development. These results are the first reported observation of a significant effect obtained with treatments of this nature established on a large experimental

20 ULTRA LOW DOSES

Fig. 3.

MODULATIONS OF EXPERIMENTAL RAT LIVER CARCINOGENESIS 21

basis. The high degree of statistical signification established was sufficient to justify their publication as a scientific fact. These results cannot be considered to be demonstrative of the activity of such high dilutions without further confirmation hopefull by other experimentalists. It should be underlined that the specifications of the treatment (nature, dilution, dosage, route of administration) were relatively arbitrarily choosen and that these parameters could probably be optimalized. Meanwhile, these results, established on a rather artificial experimental model are clearly without any medical or therapeutical signifiance. REFERENCES Cazin, J.C., Cazin, M., Gaborit, J.L., Chaoui, Boiron, J., Belon, Ph., Cherruault, Y. and Papapanayotou, C., [2], 1987, A study of the effect of decimal and centesimal dilutions of arsenic on the detention and mobilization of arsenic in the rat. Human Toxicol., 6, 315–320. Davenas, E., Poitevin, B. and Benveniste, J., [6], 1987, Effect on mouse peritoneal macrophages of orally administered very high dilutions of silicea. Europ. J.Pharmacol., 135, 313–319. Gibson, R.G., Gibson, S.L.M., Macneil, A.D and Buchanan, W.W., [4], 1980, Homeopathic therapy in rheumatoid arthritis: evaluation by a double-blind clinical trial. Br. J.Clin. Pharmacol., 9, 453–459. Peraino, C., Fry, R.J.M. and Staffeld, E., [7], 1977, Effects of varying the onset and duration of exposure to phenobarbital on its enhancement of 2 acetylaminofluoreneinduced hepatic tumorigenesis. Cancer Res., 37, 3623–3627. Pitot, H.C., Barness, L. and Goldworthy, T., [9], 1978, Biochemical characterization of stages of hepatocarcinogenesis after a single dose of diethylnitrosamine. Nature, 27, 456–458. Roberfroid, M., (in press), [1], Dietary Modulation of Neoplastic Development: Role of Fat and Fiber Content and Calorie Intake. Preventive Medicine. Shipley, M., Berry, H., Broster, G., Jenkins, M., Clover, A.L. and Williams, I., [5], 1983, Controlled trial of homeopathic treatment of arthritis. Lancet, I, 97–98. Squire, R. and Levitt, M., [8], 1975, Report of a workshop on classification of specific hepatocellular lesions in rats. Cancer Res., 35, 3214–3223. Turner, P., [3], 1980, Clinical trials of homeopathic remedies. Br. J. Clin. Pharmacol., 9, 443–444.

OVERCOMING TUMOR CELL DRUG RESISTANCE BY LOW DOSES OF RECOMBINANT TUMOR NECROSIS FACTOR AND DRUG Bonavida, B., Safrit, J., Tsuchitani, T., and Zighelboim, J. Department of Microbiology and Immunology, UCLA School of Medicine and the Jonsson Comprehensive Cancer Center, University of California at Los Angeles, CA 90024, U.S.A.

Studies in our laboratory have suggested that there exists a hierarchy of sensitivity and resistance of tumor cells to drugs, toxins, cytotoxic effector cells and cytotoxic effector molecules. Based on these studies, we have postulated that in some instances, drug resistance may be overcome by the combination of immune cytotoxic factors such as rTNF and chemo-therapeutic drugs. This hypothesis was tested in a variety of human tumor cell lines derived from the ovary, brain, lung, skin, etc. Using a combination of rTNF and drugs [e.g. adriamycin (ADR)] we were able to demonstrate that combination of rTNF and ADR reverted resistance of drug resistant lines and TNF and drug resistant lines. The concentrations of rTNF and drug used were very low as compared to the physiologic concentrations used with sensitive lines. The implications of these findings are discussed. INTRODUCTION During the last years, several advances have been made in understanding the biology of normal and malignant cells. The advent of new techniques has also advanced our understanding in the causes, prevention, and treatment of cancer. However, in the field of cancer therapeutics, the development of tumor cell resistance to conventional forms of therapy remains a major stumbling block in cancer cure. There is a definite need in the development of delective targeting cytotoxic agents to cancer cells and knowledge of mechanisms by which certain tumor cells resist the action of cytotoxic drugs and means to reverse the resistance. The introduction of biologic response modifiers to circumvent tumor cell resistance to conventional therapy has gained momentum in the last several years (Foon, 1989). The main premise of the use of biologic response modifiers is that the host immune system once activated against the tumor cells should overcome tumor cell

OVERCOMING TUMOR CELL DRUG RESISTANCE 23

resistance to conventional therapy. However, the relationship between sensitivity and resistance of tumor cells to host immune cytotoxic systems and conventional therapy has not been studied extensively. Further, overcoming drug resistance by combination treatment with drug and biological cytotoxic systems has not been considered. Recent studies from our laboratory have examined the relationship between the sensitivity and resistance of various tumor cell lines to several immune cytotoxic effector cells or factors and non-immune mediated cytotoxic agents (Bonavida, et al., 1989). Our results suggested that these drugs and immune systems may share some common pathways resulting in tumor cell killing. Tumor necrosis factor (TNF) is a polypeptide that is produced by macrophages and natural killer cells and mediates many activities one of which it is cytotoxic and/or cytotoxic to many tumor cells (Watanabe, et al., 1985). TNF has also been cloned and recombinant protein is available in large quantities for clinical trials (Pennica, et al., 1984). Because TNF-a is considered clinically alone or in combination with other agents and our findings suggesting that TNF-a and drugs may share some common mechanism of lysis, we undertook a study to examine whether treatment of tumor cells with combination of TNF-a and drugs may result in additive or synergistic activity. Further, we also examined whether the combination treatment may overcome TNF and a drug resistance of several human tumor cell lines. MATERIALS AND METHODS Tumor Cells The human ovarian cancer cell lines, PA–1, 222, A–2780, and OVCAR–3 were employed. These lines are plastic adherent cells and trypsinization is required to detach them from tissue culture flasks to make single cell suspensions. PA–1has been cultured for more than five years and 222 is a relatively fresh cell line established in our laboratory and has been cultured for less than one year. A– 2780 and OVCAR–3 were obtained from Doctor Ozols, Philadelphia. The lung carcinoma cell line 226–P59 and the melanoma line RP were obtained from Doctor Sid Golub, UCLA. Culture Medium The culture medium used consisted of RPMI–1640 (M.A. Bioproducts) supplemented with 10% heat-inactivated fetal bovine serum (GIBCO), 1% Lglutamine (GIBCO), 1% non-essential amino acids (GIBCO), 1% penicillinstreptomycin (GIBCO), and 1% fungisone (GIBCO) of a 100x stock solution.

24 ULTRA LOW DOSES

Reagents Purified human recombinant tumor necrosis factor was a generous gift obtained from Genentech, South San Francisco, U.S.A. and Smith Kline French/Beacham, Philadelphia. Cis-diamine-dichloroplatium (CDDP) was obtained from Bristol Laboratories, Div. of Bristol-Myers Company, Syracuse, New York, U.S.A., Adriamycin and 5–FU was purchased from Sigma Chemical Company, St Louis, U.S.A. Anti-TNF Antibody The hybridoma cell line was obtained from Doctor Trinchieri, Philadelphia and ascotes prepared in our laboratory (Cuturi, te al., 1987). The ascites was purified to homogeneity by protein A affinity column. The purified anti-TNF antibody (previously titrated) was added together with TNF containing medium one hour before the addition of cells to maximize the neutralizing activity. Cytotoxicity Cytotoxicity by effector cell, factors, and toxins were determined as described previously (Bonavida, et al., 1989) and Cytotoxicity by TNF and drug use done by the 51Cr release assay. Tumor cells were radiolabeled with Na2 51CrO4 (Amersham), by suspending 5×105 cells in 1 ml RPMI–1640 containing 10% FBS to which 100 µCi of 51Cr were added. The cells were incubated for 2 hours at 37°C in 5% CO2, washed three times in HBSS, and resuspended in tissue culture medium at the concentration of 2×105 cells/ml. 51Cr-labeled target cells were placed in individual wells of 96-wells round-bottomed microtiter plates (Corning Glass Works, Corning NY) and tests were done in triplicate. 0.1 ml of target cell suspension and 0.1 ml of reagent solutions (when mixed, 0.05 ml each of TNF and drugs were used) were added to each well, and incubated at 37°C in 5% CO2 air for 24 hours. After incubation, the plates were centrifuged at 200xg for 10 minutes and 100 µl of cell free supernatant was collected from each well. The 51Cr content of each supernatant was determined using a Beckman 5500 gamma counter. Maximum 51Cr released was determined from the supernatant of cells that had been lysed by the addition of 2.0% Triton and spontaneous release was determined from target cells incubated in medium alone. The mean value in triplicates was used for calculation of % specific lysis. The % specific release was calculated from the following formula: % specific release= (CPM exp−CPM spont)/(CPM max−CPM spont)×100 Flow Cytometry Tumor cell lines were trypsinized, washed, and resuspended at 2×106 cells/ml in 100 ml of fresh media (RPMI+10% FBE) or in media+verapamil at 0.5 mg/ml.

OVERCOMING TUMOR CELL DRUG RESISTANCE 25

Figure 1

100 ml of media containing 1 mg/ml DNR was then added for the noted time period to achieve a final DNR concentration of 1 mg/ml. Cells were then either immediately analyzed on an EPICS C Flow cytometer at 488 nm wavelength washed twice in a PBS and incubated at 37° for one hour in drug free media +verapamil prior to analysis. Statistics The synergistic effect of TNF and anticancer agents was statistically evaluated by the student t test, and the cytotoxicity of combination treatment was compared to the sum of cytotoxicity of each agent used alone. RESULTS I. Hierarchy of Sensitivity and Resistance of Tumor Cells to Various Cytotoxic Systems The sudden emergence of tumor cells in the body may result in either its elimination by host defense mechanisms or overcoming host response and continuous growth. Such resistant tumors are then treated with surgery, chemotherapy, radiation and combination thereof. Thus, in theory, tumor cells could succumb to one or many cytotoxic systems as diagrammed in Figure 1. Accordingly, one can reasonably assume that the acquisition of resistance by the tumor cells to cytotoxic effect by certain systems can be overriden by other nonrelated cytotoxic systems. While these assumptions focus are the basis of today’s approaches in cancer therapy, the experimental basis verifying this assumption, however, has not been studied vigorously.

26 ULTRA LOW DOSES

Thus, the relationship between the sensitivity and resistance of tumor cells to one agent with other unrelated agents was examined experimentally in a direct cytotoxicity assay. The cytotoxic effects of immune effector cells/factors, cytotoxic drugs, and bacterial toxins against a battery of human tumor cell lines of different histological types were examined recently (Bonavida, et al., 1989; Safrit, et al., submitted). The findings of these studies demonstrate that there exists a hierarchy of sensitivity and resistance of tumor cells to a variety of unrelated cytotoxic systems examined. This hierarchy is summarized for a few cell lines in Table 1 and schematically diagrammed in Figure 2. From these data, we hypothesized that there must exist a common underlying mechanism of action shared by these various cytotoxic modalities. The implications of these finding are important. For instance, it’s not possible to assume that resistance to cytotoxic drugs, for example, can be overcome by cytotoxic immune cells/ factors as is being approached in immuno-therapy. Alternatively, it is possible, under special circumstances, to overcome tumor cell resistance to two agents/ systems, by combination treatment. It is the later possibilities that were explored in the present study. Table 1. Hypothetical Sequential Pathway of Sensitivity to Drugs, Cytotoxic Cells or Factors rTNF

Macrophage

Drugs

Toxin

Macrophage

LAK

+ + + + −

+ + + + +

(superoxide) PA-1 222 222TR SKOV-3 RAJI

+ + − − −

+ + (+) −

+ + + − −

+ + + − −

II. Tumor Cell Sensitivity to Combination Treatment: Synergy observed with Recombinant Tumor Necrosis Factor (TNF-a) and Chemotherapic Drugs Studies by many investigators have examined the mechanism of drug sensivity and resistance of human cells to cytotoxic drugs used in therapy. Further, the sensivity of tumor cells to the cytotoxic effect to recombinant human TNF-a, a cytotoxic factor involved in macrophage-dependent cytotoxicity, has also been investigated by us and others. We chose to examine the effect of combination treatment of TNF-a and drug on a variety of human tumor cell lines that were either sensitive of resistant to these agents.

OVERCOMING TUMOR CELL DRUG RESISTANCE 27

Figure 2

A. Effect on Tumor Cells Sensitive to TNF and drug The cytotoxic effect of TNF and adriamycin (ADR) was tested on several cell lines and one representative experiment using the ovarian carcinoma cell line PA-1 is shown in Figure 3 and 4. The addition of a subtoxic concentration of ADR (0. 01 mg/ml) to various concentrations of rTNF was tested for cytotoxicity in an 18h 51Cr release assay. The results in Figure 3 show clearly that concentrations of rTNF and ADR that are not cytotoxic by themselves exert significant activity and synergy was also obtained at toxic concentrations of TNF. The specificity of the synergistic effect was tested by the addition of anti-TNF antibody which neutralized the TNF affect and the cytotoxicity obtained reflected that mediated by the drug alone (Figure 4). These results show clearly that TNF-a and ADR used in combination exert a synergisitc cytotoxic activity against the TNF and drug sensitive line PA-1 and other similarly sensitive lines tested. Table 2. Synergy with TNF and Adriamycin* ADR (ug/ml)

*Cell Line 222TR (Ovarian Carcinoma)

28 ULTRA LOW DOSES

Figure 3

Figure 4

B. Effect on Tumor Cells Resistant to TNF and Sensitive to Drug We next examined the effect of TNF and drug on an TNF resistant and drug sensitive ovarian carcinoma cell line 222. The TNF resistant subline 222TR was selected following culture of 222 in the present of TNF for several weeks. The TNF resistant variant was resistant to high concentrations of TNF as shown in Figure 5. The effect of TNF and ADR on 222TR was examined and the results

OVERCOMING TUMOR CELL DRUG RESISTANCE 29

Figure 5

are shown in Table 2. Clearly, the combination of TNF and ADR resulted in cytotoxicity that was synergistic. These results show that TNF resistance is overcome by combination treatment and suggest that the defect in TNF resistance is recessive. Table 3 Synergy with TNF and adriamycin against TNF and ADR resistant SKOV-3 ovarian carcinoma cell line TNF

5×10−10 5×10−9 5×10−8

Adriamycin (ug/ml) % cytotoxicity

0 2 6

100

101

0 2 8 24

2 12 12 35

C. Effect on Tumor Cells Sensitive to TNF and Resistant to Drug The human ovarian carcinoma cell line A2780 is relatively resistant to the cytotoxic effect of ADR (01.1 mg/ml). The effect of combination TNF and ADR was tested for cytotoxicity (Figure 6). Clearly, the cell line is sensitive to TNF and the mixture resulted in significant synergy even at doses of TNF that were subtoxic. These results indicate that drug resistance can be overcome by combination treatment and that the drug resistance is not a stable phenotype.

30 ULTRA LOW DOSES

Figure 6

D. Effect on Tumor Cells that are Resistant to both TNF and Drug The next series of experiments were undertaken to examine the effect of TNF and drug in overcoming resistance to both agents when used alone. The effect of TNF and ADR on the human ovarian carcinoma cell line SKOV-3 is shown in Table 3. Clearly, a significant cytotoxicity is observed with combination treatment under conditions that neither agent alone is cytotoxic. Synergy is also observed with augmented cytotoxicity as the concentrations of TNF or ADR are increased. These results demonstrate clearly that cells resistant to high concentrations of drugs can be rendered sensitive to low concentrations of TNF and drug. Overcoming drug resistance in cancer therapy is a major obstacle in the treatment of recurrent cancer and metastases. The results with SKOV-3 were not unique to the cell line as other resistant cell lines behave like SKOV-3. The findings with the ovarian carcinoma cell line are summarized in Table 4. The effect of TNF and ADR is shown in Table 4A whereby significant synergy is obtained with low concentrations of TNF and ADR. The synergy seen with ADR was not restricted to ADR as 5 FU (Table 4B) and TNF (Table 4C) show also synergy with TNF. These findings demonstrate clearly that TNF and drug can overcome resistance of ovarian carcinoma cell lines. Table 4(A). Synergy with TNF and Adriamycin* ADR (ug/ml) 10−1

100

101

1

2

7

3 1 1 2

3 2 2 4

8 9 9 18

Cytotoxicity TNF (M) 5×10−19 5×10−17 5×10−15 5×10−13

2 0 0 1

OVERCOMING TUMOR CELL DRUG RESISTANCE 31

10−1

100

101

1

2

7

5 7 9

8 14 18

39 44 43

Cytotoxicity TNF (M) 5×10−11 5×10−10 5×10−9

4 7 8

*Cell Line OVCAR 3 (Ovarian Carcinoma) Table 4(B). Cell Line OVCAR 3 5 FU (ug/ml)

TNF (M) 5×10−12 5×10−11 5×10−10 5×10−9

4 7 10 12

10−1

102

103

2

5

4

11 17 22 24

19 29 34 34

15 24 31 33

10−1

100

101

1

3

13

6 10 13 14

7 11 15 15

30 35 39 40

Table 4(C). Cell Line OVCAR 3 CDDP (ug/ml)

TNF (M) 5×10−12 5×10−11 5×10−10 5×10−9

4 7 10 12

We next examined whether the synergy was only seen with ovarian carinoma or can also be obtained with tumors of different histological types. We examined the effect of TNF and ADR on lung carcinoma (226–159) (Table 5) and melanoma RP (Table 6) cell lines. The results show synergy at concentrations of drugs that were not cytotoxic. The findings demonstrate that combination of TNF and drug results in synergy cytotoxicity with TNF and drug resistant lines of various histological types.

32 ULTRA LOW DOSES

Table 5. Synergy with TNF and Adriamycin Adriamycin (ug/ml) 10−2

10−1

100

0

0

9

0 0 0 0 0

0 1 8 1 4

15 15 16 23 41

10−2

10−1

100

101

TNF (M)

3

5

23

23

5×10−16

3 9 6 1 11

11 11 6 10 8

10 9 13 8 6

43 37 31 33 36

TNF (M) 5×10−16 5×10−15 5×10−14 5×10−13 5×10−12

1 0 0 1 0

* Lung Carcinoma (226–P59) Table 6. Synergy with TNF and Adriamycin Adriamycin (ug/ml)

5×10−15 5×10−14 5×10−13 5×10−12 * Melanoma (RP)

III. Examination on the Possible Mechanisms Involved in Overcoming Drug Resistance by TNF and drug The acquisition of drug resistance by tumor cells has been shown in many instances to result in the induction of genes for multiple drug resistance (MDR) and the overexpression of a glycoprotein 170. The glycoprotein functions as an efflux pump, and in resistant cells the drug is pumped out rapidly from the cell. We tested the effect of TNF and daunorubicin (DNR) on a drug resistant human ovarian cells ADIO, which expresses the MDR phenotype. Analysis was done by flow cytometry as the drug DNR is fluorescent and facilitates the measurement of influx and efflux of drug from the tumor cells. A representative experiment is shown in Figure 7. The figure shows that the ADR line AD10 does not retain as much DNR as the parental sensitive line A2780. The drug is pumped out rapidly from the AD10 as a function of time. Following treatment of AD10 with TNF and ADR, the drug efflux was not reduced (Figure 8). These findings demonstrate that treatment with TNF and ADR under conditions that reverse resistance do not affect the drug uptake and efflux. These findings suggest that

OVERCOMING TUMOR CELL DRUG RESISTANCE 33

Figure 7

Figure 8

the synergy seen between TNF and ADR may be mediated by mechanisms post the influx/efflux reached equilibrium. DISCUSSION We have presented evidence which demonstrates that treatment of several human tumor cell lines with combination rTNF and drugs results in synergistic cytotoxic activity. Synergy is seen with subtoxic concentrations of TNF and drug. The synergy obtained with various drugs such as adriamycin, cis-platium, and 5-FU. Tumor cells of different histologic types namely ovarian carcinoma, lung carcinoma and melanoma were sensitive to combination treatment. Noteworthy, synergy can be obtained with subtoxic low concentrations of TNF and drug.

34 ULTRA LOW DOSES

Furthermore, tumor cells that were resistant to either TNF or drug or both agents were also sensitive to combination treatment. These findings provide a new mechanism by which resistance of tumor cells to chemotherapeutic drugs may be overcome and suggest that it may have bearing for its application in clinics in the treatment of refractory tumors or metastases. Initially, our studies focused on ovarian carcinoma cell lines and we chose to investigate two anti-cancer drugs, adriamycin and cis-platium because they are frequently used in therapy of avarian cancer (Khoo, et al., 1986). TNF and anticancer drugs together at low concentrations resulted in a significant synergy and enhancement of tumor cell lysis as detected in an 18–24h 51Cr release assay. We notice a pronounced 1–2 log dose shift when TNF and drug were used in combination. The synergistic effect obtained was specific to TNF as anti-TNF antibody abrogated the effect and the use of a different cytokine (e.g. IL-2) had no effect (data not shown). The findings that TNF and drug reverse drug resistance of various tumor cells strongly suggest that either the drug enhances the cytotoxic effect of TNF or alternatively TNF overcomes drug resistance. Also possible is the fact that both agents complement each other at a site that is blocked in the pathway of each. The acquisition of drug resistance by tumor cells has been the subject of many investigations (pastan and Gottesman, 1987) and several undergoing mechanisms have been found (Moscow and Cowan, 1988). Studies by Alexander et al. (1987) have reported that TNF enhanced the cytotoxicity of several anti-cancer drugs such as adriamycin, actinomycin D, teniposide, and etoposide all targeted at DNA topoisomerase II. They examined the TNF sensitive cell line 1929 fibroblasts and did not examine other cell lines. The L1929 fibroblasts were sensitive to the anti-cancer drugs used. These authors did not find that cis-platium synergises with TNF. Our findings has extensed these studies to show that not only synergy is observed with drug sensitive lines but has also been observed with high resistant lines. The role of topoisomerase II in these synergistic effects deserves attention and further study. Ling and his co-workers, using Chinese hamster ovary cells made resistant to colchicine, were the first to show that MDR was associated with decreased intracellular drug accumulation and the first to identify the presence of an 170 kd glycoprotein involved in transport of drugs (Kartena, et al., 1983). It was possible that under conditions whereby TNF and ADR show synergistic activity, a modulation of the drug transport may explain in part the reversal of the drug resistance. We explored this possibility by testing the ADR resistance subline following treatment with ADR and drug. Under the conditions tested, we were not able to show any detectable effect on the efflux of DNR from the resistant tumor cells. These results, while preliminary in nature, suggest that the synergistic effect seen may be taken place at a site post transport of drug. The use of TNF and anti-cancer agent combination has not been used clinically as yet. Our findings that TNF and anti-cancer drugs overcome drug resistance suggest that further studies are needed to apply it clinically. Further,

OVERCOMING TUMOR CELL DRUG RESISTANCE 35

one of the major problems with chemotherapy is the severe toxicity of anti-cancer drugs when used in high concentration. Our findings showing that low doses of drugs are sufficient to mediate cytotoxicity when combined with TNF and suggest strongly that a possible anti-cancer therapy may be achieved by continuation treatment with minimal toxicity or side effects. ACKNOWLEDGEMENTS Supported in part by a grant from the Boiron Research Foundation (Lyon, France) and in part by a grant from the Concern Foundation of Los Angeles, USA. REFERENCE Alexander, R.B., Nelson, W.G. and Coffey, D.S., 1987, Synergistic enhancement by tumor necrosis factor of in vitro cytotoxicity from chemotherapeutic drugs targeted at DNA topoisomerase II. Cancer Bonavida, B., Tsuchitani, T., Safrit, J. and Zighelbolm, J., 1989, Mechanism of target cell lysis by cytotoxic cells, factors, and drugs. In: Anti-Cancer Drugs, Vol. 191, edited by H.Tapiero, J.Robert, and T.J.Lampidis (Colloque INSERM or/John Libbey Eurotext Ltd.), pp. 113–128. Cuturi, M.C., Murphy, M., Costa-Gioni, M.P. Weiswan, R., Perussia, B. and Trinchieri, G., 1987, Independent regulation of tumor necrosis factor and lymphotoxin production by human peripheral blood monocytes. J. Exp. Med., 165, 1581–1586. Foon, K.A., 1989, Biological response modifiers: The new immunotherapy. Cancer Research, 49, 1621–1639. Khoo, S.K., Hurst, T., Webb, M.J. and Mackay, E., 1986, Short-Term in vitro chemosensitivity testing of tumors of the ovary, cervix and uterus. Aust NZ J.Obstet. Gynaecol., 26, 288–294. Moscow, J.A. and Cowan, K.H., 1988, Multidrug resistance. J.National Cancer Institute, 80, 14–20. Pastan, I. and Gottesman, M.M., 1987, Multiple drug resistance in human cancer. N. England J. Medecine 316, 1388–1393. Pennica, D., Nedwin, G.E., and Hayflick, J.S., et al., 1984, Human tumor necrosis factor. Precursor structure, expression and homology to lymphotoxin. Nature, 312, 712–729. Safrit, J.T., Tsuchitani, T., Zighelbolm, J. and Bonavida, B., 1991, Is there hierarchy of sensitivity and resistance of tumor cells to cytotoxic effector cells, cytokines, drugs, and toxins? Submitted for publication. Watanabe, N., Niitsu, Y., and Neda, H., et al., 1985, Antitumor effect of tumor necrosis factor against various primarily cultured human cancer cells. Jpn, J. Cancer R. Res., (Gann), 76, 1115–1119.

EFFECT OF ACETYLSALICYLIC ACID AT ULTRA LOW DOSE ON THE INTERACTION PLATELETS/VESSEL WALL Lalanne M.C., de Sèze O., Doutremepuich C. Laboratoire d’hématologie, Faculté de Pharmacie, 3 place de la Victoire, 33076 Bordeaux cedex France.

INTRODUCTION Acetyl salicylic acid (ASA) is an old drug with always new applications. Since its discovery in 1853, numerous studies have been performed in order to identify its mechanism of action, its therapeutical potential or the right administrated doses. Discussion about the doses comes from an hypothetic balance system between the platelet prostaglandins (TXA2) and endothelial prostaglandins (PGI2). TXA2 has a proaggregant and vasoconstrictor effects (Moncada et al., 1979) sensitive in non reversible manner to low dose, and PGI2 has antiaggregatory and vasodilatator effects, which are inhibited in dose-dependent and reversible manner. Recent publications (Antiplatelets Trialist’s Collaboration, 1988) have shown that very small doses (20–40 mg per day) are sufficient to abolish completely the TXA2 avoiding the toxic effect of ASA. At low and very low doses, ASA used in arterial pathology reduces 15% of vascular mortality and 30% of thrombotic events with reduced side effects. However, no study have been performed to investigate action of ASA under a dose of 1mg/day. So, the aim of the present study was to evaluate the action of ASA treatment at ultra low dose on the effects of a vascular fragment on platelets. MATERIAL AND METHODS Drugs ASA dilution (batch n°5381, Boiron, France) used in this study was prepared as follows: 1g of pure, finely powdered ASA was suspended in 99 ml of alcohol (70°); the solution was vigourously shaken and gave the first dilution.

EFFECT OF ASA ON PLATELETS AND VESSEL WALLS 37

One milliliter of this dilution was then mixed in 99 ml of distilled water and vigourous shaken to obtain the second dilution. The process was repeated until the last dilution, i.e the fifth dilution, which contained 334.109 molecules/ milliliter solution. Michaelis buffer (pH 7.3—batch 6D197, Stago France) was used for the incubation. Isotonic saline solution (NaCl 0.9%, Delmas perfusion, France) was used too. Aggregometry One the most commonly used approaches to study anti-aggregatory drugs has been the testing of platelet responsiveness to aggregating-inducing stimuli in vitro. The model chosen for the study was the optical method described by Born (Born et al., 1963). Platelet aggregation was performed in a Chrono-LogR 500 VS aggregometer (Coultronics France). Blood was collected on healthy volunteers and drawn into a plastic tube containing trisodium citrate 3.8% (1 volume for 9 volumes of blood). Platelet count was performed using a Thrombocounter. CR (Coultronics France). Average platelet count was 197000± 44050/mm3 and average hematocrit, 38±2.8%. Platelet Rich Plasma (PRP) was obtained by centrifugation of whole blood during 10 minutes at 1000 r/min. The temperature of the aggregation module was setted to 37°C and a nickel stir bar was introduced into the test tube. The stirring speed was adjusted to 1100 r/min. Before addition of the aggregating agent (collagen—Stago France, 20 µg/ ml PRP final concentration), the passage of the light beam through the stirred platelets produced oscillations in light transmission. These oscillations were characteristic of disc-shaped platelets. Upon addition of aggregating reagent, the platelets aggregated and light transmission increased. At the peak of platelet aggregation, the large platelet clump interrupted the light beam and caused large oscillations in light transmission through the stirred suspension. One could determine the rate of platelet aggregation (velocity, expressed in % min) by measuring the tangent to the steepest slope of the aggregating tracing. The extent of platelet aggregation was represented by the maximum change in light transmission expressed in %. The delay between addition of reagent and the beginning of increase in light transmission determined the response time (expressed in seconds) The changes in light transmission were recorded and data were automatically analyzed using the MERAPR software (Coultronics France). Preparation of vascular fragments Twenty human fresh vascular fragments non used for aorto-coronary grafts were taken for this study. They were dissected free of fat and connective tissue. After

38 ULTRA LOW DOSES

carefully wash, vessels were cut into 1 cm length fragments and stored at 4°C before use. Protocols This study aimed to investigate the action of ASA in presence of vascular fragment on its ability to inhibit or potentiate platelet aggregation in PRP. Five series were established after preliminary studies and corresponded to 3 test groups and 2 controls. In the first test (Test 1), 100 µl of Michaelis buffer and 20 µl of ASA were added to 300 µl of PRP before the test. In this trial, ASA was directly used on platelets in order to determine if it presents or not an action on normal platelets. In second test (Test 2), a vascular fragment was plunged in 500 µl Michaelis buffer; incubations were performed in plastic tubes under stirring at 37° C during 10 minutes. Then, 100 µl of incubation medium and 20 µl ASA were added to 300 µl of PRP. In this series, ASA was used after incubation to detect a direct modification of vessel effects on aggregation in the tested tube. In third test (Test 3), 100 µl of ASA was mixed to 400 µl Michaelis buffer at the time of incubation of the fragment; hundred microliters of the medium was then injected in the tested tube containing 300 µl PRP. Presence of ASA at the time of incubation was to see if it may act on vessels release or not. These 3 tests were compared to 2 controls constituted as follows: In control 1, 100 µl of Michaelis buffer and 20 µl of saline solution were added to 300 µl of PRP before the tests. Michaelis buffer was used alone in order to determine basal platelet aggregation. In control 2, a vascular fragment was plunged in 500 µl Michaelis buffer; incubations were performed in plastic tubes under stirring at 37°C during 10 minutes. Hundred microliters of this incubation medium replaced the Michaelis buffer of control 1 and were tested on PRP as described above. This control verified the inhibitory effect of vascular fragment on platelet aggregation. In all case, the reacting medium was always preincubated 2 minutes before addition of aggregating reagent. Statistics Statistical analyses were carried out on PCSMR software (Deltasoft France). Results were expressed as mean±1 standard deviation of the mean. Student’s paired t- test was used to determine the significance of difference between means of control 1 by comparison with the others. A p<0.05 was taken as significant step. The three parameters of platelet aggregation were thus examined: change in curve amplitude, change in curve velocity, variation in latency time.

EFFECT OF ASA ON PLATELETS AND VESSEL WALLS 39

RESULTS AND DISCUSSION Curve amplitude Results for curve amplitude are summarized in table 1. Michaelis buffer was used as control of the dilution of the reacting medium on platelet aggregation and gave the basal aggregation. When platelets were in presence of the incubation medium of vascular fragment, aggregation was significantly decreased (p<0.001). No modification of basal aggregation could be observed after addition of ASA directly on platelets (test 1, p=NS). These results are confirming previous studies preformed ex-vivo in man (Doutremepuich et al., 1987) Table 1. Mean curve amplitude (n=20) of platelet aggregation expressed in % and results of Student paired t-test between basal aggregation (control 1) and the other trials. SERIES

MEAN AMPLITUDE ±1 S.D (%)

STUDENT t

t.TEST p

CONTROL 1 CONTROL 2 TEST 1 TEST 2 TEST 3

76.68±28.18 42.70±24.37 73.92±30.67 41.23±22.79 61.67±26.47

– 7.6987 0.1532 7.0317 3.8681

– <0.001 NS <0.001 <0.01

Presence in incubation medium of vascular factors inhibiting platelet aggregation can be assumed. In fact, Nordoy et al. (1978) have demonstrated that endothelial cells have a spontaneous inhibitory effect on collagen-induced platelet aggregation. This inhibition of platelets aggregation has often been observed in vivo by action of both PGI2 and EDRF (Hogan et al., 1988). PGI2 acts by stimulation of adenylate cyclase and augmentation of CAMP production. Endothelium, which is the major vascular site of production of prostacyclin, is involved in this phenomenon. It generates another factors modulating the interaction between platelets and vessel wall: EDRF (Radomski et al., 1987) is an humoral agent, which stimulates guanylate cyclase and increases CGMP levels (Furlong et al., 1987) and Endothelin (Thiemermann et al., 1989) is a recently characterised peptide, inhibiting ex-vivo platelet aggregation in rabbit. When the vein fragment was incubated in presence of ASA in incubation medium, an antagonization of this effect appeared and a recovery of platelet aggregation could be observed. A very statistically significant step (p<0.001) was reached when test 3 was compared with test 1 or 2. However, this recovery remained incomplete as shown by the statistical difference (p<0.01) between test 3 and control 1.

40 ULTRA LOW DOSES

Curve velocity: Results for curve velocity are summarized in table II. As observed with curve amplitude, addition of vein incubation medium (control 2) decreased curve velocity too (p=0.001). Table 2. Mean velocity (n=20) of platelet aggregation expressed in % and results of Student paired t-test between basal aggregation (control 1) and the other trials. SERIES

MEAN VELOCITY ±1 S.D (%/min)

STUDENT t

t.TEST p

CONTROL 1 CONTROL 2 TEST 1 TEST 2 TEST 3

34.32±24.12 20.17±24.94 31.76±23.38 20.39±21.025 28.67±24.15

– 4.0528 0.0396 5.0883 1.6404

– =0.001 NS <0.001 NS

S.D: Standard deviation, NS: not significant

Addition of ASA directly in PRP (test 2) was not able to modify these results. However, the presence of ASA at the time of incubation of the saphenous fragment lead to a complete recovery of this parameter (test 3 vs. control 1; p=NS). Latency time Means (±1SD) for latency time as well as results of student t-test are in table III. As shown in table III, no great variation of the latency time could be observed between the series. Only test 3 appeared statistically shortened. Table 3 Variation in curve latency time (n=20) of platelet aggregation expressed in sec. Student ttest between control 1 and the different other mixtures. SERIES

MEAN LATENCY TIME (sec)±1 SD

STUDENT t

t.TEST p

CONTROL 1 CONTROL 2 TEST 1 TEST 2 TEST 3

93.90±41.82 91.98±50.39 98.70±50.39 88.84±52.58 78.01±31.26

– 0.2448 0.2950 0.8280 2.8074

– NS NS NS <0.05

S.D: Standard deviation NS: not significant

Effects of vascular factors occur predominantly on the two majors parameters of platelet aggregation: curve amplitude and velocity.

EFFECT OF ASA ON PLATELETS AND VESSEL WALLS 41

These effects are markedly antagonized by ASA. Several studies have reported that ASA used at low doses presents a specifically effect directed on platelet thromboxane A2 synthesis, without action on vascular cyclo-oxygenase (Janes et al., 1985). However at subthreshold concentrations, ASA has no action on platelets, but rather acts on vessel wall. For instance, it is no yet known if the main mechanism really is the inhibition of vascular factors production or if ASA acts differently. So, further studies have to be performed to evaluate accurately whether way use ASA to modify vascular responses. CONCLUSION Vein fragment leds to an inhibition of platelets aggregation, which is markedly reversed after incubation in presence of ASA at ultra low doses. So, ASA at ultra low doses induces an antagonization directed against the vessels wall, by a pathway which remained to be clarified. REFERENCE Antiplatelets Trialist’t Collaboration, 1988, ASA at ultra low doses induces an antagonization directed against the vessels wall, by a pathway which remained to be clarified.

REFERENCE Antiplatelets Trialist’t Collaboration, 1988, Secondary prevention of vascular disease by prolonged antiplatelet treatment. Brit.Med.J., 296, 320–331 Born G.V.R., Cross M.J., 1963, The aggregation of blood platelets. J. Physiol. London, 168, 178–195. Buchanan M., Crozier G, Haas T., 1988, Fatty acid metabolism and vascular endothelial cell. Haemostasis, 18, 360–375. Doutremepuich C, Bousquet F, Masse A, Kuttler C, Quilichini R., 1984, Action de la prostacycline sur l’extension d’un thrombus veineux expérimental. Ann. Pharm. Française, 42, 467–471. Doutremepuich C, Pailley D, Anne M.C., De Seze O, Paccalin J, Quilichini R, 1987, Template bleeding time after ingestion of ultra low dosage of acetylsalicylic acid in healthy subjects. Preliminary study. Thromb. Res, 48, 501–504. Doutremepuich C, De Seze O, Le Roy D, Lalanne M.C, Anne M.C, 1990, Aspirin at very ultra dosage in healthy volunteers: effects on bleeding time, platelet aggregation and coagulation. Haemostasis, 20, 99–105. Furlong B, Henderson A, Lewis M, Smith J, 1987, Endothelium-derived relaxing factor inhibits in vitro platelet aggregation. Br.J.Pharmacol, 90, 687–692. Hogan J, Lewis M, Henderson A, 1988, In vivo EDRF activity platelet function. Br.J.Pharmacol, 94, 1020–1022.

42 ULTRA LOW DOSES

Janes M, Walsh J, 1985, Effect of aspirin and alcohol on platelet thromboxane synthesis and vascular prostacyclin synthesis. Thromb.Res, 39, 587–593. Moncada S, Vane J, 1979, Arachidonic acid metabolites and the interactions between platelets and blood-vessel walls. N.Enql.J.Med., 300, 1442–1447. Norvoy A, Svensson B, Schroeder C, Hoak J.C, 1978, The inhibitory effect of aspirin on human endothelial cells, Thromb.Haemost, 40, 103–110. Radomski M, Palmer R, Moncada S, 1987, Release of nitrite acid oxyde accounts for the anti-platelet activity of endothelium derived relaxing factor (EDRF): Interaction with Prostacyclin. Thromb.Res, Supply. VII, abstr. 6, pp.8. Thiemermann C, Lidbury P, Thomas R, D’Orléans-Juste P, Antunes E, Walder C, 1989, Endothelium inhibits ex vivo platelet aggregation in the rabbit. Europ.J.Pharmacol., 158, 181–182. Weksler B, Pett S, Alonso D, Ritcher R, Stelzer P, Subramanian V, Tack-Goldman K, Gay W, 1983, Differential inhibition by aspirin of vascular and platelet prostaglandin synthesis in atherosclerotic patient. New England Journal of Medicine, 308, 800–805.

II BIOPHYSICS

A QUANTUM CHEMICAL STUDY OF SOME MODEL ANTI-INFLAMMATORY COMPOUNDS: THE PREFERRED CONFORMATIONS AND THEIR ELECTROSTATIC SIMILARITIES Apurba K.Bhattacharjee Department of Chemistry, Lady Keane College, Shillong—793 001, India and S.N.Bose National Centre for Basic Sciences, Calcutta—700 064, India The stable conformations and the electrostatic properties of phenyl glycolic, αthienyl and β-thienyl glycolic acids are studied using both semi-empirical and ab initio molecular orbital methods. The minimum energy conformations of these molecules are almost identical. The molecular electrostatic potentials of these compounds are similar to each other in terms of potential contours and similarity indices. This may account for the bioisosterism of the compounds. The molecular similarities between these compounds in terms of steric and electrostatic properties, which are derivated from quantum chemical methods, may help to understand the universality of the active structures and aid in the designing of these benactyzine analogues as new anti-inflammatory agents. INTRODUCTION Substituted α-phenyl propionic or glycolic acids and their analogues are known to constitute a large family of potential anti-inflammatory agents (1). Although the success of chemotherapy in the treatment of a variety of inflammations including some kind of cancer (2) is well known, most of the chemotherapeutic agents in clinical use at present are largely ineffective against solid tumours. Therefore, continued interest is focused on the structures of α-substituted phenyl (or heterocyclic analogues) propionic or glycolic acids which are reported to be highly active against a variety of inflammations including transplantable solid tumours in mice (3, 4). The observations of Atassi et al. (3) is of specific significance as they found very striking variations in the activity of the molecules with relatively small variations in their positions and nature of the substituents. These observations suggest that stereoelectronic factors might be very important in determining their activities as well as differences in conformations between the different compounds. Although the positions around the asymmetric α-carbon in these molecules have shown to dramatically

ANTI-INFLAMMATORY COMPOUNDS 45

Figure 1

influence the pharmacological profile, little is known regarding the stereoelectronic features of these compounds. In recent years, molecular electrostatic potentials (MEPs) have proven to be powerful tools for characterizing the essential electronic features in the molecules and their stereoelectronic complementarity with receptor sites (5). Because receptors recognize stereoelectronic effects and not atoms per studies of two or three dimensional MEPs have been very effective in characterizing active sites in molecules from an electronic point of view. Thus in continuation of our studies on drug molecules (6, 7) and in an attempt to better understand the “structurefunction” relationships in these molecules, phenyl, α-thienyl and β-thienyl glycolic acids were investigated as model compounds using quantum chemical methods. The key features of these molecules were identified via the computation and display of molecular electrostatic potential maps which would predict that several other such substituted molecules should also have similar stereoelectronic properties. COMPUTATIONAL STRATEGY These molecules are relatively large, containing at least twenty atoms, therefore semiempirical SCF techniques were used first and then the salient features of the conformational maps were checked via ab initio calculations. A complete geometry optimization for each of the molecules was carried out. The potential energy surfaces of the phenyl, α-thienyl and β-thienyl glycolic acids were calculated as function of dihedral α(0(6)–C(5)–C (2)–0(1)) and β(C(9) or sulfur atom–;C(8)–C(5)–C(2)) angles (Figure 1) since the carboxyl group and phenyl (or thienyl) ring chosen were planar. The basis sets used were minimum STO-3G for phenyl glycolic acid and STO-3G* for α and β-thienyl glycolic acids. MEPs were calculated on a Connolly surface around the atoms in the molecules, using the charge distributions calculated from the molecular SCF wave function. The qi qj/r algorithm (8) was used to calculate the MEPs in order to save computer time. Most of the calculations were carried out in a VAX 11/

46 ULTRA LOW DOSES

780 computer at the “Institut de Topologie et de Dynamique des Systemes de l’Université Paris 7”. The GEOMO (9) and Gaussian 82 (10) packages were used for the calculations. RESULTS AND DISCUSSIONS (a) Phenyl glycolic acid: the conformational map of the compound at the CNDO/ 2 level of calculations is shown in Figure 2. The energy profile for rotation of the carboxyl group shows that the minimum energy conformation is the one with the hydroxyl group at the α-carbon in the same plane as the carboxyl group. The preferred geometry indicates a better stabilization by hydrogen bonding with the carboxyl (Figure 3A) than with the hydroxyl group (Figure 3B), the difference in energy between the conformers is about 2 kcals per mole. The net charge densities on the hydroxyl hydrogens (which are hydrogen bonded) are 0.204 for A and 0.215 for B, respectively, in agreement with the well known documentation of stronger bonding in A i.e., with the carbonyl oxygen. The rotation of the ring (i.e., the dihedral angle β) shows a global minimum at 60°, the phenyl ring and the Cα -H bond are coplanar. The other interesting feature of the energy profile is the stronger barrier to rotation about α, (3 kcals) when α= 180° than α=0° where the barrier was found to be 2 kcals per mole. This difference could be attributed to the probable interaction between the pi orbitals of the phenyl ring and the lone pairs of the hydroxyl oxygen rather than with the carbonyl oxygen of the carboxyl group. (b) α-thienyl glycolic acid: the preferred conformation for this molecule was found to be at α= 0° and β=120° i.e., the ring in the plane with the Cα -OH bond (the sulfur atom of the ring is nearer to the Cα -hydroxy moiety). Like the preceeding compound, a local minimum was also found at α=180° with an energy 3.7 kcals greater than the preferred conformer. Rotation about β (the rotation of the thienyl ring) was observed to be highly flexible, the difference in energy between the highest and the preferred conformer being 1.0 kcal. (c) β-thienyl glycolic acid: the conformational map of the compound is shown in Figure 4. The energy profile suggests that the preferred conformation is identical as the α-thienyl glycolic acid, i.e., α= 0° and β=120°. However, the barrier about α-axis was found to be 2.7 kcals and a local minimum of 0.6 kcals was observed at α=180°. The comparative flexibility of rotation about β-axis for both the thienyl compounds probably accounts for a balance between the steric factors due to bulkier sulfur atoms and the electronic factors due to sulfur-oxygen interactions. Electrostatic Potentials Molecular electrostatic potential maps (MEPs) were generated using ab initio level (STO-3G & STO-3G*) quantum chemical calculations which were carried

ANTI-INFLAMMATORY COMPOUNDS 47

Figure 2

out for the conformational studies. The interaction of a positive point charge with a molecule can be written in the LCAO framework in the form of equation (i) ....... (i) in which the R(r) term corresponds to the nuclear repulsion and A(r) to the electronic attraction. Pmn is the element of the density matrix associated to the AO’s. The attraction integral between the AO’s ( Ψm& Ψn) orbitals and a positive charge is given by equation . . . . . . . (ii) For medium and large size molecules the number of these integrals to be calculated is very large. This number could be reduced using Mulliken’s approach for which each integral is expressed as a function of the overlap matrix Smn between the AO’s Ψm and Ψn as

48 ULTRA LOW DOSES

Figure 3

(m/r−1/n)=1/2 Smn [(m/r−1/m)+(n/r−1/n)] Combining (i) and (iii),

. . . . . . . (iii)

....... (iv) The “gross overlap population” Dmm of the ADm, that is obtainable from the eigenvectors of the Fock matrix of the SCF-LCAO calculations, is given by equation (v) . . . . . . . (v) The MEP uses the orthogonalised MO functions (LCAO of the STO functions) for the optimized conformations of the energy minimum coming from the STO-3G & STO-3G* level calculations.

ANTI-INFLAMMATORY COMPOUNDS 49

Figure 4

The MEPs have been calculated for all the molecules and examined on a DEC VT290 colour graphics terminal. The MEPs were obtained with a slightly modified version of DENPOT (QCPE 360) and were generated in planes parallel

50 ULTRA LOW DOSES

Figure 5

to the aromatic ring, viewing compounds from above and below the aromatic ring plane for each compound. Although potentials were coded in various colours, the MEPs presented here are in black and white. The MEP minima in all the compounds are located near the carbonyl oxygen and the hydroxyl oxygen of the carboxyl group. The other minima near the glycolic hydroxy oxygen are noticeably affected due to hydrogen bonding with the carbonyl oxygen. The MEPs of all these compounds are found to be similar to each other in terms of their contour are found to be similar to each other in terms of their contour patterns (Figure 5). Stereo drawings of electrostatic potentials projected onto the van der Waals surfaces of phenyl glycolic acid is presented in Figure 6.

ANTI-INFLAMMATORY COMPOUNDS 51

Figure 6

Electrostatic Similarities The electrostatic molecular similarity of two molecules, X & Y may be quantified by comparing their electrostatic potentials, ex & ey and calculating an

52 ULTRA LOW DOSES

index of similarity, Rxy, as shown in the following equation:

The index Rxy takes values in the range −1 to 1, with Rxy=1 indicating perfect similarity of the species being compared. The concept of simularity index in terms of electron densities introduced by Carbo et al (11) is equivalent to this definition. The parameter Rxy could be computed using a program which calculates the similarity of two molecules from their electrostatic potentials within a distance of 3 A from the van der Waals edge of each molecule. In this method the space witin the van der Waals surfaces of the two molecules is excluded from the calculation in order to avoid singularities. The molecules I, II and III are superimposed upon each other using least square fitting of nuclei method (12) and the values of Rxy are listed as a matrix in Table 1. The similarity indices are in the range of 0.65 to 0.87 which is sufficient to account for the bioisoterism of these molecules. I

II

III

1.0

0.65

1.00

0.87

0.65

1.00

Comparison with crystallographic results The preceeding computational results could be related to the crystallographic observations keeping in mind the constraints of the two approaches. However, reasonable consistancy of the calculated results were observed with those of the crystallographic ones. The salient features are: (a) Although most crystallographic results on similar molecules refer to esters, the common observation is the planarity of the group although we have chosen acids for our study where the carboxyl group is planar and stabilized due to hydrogen

ANTI-INFLAMMATORY COMPOUNDS 53

bonding, (b) Another similarity in this study is the observation of coplanarity of the carboxyl group with the glycolic hydroxyl group, an observation reported in the crystallographic study of hexapyronium bromide (14). (c) Regarding the conformation of the ring, literature results show varied orientations. The data reported for glycopyronium bromide the ring and the glycolic hydroxy group were found to be coplanar (15). This is also our observation CONCLUSIONS The MO calculations at both CNDO/2 and ab initio levels on phenyl glycolic and and–;thienyl glycolic acids are in substantial accord as to the preferred conformer and show qualitative agreement for the barrier heights. The results are also consistent with those of the crystallographic data on related compounds. It is observed for all the molecules investigated that the carboxyl part and the C–glycolic hydroxyl part are coplanar independent of the substitution pattern at the other C—positions and the stability is due to intramolecular hydrogen bonding. The stabilization is observed to be better favoured when the C–OH (glycolic) group is nearer to the carbonyl part than that of the hydroxyl part of the carboxyl fragment of the molecules. For all these glycolic acids the preferred conformer corresponds to the coplanarity of the aromatic part with the C–OH (glycolic) bond except for the phenyl substituted one where the preferred form was found to correspond to the coplanarity of the phenyl ring and the C–H bond. The electrostatic potentials of the molecules are found to be similar to each other in terms of their contours patterns. The similarity indices were found to be in the range 0.65 to 0.87 which may be considered to be sufficient to account for the bioisosterism of these molecules. Considering the above results, a model of molecular recognition of these molecules is possible in which the electrostatic similarities as well as the size and shapes of the molecules would be crucial. Thus the detailed analysis of conformations and the MEPs derived from the ab initio calculated structures will be useful in predicting other such compounds that might have anti-inflammatory properties. ACKNOWLEDGEMENTS The author expresses his gratitude to Professors J.E.Dubois, J.P.Doucet, (Mme) A.Cosse-Barbi and (Mme) C.Mercier of the “Institut de Topologie et de Dynamique des Systèmes de l’Université Paris 7” for their continuous help and encouragement during the tenure of the work. The High Level Fellowship award (postdoctoral) from the Government of France is gratefully acknowledged.

54 ULTRA LOW DOSES

REFERENCES Atassi, G.A., Briet, P., Berthelon, J.J. and Collonge, T., [3], 1985, Eur J Med Chem— Chim Theo, 5, 393. Baker, R.W., Dutta, N. and Pauling P.J., [15], 1973, J.Chem Soc Perkin Trans, 2, 1963. Bhattacharjee, A.K., [6], 1990, Proc. Indian Acad Sci. (Chem. Sci.), 102, 159. Bhattacharjee, A.K., [7], 1990, Bull. Chem. Soc., Japan, Communicated. Binkley, J.S., Raghavchari, K., De Frees, D.J., Schlegel, H.B., Whiteside, R.A., Fluder, G., Frisch, J.M., Seeger, R. and Pople, J.A., [10], 1982, “Gaussian 82 Release A”, Carnegie-Mellon University, Pittsburg, U.S.A. Burger’s Medicinal Chemistry, [1], 1981, 4th ed., Wiley, New York, Vol. 3, pp 1237–1241. Carbo, R., Leyda, L. and Arnau, M., [11], 1980, Int. J Quant Chem, 17, 1185 De Vita, V.T., Hellman, S. and Rosenberg, S.A., [2], 1985, 2nd ed (Lippincott, Philadelphia), in: Cancer: Principles and Practice of Oncology. Guy, J.J. and Hamor, T.A., [13], 1973, J Chem Soc Perkin Trans, 2, 1875. Guy, J.J. and Hamor, T.A., [14], 1974, J Chem Soc Perkin Trans, 2, 1126. Mackay, A.L., [12], 1984, Acta Crystalloqr, A40, 165. O’Dwyer, P.J., Schoemaker, D., Zaharbo, D.S., Grieshaber, C., Corbett, T., Valeriote, F., King, S.A., Cradock, J., Hoth, D.F., Leyland-Jones and Plowman, J., [4], 1987, Cancer Chemother Pharmacol, 19, 6. Politzer, P. and Dalker, K.C., [8], 1981, “The force Concepts in Chemistry”, ed Van Nostrand Reinhold Company , New York. Schmidling, D., [9], 1978, The GEOMO program system, No. 350 QCPE, 11, 350. Tayar, N.E., Carrupt, P.A., Waterbeemd, H.V. and Testa, B., [5], 1988, J Med Chem, 11, 2072.

INFLUENCE OF SEVERAL PHYSICAL FACTORS ON THE ACTIVITY OF ULTRA LOW DOSES Cazin J.C., Cazin M., Chaoui A., Belon P.* Laboratoire de Pharmacologie, Faculté de Pharmacie 3 Rue du professeur Laguesse, 59000 Lille, France * Fondation Française pour la Recherche en Homeopathie, 20 Rue de la Liberation, 69110 Sainte-Foy-Lès-Lyon, France

INTRODUCTION Previous studies had already demonstrated the influence of ultra low doses of arsenate salts on the retention and elimination of this same compound (Lapp, 1955), (Boiron, 1962), (Mouriquand, 1961). We took up these studies again using a tracer since this is a precise method for evaluation. Our main purpose was to determine an experimental protocol to permit studies which are rapid, reliable, reproductible and statistically valid. Beforehand, we carried out a series of pharmacokinetic studies on mice and rats (Brunet, 1982), (Brunet, 1983), (Brunet, 1984), (Gaborit, 1987), as these animals show very different pharmacokinetic profiles (Dutkiewicz, 1977), (Fowler, 1975), (Klaasen, 1974), (Odanaka, 1980), (Vahter, 1980). Through these studies we observed that mice do not seem to lend themselves well to trial studies on the retention and possible mobilization of arsenic under the effects of ultra low doses of arsenious anhydride. Indeed we did not observe sufficient fixation of the toxin in any particular organ, and it appeared that most of the administered dose was eliminated within 48 hours. On the other hand, for the planned study, rats offered numerous additional possibilities, since blood appears to be one of the preferential areas of fixation for arsenic, and the blood concentration remains high for long periods of time (blood peak after 12 hours). Moreover, urine and feces are the most important channels for eliminating the toxin. Our studies were thus carried out on these animals and consisted, in the first half, of developing an experimental protocol to allow us, in the second half, to study the influence of certain physical and chemical parameters. We were thus able to carry out the following studies: 1) Effects of a single injection of an ultra low dose of arsenious anhydride,

56 ULTRA LOW DOSES

2) Effects of three injections of ultra low doses of arsenious anhydride, These two preliminary studies allowed us to determine the experimental protocol to be applied to the following experiments: 3) Influence of heat on the action of the As2O3 at the 7th centesimal dilution. 4) Influence of preparing the dilution under nitrogen blanketing. * It does indeed seem that the preparation of infinitesimal dilutions leads to a “physical state” of the solvent, specific to the basic substance. The presence of oxygen at the time of preparation may play a role in obtaining this “physical state” (Boiron, 1972). During this experiment a comparative study was carried out on the effects of ultra low doses prepared either in the presence or absence of oxygen (under nitrogen blanketing). 5) Influence of the chemical composition of the dilutions. In the previous experiments, we used arsenious acid as a radioactive tracer, and arsenious anhydride as an ultra low dose. In the fourth series of experiments we probed the influence and kinetics of action of dilutions of arsenious acid on the same radioactive tracer. 6) Kinetics of action of 14 different dilutions of arsenious acid. MATERIALS AND METHODS A—General features of all the experiments Young male Wistar rats weighing 70+grams each, were split, at random into groups of 20 to 30 animals (cf infra). By means of esophageal intubation they were given a single oral dose of 10 mg/Kg of arsenious anhydride (As2O3) and a tracer dose of 100 µCi/Kg of 73As in the form of arsenious acid H3AsO3 (Amersham, England). The time of this administration defined to for each animal. The products were put into suspension in 5% gum arabic and were administered to the trial animals in volumes of 0.5 ml per 20 g of body weight. The ultra low doses of arsenic were administered intraperitoneally in volumes of 1 ml per rat: these ultra-low doses were prepared by dilution of 1 gram of the studied product (either arsenious anhydride or arsenious acid) in 99 ml distilled water. This was then subjected to a brief period of vigorous shaking (potentization) to produce the first centesimal dilution (1 C). Repeating the same operation and starting from this first centesimal produces the second centesimal (2 C); this process is repeated to give different desired centesimal dilutions. An equivalent quantity of water potentized under the same conditions was administered to the control rats. As soon as the defined ultra low dose was administered, the the rats were put into individual metabolism cages and were given food and water according to need. The urine and fecal matter, automatically separated by the metabolism cages, were collected, weighed, then taken as samples to be measured for radioactivity. This was done for all the experiments.

INFLUENCE OF SEVERAL PHYSICAL FACTORS

57

Blood was taken by cardiac puncture and was collected in dry tubes. This was done for experiments 1, 2, 3, 5 and 6. The radioactivity of the samples was measured in an automatic gamma counter (intertechnique, model C.G.4000, Cristal 3 inches). All findings were corrected for counting effectiveness, background noises and radioactive decay. These findings were expressed in micrograms of arsenic per 100 mg of sample treated or per rat. To express these data in d.p.m. (disintegration per minute), a 22.200 coefficient multipler (1 µg=22.200 d.p.m.) needed simply to be applied. Administering the dilutions, and killing the animals were done according to two experimental protocols (Figure 1): – experiments 1, 3–6, one injection at T0+12 hours ; this time corresponds to the blood peak (C max) in the pharmacokinetics of arsenious anhydride per os in rats (13). – experiments 2, three injections at T0+36 hours. B—Special characteristics adapted to each experiment . Experiment 1: Effect of a single injection of a 7th centesimal of arsenious anhydride. 30 rats were treated with 1 ml of a 7th centesimal (7C corresponding to a theoretical concentration of 10–14g/ml) of arsenious anhydride and 30 control rats with potentized water (corresponding to distilled water treated in similar manner to the arsenious anhydride dilution). The injection was given at T0+ 12 hours. The effect was measured eight hours later (T0+20 hours). . Experiment 2: Effect of three injections of a 7th centesimal of arsenious anhydride. 25 rats were given three injections of a 7th centesimal of arsenious anhydride (cf supra) at T0+ 12 hours, T0+24 hours, T0+36 hours. The animals were killed at T0+7 days and the arsenious elimination was measured during the period of: T0+ 12 hours to T0+7 days. The control group consisting of 25 rats, was treated under the same experimental conditions using potentized water (cf supra) instead of arsenious anhydride 7th centesimal. . Experiment 3: influence of heat on the action of a 7th centesimal of arsenious anhydride. We measured the effects of a single injection of three different products on three groups, of 30 rats each: – group A: treated with potentized water (cf supra) – group B: treated with 7th centesimal of arsenious anhydride (cf supra) – group C: treated with 7th centesimal of arsenious anhydride which, immediately after preparation, has been heated to 120°C in autoclave for 30 minutes. . Experiment 4: Influence of preparing the dilution under nitrogen blanketing; Three groups, of 24 rats each were used. The rats were given: we used the three following solutions: – group A: a “standard” 7th centesimal of arsenious anhydride (i.e. prepared as described for experiment 1)

58 ULTRA LOW DOSES

– group B: a 7th centesimal arsenious anhydride, entirely prepared under nitrogen blanketing (absence of oxygen) – group C: potentized water, also prepared under nitrogen blanketing. The effects on blood arsenical concentration and on arsenical elimination were determined at T0+20 hours. . Experiment 5: Influence of the chemical composition of the dilutions. Three groups, of 25 rats each were used. The rats were given: – group A: a 7th centesimal of arsenious anhydride (As2O3) – group B: a 7th centesimal of arsenious acid (H3AsO4) – group C: potentized water. The effects on blood arsenical concentration and on arsenical elimination were determined at T0+20 hours. . Experiment 6: Kinetics of action of 14 different dilutions of arsenious acid. The effects of different dilutions of arsenious acid compared to the effects of potentized distilled water were studied on 15 groups of rats, of 21 animals each. The following dilutions were studied: 5C, 7C, 9C, 11C, 13C, 15C, 17C, 19C, 21C, 23C, 25C, 27C, 29C, 31C. RESULTS Measures are expressed in µg of arsenic, either per rat or per 100 mg sample. . Experiment 1: Effect of a single injection; On blood:

A significant difference is noted between the concentrations of arsenic in the blood of the rats treated with the arsenious anhydride 7C, and the concentrations of arsenic in the blood of the rats treated with the potentized water. 8 hours after the ultra low dose was administered, the rats having received the arsenious anhydride 7C showed an average arsenical blood concentration of 8.65 +0.46 µg/100 mg of blood, as opposed to 9.15+ 0.59 µg As/100 mg of blood for the rats treated with the potentized water. This difference is significant (p<0.01). On elimination:

For rats treated with the arsenious anhydride 7C, the increased elimination of arsenic (figure 1) for the period after its administration (12 hours, 20 hours), is in agreement with the results obtained concerning blood: the average elimination of arsenic is 21.98+2.13 µg per animal, in the group treated with the arsenious anhydride 7C whereas it is 13.78+ 1.98 µg per animal in the potentized water treated group. This difference is significant (p<0.001 determined both with the Student t-test and Fisher-Snedecor). . Experiment 2: Effects of three injections. On blood:

INFLUENCE OF SEVERAL PHYSICAL FACTORS

59

Figure 1. Total elimination of the arsenic (urine and fecal matter—averages of 30 rats), measured at T0+20 hrs. **** p<0.001

The average concentration of arsenic in the blood of the 25 rats treated with 3 injections of 1 ml of arsenious anhydride 7C (after 7 days) is 6.25+0.62 µg/100 mg of blood as opposed to 6.91+0.81 µg/100 mg of blood, on the average, for the potentized water treated group. According to the Fisher-Snedecor test and Student t-test this difference is significant p<0,01. On elimination:

The total of arsenious elimination, both urinary and fecal (figure 2) shows that: – the rats treated with the arsenious anhydride 7C eliminated on the average 97 +11 µg of arsenic for the period T0+12 hours to T0+7 days. – for the same period, the rats treated with the potentized water eliminated on the average 59+9 µg of arsenic. The difference is significant: p<0.001. The difference in elimination is mainly observed during the first period i.e. T0+36 hours. . Experiment 3: Influence of heat. On blood:

The average concentration of arsenic in the blood of the 30 rats treated with standard (non-heated) arsenious anhydride 7C is 9.14+0.43 µg/100 mg of blood (group B), whereas it is 10.27+0.73 µg/100 mg on the average for the group of 30 rats treated with the potentized water (group A), and 9.45+0.89 µg/100 mg for the group of 30 rats treated with heated (120°C for 30 min) arsenious anhydride 7C (group C). According to the Fischer-Snedecor test and the Student t-test the differences between group A and group B are significant, whereas the differences between group A and group C, on the one hand, and group B and group C, on the other hand are not significant.

60 ULTRA LOW DOSES

Figure 2. Total elimination of the arsenic (urine and fecal matter). Values cumulated from one date to the next. **** p<0.001 On elimination:

The total of the fecal and urinary elimination of arsenic, for the period from T0 +12 hours to T0+20 hours, is respectively: – for group A 21.37+1.52 µg/rat – for group B 27.57+2.56 µg/rat – for group C 23.97+2.79 µg/rat. The differences between group A and group B, and the differences between group B and group C are significant p<0.01 in both cases. The differences between group A and C are not significant. At least we can observe that heating arsenious anhydride 7C for 30 minutes at 120°C strongly diminishes its action. . Experiment 4: Influence of preparing the dilution under nitrogen blanketing. On elimination:

The total of the fecal and urinary elimination of arsenic, for the period from T0 +12 hours is respectively: – for group A (arsenious anhydride 7C) 26.54+3.22 µg/rat – for group B (arsenious anhydride prepared under nitrogen blanketing) 22.56 +2.82 µg/rat – for group C (distilled water potentized under nitrogen blanketing) 24.30+2. 44 µg/rat; According to the student t-test the difference are significant: – group A versus group B p<0.001 – group A versus group C p<0.01 – group B versus group C p<0.05. . Experiment 5: Influence of the chemical composition of the dilutions. On blood:

INFLUENCE OF SEVERAL PHYSICAL FACTORS

61

The average concentration of arsenic in the blood of the 25 rats treated with the arsenious anhydride 7C (group A) is at T0+20 hours: 9.68+0.79 µg/100 mg of blood, for the 25 rats treated with the arsenious acid 7C (group B) it is 9.60+0. 67 µg/100 mg of blood, for the 25 rats treated with the potentized water (group C) it is 10.36+0.85 µg/100 ml of blood. According to the Student t-test the difference of group A versus group B is non-significant, whereas the difference of group A versus group C is significant, p<0.01, as is the difference of group B versus group C p<0.01. On elimination:

The total of the fecal and urinary arsenic elimination for the period from T0 +12 hours to T0+ 20 hours is: – for group A 21.12+1.97 µg/rat – for group B 23.04+2.56 µg/rat – for group C 19.70+2.19 µg/rat. According to the student t-test the differences observed are significant: – group A versus group C p<0.02 – group B versus group C p<0.001 – group A versus group B p<0.05. Therefore arsenious acid 7C is more active than arsenious anhydride 7C on the elimination of arsenic in arsenical-intoxicated rats. . Experiment 6: Kinetics of 14 different dilutions of arsenious acid. On blood:

The mean averages of the blood arsenic concentrations in the rats treated with the different arsenious acid dilutions are lower than the mean concentrations in the rats treated with the potentized water. The decreases vary according to the dilutions – from 3.2%, which is not significant, for arsenious acid 31C, – to 7.2%, which is significant, (p<0.05) for arseniuos acid 7C. Two others dilutions show a significant decrease of blood arsenical concentrations: for 5C the decrease is 6.9% p<0.05, for 7C the decrease is 6.4% p<0.05. All other dilutions induce a decrease in the blood arsenical concentration but none is significant. On elimination:

Each dilution (5, 7, 9, 11, 13, 15, 17, 19, 21, 23, 25, 27, 29, 31) induces an increase of the arsenical elimination. Figure 3 expresses the arsenical elimination (fecal+urinary) for each of these dilutions.

62 ULTRA LOW DOSES

Figure 3. Variations in total elimination of arsenic (urine and fecal matter), per rat, expressed as percentages in relation to potentized water.

The results are given in percentages of elimination observed when rats were treated with the potentized water (100% being the arsenical elimination obtained in this case). Figure 3 shows that: – arsenious acid 7C gives an increase of 27% p<0.001 – arsenious acid 5C gives an increase of 24% p<0.001 – arsenious acid 17C gives an increase of 20% p<0.001 – arsenious acid 9C gives an increase of 18% p<0.001 – arsenious acid 13C gives an increase of 17% p<0.01 – arsenious acid 11C gives an increase of 16% p<0.01 – arsenious acid 19C gives an increase of 13% p<0.02 – arsenious acid 15C gives an increase of 11% p<0.05 All others dilutions give an increase which is not significant. DISCUSSION The results obtained lead to the following remarks: – The action of the arsenious anhydride 7C on the mobilization and retention of arsenic in rats is well demonstrated here, regardless of the protocol chosen. This confirms the findings obtained by other studies. – The simplified protocol of the first experiment leads, in the long run, to results which are closely comparable to those previously published (Cazin, 1987). – In the second experiment, the first two injections (T0+12 hours and T0+24 hours), were able to show a marked difference between the effects induced by

INFLUENCE OF SEVERAL PHYSICAL FACTORS

63

the arsenious anhydride 7C and those induced by the potentized water. Beyond T0+36 hours, the quantities of arsenic eliminated by the two groups of rats are constant and evolve in comparable fashion. – The results obtained with these two protocols show that the effects of arsenious anhydride 7C are mainly detected in the blood and in elimination for the first 24 hours after the radioactive arsenic is administered, and under the action of a single injection of the dilution carried out 12 hours after that administration. – Heating arsenious anhydride 7C at 120°C for 30 minutes diminishes, even nullifies the action of the product. – Preparing the dilution of arsenious anhydride under nitrogen blanketing hampers its effects on the elimination and retention of arsenic in rat. – Arsenic acid 7C is more active than arsenious anhydride 7C; both are more active than potentized water. – The study of the kinetics of action of 14 dilutions of arsenious acid demonstrates that the 7C dilution is the most active. This action diminishes as the dilutions are increased, a second peak of activity appears at 17C, and the activity of the following dilutions decreases thereafter. ACKNOWLEDGEMENTS We sincerely thank: Mr Jean Boiron for his advice and collaboration Mrs Sheila Adrian for her help in rereading and correcting the English of this article. REFERENCES Boiron, J. and Cier, A., 1962, Elimination provoquée et spécificité d’action des dilutions infinitésimales d’éléments toxiques. Annales Homéopathiques Françaises, 10, 789–795. Boiron, J. and Cier, A., 1972, Premières recherches sur l’inactivité des dilutions préparées sous azote. Annales Homéopathiques Françaises, 4, 297–301. Brunet, C., Luyckx, M. and Cazin, M. , 1982, Etude pharmacocinétique de l’anhydride arsénieux chez la souris. Toxicological European Research, IV n°4, 175–179. Brunet, C., Luyckx, M. and Cazin, M. , 1983, Etude de la distribution topographique de l’anhydride arsénieux chez la souris gestante par autoradiographie macroscopique. Toxicological European Research, V n°2, 55–61. Brunet, C., Luyckx, M. and Cazin, M., 1984, Etude pharmacocinétique comparative de l’anhydride arsénieux chez le rat et la souris. Journal de Toxicologie et de Médecine, 4 n°4, 293–303. Cazin, J.C., Cazin, M., Gaborit, J.L., Chaoui, A., Boiron, J., Belon, P., Cherruault, Y., and Papapanayotou, C., 1987, A study of the Effect of Decimal and Centesimal Dilutions of Arsenic on the Retention and Mobilization of Arsenic in the Rat. Human Toxicology, 6, 315–320.

64 ULTRA LOW DOSES

Dutkiewicz, T., 1977, Experimental Studies on Arsenic Absorption Routes in Rats. Environmental Health Perspective, 19, 173–177. Fowler, B.A., 1975, Toxicology of Environmental Arsenic. In: Advances in Modern Toxicology. II. Toxicology of Trace Elements, Washington, Edited by M.Mehlamn and R.A.Goyer, (Washington Hemisphere), pp. 79–122. Gaborit, J.L., 1987, Contribution à la recherche de modèles biologiques chez les mammifères permettant de visualiser l’action de différents facteurs physiques sur l’activité des dilutions infinitésimales. These de Doctorat de 3° cycle pharmacologie, Lille. Klaassen, C.D., 1974, Biliary Excretion of Arsenic in Rats, Rabbits and Dogs. Toxicology Applied to Pharmacology, 29, 447–457. Lapp, C., Wurmser, L. and Ney, J., 1955, Mobilisation de l’arsenic fixé chez le cobaye sous l’influence de doses infinitésimales d’arséniate de sodium. Thérapie, 10, 625–638. Mouriquand, G. , Cier, A., Boiron, J., Edel, V. and Chighizola, R., 1961, Rétention et mobilisation de toxiques exogènes chez le pigeon. Essais de doses infinitésimales de ces mêmes éléments et variations concomitantes de l’indice chronologique vestibulaire. Comptes-Rendus de l’Académie des Sciences de Paris, 24 Mai 1961. Odanaka, Y., Matano, O. and Gotos, S., 1980, Biomethylation of Organic Arsenic by the Rat and some Laboratory Animals. Bulletin of Environmental Contamination and Toxicology, 24, 452–459. Vahter, M. and Morin, H., 1980, Metabolism of Arsenic-74 Labeled Trivalent and Pentavalent Inorganic Arsenic in Mice. Environmental Research (U.S.A), 21, 446–457.

III BIOCHEMISTRY—TOXICOLOGY

URANYL NITRATE INDUCED CORPUSCULAR DERANGEMENT IN RAT, AN EARLY INDICATION OF RENAL DYSFUNCTIONING Mohan Gojer and Vijay Sawant Laboratory of Animal Physiology, Shivaji University, Kolhapur-416004, Maharashtra, India

Erythrocyte Structure has been found to be greatly influenced during the pathogenic progression of uranyl nitrate (UN) induced acute renal failure (ARF). A marked anisocytosis and poikilocytosis during different phases of ARF is characteristic of UN induced ARF. During early initiation phase (2 hr) reticulocytes emerged in blood smear, followed by remarkable transformation of erythrocytes into echinocytes in late initiation phase (8 hr). It is characteristic of hemolytic or macrocytic anemia. The maintenance phase is marked by formation of cytoplasmic bridges among erythrocytes. Such structural alteration is analogus to B 12 deficiency and liver diseases. The corpuscular derangement occured during progression of ARF is an early indication of renal dysfunctioning. The etiology of erythrocyte derangement is discussed. INTRODUCTION Uranyl nitrate (UN) has become a target of several investigators due to its marked effect on renal morphology anf physiology and reported as occupational health hazard. (Bencosme, et al, 1960; Flamenbaum, 1973) With respect to renal pathopysiology and histopathology much reports have been published. (Yano, and Ueshima, 1982; Flamenbaum, W. 1983; Desai and Sawant, 1988) Behaviour of serum tri acyl glycerol hydrolase (EC 3:1:1:3) activity has been reported to be a sensitive diagnostic tool indicating the prognosis of acute renal failure (ARF). Changes have also been reported in the hematological profile of rats as a result of UN induced ARF. In these studies it was observed that initiation of ARF had marked effect on hemoglobin content and blood cell count which resulted in macrocytic hypochromic anemia; characteristic of UN induced ARF. A diagnostic approach towards UN induced ARF was then concentrated on kidney and bone marrow being chiefs sites of accumulation. Nevertheless, recently hemolytic syndrome/hypoxia induced as a result of UN intoxication has drawn

URANYL NITRATE INDUCED CORPUSCULAR IN RAT 67

much attention towards resultant alteration in erythrocyte and its correlation with ongoing progression of ARF. An early indication of renal dysfunction, as early as 48 hr post UN administration has been reported by subsequent change in erythrocyte structure, (Meola, etal 1982). The present communication reports that erythrocyte membrane abnormality begins as early as 8 hr post UN administration, revealed at light microscopic level. The etiology of erythrocyte membrane abnormality during pathogenic progression of ARF prior to the clinical manifestation of ARF is discussed. MATERIALS AND METHOD Animals and Administration of UN Adult male albino rats (Rattus norvegicus) weighing 200–225 g had free access to pellet feed and water ad libitum. Animals were divided in two groups, each group consisting of 5 animals. Uranyl nitrate was dissolved in isotonic saline in a concentration such that each animal in group B received 1 ml intraperitonial solution of 10 mg/kg UN. Control group A received an equivalent volume of 0.9% saline. Animals were sacrified after a desired time interval during early initiation phase (2 hr), late initiation phase (8 hr) and the maintenance phase (24 hr) respectively. Abnormal Erythrocyte Counting Reticulocyte count was performed using new methylene blue stain in 5 samples and recorded as cells per percent normal erythrocytes. Abnormal erythrocyte count was performed in 5 samples and reported as percent cells present in total population. Erythrocyte morphological studies were preformed on light microscope using oil immersion technique (1000 x). A series of photographs were taken from the same field on Karl Zeiss photographic unit. Total BUN Creatinine and Calcium estimation Total blood urca nitrogen, creatinine and calcium from serum of rat were determined by Erba-chem-5 semi-auto analyser. RESULTS Corpuscular Derangement Uranyl nitrate induced ARF in experimental models-revealed certain specific morphological alterations which are depicted in figures 1, 2 and 3 respectively.

68 ULTRA LOW DOSES

Figure 1

Figure 2

It can be clearly seen in these photographs that uranyl nitrate induced ARF exhibits sequential changes in erythrocyte membrane morphology. The onset of ARF (early initiation phase—2 hr) begins with appearance of reticulocytes (RC) in peripheral blood smear (Fig.1). The phenomenon being the earliest change in erythrocyte profile. During late initiation phase (8 hr) the reticulocyte number decreased and erythrocytes with crenation on outer surface (echinocytes—EC) appeared. The projections on—outer surface were irregular and had unequal diameter (Fig.2). However, the echinocytes were not unformly present throughout the sample. The crenated cells—resumed their original discoid shape on addition of normal plasma to the smear. During the maintenance phase (24 hr) the crenation on outer surface of erythrocytes increased in length and cells were fused to each other by

URANYL NITRATE INDUCED CORPUSCULAR IN RAT 69

Figure 3

cytoplasmic bridges (CB). The echiocytes with cytoplasmic bridges were seen in patches and not invariability seen throughout the smear. The percentage of abnormal erythrocytes and reticuloctytes is shown in Table 1. UN Treatment

Percentage of Abnormal Erythrocytes

Percentage of Reticulocytes

2 hr UN 8 hr UN 24 hr UN

1.20±0.015 4.75±0.022 6.95±0.044

7.45±0.013 4.33±0.016 3.90±0.012

Values are mean±SE of 5 animals

The blood urea nitrogen level and creatinine clearence showed a gradual onset of renal failure in the experimental model. The serum calcium level showed a gradual depletion as the renal dysfunctioning progressed (Table 2). Treatment

BUN mg %

Creatinine mg %

Total Calcium mg %

Normal 2 hr UN 8 hr UN 24 hr UN

35±1.52 42±0.28 48±1.02 54±0.89

1.3±0.02 1.49±0.007 1.8±0.02 1.9±0.02

10.0±0.096 8.04±0.032 7.6±0.025 5.17±0.042

70 ULTRA LOW DOSES

DISCUSSION During early initiation phase (2 hr) of ARF, appearance of reticulocytes is a marked event. Reticulocytosis observed at this stage indicates initiation of hemolytic anemia and increased erythropoietic activity. The hypochromic anemia reported at this stage (Gojer and Sawant, 1989) may not be entirely accountable to the direct detrimental effect of UN on erythrocyte membrane, as immediate response of erythroid marrow to subsequent erythrocyte destruction is the erythropoietin related shift of marrow reticulocytes into circulation. It is noteworthy that possibility of dyshematopoietic effect in early initiation phase have been reported to be associated with anemia developed by certain metals (Berlin and Friberg, 1960; Berlin et al, 1961). At late initiation phase (8 hr) erythrocyte structure exhibited a peculier change. Such structural abnormality is referred to as “echinocytes” due to their reversability into normal shape by addition of normal plasma. Formation of echinocyte is evident by certain chemical compounds such as fatty acids, bile acids, lysolecithin, barbiturate dipyridamole, phenylbutazone and salicylates (Mohandas et al, 1980). (Gojer et al, 1985) have reported a significant elevation in the free fatty acid level at this stage, which may result into formation of echinocytes. The elevated level of free fatty acids is persistant during the maintenance phase also, hence the elevated serum fatty acid level could be the intrinsic causative factor for echinocyte formation. During the maintenance phase (24 hr) echinocytes and cytoplasmic bridges become much prominant with the development of macrocytic hypochromic anemia. The condition referred to as fatty liver has been reported at this stage, (Desai and Sawant, 1987), where a fatty degeneration of liver was observed, may result into structural alteration of erythrocytes as reported in B 12 deficiency and liver diseases. In addition to the elevated serum free fatty acid level, intrinsic factor such as calcium is also responsible for structural change in erythrocytes. Weed and Chailly, (1973) have suggested that morphological transformation of erythrocytes is associated with an increased calcium permeability. The observed erythrocyte membrane abnormality in ARF experimental model may be associated with an indirect effect of uranium on intrinsic factors such as increased membrane permeability to calcium and energy depletion of erythrocytes. Hence, the present study points out the possibility that erythrocyte membrane abnormality may have been caused by elevated free fatty acid level and calcium depletion due to UN induced ARF. However, it is noteworthy that the entire erythrocyte population does not show the abnormality. It is apparent from the percentage of abnormal erythrocyte that older erythrocytes are more vulnerable for membrane transformation, as the duration increased, more and more erythrocytes tend to show abnormality. It is possibility due to the fact that erythrocytes are known to have a definite life span. In conclusion, there is a definite correlation between the duration of UN administration and extent of alteration in erythrocyte structure. It should be noted

URANYL NITRATE INDUCED CORPUSCULAR IN RAT 71

that—echinocyte formation is marked as early as 8 hr post UN administration, much earlier than the findings of Meola et al (1982). The fusogenic property of uranyl nitrate is not yet been established, nevertheless like uranyl acetate (Mujumdar, et al 1980) it may exhibit similar effect as evident in 8 hr post UN administration. Thus, definitive changes occured in erythrocyte structure is much earlier than the clinical and histopathological manifestation of ARF (Desai and Sawant, 1988; Meola, et al 1982). Hence, it could be much sensitive diagnostic tool for early acute renal dysfunction. REFERENCES Bencosme, A.L., Wills, J.H., Newman, Wm.F., Lan, T.H., Haven, I., Newman and Roberts, 1960, American Medical Associations Archives of Pathology, 69, 479. Berlin, M. and Friberg, L., 1960, Archives of Environmental Health, 1, 478. Berlin, M., Freoer, M. and Lingeg, B., 1961, Archives of Environmental Health, 3, 176. Desai, D. and Sawant, V., 1987, Ph.D, Thesis, 98. Desai, D. and Sawant, V., 1988, Indian Journal of Comparative Animal Physiology, 6(2), 144–149. Flamenbaum, W., 1973, Archives of International Medecine, 131, 911–928. Flamenbaum, W., Kaufman, J., Chopra, S. and McNeil, J.S. , 1983, In: Cellular Pathobiology of Human Diseases, Edited by G.Fisher (Academic Press, New-York). Gojer, M. and Sawant, V., 1985, Indian of Physiology and Pharmacology, 29(2), 96–102. Gojer, M., Desai, D. and Sawant, V., 1989, Journal of Environmental Biology, 10(1), 35–41. Meola, S.M., Rowe, L.D., Lovering, S.L. and Moore, G., 1982, Scanning Electron Microscopy, 3, 1229–1235. Mohandas, N. , Weed, R.I. and Bessis, M., 1980, In: Trump B.F.; Arstila A.U., (eds), Pathobiology of Cell Membranes, Academic Press, 41–91. Mujumdar, S., Baker, R.F. and Kalro, V.K., 1980, Biochemica Biophysica Acta, 598, 411–416. Weed, R.I. and Chailley, B., 1973, In: Bessis M, Weed R.I., Leplong, P.F., (Eds), Red Yano, I. and K.Ueshima; 1982, Wakayama Medical Report, 25, 75–85.

DEGRANULATION OF MESENTERIC MAST CELLS AS “SPOT TEST”—IN TOXICOLOGY Rathinam, K., Mohanan, P.V. and Lizzy Michael Division of Toxicological screening of Materials, Sree Chitra Tirunal, Institute for Medical Sciences and Technology, Biomedical Technology Wing, Poojapura, Trivandrum-695 012, India

INTRODUCTION Mast Cells (MC) are the permanent resident of connective tissues throughout the body more around the blood vessels and nerves (Mariano, 1958). They are the shock troops of the immune system pharmacologically important substances such as histamine, heparin and serotonin and produced a variety of local tissue changes as a result of degranulation and/as cellular disruption (Fulton etal, 1957, Boscila, 1963). The degranulative potential of the extract of Epoxy Polymer material and baryum Methacrylate—Monomer (Bame) has already been demonstrated in Rat Mesenteric Mast Cell and chicken mesenteric Mast cell (Rathinam, etal 1990). It has been suggested that Disodium Chromo Glycolate (DSCG) prevents the histamine release. DSCG in vitro prevent histaminerelease from MC. caused by compound 48/80. The drug appears to have a direct protective effect on Mast Cell. The degree of degranulation is found to be the function of concentration of Toxic agent exposed. (Rathinam, et al 1990). The present study highlights the importance of the in vitro MC. system toxicology. METHODS Overnight fasted Wistar rats of either sex having body weight between 220 and 240 g. were sacrificed by cervical dislocation. The intact gut mesentery was collected in a kidney tray having Ringer-Locke glucose solution. The intertime with Mesentery of chicken was received from the local slaughter house in Ringer-Locke glucose solution after the removal of all intestinal contents. Under sterilised condition the rat’s and chicken’s gut mesentery was cut in to small bits. The experiment was arranged as 4 sets each with 3 boiling tubes having 10.0 ml. of Ringer-Locke glucose solution. In the first set having different

DEGRANULATION OF MESENTERIC MAST CELLS 73

concentrations (0.1, 0.3, 0.5 ml) of exposy-polymer material’s extract (causing mortality in mice) with 6 rat mesenteric bits in each tube, second set having different concentrations of (0.5, 1.0, 1.5 ml) of Bame (systemically toxic to mice) with 6 rat mesenteric bits in each tube, third set is with different concentrations of Bame (0.5, 1.0, 1.5 ml) with 6 chicken mesenteric bits each and forth set with Ringer-Locke and mesenteric (chicken, rat seperate) bits served as controls (Norton, 1954). All the tubes were incubated for 10 minutes at 37°C. Consequently the bits were kept on a clean glass slide, dried, stained and counter-stained with toludine blue and light green stains respectively (Gupta et al, 1968). Microcopic evaluation of Mast Cell for degranulation was counted ie total hundred cells were counted and the number of degranulated cells were noted and expressed as percentage of degranulated cells. RESULTS AND CONCLUSION The results of our experiments were analysed statistically using Student ‘t’ (Table I). Table 1. Degranulative effects of Epoxy and Bame extracts in Rat and Chicken Mesenteric MC. No. Animal

Test

Dosage Control

% of degranulati on in test Mean±SE

% of P value of degranulati test on in Control

1

Rat (6),

Epoxy extract

0.1 ml 0.3 ml 0.5 ml

10.0 ml RingerLocke

31.8± 4.94

P<0.001 P<0.001 P<0.001

2

Rat (6)

Bame

.5 ml 1.0 ml 1.5 ml

10.0 ml RingerLocke

24.3± 4.25

P<0.001 P<0.001 P<0.001

3

Chicken (6)

Bame

.5 ml 1.0 ml 1.5 ml

10.0 ml RingerLocke

52.3± 2.36 84.3± 7.53 91.5± 2.09 47.8± 3.78 50.0± 6.05 59.5± 6.42 74.8± 2.37 76.0± 3.89 80.5± 5.67

28.16+ 2.69

P<0.001 P<0.001 P<0.001

Number in Parenthesis indicates number of mesenteric bits used. S.E.: Standard Error.

74 ULTRA LOW DOSES

The results proved that the toxic extract of epoxy material on rat mesenteric MCs and bame on rat and chicken mesenteric MCs have significantly (p<0.001) included degranulation. It has also been observed that the concentrations of chemicals/drugs bear a linear dose-response relationship. The in vitro MC technique provides a reproducibility, economy and speed as evidenced by our own study (unpublished data). Further our current study will be of appropriate approach towards this direction. REFERENCES Boseila, A.W.A., 1983, Hormonal influence on blood and tissue basophilic granulocytes. Annuals of New York Academic Sciences, 103, 394. Fluton, G.P., Maynard, F.L., Riley, J.F. and West, G.B., 1957, Humoral aspects of tissue Mast cells. Physiological Review, 37, 221. Gupta, R.K. and Skelton, F.R., 1968, The role of Mast cell in Cadmium chloride induced injury in mature rat testis. Archives of Pathology, 85, 89–93. Linda Gamlin, 1989, Mast cells move to centre stage. New scientist, IPC Magazine LTD, England 44. Mariano, Sh. Di Fiore, 1958, Atlas of Human-Histology Vol.1, edited by Lea and Febiger, Philadelphia, pp. 26–27. Norton, S., 1954, Quantitative determination of Mast cell fragmentation by Compound 48/ 80. British Journal of Pharmacology, 9, 494. Rathinam, K., Mohanan, P.V. and Lizzy Michael, 1990, In vitro procedure using rat mesenteric Mast cell for toxicity screening . Accepted for publication in Indian Journal of Pharmacology. Rathinam, K., Mohanan, P.V. and Lizzy Michael, 1990, Effect of Barium Methacrylate (Monomer) on chicken mesenteric Mast cell. Indian Journal of Pharmacology, 22, 182–183.

IV CELL BIOLOGY

SIMULTANEOUS MEASUREMENT OF OXIDATIVE METABOLISM ADHESION OF HUMAN NEUTROPHILS AND EVALUATION OF MULTIPLE DOSES OF AGONISTS AND INHIBITORS P.Bellavite, S.Chirumbolo, A.Signorini, I.Bianchi and *P.Dri Istituto di Chimica e Microscopia Clinica, Università di Verona, Ospedale Policlinico, 37134 Verona, Italy and *Istituto di Patologia Generale, Università di Trieste A testing model for the detection and evaluation of multiple doses of neutrophil agonists and antagonists is described. Incubations are performed in microtiter plates, where oxidative metabolism is measured as superoxide anion production and in the same assay system the adhesion is measured as activity of acid phosphatase. The test is simple to handle and is highly reproducible, and, therefore, it appears particularly useful for drug screening. Dose-response curves of various cell stimulants and inhibitors showed that the lowest effective doses ranged between 10–6 and 10–10 M according to the agents employed. Adhesion appeared to be more sensitive to low doses than superoxide production. INTRODUCTION Neutrophil granulocytes play a key role in the host defence against bacterial infections and also in other pathologic processes related to inflammatory reactions. When pathogenic agents invade a tissue, neutrophils are recruited from the bone marrow and the blood compartment into the inflammed area. At the inflammatory site, neutrophils and endothelial cells express membrane anchoring molecules and receptors that mediate the adhesion of the leukocytes to the vessel wall. Neutrophils then orientate and migrate through a gradient of chemotactic factors towards the centre of the inflammation, where they interact with the etiopathogenic agent. At this stage, phagocytosis of foreign particles, release of lysosomal granule constituents and activation of the oxidative metabolism, with production of superoxide anion (O2-), hydrogen peroxide and other toxic oxygen derivatives rapidly occur. All these events are regulated by a variety of mediators, some of which are of exogenous origin, like vegetable or bacterial compounds (e.g. phorbol esters, lectins, peptides, lipopolysaccharides, toxins, etc.), while others are produced endogenously, like complement fragments, kinins, cytokines, neuropeptides,

MEASUREMENT OF OXIDATIVE METABOLISM ADHESION 77

growth factors, eicosanoids, platelet-activating factor, adenine nucleotides, etc. Most of these factors act on specific membrane receptors and may affect neutrophil functions essentially in three ways: 1) by directly triggering one or more cell responses such as adhesion, chemotaxis, phagocytosis, respiratory burst, secretion of lysosomal constituents, b) by priming the neutrophils, that is by including in the cell a state of enhanced responsiveness to a subsequent stimulus, c) by inhibiting, or de-sensitizing, the neutrophil, whose functions may result dampened. Activation, priming and inhibition are subtly controlled by physiological mechanisms and could be the target, at least in theory, of pharmacological manipulation. Priming and activation are required and desirable during microbial infections and should be exogenously reinforced particularly in immunocompromised hosts, while inhibition and de-sensitization may be necessary in order to prevent the toxic and damaging effects of neutrophils during immunopathologic processes, intravascular aggregation, post-ischemic tissue injury, etc. [1–3]. The mechanisms of action of various compounds active on neutrophils may be better investigated using methods that evaluate more than one function. In this report we describe a novel microplate assay system that allows the simultaneous evaluation of the oxidative metabolism (measured as superoxide anion release) and of the adhesion of neutrophils exposed to a variety of agents. Several classic leukocyte stimulants were tested: the bacterial peptide N-formyl-L-methionyl-Lleucyl-L-phenylalanine (fMLP), the lectin Concanavalin A (Con A), the active principle of Croton oil phorbol-12-myristate-13- acetate (PMA), and yeast particles in the form of serum-treated zymosan (STZ). We also tested the effects of endotoxin, the lipopolysaccharide (LPS) component of the cell wall of Gram negative bacteria, a compound that has been previously shown to act as a powerful priming agent [4,5]. Finally, we report data of experiment where the cells have been treated with inhibitors before stimulation. The inhibitors utilized in this study were N-ethyl-maleimide (NEM), a toxic and irritating compound that strongly interacts with functional sulfhydryl groups [6], adenosine, a molecule with many physiological roles including effects on leukocytes [7], and corticosteroids, that are largely used as anti-inflammatory agents and that in previous papers have been shown to inhibit the leukocyte metabolism [8,9]. METHODS Materials fMLP zymosan, PMA, Triton X-100, NEM, LPS (from E. Coli, serotype n. 026.B6) were purchased from Sigma Chem. Co., St-Louis, M.O.Cytochrome c and adenosine were from Boehringer, Mannheim, F.R.G.. Concanavalin A was Vector Lab. Inc., Burlingame, CA. Human copper-zinc superoxide dismutase (SOD) was a gift of Prof. J.V.Bannister (Cranfield Institute of Technology,

78 ULTRA LOW DOSES

Bedford, U.K.). Fetal bovine serum (FBS) was from Flow Laboratories. Serum was inactivated by incubation at 56°C for 1 h. Percoll was from Pharmacia, Uppsala. Sterile 96-well microtiter plates with flat bottom wells were of type Linbro, from Flow Laboratories. Hank’s balanced salt solution (without calcium and magnesium) (HBSS) was from Gibco Ltd, Paisley, Scotland. Other materials and reagents were of the highest purity available. In order to avoid any contamination that could cause artifactual activation or priming of the cells, sterile solutions and disposible plasticware were used throughout all the experimental procedures, that were carried out, when possible, under a laminar flow hood. Reagents were prepared using clinical apyrogen water or 0.9% NaCl solutions. Human neutrophils were prepared from EDTA-anticoagulated blood by centrifugation over Percoll gradients. Percoll was diluted with distilled water and then with 1/10 volume of 10 x phosphate-buffered solution (PBS) in order to have the desired concentration inl x PBS (KH2PO4 0.144 g/liter, NaCl 9 g/liter, Na2HPO4 0.795 g/liter, pH 7.4). Starting from 40 ml of blood, two 50 ml sterile plastic tubes were layered (from bottom) with: 15 ml of 73% Percoll, 15 ml of 62% Percoll, 20 ml of blood. The tubes were then centrifuged for 20 min at 1800 RPM with a Sorvall T6000B centrifuge. Neutrophils were recovered as a broad band at the 73%–62% interfacie, were diluted with 1 volume of PBS, and centrifuged for 10 min at 1200 RPM. The pellet was usually slightly contaminated by erythrocytes, that were lysed by a brief hypotonic shock: cells were suspended in 20 ml of 0.2% NaCl for 20 seconds, then the isotonicity was restored by addition of 20 ml of 1.6% NaCl and right pH was restored by addition of 10 ml of PBS. The cells were then centrifuged for 10 min at 1200 RPM, washed once with 50 ml of PBS, then finally suspended in HBSS supplemented with 5 mM glucose, 0.5 mM Cacl2, 1 mM MgSO4, 0.2% human serum albumin (solution H-GCMA). Cells were counted and the concentration was usually brought to 2.7×106/ml with the same suspension buffer, the yield ranged from 6 to 11×107 cells, and more than 98% of purified cells were neutrophils. Dilutions of stimulants and inhibitors Stock solutions of compounds used in this work were prepared and stored as follows: fMLP (10–4M) and PMA (10–4M) were dissolved in dimethylsulfoxide and stored at −70°C. Zymposan was suspended in HBSS supplemented with 0.5 mM CaCl2 and 1 mM MgCl2, was opsonized with 50% pooled human sera for 30 min at 37°C, was centrifuged and washed with the same buffer, then was suspended in 0.9% NaCl at the concentration of 20 mg/ml. Aliquots of this suspensions were frozen at −20°C and thawed the day of the experiment. Concanavalin A (10–4M, using the mol. weight of 104,000), NEM (3×10−2M), hydrocortisone (10–3M) were dissolved in 0.9% NaCl and used the day of the experiment. LPS (2×10−4M, using a mean mol. weight of 10,000) was dissolved

MEASUREMENT OF OXIDATIVE METABOLISM ADHESION 79

in HBSS and stored at +4°C. Dexametasone (4×10−3M) was dissolved in ethanol and stored at +4°C. When indicated, dilutions were done as follows: the stock solution was diluted with 0.9% NaCl to a concentration exactly 4 times higher than that required for the maximal dose to be employed in the planned experiment. This has been done because each agent becomes diluted 4 times in the microwell incubation mixture. This maximal dose was then used for preparing a further series of either 2x or 10x dilutions until the lowest dose required. for dilutions, 16×100 mm polyethylene disposable tubes were used. 2x dilutions were made by adding 0.5 ml of the more concentrated solution to 0.5 ml of 0.9% NaCl. 10x dilutions were made by adding 0.2 ml of the more concnetrated solution to 1.8 ml of 0.9 NaCl. Immediately after dilution, each solution was stirred for 20 seconds with a Vortex tube mixer. Diluted solutions were used only the day of the experiment. Coating the microplates Preliminary experiments showed that neutrophils incubated in microplates spontaneously adhere to the bottom of the well in less than 30 min and produce considerable amounts of superoxide. This prevents the possibility to study the cell adhesion and metabolism upon addition of specific stimuli. Nonspecific activation was totally abolished by coating the microplate wells with serum proteins. In order to use safe, sterile and standardized serum preparations, fetal bovine serum (FBS) was used for this purpose. 100 µl of 50% FBS (diluted with PBS) were delivered in each well and the plate was incubated for at least 1 h at room temperature. Immediately before use, the plates were washed three times with PBS by using an automatic plate washer (Easy Washer 2, SLT Labs Instruments). Assay of superoxide production and of adhesion In synthesis, superoxide anion was measured by the SOD-inhibitable reduction of ferricytochrome c. At the end of the incubation, the assay mixture and unhadherent cells were washed out, while remaining adherent cells were quantitated by enzymatic assay. The 96-well microtiter plates were prepared according to various schemes and combinations, depending on the test assay to be carried out (e.g. various incubation times or various concentrations, etc.), on the number of compounds and of controls to be tested. After preliminary experiments, when multiple concentrations of stimulants or of inhibitors were tested, the scheme that was adopted is the following: Rows B, D, F, contained 25 µl of test compounds. Therefore 12 different doses in triplicate were tested for each plate. Rows C, E, G, contained 25 µl of 0.9% NaCl, or of dilutions of the solvent used in the preparation of the test compound. In row H, four wells were used as blank without cells (H9-H12), and in the other wells a known number of cells in a

80 ULTRA LOW DOSES

small volume (5–10 µl containing 2–20× 104 neutrophils) was added after washing. These cells were used as standard for the calculation of the % of adhesion. Row A was used when other controls were required, i.e. assays in the presence of SOD (25 µl of 0.9% NaCl with or whitout test compound containing 0.5 mg/ml SOD) or blanks of the test compound plus cytochrome c (to check for possible nonspecific reduction of cytochrome c by test compound). These control experiments demonstrated that in all the experimental conditions employed, the inclusion of SOD totally inhibited cytochrome c reduction, while it did not affect the adhesion of neutrophils. Therefore, SOD was not currently included as control in all dilutions of agonists and of inhibitors, and the reduction of cytochrome c was taken as the measure of superoxide production. This greatly increased the versatility of the test without affecting its precision. The plate was brought to 37°C, then 75 µl of pre-warmed neutrophil suspension were added to each well, except blanks. By using automatic pipette, this operation required about 60–90 seconds. Usually 2× 105 cells/well were added, because control experiments demonstrated that this is the optimal cell number to have good sensitivity in the range of cell concentrations where both superoxide and adhesion were linear (between 5×104 and 3×105 cells/well). At this point two different schemes were used, according to the object of the experiment: scheme 1. In experiments where multiple stimulants or multiple doses of stimulants were tested, the reaction was started by addition of neutrophils plus cytochrome c. Just before delivering to the plate, the cell suspension was supplemented with a little volume of concentrated solution of ferricytochrome c, in order to obtain a final concentration of 0.15 mM. The plate was then incubated at 37°C in cell culture incubator (5% CO2) for the desired time, then measured as detailed below. Scheme 2. In experiments where priming agents or inhibitors were tested, the neutrophils were dispensed in the plate without cytochrome c, then incubated for the indicated period of time at 37°C. After this pre-incubation, the wells were rapidly supplemented with 50 µl of a solution H-GCMA that was pre-warmed at 37°C and contained 0.45 mM ferricytochrome c (final concentration: 0.15 mM) plus the stimulatory agent 3x concentrated with respect to the final concentration in the assay mixture. The reaction was then incubated as describedin scheme 1. In all the procedures, care was taken to avoid cooling of the plate when it was taken from the incubator for filling or reading. At the end of the incubation, the plates were rapidly transferred into a microplate reader (Reader 400, SLT Labs Instruments) and the reduction of cytochrome c was measured by reading the plates at 550 nm using 540 nm as reference wavelenght to avoid interferences due to light scattering. Quantitation of absorbance of standard amounts of reduced and oxidized cytochrome c established that in this system the reading of 1 nmole of reduced minus oxidized cytochrome c (that is 1 nmole of 02-) is 0.037 0.D. Reading the plate took about 10 seconds.

MEASUREMENT OF OXIDATIVE METABOLISM ADHESION 81

Immediately after reading cytochrome c reduction, the plate was transferred to the automatic washer and subjected to two cycles of washing with PBS at room temperature. Washing was carefully calibrated because this was essential in order to obtain an optimal sensitivity and reproducibility of adhesion measurements. Each cycle was done as follows: emptying by rapid aspiration, filling by slight jet for 1.5 seconds, soaking for about 30 seconds, emptying by aspiration. The overall washing procedure took about 90 seconds. Adherent cells were quantitated by measuring the membrane enzyme acid phosphatase. 75 µl of 0.1 M acetate buffer, pH 5.3, containing 0.2% Triton X-100 were dispensed into the wells. After 5 min at room temperature 75 µl of the 0.1 M acetate buffer, pH 5.3 containing the substrate 10 mM p-nitrophenyl-phosphate were added. The reaction was incubated at room temperature for 20 min., then it was blocked by addition of 100 µl of 2 N NaOH. The p-nitrophenol produced by the phosphatase reactions was measured with the microplate reader at 405 nm. The % of adherent cells was calculated on the basis of a standard curve obtained with a known number of cells. RESULTS Optimal experimental conditions and kinetics The first series of experiments was done in order to establish the optimal conditions for exploring metabolic and adhesion functions of neutrophils in our assay system. We found that the reduction of cytochrome c (measurement of superoxide production) and the acid phosphatse reaction (measurement of adhesion) are linear in a range of cell concentrations between about 5×104 and 3×105 neutrophils/well. The subsequent experiments were therefore performed by using 2×105 cells/well. This sensitivity allowed to carry out experiments on two-three plates starting from 40-ml blood samples. Studies on low and ultra-low doses require high precision and accuracy of the test. Table 1. Precision of microplate assay for neutrophil adhesion and superoxide production. Pre-warmed cells added to several random wells of the microplate containing stimulants or 0.9% NaCl (see scheme 1 of the methods). Zymposan: 0.4 mg/ml; fMLP: 10–7 M. 02–: nmoles/106 neutrophils; adhesion : % of total. Test

24 identical wells

48 identical wells

Stimulant

None (0.9% NaCl) Opsonized zymosan (60 min)

Results Mean

S.D.

C.V.

O2−

0.8

0.3

37.5

ADHESION O2−

1.7 11.5

0.4 0.9

23.5 7.8

82 ULTRA LOW DOSES

Test

36 identical wells

Stimulant

fMLP (30 min)

Results Mean

S.D.

C.V.

ADHESION O2−

48.9 7.6

2.8 1.1

5.7 14.4

ADHESION

25.6

3.2

12.5

Table 1 reports data on superoxide production and adhesion of multiple replicate assays carried out in random wells of the same plate. It can be seen that in the absence of stimulants the spontaneous activation is extremely low. This is particularly important when very low doses of stimulants and bordeline responses have to be evaluated. When the cells were challenged with either a phagocytosable agent (opsonized zymosan) or a soluble stimulant (the bacterial tripeptide fMLP), both superoxide anion production and the adhesion were markedly stimulated. Comparing the stimulated with the resting cells demonstrates that the assay is very sensitive in order to reveal specific activation. Standard deviations and variation coefficients were very low, demonstrating that the test is sufficiently precise and reliable. Comparing the results obtained with stimulated cells by using different blood samples in different days showed that inter-individuals variations are in the range of the 17–63% for superoxide production and in the range of 14–39% for adhesion according to the different stimulants used (data not shown). Therfore the test is more suitable for multiple tests on the same cell preparation than when used for studies on the responses of different subjects. Figure 1 shows the time-course of the superoxide production and the contemparaneous adhesion of neutroplis. All the stimulants employed induced both responses in neutrophils, but differences were noted in the kinetics of the two phenomena. With PMA as stimulant, adhesion was quite precocious with respect to metabolic activation, but at 20 min the adhesion reached a plateau, then slightly declined. The superoxide production with PMA was linear until 60 min. With fMLP, a very rapid initial burst of superoxide production occured in the first five minutes, then oxidative metabolism slowed down and continued linearly. The corresponding adhesion was very slow, reaching a maximum of 20% adherent cells 60 min from the stimulation. Opsonized zymosan induced, after a 5 minutes lag, sustained metabolic response, while in the same conditions the adherence was low and completed in 20 min. In the presence of the lectin Concanavalin A neutrophils behaviour was very different with respect to the other stimulants. In fact, adhesion was rapid and almost complete in 20 minutes, while the oxidative metabolism was appreciable 20 minutes of incubation and was maximum after 60 minutes. At 10 minutes, about 40% of the cells were adherent, in the absence of any superoxide anion production.

MEASUREMENT OF OXIDATIVE METABOLISM ADHESION 83

Fig.1. Time-course of superoxide production and adhesion of neutrophils in response to various stimulants. Solid line: superoxide; dotted line; adhesion. Assays carried out according to scheme 1 described in Methods.

Dose-response curves The different kinetic patterns of adhesion and of superoxide production prompted us to study the same cell responses with different doses of stimulants. By using PMA as stimulant (fig.2), the maximum superoxide production was obtained with doses higher than 10–7 M, while adhesion peaked at 5×10–9. E.D. 50 schifted to the right in the adhesion test with respect to superoxide. Fig.3 shows that with Con A as stimulant the peak of adhesion was obtained with 10–7M, and the E.D. 50 was of about 10–8M, whereas the peak of metabolic activation was optimal at 10–6M, and the ED50 was of about 10–7M. Similar differences were found in two other independent experiments. These results indicate that adhesion is not directly and automatically linked to activation of the respiratory burst, and that, at least with Con A as stimulant, adhesion is amore sensitive parameter for measuring a neutrolphil biological response. From fig. 3 it can be seen that no activities were elicited by doses between 10–9 and 10–20M. Fig.4 shows the dose-response of the stimulation induced by fMLP. In this case the peak (10–7M) and the ED50 (about 4×10−8M) were similar for both superoxide production and adhesion. Careful inspection of the results of adhesion showed that a small percentage of cells were adherent even with 10–9M fMLP.

84 ULTRA LOW DOSES

Fig.2. Activation of neutrophils by decreasing doses of PMA. In this experiment, cells were pretreated with 2×10−8 M, fMLP, for 5 min, then stimulated with PMA for 1 h.

The difference between data with 10–9M fMLP and control matched undstimulated cells was significant with p<0.05 (paired student t test). A similar small peak in this “ultra low dose” range was obtained in another experiment, but it was not reproduced in a further one. Figure 4 also reports the results of parallel assays carried out using cells that had been previously treated with a low dose of fMLP. This treatment caused the cells to exhibit a slightly higher basal superoxide production, but their oxidative metabolism was totally unaffected by a subsequent stimulation by any dose of fMLP. This is probably due to receptor desensitization [10]. Also the capability to adhere to surfaces was desensitized, although not to the same extent as metabolic activity. In fact, with 10–6M fMLP as a second stimulus, at least 20% of the cells were adherent. Therefore, in this experiment a condition of activation of adhesion without activation of superoxide release is described. This further indicates that the two events are possibly controlled by mechanisms that are different at least in part.

MEASUREMENT OF OXIDATIVE METABOLISM ADHESION 85

Fig.3. Activation of neutrophils by decreasing doses of Con A. Incubation time: 60 minutes.

Since some papers that reported “ultra-low dose” effects of stimulants and inhibitors on basophil leukocytes mentioned that in order to have the maximum effect of these solutions it was necessary to subject them to a process of vigorous succusion [11–13], we performed experiments where the neutrophils were challenged with fMLP dilutions prepared in different ways. However, we did not find significant differences in the dose-response curves when the dilutions were followed by vigorous stirring (20 seconds with Vortex) or when the tubes containing the dilution were gently tapped for two-three seconds. Similar results were obtained by measuring adhesion (data not shown). Printing by endotoxin It is well known that endotoxin (LPS) primes various activities of neutrophils and macrophages [4,5]. This enhancement of function may be of relevance also

86 ULTRA LOW DOSES

Fig.4. Activation of neutrophils by decreasing doses of fMLP. Normal neutrophils and neutrophils pretreated for 10 min with 2×10–8M fMLP were added to microwells containing the indicated doses of fMLP and the superoxide production and adhesion were measured as described in methods (scheme 1). Hollow symbols are control cels, unstimulated, with 99% confidence limits (dotted lines).

in vivo for host defenses. We have, therefore, studied the adhesion of neutrophils and their metabolism after a treatment with different doses of LPS. Neutrophils were incubated for 1 h with LPS, then various stimulants were added (figure 5). LPS treatment in the absence of further stimulation induced a very small burst of superoxide production and did not induce adhesion. However, the priming effect was very intense: fMLP and Con A induced in primed cells a superoxide production that was six and three fold higher than control cells respectively. The adhesion response to fMLP was also primed by LPS, although to a lesser degree than O2-production. Con A-induced adhesion was not primed, probably because it was already maximal. Unespected and previously not reported findings were revealed by using opsonized zymosan as

MEASUREMENT OF OXIDATIVE METABOLISM ADHESION 87

Fig. 5. Priming of neutrophil responses by LPS (10–8M, 1 h). After stimulation, cells were incubated for 40 minutes.

stimulus: the oxidative and adhesion functions were not increased by LPSpretreatment, excluding the small effect of the LPS alone. This indicated that, at least in these experimental conditions, priming does not involve increase of the effector systems necessary for 02- or phagocytosis (e.g. NADPH oxidase and contractile apparatus respectively), but involves other regulatory steps that are probably more proximal to the receptor and membrane transduction systems. The changes of cell activation apparatus that are responsible for priming by LPS are under investigation, but it is conceivable that these changes are more “subtle” and easy to be triggered by low doses. Therefore we looked for possible low-dose and ultra-low dose effects of priming by LPS. The data of figure 6 show that LPS pretreatment primed the cells to respond to fMLP at doses as low

88 ULTRA LOW DOSES

as 10–8M, but the effect decreased rapidly at further dilutions and became undetectable below 10–10M. The adhesion appeared to be slightly more sensitive to priming than the superoxide production. In any case, since 10–8M corresponds to approximately 0.1 µg/ml, it can be speculated that in 5 liters of blood a similar effect on circulating neutrophils could occur in vivo by a dose as low as 0.5 mg of endotoxin. Effects of inhibitors The assay described in this paper could be very useful for screening of drug effects on two peculiar leukocyte functions. Besides the inhibitory agents used as tools for the study of neutrophil biochemistry, also the drugs used in clinical as anti-inflammatory agents are currently characterized in vitro models. Here we report some experiments using three compounds that have been shown to have antagonistic activity in various functions of neutrophils (excluding adhesion, that has been studied very rarely). This approach was chosen in order to check the validity of the method for this kind of research and also to look for possible inhibition effects in the ultra-low dose range. The sulfydryl reagent n-ethyl maleimide was found to be an extremely potent inhibitor of both the evaluated leukocyte functions, that were totally blocked at doses of 10–5M irrespective of the stimulatory agent utilized (fig.7). However, by examing the various dose-response curves, we noted some differences: First, the stimulation by fMLP was more sensitive to inhibition than the stimulation by zymosan. Second, with STZ as stimulus, the adhesion was blocked at doses of NEM (5×10–6M) that were much less inhibitory on the superoxide production. These data indicate that selected functions (adhesion) and selected receptorspecific responses (e.g. activation by fMLP) are more dependent on the integrity of functional sulfhydryl groups. Fig. 8 shows the effect of a different type of inhibitor, the nucleotide adenosine, that interacts with specific membrane receptors and has been shown to inhibit the superoxide anion production but not the aggregation of neutrophils [7, 14–15]. It has been suggested that adenosine exerts its inhibitory effects probably by increasing the intracellular levels of CAMP [16]. In our conditions, adenosine inhibited both superoxide generation and adhesion induced by fMLP, with similar dose-response curves. Finally, we tested the effects of corticosteroids such as dexametasone and hydrocortisone at doses between 10–5M and 10–15M. Contrary to what reported by others [8, 9] these agents were totally ineffective on the considered neutrophil functions, even at doses much higher than those, used in corticosteroid therapy (data not shown).

MEASUREMENT OF OXIDATIVE METABOLISM ADHESION 89

Fig.6. Dose-response of the priming effect of LPS. Neutrophils were pretreated for 1 h with various doses of LPS, then they were stimulated with 10–7M fMLP, according to the scheme 2 of the methods. Filled triangles: pretreatment with LPS, stimulation with fMLP; open triangles: control cells pretreated with dilutions of dimethylsulfoxide in 0.9% NaCl, stimulation with fMLP; closed squares: pretreatment with LPS, no stimulation.

DISCUSSION We have described a testing system for the detection and evaluation of stimulatory and inhibitory compounds of neutrophils. The assay is sensitive and versatile, allowing to perform a two-three plates experiment starting from a relatively small amount of blood (40 ml), that can be easily obtained as suplemental sample from blood donors. for each plate, up to 12 different doses of a test compound may be evaluated in triplicate with controls no blanks. This test is based on a microplate assay for superoxide production by leukocytes, similar to others that have been previously reported [17]. We have established experimental conditions that allow to extend the application of the assay to the simultaneous measurement of the adhesion of the neutrophils. Adhesion is an important leukocyte function, that is the objet of active investigations [18], because it is involved in various steps of their activity including the regulation of the marginated and circulating pools, extravasation, movement into the connective tissue, and phagocytosis. Genetic diseases of cell adhesion mechanisms cause serious impairment of the defenses against bacterial infections [19]. It has been recently demonstrated that the adhesion capability of neutrophils (and of other leukocytes) is not stable, but is modulated by the expression of anchoring systems on the surface of the cell, that, in turn, is controlled by extracellular compounds such as bacterial products, inflammation mediators and cytokines [18, 20]. The data here reported confirm that measurement of cell adhesion is a sensitive and reliable parameter of the

90 ULTRA LOW DOSES

Fig.7. Effects of NEM on neutrophil functions. Cells were pretreated for 1 h with NEM, then stimulated with 10–7M fMLP or 0.2 mg/ml opsonized zymosan for 30 min.

activation state of neutrophils, that may be either induced or inhibited by extracellular signals. Besides to the saving of time and materials, simultaneous measurement of two cell functions offers important advantages over existing methodology. This work provides several examples of the fact that a test compound may differentially affect the two considered cell functions with respect to both kinetics and dosedependence. The reason of these differences lies in the complex—and in part still unknown—biochemical events that regulate the receptor sensitivity and the post-receptor signalling pathways. Analysis of each of the experimental findings here reported is outside the scope of this paper, that is methodologically-oriented. In general, the following indications emerge from our findings: a) stimuli that activate the oxidative metabolism also activate the adhesion, but the kinetics of the two phenomena

MEASUREMENT OF OXIDATIVE METABOLISM ADHESION 91

Fig.8. effects of adenosine on neutrophil functions. Normal neutrophils and neutrophils that were pretreated with 10–8M LPS were treated for 10 min with adenosine, then were stimulated with 10–7M fMLP for 20 min.

vary according to the stimulus, b) the two considered responses may be differentially affected by agonists or inhibitors, c) adhesion may occur in the absence of superoxide production: low doses of stimulants such as Con A (fig. 3) or high doses of fMLP in desensitized cells (fig.4) can induce adhesion without activation of the respiratory burst, d) adhesion is not a pre-requisite for metabolic activation and for the increased responses observed after priming by LPS. The above conclusions are in agreement with the view that the various signaltransduction pathways for the different leukocyte responses may be experimentally separated and analysed. So far, the relationship between adhesion and metabolic activation has received little attention, also because methods for the simultaneous quantitation of the two parameters are not available. The classic view that adherence to substrates causes spreading of the cell in the attempt to phagocytose them (“frustrated phagocytosis”) and therefore causes activation of the respiratory burst should be reviewed. This simultaneous activation

92 ULTRA LOW DOSES

undoubtely occurs when neutrophils adhere to uncoated plastic wells, or to surfaces coated with immuno-complexes, but we have here demonstrated that on a more physiological substrate (in this case serum proteins) and with particular agonists (in this case Con A, but it is conceivable that other lectins can cause similar possible that Con A, at least at low doses, binds toreceptors that mediate adhesion, but not the burst, in a manner similar to that described for C3b [21]. The method here described was adopted to investigate the effects on neutrophil functions of some selected compounds usqed at ultra-low doses, that is at doses several orders of magnitude lower than those that are expected to have any effect on the basis of previous knowledge. The rationale of this approach is based on the fact that recent papers reported that solutions of stimulants and inhibitors that had been prepared acceding to special dilution procedures (very close to those employed by homeopathic pharmacopea) exhibited a biological activity even in the ultra-low dose range [11–13, 22–24]. In particular, it was reported that in the first dilution steps the activity decreased, as expected, but further dilutions caused the re-appearance of the activity, that was therefore attributed to some unknown “meta-molecular” [13] biological effect. These paradoxical observations raised a number of questions and also controversies based on methodologic considerations [25–27]. However, in our opinion, these reports were of interest because pointed to the possible existence of previously unrecognized biophysiacl phenomena and because they documented the effort of setting-up laboratory models of investigation of ultra-low dose pharmacology that, as it is well known, has a wide application in clinical practise in a number of countries. We therefore did not consider “unbelievable” [25] this approach and we tried to reproduce, using our model system, similar effects. The experiments here reported show that we were not able of obtaining ultralow dose effects on neutrophil superoxide production and adhesion. Active doses were very low, especially in the adhesion tests, but these doses probably can not be defined as ultra-low in the sense that they might exert their action through some meta-molecular effects. However, these are only preliminary studies, and it can not be excluded that by using different preparations of test compounds or a different experimental design, such effects could be found. For example, the following points are opened to further investigations: a) the establishment of optimal methods of dilution and succussion of test compounds, b) the testing of other compounds and also of commercial drug preparations, c) the exploration of effects of higher dilutions, beyond those utilized in this work, d) the study of other biological and biochemical events that could be affected by ultra-low doses of agonists or antagonists, such as changes of ionic fluxes, of membrane permeability, of gene expression, etc. ACKNOWLEDGEMENTS This work was supported by grants from University of Verona and from Guna S.r.l. (Milano).

MEASUREMENT OF OXIDATIVE METABOLISM ADHESION 93

REFERENCES Anderson, D.C. and Springer, T.A., [19], 1987, Leukocyte adhesion deficiency: an inherited in the Mac-1, LFA-1, and p150, 95 glycoproteins. Annu. Rev. Med., 38, 175–194. Babior, B.M., [1], 1984, Oxidants from phagocytes: Agents of defense and destruction. Blood, 64, 959–966. Benveniste, J., [27], 1988, Nature, 334, 291. Cain, J.A., Newman, S.L. and Ross, G.D., [21], 1987, Role of complement receptor type three and serum opsonins in the neutrophil response to yeast. Complement, 4, 75–86. Coates, T.D., Wolach, B., Tzeng, D.Y., Higgins, C., Baehner, R.L. and Boxer, L.A., [9], 1983, The mechanism of action of the anti-inflammatory agents dexametasone and Auranofin in human polymorphonuclear leukocytes. Blood, 62, 1070–1077. Cronstein, B.N., Kramer, S.B., Weissman, G. and Hirshhorn, R., [7], 1983, Adenosine: A physiological modulator of superoxide anion generation by human neutrophils. J. Exp. Med., 158, 1160–1177. Cronstein, B.N., Daguma, L., Nichols, D., Hutchison, A.J. and Williams, M., [15], 1990, The adenosine/neutrophil paradox resolved: Human neutrophils possess both A1 and A2 receptors that promote chemotaxis and inhibit 02-generation, respectively. J. Clin. Invest., 85, 1150–1157. Davenas, E., Beauvais, F., Amara, J., Robinson, M., Miadonna, A., Tedeschi, A., Pomeranz, B., Fortner, P., Belon, P., Sainte-Laudy, J., Poitevin, B. and Benveniste, J., [13], 1988, Human basophil degranulation triggered by very dilute antiserum against IgE. Nature, 333, 816–818. Davenas, E., Poitevin, B. and Benveniste, J., [22], 1987, Effect on mouse peritoneal macrophages of orally administered very high dilutions of silica. Eur. J. Pharmacol., 135, 313–319. Doucet-Jaboeuf, M., Guillemain, G., Piechaczyk, M., Karouby, Y. and Bastide, M., [23], 1982, Evaluation de la dose limite d’activité du facteur thymique serique. C.R. Acad. Sc., Paris, 295, 283–286. Elliot, K.R.F., Miller, P.J., Stevenson, H.R. and Leonard, E.J., [16], 1986, Synergistic action of adenosine and fMET-LEU-PHE in raising cAMP content of purified human monocytes. Biochem. Biophys. Res. Commun., 138, 1376–1382. Guthrie, L.A., McPhail, L.C., Henson, P.M. and Johnston, R.B., [5], 1984, Priming of neutrophils for enhanced release of oxygen metabolites by bacterial lipopolysaccharide. Evidence for increased activity of the superoxide-producing enzyme. J. Exp. Med., 160, 1657–1671. Goldstein, I.M., Roos, D., Weissmann, G. and Kaplan, H.B., [8], 1976, Influence of corticosteroids on human polymorphonu- clear leukocyte function in vitro: reduction of lysosomal enzyme release and superoxide production. Inflammation, 1, 305–315. Harish, G. and Kretschmer, M., [24], 1988, Smallest zinc quantities affect the histamine release from peritoneal mast cells of the rat. Experientia, 44, 761–762. Horejsi, V. and Bazil, V., [20], 1988, Surface proteins and glycoproteins of human leukocytes. Biochem. J., 253, 1–26. Kaku, M., Yagawa, K., Nagao, S. and Tanaka, A., [4], 1983, Enhanced superoxide anion release from phagocytes by muramyl dipeptide or lipopolysaccharide. Infect. Immun., 39, 559–564.

94 ULTRA LOW DOSES

Maddox, J., Randi, J. and Stewart, W.W., [26], 1988, “High-dilution” experiments a delusion. Nature, 334, 287–290. Metcalf, J.A., Gallin, J.I., Nauseef, W.M. and Root R.K., [17], 1986, Laboratory manual of neutrophil function, Raven Press, New York. Poitevin, B., Aubin, M. and Benveniste, J., [11], 1986, Approche d’une analyse quantitative de l’effet d’apis mellifica sur la degranulation des basophiles humains in vitro. Innov. Tech. Biol. Med., 7, 64–68. Poitevin, B., Davenas, E. and Benveniste, J., [12], 1988, In vitro immunological degranulation of human basophils is modulated by Lung histamine and Apis mellifica. Brit. J. Clin. Pharmacol., 25, 439–444. Pool, R., [25], 1988, Unbelievable results spark a controversy. Science, 241, 407. Simkowitz, L., Atkinson, J.P. and Spilberg, I., [10], 1980, Stimulus-specific deactivation of chemotactic factor-induced cyclic AMP response and superoxide generation by human neutrophils. J. Clin. Invest., 66, 736–747. Springer, T.A., [18], 1990, Adhesion receptors of the immune system. Nature, 346, 425–433. Ward, P.A., Cunningham, T.W., Walker, B.A.M. and Johnson, K.J., [14], 1988, Differing calcium requirements for regulatory effects of ATP, ATP-gammas and adenosine on 02-response of human neutrophils. Biochem. Biophys. Res. Commun., 154, 746–751. Ward, P.A., Warren, J.S. and Johnson, K.J., [2], 1988, Oxygen radicals, inflammation, and tissue injury. Free Rad. Biol. Med., 5, 403–408. Weiss, S.J., [3], 1989, Tissue destruction by neutrophils. N. Engl. J. Med., 320, 365–376. Yamashita, T., [6], 1983, Effect of maleimide derivatives, sulfhydryl reagents, on stimulation of neutrophil superoxide anion generation with concanavalin. A. FEBS Lett., 164, 267–271.

NON-STIMULATORY CONCENTRATIONS OF CONCANAVALIN A CAN MODULATE SUBSEQUENT STIMULATION BY THE SAME MITOGEN D.P.Eskinazi, D.D.S., Ph.D1 and A.Molinaro2, M.D. 1Columbia,

MD, 2Loma Linda University, Loma Linda, CA

We evaluated whether concentrations of biologically-active substances too low to induce a direct effect can modulate a response induced by a higher concentration of the same substance. We took as experimental model the concanavalin A (Con A)-dependent mitogenesis of mononuclear cells as assessed by 3H-thymidine incorporation. In seven consecutive experiments, we observed that pre-treatment with non-mitogenic doses of Con A modulated the proliferation induced by suboptimal stimulatory concentrations of Con A. INTRODUCTION Recent work has confirmed that various biologically-active factors can be effective at very low concentrations. Thus, some lymphokines can exert an effect at concentrations as low as 10–15 M (Klein, 1990) and endogenous opioids can enhance cellular cytotoxicity at similar concentrations (Sibinga and Goldstein, 1988). Further, it has been shown that two or more factors such as lymphokines can act in synergy (Durum and Oppenheim, 1989). In this context, factor concentrations too low to produce a direct effect could still modulate the response to stimulatory concentrations of synergistic substances. The present studies were designed to extend this concept to self-modulation and evaluate whether a substance too low to promote direct effects could still modulate effects induced by higher concentrations of the same substance. MATERIALS AND METHODS Isolation of Peripheral Mononuclear Cells Ten to 100 ml of freshly drawn blood was diluted 1:2 in RPMI 1640. Two volumes of diluted blood were layered onto one volume of Ficoll-Paque (Pharmaceutical

96 ULTRA LOW DOSES

LKB, Uppsala, Sweden). After centrifugation at 400×g at room temperature for 30 minutes, the cells at the plasma/Ficoll interface were collected, diluted in 10 volumes of RPMI 1640, and centrifuged for 200×g at room temperature for 10 minutes. The pelleted cells were washed two more times in a similar fashion, and finally resuspended in RPMI 1640 supplemented with heat inactivated fetal bovine serum (10%), penicillin (50 U/ml) and streptomycin (50 µg/ml). Mononuclear Cell Stimulation Assays 1×105 cells in 180 µl of medium were pipetted into wells of a microtiter culture plate containing 20 µl of appropriate modulatory concentrations of Con A and cultured at 37°C for one hour in a 5% CO2 humidified chamber. Then 20 µl of either of four stimulatory concentrations of Con A was added to triplicate wells to provide final concentrations of 3.0 to 0.1 µg/ml and the plates were reincubated. After three days, 1 µCi of 3H-Thy in 20 µl of RPMI 1640 was added to each well. After incubation at 37°C for four hours, the cells were harvested on glass strips and the 3HR-Thy incorporation was counted in a beta scintillation radioactivity counter. Statistical Analysis Data were subjected to the analysis of variance (ANOVA). RESULTS The effectiveness of preincubations with dilutions of Con A to modulate the response of peripheral blood mononuclear cells to subsequent stimulatroy doses of Caon A was evaluated. In each experiment four stimulatory doses of Con A were used; prior to being incubated with each of the stimulatory doses, cells were preincubated either with medium not containing Con A or containing Con A concentrations ranging from 1×10–3 to 3×10–7 µg/ml. In experiments using doses of Con A inducing maximal stimulation of the lymphocytes, we found no modulatory effect of the preincubations. In seven successive experiments using one or more doses of Con A inducing submaximal proliferation, we found that preincubation with dilutions of Con A resulted, within eah experiment, in a statistically significant modulation of the proliferation, with p values ranging between 0.05 and 0.0005 depending on the experiment (Fig.1). To test for a reproducible operator-dependent artifact, in one of these experiments, the blood sample was divided in two equal aliquots that were handled independently by two technicians. Thus in this experiment, cell isolation, stimulatory Con A dilutions, modulatory dilutions, and experimental set up were prepared in parallel but independently. In the microtest plates handled by either technician, two of the stimulatory doses of Con A resulted in submaximal stimulation. For each of those doses, the level of unmodulated stimulation was

NON-SIMULATORY CONCETRATIONS OF CONCANAVALIN A MODULATION 97

Fig.1. Prestimulation with low concentrations of Con A can significantly modulate subsequent Con A stimulation. This experiment is representative of seven consecutive experiments. “30%”, “10%”, “3%” and “1%” refer to percentages of the Con A concentration inducing maximal proliferation (3 µg/ml). One or more of the 30, 10, 3 or 1% dilutions produced measurable suboptimal stimulation. Preincubations were carried out with serial threefold dilutions of a 1×10–3 µg/ml Con A solution.

somewhat dissimilar in the microtest plates handled by one versus the other technician; the level of modulation observed was concomitantly different. For the second stimulatory dose of Con A that induced submaximal stimulation, the level of proliferation was quite similar in the microtest plates handled by either technician; the level of modulation was concomitantly very similar (Fig.2). To test for a reproducible design-dependent artifact, an experiment was conducted with alternating control (i.e., preincubation not containing modulatory doses of Con A) and experimental (i.e., preincubation containing modulatory doses of Con A) triplicates. Proliferation in control wells was observed to be significantly lower than in the experimental ones (p<0.005) (Fig.3).

98 ULTRA LOW DOSES

Fig.2. Experiment conducted to evaluate the possibility of operator-dependent artifacts. Two technicians handled in parallel aliquots of the same blood sample (see text). Data obtained by either operator are presented in the left and right panels, respectively. Modulatory and stimulatory solutions of Con A were prepared as indicated in Fig.1.

DISCUSSION Our data are compatible with the hypothesis that dilutions of a given substance can modulate the biological effect(s) that this same substance exhibits at higher concentrations. In seven consecutive experiments, we observed that the variations in thymidine incorporation induced by preincubation with modulatory concentrations of Con A were significantly higher than the experimental variation for this particular experiment. While the statistical analysis indicated that our data cannot be attributable to chance alone, the observed modulation was relatively low and was not always consistent from experiment to experiment. Both limitations have been experienced by others laboratories (Zighelboim, Pomeranz, personal

NON-SIMULATORY CONCETRATIONS OF CONCANAVALIN A MODULATION 99

Fig.3. Experiment conducted to evaluate the possibility of design-dependent artifacts. Either medium or medium+modulatory Con A solutions were added in alternate triplicate wells throughout the microtiter plates. Preincubations were carried out with serial tenfold dilutions of 1×10–3 µg/ml Con A solution.

communications) for similar types of experiments, albeit in different experimental models. However, when the same blood sample was handled by two operators independently the modulatory effect was observed in both experiments, for the level of Con A (3%) that gave comparable levels of stimulations in cultures handled by both technicians, the modulations were quite parallel. Further, the experiment utilizing alternating controls and experimental triplicates suggests that the observed modulation is not simply due to a reproducible flaw in the experimental design. It is known that reproducibility may be a problem in this assay: mononuclear cell proliferation in response to stimulatroy doses of Con A (or other mitogens) is itself quite variable. This problem has been addressed by routinely performing mitogen stimulation at several concentrations to ensure that the maximal stimulation is obtained. Thus, lack of reproducibility can stem from the fact that Con A stimulation has a relatively high coefficient of variation compounded by high coefficient of variation of the modulatory effect itself. Notably, the reproducibility of subsequent experiments was poorer. It is likely that part of the problem is due to inherent limitations of the experimental system as suggested by observations made by others (Pomeranz, Zighelboim, personal communications). In laboratory science, one strives to obtain clear-cut data, and modulation is expected to result in considerable magnification or inhibition. Even in clinical

100 ULTRA LOW DOSES

practise, the trend is to utilize ever more potent pharmacological agents in increasingly higher concentrations (one looks for the “maximum tolerated dose”). There are, however, data indicating that very small concentrations can be very active: pheromones, endorphins, some lymphokines are all active at 10–15 M (1–3). In fact, it has been shown that lower doses are sometimes far more powerful than higher ones. Thus, in an experimental model, it has been shown recently that a chemotherapeutic agent was more effective at much lower doses than previously utilized (Mokir and Dray, 1987). These data may be akin to the ones presented here. The limitations facings the study of biological effects of low concentrations are probably inherent to the nature of effects observed rather than to the experimental system. However, they can be addressed by using a refined methodology. REFERENCES Durum, S.K. and Oppenheim, J.J., 1989, Macrophage-derivated mediators: IL-1, TNF, IL-6, IFN, and related cytokines. In Fundamental Immunology, New York, NY, edited by W.E.Paul (New York, NY: Raven press). Klein, J., 1990, Immunology, (Boston, MA: Blackwell Scientific Publications), pp. 227. Mokir, M.B. and Dray, S., 1987, Interplay between the toxic effects of anticancer drugs and host antitumor immunity in cancer therapy. Cancer Investigation, 5, 31–38. Sibinga, N.E.S. and Goldstein, A., 1988, Opioid peptides and opioid receptors in cells of the immune system. Annual Review Immunology, 6, 219–249.

BIOLOGICAL ACTIVITY OF ULTRA LOW DOSES: I/EFFECT OF ULTRA LOW DOSES OF HISTAMINE ON HUMAN BASOPHIL DEGRANULATION TRIGGERED BY D.PTERONYSSINUS EXTRACT Sainte-Laudy, J.*, Sambucy, J.L.*, and Belon, P.** * Centre d’Etudes et de Recherches en Immunologie, 20 Rue de la Pompe, 75016 Paris, France ** Institut BOIRON, 20 Rue de la Liberation 69110 Sainte FoyLès-Lyon, France INTRODUCTION The use of human basophil degranulation by an optical method as an in vitro model bring two kinds of difficulties: – the necessity to demonstrate the statistical reliability of the model, knowing the lack of publications in the scientific literature; this latter concerning mostly the diagnostical aspect of the model (Shelley 1961). – the scientific fact itself leading to the conclusion that solutions containing theoretically no molecules of the effectors may have a reproducible and specific biological activity. In first we so standardized this in vitro model for its use in pharmacology (Sainte-Laudy 1982) and also mathematically modelized the degranulation curve (Sainte-Laudy 1987). In fact the studied phenomenon commonly called degranulation is more exactly the loss of tinctorial affinity of the intracytoplasmic granules for specific dyes such as toluidine blue and alcian blue. By experience the main interest of this model is to be more sensitive to the effect of various exogenous effectors than the histamine release. In the present case of ultra low doses of histamine the first results obtained were in favour of the use of the optical method, the results of the histamine release being non significant. The existence of biological activity of ultra low doses has been reported by more than five hundred references in homoeopathic literature, e.g. (Boiron 1952), (Boiron 1962), (Sainte-Laudy 1986). By “ultra low doses” we mean here dilutions in which the presence of the effector is less than one molecule per liter. Over the past few years results in favour of this activity have been published, in scientific literature such as:

102 ULTRA LOW DOSES

– Enhancement of Platelet Activating Factor release by preincubation of mouse peritoneal macrophages with highly (1.66 10–9M and 1.66 10–17M) diluted Silicea (Davenas 1987); – Increased excretion of Arsenic by highly (from the 5th to the 15th centesimal) diluted Arsenic, in Arsenic-intoxicated rats (Cazin 1987); – Inhibition of IgE-Triggered human basophil degranulation by LungHistamine and Apis mellifica (Poitevin 1988); (Lung-Histamine is an extract from guinea-pig lungs after anaphylactic shock, and Apis mellifica, an extract from honey bee); – Stimulation of human basophil degranulation with highly (from the 5th to the 60th centesimal) diluted anti-IgE (Davenas 1988). – Clinical improvement of hay fever by treatment with a grass pollen extract diluted to the 30th centesimal (Reilly 1986). We present here results obtained with the first protocol used. Passive sensitization of human basophils followed by the degranulation triggered by the specific allergen (house dust mite). We studied and compared the activity of ultra low doses of histamine and histidine and also the activity in this model of an antiH2 (cimetidine) and a specific enzyme (histaminase). MATERIALS AND METHODS A/ Preparation of the Basophil Suspension 20 ml of venous blood were taken (on 5 UI/ml heparin, 5mM EDTA) from 12 healthy donors. As described by (Pruzansky 1988), the plasma was enriched in total leukocytes by sedimentation at 1 g, washed three times in buffer A (20mM hepes, 130mM NaCl, 5mM KCl, 5mM glucose, pH 7.4) and then centrifuged (500g×10 min). The cell pellets were washed with buffer A, centrifuged (500g×10 min), mixed, and incubated in a lactic acid buffer (120mM lactic acid, 5mM HCl, pH 3.5) for 5 min at 20+2°C. All the buffers were prepared with distilled water. The leukocytes were washed again using the same buffer and then incubated 3 h at 4°C with the specific IgE-rich sera diluted to the 1/20th. The specific IgE-rich sera pool was obtained from 10 patients with atopic dermatitis and sensitized to house dust mite (D.Pteronyssinus), showing positive skin tests and a level of specific IgE higher than class 3 measured by RAST (Phadezym-RAST Pharmacia). All the experiments were performed using this same pool. After another wash with buffer A, the sensitized basophil suspension was adjusted to 1500+ 200 basophils/mm3.

BIOLOGICAL ACTIVITY OF ULTRA LOW DOSES:I 103

B/ Preparation of the Dilutions of Effector and Pre-incubation with the Sensitized Basophils We used several effectors: histamine, histidine, cimetidine, histaminase. – Histamine dilutions (Sigma histamine hydrochloride) were prepared in distilled water, and in disposable polystyrene flaks (CML) and kept at 4°C, from a first molar solution, by serial dilutions to the hundredth, with vigorous shaking (150 times in 10 s on an automatic apparatus). In order to make them isotonic, 9 volumes of histamine dilution were added before each experiment to 1 volume of buffer C (200mM hepes, 1300mM NaCl, 50mM KCl, 50mM glucose, pH 7.4). – Histidine dilutions (histidine hydrochloride Sigma) were prepared as described for histamine. – 20 µl cimetidine (SKF) diluted to 10–5M in buffer A was added to 20 µl of the preincubation medium. – Histaminase (Sigma diamine-oxydase) diluted to 0.01 mg/ml in buffer A was mixed v/v to each histamine dilution and then all were incubated at 37°C for 1 hour. 20 µl of each dilution of effector (histamine or histidine) was mixed with 20 µl of the cell suspension and incubated at 20+2°C for 30 min in 1 ml disposable polystyrene tubes (CML). C/ Basophil Degranulation The allergen extract (Dome/Hollister-Stier freeze-dryed alpha fraction) was diluted to 10–4mg/ml in buffer B (20mM hepes, 115mM NaCl, 5mM KCL, 5mM glucose, pH 7.4) from a 10mg/ml stock solution kept at −80°C. 20 µl of this solution were mixed 20 µl of cell suspension in microtiter plate wells (CML) and then incubated at 37°C for 20 min. For every histamine dilution an allergen-free control was performed in parallel. D/ Basophil Staining and Counting 140 µl of toluidine blue solution (750mg/l Sigma toluidine blue, 25 % ethanol, pH 3.9) were added to every well and then incubated for 30 min at 20+2°C. The plates could be covered with paper (Dynatech plate sealer) and stored at 4°C. Under these conditions the readings could be delayed for several days. After gentle homogenization by drawing-off and blowing back with a micropipette, the basophil suspensions were distributed in the hemocytometers and after 2 to 3 min of sedimentation in a moist chamber the basophils appeared reddish blue on a light blue background. The basophils were then counted once. Cell distribution in the hemocytometer had to be regular (absence of aggregates).

104 ULTRA LOW DOSES

E/ Processing of the Data All the results obtained with this protocol have the same structure i.e. two basophil counts foe each tested effector dilution: – a first row corresponding to the controls containing the effector without allergen and another control effector free, – a second row with the effector dilution and with the allergen. For each experiment we used onleukocyte pool and 16 dilutions of either histamine or histidine (5 to 20th dilution to the hundredth) and another one effector free. Ten experiments were performed with 56 dilutions of histamine (5 to 60th dilutions to hundredth) and 10 different leukocytes pools. Having previously demonstrated that the basophil counts followed a normal distribution (Sainte-Laudy 1987) we checked that for the experiments performed the two rows of counts had the same variance, so we used the method of the comparison of the means by the t-Student test. The relative activity of the different effectors was compared by splitting the histamine dilutions series into two groups: – group A corresponding to dilution 5C to 12C equivalent to 10–10M and to 10–24M respectively, – group B corresponding to dilution 13C to 20C equivalent to 10–26M and to 10–40M respectively. This was done in order to compare the activity of dilutions in which molecular presence of the effector is probable with the activity of dilutions in which molecular presence is highly improbable. For these two groups of histamine dilutions we compared the mean basophil counts. All reagents were prepared just before each experiment which took two days on an average. The results were also expressed in mean percentages of inhibition+S.E.M. for the n experiments. The percentage of inhibition was defined as wherein: – A is the number of stained basophils for the allergen and effector free control – B is the number of stained basophils for the effector free control but with the allergen – C is the number of stained basophils for allergen free control with the effector – D is the number of stained basophils with the effector and with the allergen.

BIOLOGICAL ACTIVITY OF ULTRA LOW DOSES:I 105

RESULTS A/ Statistical Reliability of the methods 1) Passive sensitization

To demonstrate the reliability of the passive sensitization step we calculated the intra-assay coefficient of variation on 21 experiments. The mean percentage of degranulation was 74.3+4.5% corresponding to a coefficient of variation of 6%. All the experiments were performed with the same IgE reach plasma kept at −20°C. 2) Statistical distribution of the counts

The statistical reliability of the method to assess degranulation basophil has often been discussed but very little data are available in the literature (SainteLaudy 1987), (Gérard 1981) (Petiot 1981). The different authors concluded that the distribution of the basophil counts is normal. In our experimental conditions the coefficients of variation of ten successive series of 16 basophils counts ranged from 4% to 7.5% for the allergen-free control in which the number of counted basophils ranged from 50 to 65. The mean coefficient of variation for the ten experiments was 5.5%. We then controlled coefficients of variation in 10–5M cimetidine (cf infra) which was taken as a negative control. As expected, the calculated coefficients of variation were higher than those found for the allergen-free controls (3% to 14. 9%) the number of counted basophils being much lower (between 10 and 25). Nevertheless the mean coefficient of variation (11.8%) remained within generally accepted limits for pharmacological studies based on cellular tests. B/ Effect of Histamine and Histamine Dilution on Basophil Degranulation In eight sucessive experiments we compared the activity of dilutions of histamine and histidine on the 8 different leukocyte pools. To evaluate the significance of a single experiment we took into account the absolute basophil counts. For each experiment, the inhibition of basophil degranulation by the dilutions of histamine was statistically significant. (0.05>p>0.001) in 5 out of 8 experiments for group A of dilutions (5C–12C) and in 6 out of 8 for group B of dilutions (13C–20C). The dilutions of histidine showed no activity. Since inhibition by histamine took place for similar dilutions from experiment to experiment, we calculated the mean curve for the eight experiments. This

106 ULTRA LOW DOSES

Figure 1. Inhibition of basophil degranulation by successive dilutions to the hundredth of histamine. Reversibility of this inhibitory activity by the addition of histaminase (0.01 mg/ ml) to each histamine dilution. The results are expressed in mean percentages of inhibition ±S.E.M. on ten experiments (histamine alone=closed bars; histamine+histaminase=open bars). The statistical difference of the percentages of inhibition are indicated: * p<0.005; ** p<0.01; *** p<0.001.

curve peaks at 6C and 17C for dilutions of histamine, the inhibition of basophil degranulation reaching 75% and 69% respectively. C/ Effect of Histaminase on the Inhibition of Basophil Degranulation by Dilutions of Histamine After having individually treated for each experiment the sixteen dilutions of histamine ranging from 5C to 20C with histaminase before preincubation with the sensitized leukocytes, the effect of these dilutions were compared to the effect of the untreated dilutions of histamine. For group A dilutions, the preincubation of the histamine dilutions with histaminase significantly suppressed (0.05>p>0.001) the inhibition by dilutions of histamine in 8 out of 10 experiments, whereas no effect was observed for group B of dilutions. The mean percentages of inhibition are presented (Figure 1), they peak at dilution 7C and 18C for the untreated dilutions of histamine (8/ 10 significant inhibition) and only at dilution 18C for the histamine dilutions previously treated with histaminase. The statistical difference between the two peaks in group B was not statistically significant.

BIOLOGICAL ACTIVITY OF ULTRA LOW DOSES:I 107

Figure 2. Inhibition of basophil degranulation by successive dilutions to the hundredth of histamine. Reversibility of this inhibitory activity by the addition of cimetidine (10−5 M) to the preincubation medium. The results are expressed in mean percentages of inhibition ±S.E.M. on ten experiments (histamine alone=closed bars; histamine+cimetidine=open bars). The statistical difference of the percentages of inhibition are indicated: * p<0.05; ** p<0.01; *** p<0.001.

D/ Effect of Cimetidine on Inhibition of Basophil Degranulation by Histamine Dilutions Cimetidine was added at a final concentration of 10–5M in the incubation medium containing the histamine dilutions and the sensitized leukocytes. As compared to the cell suspensions without cimetidine the number of basophils counted after addition cimetidine was significantly lower (0.05>p>0.01) in 9 out of 10 in group A, and 8 out of 10 in group B. The mean curve of the ten experiments peaks at dilution 6C and 17C in the absence of cimetidine and shows no activity for any of the histamine dilutions in the presence of 10–5M cimetidine (Figure 2). E/ Time-effect phenomenon In two experiments we have compared the mean basophil counts corresponding to 7 successive dilutions of the mite extract and one allergen free control for preincubation delays of 0, 3, 6, 9, 12, 15, 20 minutes for the first experiment; and 0, 10, 15, 20, 25, 30 minutes for the second one.

108 ULTRA LOW DOSES

Simultaneously we compared the activity of the 6th and 17th dilutions (to the hundredth) of histamine. As presented in table I and II we observed a progressive decrease of the standard error in parallel with an increase of the mean basophil counts almost reaching the mean basophil counts of the controls. The difference of the means is significant after 15 minutes of preincubation for the 6th dilutions of histamine and after 20 minutes of preincubation, of the 17th dilutions of histamine. Table I Time-effect phenomenon: First experiment; different preincubation times were checked on the same basophil pool for 2 dilutions of histamine (100−6 and 100−17); NS=non significant, *=0.02
0 33.4

3 32.8

6 32.7

9 36

12 41.1

8.6 33.7

8.7 32.5

9.3 35.4

8.7 4.5 37 40.7

3.9** 2.9*** 43.5 43.8

10.2

9.5

9.4

7.9

4.6

8.8

15 44.7

20 46.2

4*

Table II Time-effect phenomenon: Second experiment; different preincubation times were checked on the same basophil pool for 2 dilutions of histamine (100−6 and 100−17); NS=non significant, *=p<0.05. Preincubation time in minutes 100−6 dilution of histamine: mean basophil counts standard error 100−17 dilution of histamine: mean basophil counts standard error

0 35.8

10 40.6

10.8 39.1 10

15 48

20 46

25 46.5

30 46.1

9.1 5.4* 39.4 42

4.6* 48

3.4* 45.8

3.2* 47.4

6.9

4.3*

2.4*

3.5*

8.6

F/ Inhibition of basophil degranulation by ultra low doses of histamine from the 5th to the 59th centesimal dilution On 10 experiments, we found five peaks of inhibition centered on 7C, 17C, 27C, 39C, 51C, without any significant decrease of the maximum inhibitory activity (Figure 3).

BIOLOGICAL ACTIVITY OF ULTRA LOW DOSES:I 109

Figure 3. Inhibition of basophil degranulation by successive dilutions to the hundredth of histamine from the 5th to the 59th. The results are expressed in mean percentages of inhibition±S.E.M. on ten experiments. The statistical significance is indicated: * p<0.05; ** p<0.01; *** P<0.001.

DISCUSSION Mixing group A and group B dilutions of histamine have a statistical biological activity on basophil degranulation in 40 out of 46 experiments. So we have to point out the high frequency of this effect and the reproducibility of the situation of the optima centred on the 7th and the 17th dilutions; these inhibition peaks are found also for higher histamine dilutions up to the 51th dilution. This inhibitory activity is suppressed by an H2 antagonist (cimetidine and by a specific enzyme histaminase) for the first group of dilutions (group A), but this enzyme is inactive on the second group of dilutions (group B), may be due to the use of an inadequate enzyme concentration. The reversibility of the inhibition effect by an H2 antagonist is in favour of the participation of the H2 receptors; but it is paradoxal to think in terms of molecular biology when the oretically there are no molecule of the effector in some of the active dilutions tested. Concerning the specificity of the effect observed we have also to point out that histidine has no activity and that a preincubation time of at least 15 minutes is necessary to make the phenomenon appear. However the “in vitro” model used here presents several difficulties as the necessity of a long training for counting the basophils stained with toluidine blue and also the existence in the protocol of a passive sensitization step needing a specific IgE rich plasma pool and a relatively large volume of fresh human blood. Therefore, faced with the necessity of diffusing a protocol, we set up a simplidied one based on direct stimulation of human basophils with a commercially available anti-IgE. This method was compared to histamine release (Sainte-Laudy 1990); the results obtained for ultra low doses of histamine are presented in the second part of this paper.

110 ULTRA LOW DOSES

ACKNOWLEDGEMENTS We sincerely thank: Mr Jean Boiron for this advice and collaboration Mrs Sheila Adrian for her help in rereading and correcting the English of this article. REFERENCES Boiron, H. and Devraigne, M., 1952, Etude expérimentale de l’activité biologique de dilutions homéopathiques de folliculine. Gynécologie Pratique, 3, 3–9. Boiron, J. and Cier, A., 1962, Elimination provoquee et spécificité d’action des dilutions infinitésimales d’éléments toxiques. Annales Homéopathiques Françaises, 10, 789–795. Cazin, J.C., Cazin, M. , Gaborit, J.L., Chaoui, A., Boiron, J., Belon, P., Cherruault, Y. and Papapanayotou, C., 1987, A study of the effect of decimal and centesimal dilutions of Arsenic on the retention and mobilization of Arsenic in the rat. Human Toxicology, 6, 315–320. Davenas, E., Beauvais, F., Amara, J., Oberbaum, M., Robinson, B., Miadonna, A., Tedeschi, A., Pomeranz, B., Fortner, P., Belon, P., Sainte-Laudy, J., Poitevin, B. and Benveniste, J., 1988, Human basophil degranulation triggered by very dilute antiserum against IgE. Nature, 333, 816–820. Davenas, E., Poitevin, B. and Benveniste, J., 1987, Effect on mouse peritoneal macrophages of orally administered very high dilutions of Silicea. European Journal of Pharmacology, 135, 313–319. Gerard, H., Legras, B. and Moneret-Vautrin, D.A., 1981, Le test de dégranulation des basophiles humains (TDBH). Interet d’une leucoconcentration et du calcul statistique applique au taux de dégranulation. Pathologie Biologie, 29, 137–141. Moore, J.E. and James, G.E., 1965, Proceedings of the Society for Experimental Biology and Medecine, 82, 601. Sainte-Laudy, J;, Belon, P. and Halpern, G., 1982, Homeopathy, effect of Histaminum on “in vitro” basophil dégranulation. In: Abstracts of the XI International Congress of Allergy and Clinical Immunology, London, p 338. Petiot, J.F., Sainte-Laudy, J. and Benveniste, J., 1981, Interpretation du resultat d’un test de dégranulation des basophiles humains. Annales de Biologie Clinique, 39, 355–359. Poitevin, B., Davenas, E. and Benveniste, J., 1988, In vitro immunological dégranulation of human basophils is modulated by Lung histamine and apis mellifica. British journal of Clinical Pharmacology, 25, 439–444. Pruzansky, J.J. and Patterson, R., 1986, Binding constants of IgE receptors on human blood basophils. Immunology, 58, 257–262. Reilly, D.T., Taylor, M.A., McSharry, C. and Aitchison, T., 1986, Is homeopathic a placebo response? Controlled trial of homeopathic potency with pollen in hayfever as model. The Lancet, October 18th, 881–886. Sainte-Laudy, J., 1987, Standardization of basophil degranulation for pharmacological studies. Journal of Immunological Method, 98, 279–282.

BIOLOGICAL ACTIVITY OF ULTRA LOW DOSES:I 111

Sainte-Laudy, J. and Belon, P., 1986, Application de modèles d’hypersensibilité du type I à l’étude de l’activité de dilutions hahnemanniennes des médiateurs de cette hypersensibilité. Homéopathie, 3, 40–44. Sainte-Laudy, J. and Henocq, 1990, Reactivity of human basophils to anti-IgE and protein A in atopic dermatitis. Agents and Action, 30, 250–253. Shelley, W. and Juhlin, L., 1961, A new test for detecting anaphylactic sensitivity: the basophil reaction. Nature, 191, 1056.

BIOLOGICAL ACTIVITY OF ULTRA DOSES: II/EFFECT OF ULTRA LOW DOSES OF HISTAMINE ON HUMAN BASOPHIL DEGRANULATION TRIGGERED BY ANTI-IgE Sainte-Laudy, J.* and Belon, P** * Centre d’Etudes et de Recherches en Immunologie, 20 Rue de la Pompe, 75016 Paris, France ** Institut BOIRON, 20 Rue de la Liberation, 69110 Sainte Foy-Lès-Lyon, France

INTRODUCTION Faced with the necessity of a wide circulation of the mode, we set up a second protocol based on: – commercially-available reagents, – direct stimulation of human basophils by an anti-IgE in order to avoid difficult passive sensitization – the use of another dye, alcian blue, which allows rapid and clear-cut basophil count without time-consuming training (Pruzansky 1986), (Gilbert 1975). We also wanted, by using this protocol, to answer one of the main criticisms formulated against our previous results i.e. the necessity of counting basophils “blind”. We are presenting here the results obtained with two different protocols for series of histamine dilutions compared to the same dilutions of pure solvent as controls, and for 16 repeated basophils counts. MATERIALS AND METHODS I/ Samples Selection For each experiment, 20 ml of venous blood where taken, on 5 UI/ml heparin, 5mM EDTA, from 12 subjects. First the blood was enriched with leukocytes by sedimentation progressed, the leukocyte-rich plasma was collected and then washed three times with buffer A (20mM hepes, 130mM NaCl, 5mM KCL, 5mM glucose, pH 7.4).

BIOLOGICAL ACTIVITY OF ULTRA LOW DOSES:II 113

20 µl of the leukocyte-rich plasma were mixed with 20 µl of 1, 0.2 and 0.04 pg/ml solutions of a polyclonal anti-IgE (ATAB Laboratory IgG, fraction affinity purified) diluted in buffer B (20mM hepes, 115mM Nacl, 5mM KCL, 10mM CaCl2, 2mM mgCl2, 5mM glucose, 0.1% BSA, pH 7.4). After a 20 min incubation period at 37°C, basophils were stained as described below. We kept only cell suspensions corresponding to a percentage of degranulation higher than 60%. The percentage of degranulation was defined as wherein: – C is the number of stained basophils for the anti-IgE-free control – T is the number of stained basophils for a given anti-IgE concentration. Degranulated basophils are not stained under these conditions and cannot be differenciated from the other polymorphonuclears. At least three suspensions (in fact 3 to 6) were then mixed and adjusted to 1500 +200 basophils/mm3, by dilution in buffer A. II/ Testing of Histamine Dilutions A/ Pre-incubation of the Basophils with the Effector Dilutions Histamine (Sigma histamine hydrochloride) dilutions were prepared in distilled water by successive dilutions to the hundredth, then shaken vigorously (vortexed for at least 10 s). In order to make them isotonic, 9 volumes of histamine dilution were added to 1 volume of buffer C (200mM hepes, 1300mM NaCl, 50mM KCl, 50mM glucose, pH 7.4). Distilled water controls underwent exactly the same treatment as the histamine dilutions so as to prepare control dilutions. 20 µl of each dilution of histamine was mixed (after having been coded out at random) with 20 µl of the cell suspension, and incubated at 20+2°C for 30 min. B/ Basophil Degranulation The cell suspensions were distributed into wells of U-shaped polystyrene strips (Nunc Lab Removawells), in order to avoid possible zone phenomena in the microtitration plates. In order to focus on the dilutions corresponding to the maximum effect of the anti-IgE and to make it possible to apply the t-Student test to one particular IgE dilution, we set up two protocols: – Protocol A: Two anti-IgE free blanks containing the tested dilutions and 6 successives dilutions of the anti-IgE at 5 µg/ml with a dilution ratio of 5.

114 ULTRA LOW DOSES

– Protocol B: An anti-IgE-free blank and three dilutions of the anti-IgE (5, 1, 0.2 µg/ml) containing the tested histamine dilutions; each histamine dilution being tested 16 times. For both protocols (A and B), a distilled water control test was performed simultaneously on the same basophil suspension. For protocol B distilled water controls underwent the same number of dilutions as the related histamine dilutions tested. C/ Basophil Staining and Counting 120 µl of alcian blue solution prepared as described by Gilbert and Orstein (13) (8.5 µg NaCl, 444 µg EDTA-tri-Na, 380 µg cetylpyridinium chloride, 3.5 g lanthanum chloride, 715 µg alcian blue, 1 ml Tween-20, 56 ml HCl IN, H2O up to 1 litre) were added to each well. The strips were incubated for 30 minutes at 20 +2°C. The strips were covered with paper (dynatech plate sealer) and stored at 4°C. Under these conditions readings could be delayed for several days. After gentle homogenization, the basophil suspensions were distributed into the hemocytometers and after 2 to 3 minutes of sedimentation in a moist chamber the basophils appeared turquoise blue on a hyaline background. Basophils were then counted. Cell distribution in the hemocytometer had to be regular (absence of aggregates). For each well a minimum of 60 basophils was counted for the total surface of the hemocytometer. D/ Processing of Data All statistical analysis was carried out on the basophil counts. For each series of basophil counts we compared the means by the t-Student test, the variance of the counts being statistically equivalent. RESULTS Results obtained with protocol A:

8 sucessive experiments were performed with one distilled control and with the 6th, 7th, 17th, 18th dilutions of histamine to the hundredth. Out of 24 series of basophil counts (4 dilutions tested×8 experiments) we obtained only 6 significant results, but we systematically observed for all experiments, mean basophil counts for the histamine dilutions which were higher than those of the distilled water controls. The higher difference between the histamine dilutions and the distilled water controls was observed for the 4 dilutions of histamine tested with the 1 µg/ml

BIOLOGICAL ACTIVITY OF ULTRA LOW DOSES:II 115

concentration of the anti-IgE. Statistical analysis of the basophil counts for this IgE concentration and for the 8 experiments was also non-significant, mostly due to the dispersion of the basophil counts in the controls. So we set up the protocol B using fewer anti-IgE dilutions but increasing the number of histamine dilutions or repeating the experiments for the same histamine dilution. Results obtained with protocol B:

Three consecutive experiments were performed on three different basophil pools. Statistical significance (Table I) shows that for each of the three experiments, at least two different anti-IgE dilutions were statistically active. Moreover the different anti-IgE dilutions were not equally active. Table I Anti-IgE concentration

5 1 0.2

µg/ml µg/ml µg/ml

Significance of experiment: 1

2

p<10−2

p<10−2

p<10−3 p<10−3

p<10−3 p<10−3

3 NS p<10−3 p<10−2

We then repeated this protocol for two anti-IgE concentrations (1 and 0.2 µg/ ml) with four histamine dilutions (15C, 16C, 17C, 18C) and one distilled water control. This experiment was repeated 16 times. The difference was not significant for the mean basophil counts related to the highest anti-IgE concentration (1 µg/ml), but it was significant for the four histamine dilutions with the lowest anti-IgE concentration (0.2 µg/ml), respectively for: 15C 16C 17C 18C

p<10−3 p<10−3 p<10−3 p<10−2

DISCUSSION These experiments, carried out “blind”, are in favour of the biological activity of ultra low doses of histamine. The use of the described protocol makes it similar results. The magnitude of this effect which depends on the concentration of the stimulants must to be taken in account, as under similar conditions the use of a wrong concentration may completely cancel this effect. A simple “plate artefact” was excluded due to the use of disposable strips.

116 ULTRA LOW DOSES

The question of how the organism reacts to the effect of such dilutions has been confronted in our study since, throughout it, we were able to demonstrate the activity of high dilutions on the tinctorial affinities of intra-cytoplasmic granules of basophils. Nevertheless, we were not able to correlate this effect with histamine release by these same cells. To conclude, we believe that there is no acceptable hypothesis today to cover the mode of action of high dilutions. For this reason we are developing this easily-reproducible new protocol and we will continue our experiments. ACKNOWLEDGEMENTS We sincerely thank: Mr Jean Boiron for his advice and collaboration Mrs Sheila Adrian her help in rereading and correcting the English of this article. REFERENCES Gilbert, H.S. and Ornstein, L., 1975, Basophil counting with a new staining method using alcyan blue. Blood, 42, 279–286. Pruzansky, J.J. and Patterson, R., 1986, Binding constants of IgE receptors on human blood basophils. Immunology, 58, 257–262.

V CLINICAL PHARMACOLOGY

A STUDY OF THE EFFECTIVENESS OF ULTRA LOW DOSES OF COPPER IN THE TREATMENT OF HEMODIALYSISRELATED MUSCLE CRAMPS E.Hariveau*, P.Nolen** and A.Holtzscherer*** * Institut BOIRON, Sainte-Foy-lès-Lyon, France ** Hemodialysis Department C.H.R.U. de Poitiers, France *** Homeopath at Pont-du-Casse, France Twenty subjects under hemodialysis for an average of 6.7 years were recruited to compare the efficacy of ultra low doses of Copper (2×10−32M) to a placebo in the resolution of hemodialysis-related muscle cramps. Cramps were eliminated in 90% of the subjects with Copper, and in 35% of the subjects with the placebo. INTRODUCTION The frequency of hemodialysis-related muscle cramps is evaluated differently by authors. Rosa et al (1980) consider that the cramps stabilize after the first 13 dialysis sessions; Panadero-Sandoral et al (1980) consider the number of patients subject to cramps to be 26%, and the number of patients with a specific tendency to cramps to be 38%. Different treatments have been suggested. Panadero-Sandoral et al (1980) obtained good results with quinine sulfate. Neal et al (1981), then Sherman et al (1982) found effective action in an aqueous solution of 50% dextrose. Bellinghieri et al (1983) were able to reduce the frequency of cramps with LCarnitine. On the other hand, Marcen et al (1988), obtaining good results on all dialysis-induced symptoms by lowering the temperature of the dialysis liquid, were unable to obtain any results on muscle cramps Hemodialysis-related muscle cramps therefore remain a syndrome which does not react well to classical treatments (massaging, placing a warm towel on the cramped area, administering hypertonic salt solution). A physician working in the Poitiers hemodialysis department suggested testing very weak doses of Copper, a medicine recommended in homeopathy for treting cramps. The following experimentation was therefore understaken.

A STUDY OF THE EFFECTIVENESS OF ULTRA LOW DOSES 119

MATERIALS AND METHODS A crossover, double blind controlled trial, made it possible to recruit 20 patients: 9 women and 11 men, average age 54.2+18.3 years (35–77 years old). These patients, under hemodialysis for an average of 6.7+4.0 years (10 months to 13 years) underwent 3 sessions per week. The apparatus used was a Gambro AKIO dialyzator with ultrafiltering control. Blood flow during the sessions was about 250 ml/mn and the solution flow was set at 500 ml/mn. The length of a session varied from 4 to 5 hours. The cramps generally occur during the second half of dialysis, and more particularly during the last hour. Their frequency is such that the average interval between 2 dialysis sessions with cramps is 12.4+11.4 days (2 to 42 days). During 2 successive dialyses with cramps, the patients were given, in random sequence, either Cuprum metallism 15 C, or a placebo. Cuprum metallicum 15 C is a homeopathic medicine prepared from Copper metal, following a procedure described in the French Pharmacopoeia. After 3 successive triturations and dilutions to the 1/100th, in lactose, the preparation is successively diluted 12 times to the 1/100th in a mixture of water and alcohol (60% v/v). Each of the dilutions in the liquid medium is followed by a potentization (150 vigorous shakings in 7.5 seconds). According to theoretic calculation, this final dilution contains 2 ×10–32 moles. This dilution is next impregnated onto small pellets (saccharose-lactose) having a diameter of 1.2 mm. These medicated pellets are packaged in unit-doses of 200 pellets (1 g.). The placebo is composed for inert pellets. Cuprum metallicum is traditionally prescribed for violent cramps in the calves and feet, and for violent spasmodic pain which begins and ends suddenly. A first unit-dose of small pellets (to be left to slowly melt under the tongue) was administered at the onset of the cramp. If the cramp did not stop, a second unitdose was administered after 3 minutes. The effectiveness of the treatment, estimated when the cramp stopped, was evaluated 3 minutes after administration of the first dose, then 3 minutes after administration of the 2nd dose. Statistical analysis used the chi-2 test to assess the results expressed in number of patients. RESULTS Under active treatment (Cuprum 15 C), for 12 subjects out of 20 (60%), cramps stopped within 3 minutes. Among the 8 other subjects, a second dose of Cuprum 15 CH made it possible for cramps to be stopped in less than 3 minutes, for 6 subjects (75%). Under placebo treatment, for 3 subjects out of 20 (15%), cramps stopped within 3 minutes. Among the 17 other subjects, a second dose made it possible for cramps to be stopped in less than 3 minutes, for 4 patients (23%). The 13 remaining subjects were unrelieved by the treatment.

120 ULTRA LOW DOSES

Table 1: Effect of the treatment after the 1st dose (number of patients)

success failure

Cuprum 15 C

Placebo

Statistical Analysis

12 8 20

3 17 20

X2=8.6 p<0.01

Table 2: Effect of the treatment after the Ist dose (number of patients)

success failure

Cuprum 15 C

Placebo

Statistical Analysis

6 2 8

4 13 17

X2=4.7 p<0.05

Statistical analysis carried out using the Chi-2 test shows significantly superior effectiveness for Cuprum 15 C, compared to a placebo after the 1st dose (p<0. 01), as well as after the 2nd dose (p<0.05). To conclude, the overall effect of Cuprum 15 C, at the onset of a cramp, gave relief to 18/20 patients (90%), whereas the placebo only provided relief to 7/20 patients (35%). This difference is statistically significant (p<0.001). DISCUSSION This study made it possible to point out the therapeutic effectiveness of a homeopathic medicine, Cuprum 15 CH, which contains, according to the Avogadro formula, 10–8 atoms of Copper. The homeopathic approach usually consists of individualizing treatment depending on the specific rectional mode of each patient. In the case of cramps brought on by hemodialysis, we find ourselves facing a model of clinical pharmacology wherein common etiology made it possible to give patients the same medicine, which is characteristic to the symptomatology presented. The fact that most of the patients were under dialysis 2 to 3 times a week for several years made it possible to give credibility to their self-evaluation of pain. In addition, the fact that each patient proof on himself the two therapeutic sequences reinforced the dependability of the comparison between two treatments. The difference in therapeutic effectiveness between Cuprum 15 C and the placebo is great enough to be statistically significant although only 20 patients were concerned. We attempted to have this trial reproduced in another dialysis service. Certains teams did not want to take up the challenge, others told us that they had no cramping problems with their hemodialysis techniques. This study appeared to us to be interesting for several reasons: it made it possible to show the therapeutic effectiveness of a high-dilution homeopathic

A STUDY OF THE EFFECTIVENESS OF ULTRA LOW DOSES 121

medicine while both following scientific study design criteria and also the homeopathic prescribing approach. REFERENCES Bellinghieri, G., Savica, V., Mallamale, A., Di Stefano, C., Console, F., Spagnoli, L.G., Villaschi, S., Palmieri, G. , Corsi, M. and Maccari, F., 1983, Correlation between increased serum and tissue L-Carnitine levels and improved muscle symptoms in hemodialysed patients. Am. J. Clin. Nutr., 38(4), 523–531. Marcen, R., Quereda, C., Orofino, L., Lamas, S., Tervel, J.L., Matesanz, R. and Ortuno, J., 1988, Hemodialysis with low-temperature dialysate: a long-term experience. Neuphron., 49(1), 29–32. Neal, C.R., Resnikoff, E. and Unger, A.M., 1981, Treatment of dialyse-related muscle cramps with hypertonic dextrose. Arch. Intern. Med., 141(2), 171–173. Panadro-Sandoval, J., Perez-Garcia, A., Martin-Abad, L., Piqueras, A., Garces, L., Chacon, J.C., and Cruz, J.M., 1980, Accion del sulfato de quinina sobre la incidencia en la aparicion de calambres musculares en las hemodialisis. Med. Clin. (Barc), 75 (6), 247–249. Rosa, A.A., Fryd, D.S. and Kjellstrand, C.M., 1980, Dialysis symptoms and stabilization in long-term dialysis. Practical application of the Cusum plot. Arch. Intern. Med., 140 (6), 804–807. Sherman, R.A., Goodling, K.A. and Eisinger, R.P., 1982, Acut therapy of hemodialysisrelated muscle cramps. Am. J.Kidney. Dis., 2(2), 287–288.

ACTION OF ASPIRIN AFTER INGESTION AT ULTRA LOW DOSES IN HEALTHY VOLUNTEERS Lalanne M.C., De Seze O., Le Roy D., Doutremepuich C., Boiron J. Laboratoire d’Hématologie, 3 Place de la Victoire F33076 Bordeaux Cedex, France

INTRODUCTION Acetylsalicylic acid (ASA) was synthesized by Gehardt in 1853 and included into clinical medicine in 1899. Its fate in man was not elucidated until the last ten decade and some aspects of its metabolism still remain unresolved. Doses usually used to prevent arterial thrombosis are between 30 and 600 mg daily. At such doses, ASA prolonges the bleeding time by inhibition of the platelet prostaglandin derivative thromboxane A2, which is a potent vasoconstrictor and platelet aggregator (Moncada et al., 1979). This inhibition cannot be reversed since platelets lack nuclei and cannot regenerate new enzymes. No investigation has been performed at doses below 1 mg. Therefore, the aim of the present study was to determine the effect of ASA at ultra low doses (
ACTION OF ASPIRIN AFTER INGESTION 123

Healthy volunteers Young healthy male (n=20) were included in this double blind trial. Male were chosen to avoid sex variation (Bain et al., 1980; Buchanan et al., 1983). Mean age was 26.6±5.7 years. The subjects had no ASA allergy and no coagulation disorders. No medication interfering with platelet function or coagulation was taken for at least a week before the study. An informed consent was obtained for each patient included in the study. Experimental design The volunteers were randomly divided in 2 groups. Ten subjects received ASA dilution, the 10 other distilled water according to the randomization by method of random permuted blocks (Pocock et al., 1983) Ingestion way was sublingualy and corresponded to time TO. Tested solutions were stored in identical glass vials in order to avoid eventual idiosyncratic response. Measurement of bleeding time was effected every two hours (table 1) by the Ivy method (Ivy et al., 1940) standardized using a SimplateR device (General diagnostics) (Samama et al., 1986). A catheter inserted into an antecubital vein on the opposite arm gave an open line allowing the collection of blood without occlusion. Blood was sampled on natrium citrate 3.8% (1 volume for 9 volumes blood) every hour for 6 hours and centrifugated for further investigations. Table 1. Protocol of sampling and bleeding time TO TO+0.5H TO+1H TO+2H TO+3H TO+4H TO+5H TO+6H

Bleeding time

+

Bleeding time

+

Bleeding time

+

Bleeding time

+

Blood collection Blood collection Blood collection Blood collection Blood collection Blood collection Blood collection Blood collection

Platelet aggregation using the optical method of Born (Born et al., 1963) was performed on a Chrono LogR aggregometer 500 VS (Coultronics France). Platelet aggregometer 500 VS (Coultronics France). Platelet rich plasma (PRP) obtained by centrifugation of citrated blood (10 minutes at 1000 rpm) was maintained under stirring at 37°C. After 2 minutes incubation, 20 ul of aggregating reagent were added to PRP inducing platelets aggregation. Reagents used were collagen (25 ug/ml PRP final concentration, Stago France) and ADP (1 or 2 uM final concentration, Stago France).

124 ULTRA LOW DOSES

The changes in light transmission were recorded using the MerapR software (Coultronics France) and the aggregation parameters were automatically given by the computer. They were the platelet aggregability expressed in %, the velocity (or maximal slope) expressed in %/minutes and the response time (delay between addition of reagent and beginning of the aggregation) expressed in seconds. Coagulation tests performed on plasma. Plasma was immediatly frozen and stored at −20°C for coagulation tests: APTT (CK PrestR Stago France), Thrombin Clotting Time (ThrombaseR Houde France), Prothrombine Time (NeoplastineR Stago) and Fibrinogen level (FibriprestR Stago). Statistical Analyses The analysis were performed on PCSMR Software (Deltasoft France) on two steps: – Determination of mean+standard deviation – Multivariate test: analysis of variance with repeated measures. Assays were observed n fold (n corresponding to the various collecting times) and corresponded to the quantitative factor. The tested drugs were the qualitative factor. Variations beween treatments (interpersonal factor) or between times (interperiod variations) or interactions between time and treatment were determined. This type of analysis allows avoidance of observations not caused by the experiment. The step for significance was p<0.05.

RESULTS Bleeding time The results of bleeding time are given in figure 1. The variance analysis showed a statistically significant bleeding time reduction (p<0.05) in the aspirin group as compared to the control group. This analysis was precized using a non parametric test and the statistical difference was at the time TO+2H (p< 0.05). No blood dilution was observed owing to no variation appeared in the hematocrit whatever the time of collection in both aspirin and placebo groups. Coagulation tests Results of the analysis of variance of the coagulation tests are reported in table 2.

ACTION OF ASPIRIN AFTER INGESTION 125

Figure 1. Bleeding time of both groups (Placebo and ASA) according to time in sec. Table 2: Results of the analysis of variance with repeated measures for the coagulation tests (F: factor of the analysis of variance). Tests

APTT PT Fib. TCT

Intertreatment variation

Interperiod variation

Interaction time/treatment

F

P

F

P

F

P

0.0478 0.6250 0.1962 8.5514

NS NS NS <0.01

0.4919 1.3287 0.9851 4.6973

NS NS NS <0.01

3.0388 0.5061 0.5530 1.8839

<0.01 NS NS NS

PT: Prothrombin time level TCT: Thrombin clotting time signifiant

Fib.: Fibrinogen NS: not

Whatever the period, aspirin thrombin clotting time is statistically higher than in the control group. Platelet aggregation on PRP Mean platelet counts were at TO, 254 500±61 460/mm3 for the placebo group and 241 300±61 060/mm3 for the aspirin group. At time TO+6H, they were 245 500 ±55 880/mm3 for the placebo group and 248 200±68 810/mm3 for the aspirin group. When reagent was ADP (1 uM), the amplitude curve showed an increase at T1H in the placebo group 33.8± 15.0% as compared to the aspirin group 21.7±8. 5%

126 ULTRA LOW DOSES

For the other times, the values varied from 22.4± 8.6% to 25.5±11.1% in the placebo, and from 19.3 ±4.8% to 29.6±11.8% in the aspirin group. Velocity curve gave at T1H: placebo 30.6±15.0%/min and aspirin 17.5±7.0%/min (p<0. 05). For the other times, velocity was between 18.9±6.8%/min and 22.1±10.9%/ min in the placebo group, and between 17.5±3.8%/min and 23.5±8.2%/min in the aspirin group. The variance analysis with repeated measures showed difference of response between time in amplitude curve (p<0.05). The same time interaction and timedrug effect interaction in velocity curve was noted (p<0.05). The response time was included between 9.9±6.5 sec and 12.5±5.8 sec in the placebo group and between The response time was included between 9.9±6.5 sec and 12.5±5.8 sec in the placebo group and between 9.2±2.3 sec and 13.8±5.2 sec in the aspirin group. The differences between the two groups were not statistically significant (p=NS). When reagent was ADP (2 uM), the amplitude curve was different according to time: at T1H, placebo (54.8± 24.6%) was higher than aspirin (36.5±13.9%); however; at T4H, the aspirin group (52.3±18.9%) was higher than placebo (35.2 ±12.9%). Velocity was different between the two groups at T1H: placebo 39.7±16.5%/ min and aspirin 25.8± 10.5%/min. For the other times, velocity varied from 27.2 ±12.3 to 31.8±12.5%/min for placebo and from 25.±11.0 to 34.0±10.4%/min for aspirin. The amplitude or velocity curves was different between treatments, but a timetreatment interaction was observed. The response time was included between 8.8±2.9 and 11.0±3.5 sec in the placebo group and between 9.4±4.2 and 11.1±3.9 sec in the aspirin group. The differences between the two groups were not statistically signifiant (p=NS). When the reagent was collagen (25 ug/ml), the amplitude curve showed certain differences according to drugs and to test times. The aspirin group was higher than the placebo group at T2H and T4H. Velocity was higher at T1H in the placebo group and, at T2H and T4H in the aspirin group. The variance analysis with repeated measures of amplitude or velocity curves showed no difference between treatments or between periods, but an important interaction time-treatment (respectively p <0.01 and p<0.05) existed. The response time did not statistically differ between the two groups: in the placebo group, it varied from 104.0±29.3 to 120.0±23.5 sec and in the aspirin group from 90.7±22.4 to 104.0±23.0 sec. DISCUSSION The results obtained in this study are in agreement with those previously described with aspirin used at ultra low dosage (Doutremepuich et al., 1987) demonstrating that aspirin statistically reduced bleeding time. This phenomenon is quite new and for this reason the present study was performed in a controlled

ACTION OF ASPIRIN AFTER INGESTION 127

trial with very rigourous conditions. Our work was performed only on men in order to standardize the experimental model. Only one test was modified among the coagulation tests: Thrombin clotting time (TCT) was always higher in the placebo group. This may be due to the release of a glycosaminoglycan from the endothelial cells in presence of aspirin at ultra low dose. On the other hand, platelet aggregation was modified according to time. An important interaction (p<0.01 and p<0.05) appeared between time and product action for both amplitude and velocity. However, the time-treatment interaction could be sufficient to mask the drug effects. So, further studies have to be performed in order to avoid these interactions and to precise these results. In the main, such observations are opposite to those made with normal dosage (Buchanan et al., 1987; O’Brien et al., 1968). When aspirin is ingested at high or low doses (50–500 mg) it induced a bleeding time increase (Waldemar et al., 1988) as well as an inhibition of platelet aggregation (Buchana et al., 1983). In the contrary in this study, bleeding time was reduced and no statistical change could be observed in platelet aggregation in vitro. The reduction of bleeding time observed was limited in time, but such transient effect of aspirin has previously been described in the literature (De Gaetano, 1988). However, further studies have to be performed to determine by which way aspirin used per os at ultra low dose acts on hemostasis. CONCLUSION Aspirin used at ultra low dose has no effect on platelet aggregation. It induces an effective, but transient reduction of bleeding time, which could be interesting for reduce or palliate post-operative bleeding. REFERENCE Bain B., Forster T. 1980, A sex difference in the bleeding time. Thromb. Haemostas., 43, 131–132. Born G.V.R., CROSS M.J. 1963, The aggregation of blood platelets. J. Physiol. London, 168, 178–195. Buchanan M.R., Rischke J.A., Butt R., Turpie A.G.G., Hirsh J., Rosenfeld J. 1983, The sex-related differences in aspirin pharmacokinetics in rabbits and man and its relationship to antiplatelet effects. Thromb. Res., 29, 125–139. Buchanan M.R., Hirsh J. 1986, Effect of Aspirin on hemostasis and thrombosis. New Engl. Reg. Allergy Proc., 7, 26–32. Cerletti C., Latini R., Del Maschio A., Galetti F., Dejana E., De Gaetano G. 1985, Aspirin Kinetics and inhibition of platelet thromboxane generation. Atherosclerosis, 13, 67–77. De Gaetano G., Cerletti C., 1988, Prolongation of bleeding time by aspirin: A dual mechanism? Thromb. Res., 50, 907–912.

128 ULTRA LOW DOSES

Doutremepuich C., Pailley D., Anne M.C., De Sèze O., Paccalin J., Quilichini R., 1987, Template bleeding time after ingestion of ultra low dosage of acetylsalicylic acid in healthy subjects. Preliminary study. Thromb. Res., 48, 501–504. Doutremepuich C., De Sèze O., Anne M.C., Hariveau E., Quilichini R., 1987, Platelet aggregation in whole blood after administration of ultra low dosage acetylsalicylic acid in healthy volunteers. Thromb. Res., 47, 373–377. Ivy A.C., Nelson D;, Bulcher G., 1940, The standardization of certain factors in the cutaneous “venostasis” bleeding time technique. J. Lab. Clin. Med., 26, 1812–1822. O’Brien J.R., 1968, Effect of anti-inflammatory agents on platelets. Lancet, 1, 894–895. Pocock S.J., 1983, Clinicals trials. A practical approach . edited by J.Wiley and Sons Ed. (Chichester). Samama M., Charpentier M.C., 1986, Le temps de saignement par la méthode d’Ivyincision à I’aide du dispositif SimplateR. Feuillet de Biologie Clinique, 150, 27–36. Waldemar G., Petersen P., Boysen G., Knudsen J.B., 1988, Inhibition of platelet function by low dose plain and micro-encapsulated acetylsalicylic acid. Thromb. Res., 50, 265–272.

Index

acetylaminofluorene 20 acetylsalicylic acid 46, 48, 49, 50, 51, 146, 147, 149 acidophilic cells 26 acid phosphatase 91, 97 acid phosphatase reaction 97 actynomycin D 43 adenine nucleotides 92, 110 adenosine 93, 107, 108 adenosine biphosphate 150, 151 adriamycin 31, 35, 36, 37, 38, 39, 42, 43 allergen 126, 128, 131 anisocytosis 80 anti-cancer drugs 42, 43 anti-IgE 125, 133, 136, 137, 138, 139 anti-inflammatory agents 55 anti-inflammatory compounds 55 anti-TNF antibody 31, 35, 36, 37, 38, 39, 42, 43 aorto-coronary grafts 48 apis mellifica 125 arg-vasopressin 16 arsenate salts 68 arsenic 68, 72, 74, 75, 77, 125 arsenious acid 69, 71, 72, 76, 77 arsenious anhydride 68, 69, 70, 71, 72, 73, 74, 75, 77 arterial pathology 46 arterial thrombosis 146 aspirin 146, 150, 151, 152

B 12 vit. deficiency 85 bacterial infections 91 barbiturate dipyridamole 84 basophil 124, 125, 126, 127, 128, 129, 130, 131, 133, 136, 137, 138, 139, 140 basophil leukocytes 104 benactyzine 55 bile acids 84 bioamins 13 bleeding time 147, 149, 152 blood urea nitrogen 82, 84 blood vessels 87 calcium 82, 84 cancer 19, 32, 39, 42, 55 cancer cells 30 cancer therapy 39 carcinogen 19 carcinogenesis 19, 20, 22 carcinogenic agent 21, 26 carcinoma 22, 40, 42 chemoprevention 19 chemotherapeutic drugs 29, 34 chemotherapy 43, 55 cholinesterase 12 cimetidine 125, 128, 130, 133 cis-diamine-dichloroplatium 31 cis-platium 42, 43 coagulation 147, 148, 149, 150, 152 colchicine 43 collagen 47, 148, 151 colon 19

B 12 vit. 80, 85 129

130 INDEX

complement fragments 92 concanavalin A 92, 100, 102, 104, 105, 111, 112, 116, 117, 118, 119, 120, 121 connective tissues 87, 110 copper 141, 142, 144 corticosteroids 93, 108 creatinine 82, 84 cuprum 6 cuprum 4CH 5, 7, 8, 9, 11 cuprum 15CH 143, 144 cuprum metallicum 15Ch 142 cyclo-oxygenase 52 cytochrome C 96, 97 cytokine 43, 92 cytotoxic effector cells 29 cytotoxic effector molecules 29 daunorubicin 41 D-D vasopressin 16 defensin 13, 14, 16 degranulation 87, 89 dermatitis 126 dexametasone 108 dextrose 141 D-G vasopressin 16 dialysis 144 dihedral 56 DNA topoisomerase II 43 dopamin 16 drug resistance 29 duodenal motricity 11 echinocytes 80, 83, 84, 85 EDRF 50 eicosanoids 92 electrostatic potentials 60, 64, 65 endorphins 123 endothelial cells 50, 91, 152 endothelin 50 endotoxin 104, 105, 106, 107, 108 enkephalins 16 erythrocyte 80, 83, 84, 85, 94 fatty acids 84 fibroblasts 43 formyl-methionyl-leucyl-phenylalanine 92, 98, 100, 103, 104, 105, 106, 107, 108

glycoprotein 170, 41 glycosaminoglycan 152 gram negative bacteria 93 granulocytes 91 growth factors 92 guanylate cyclase 50 H2 antagonist 133 H2 receptors 133 hemodialysis 141, 142, 143, 144 hemoglobin macrocytic hypochromic anemia 81 hemolytic anemia 80, 84 hemolytic syndrome 81 hemostasis 146 heparin 87, 136 hepatocarcino-genesis 26 histaminase 125, 126, 129. 130, 133 histamine 124, 125, 126, 127, 128, 129, 130, 131, 132, 133, 136, 137, 138, 139, 140 histidine 125, 126, 127, 129 hydrocortisone 108 hydrogen peroxide 92 IgE 126, 128 immune cytotoxic factors 29 immune system 87 inflammatory reactions 91 interleukin 1, 16 intestinal transit 5, 6, 7, 11 kinins 92 L-Carnitine 141 lectins 92, 112 leukocytes 91, 92, 93, 107, 110, 112, 125, 127, 136 lipopolysaccharides 92, 93 liver 19, 20 lymphatic flow 17 lymphatic vessel 14, 16 lymphocytes 117 lymphokines 116, 123 lysolecithin 84 lys-vasopressin 16

INDEX 131

macrocytic anemia 80 macrocytic hypochromic anemia 85, macrophage 30, 34, 104 mammary glands 19 mast cells 87, 88 melanoma 40, 42 mesentery 13 metastases 39, 42 microcarcinomas 26 mitogen 116 molecular electrostatic potentials 55, 56 molecular orbital 55 mononuclear cells 116, 117, 121 multiple drug resistance 41, 43 muscle 141 NADPH oxidase 106 natural killer cells 30, neostigmine 5, 6, 7, 8, 9 nerves 85 N-ethyl-malcimide 93, 107, 109 neuropeptides 92 neurophisins 17 neutrophil 91, 92, 93, 94, 96, 97, 98, 99, 101, 102, 104, 106, 107, 108, 110, 112 noradrenaline 16 oxydative metabolism 91, 92, 100, 103, 111 oxytocin 16 phagocytosis 112 phenobarbital 20 phenobarbital 9CH 21, 23, 24 phenol-sulfonephtalein 5 phenylbutazone 84 phenyl acids 56 phenyl glycolic acids 55, 57, 65 phenyl propionic acids 55 pheromones 123 phorbol esters 92 phorbol-myristate-acetate 92, 100 platelet 46, 47, 52, 146, 147, 148, 150 platelet activating factor 92, 125 platelet aggregation 47, 49, 52, 147, 150, 152 p-nitrophenyl-phosphate 97

poikilocytosis 80 polymorphonuclears 137 prostacyclin 50 prostaglandin 146 quinine sulfate 141 recombinant tumor necrosis factor 34 regulatory peptides 14 renal dysfunctioning 80 respiratory burst 112 reticulocytes 80, 81, 83, 84 reticulocytosis 84 salicylates 84 salsolinol 16 serotonin 87 substance P 16 sulfur atom 56 superoxide 96, 104 superoxide anion 91, 92, 98, 100, 107 superoxide dismutase 96 superoxide generation 108 superoxide production 95, 97, 98, 99, 100, 103, 110, 111, 113 thienyl acids 55, 56 thienyl glycolic acids 55, 56, 58, 59, 65 thromboxane A2 46, 52, 146 thymidine 116, 119 thymoptin 16 thyroliberin 13, 14 thyrotrophin-releasing hormone 14 toluidine blue solution 126 toxins 29 tufts 13 tuftsin 14, 15, 16 tumor 19, 26, 40, 42, 55 tumor cells 29, 30, 32, 33, 34, 35, 36, 38, 42 tumor necrosis factor 29, 30, 31, 35, 36, 37, 38, 39, 42 uranyl acetate 85 uranyl nitrate 80, 81, 82, 83, 85 vascular fragments 48, 49

132 INDEX

vessel wall 46, 52, 92 zymball gland 21, 22, 26 zymposan 98