THE EFFECTS OF NEOMYCIN UPON TRANSMITTER RELEASE AND ACTION 2

THE JOURNAL OF PHARMAOLOGY AND EXPERIMENTAL THERAPEUTIS opyright 1977 by The Willims & Wilkins o. Vol. 2, No. 3 Printed in U.S.A. THE EFFETS OF NEOMYIN UPON TRANSMITTER RELEASE AND ATION 2 J. M. WRIGHT3 AND B. OLLIER Deprtment of Phrmcology nd Therpeutics, McGill University, Montrel, Quebec, nd Accepted for publiction September 23, 1976 ABSTRAT WRIGHT, J. M. AND B. OLLIER: The effects of neomycin upon trnsmitter relese nd ction. J. Phrmcol. Exp. Ther. 2: 576-587, 1977. These experiments were designed to determine the site nd mechnism of ction of neomycin nd streptomycin on cholinergic trnsmission. These gents depressed the response of rt diphrgm preprtions to phrenic nerve stimultion nd to injected cetylcholine (Ah); however, equi-effective neuromusculr blocking concentrtions ofneomycin (6 x 1 M), streptomycin (1.2 x 1 M) or d-tubocurrine (6.5 x 1 M) reduced the muscle response to injected Ah to 54, 27 nd 15% ofcontrol, respectively, suggesting tht neomycin nd streptomycin hve presynptic effect. This finding ws confirmed by mesuring Ah relese from the diphrgm during phrenic nerve stimultion; neomycin (6 x 1 M) nd streptomycin (1.2 x 1 M) depressed Ah relese to 29 nd 41% of control, respectively. In the ct superior cervicl gnglion, neomycin (2 x 1 M) blocked gnglionic trnsmission, did not reduce the response of gnglion cells to injected nicotine nd depressed Ah relese during pregnglionic nerve stimultion to 61% of control in norml (2.5 mm) medium nd to <1% of control in low (.5 mm) medium. The incresed ccumultion of 45 induced in rt isolted gngli by pregnglionic nerve stimultion ws not chnged by d-tubocurrine (2 x 1 M), but ws bolished by neomycin (2 x 1 M). It is concluded tht neomycin blocks Ah relese by blocking the influx of necessry for trnsmitter relese. This conclusion suggested tht neomycin should block nordrenline relese, nd this ws shown using the nococcygeus preprtion from the rt. Respirtory rrest ssocited with ntibiotic therpy ws first reported by Pridgen (1956), nd since tht time mny other reports hve ppered (reviewed by Pittinger et t., 197). Most cses hve involved the use of one of four minoglycoside ntibiotics (neomycin, streptomycin, dihydrostreptomycin or knmycin), Received for publiction April 9, 1976. This work ws supported by grnt from the Medicl Reserch ouncil of nd. 2 An bstrct of portion of this work ppered in Phrmcologist (16: 285, 1974). Supported by Fellowship from the Medicl Reserch ouncil of nd. Send reprint requests to: B. ollier, McIntyre Building, 3655 Drummond, Montrel H3G 1Y6. nd prlysis ws lmost lwys ssocited with the use of the ntibiotics in combintion with other neuromusculr depressnts or in ptients with pre-existing neuromusculr disese. In nimls, minoglycosides cn block neuromusculr trnsmission, nd this effect is potentited by other neuromusculr blocking gents (reviewed by Pittinger nd Admson, 1972). It is not yet cler whether these ntibiotics block neuromusculr trnsmission by presynptic ction to reduce trnsmitter relese or by postsynptic ction to reduce trnsmitter ction. In erly studies, the neuromusculr blocking properties of streptomycin, neomycin, knmycin nd dihydrostreptomycin ppered to 576

1977 NEOMYIN AND TRANSMITTER RELEASE 577 resemble those of mgnesium (Vitl Brzil nd orrdo, 1957; Pittinger et t., 1958; Timmermn et t., 1959), s if the gents ffected cetylcholine (Ah) relese. Elmqvist nd Josefsson (1962) demonstrted tht neomycin hd postsynptic blocking ction, s indicted by decresed sensitivity of denervted diphrgm to Ah, nd by decrese in size of miniture end-plte potentils (m.e.p.p.) recorded from the end-plte region of innervted muscle; they lso demonstrted presynptic blocking ction, s indicted by blockge of the incresed m.e.p.p. frequency induced by rised potssium. The presynptic block by neomycin ws ntgonized by clcium, but the postsynptic block ws not. Vitl Brzil nd Prdo-Frnceschi (1969) lso demonstrted pre- nd postsynptic blocking effects of neomycin nd gentmicm, nd they suggested predominnt presynptic effect becuse the drugs inhibited the mount of Ah collected from the stimulted phrenic nerve-diphrgm preprtion of the rt. This ction ws ntgonized by clcium, nd Vitl Brzil nd Prdo-Frnceschi suggested tht neomycin competes with the clcium tht is necessry for Ah relese. More recently, however, Dretchen et t. (1972) hve reported tht neuromusculr blocking concentrtions of neomycin, streptomycin nd knmycin did not ffect Ah relese from the frog neuromusculr junction, suggesting postsynptic blocking ction of the drugs. In nother study of the effects of streptomycin on the frog srtorius muscle, Dretchen et t. (1973) demonstrted tht this drug blocked neuromusculr trnsmission by postsynptic ction becuse m.e.p.p. mplitude nd end-plte sensitivity to iontophoreticlly pplied Ah ws decresed in the presence of streptomycin; in contrst, similr concentrtions of the drug cused incresed nerve terminl ctivity s judged by n increse in frequency of m.e.p.p.s. The present experiments were done to clrify the site of ction of neomycin, the most potent of the minoglycosides (Vitl Brzil nd orrdo, 1957), nd to determine, if possible, its mechnism of ction. Methods Medi. Krebs solution (118 mm N1, 4.7 mm K1, 2.5 mm l2, 1.2 mm KH2PO4, 1.2 mm MgSO4. 7H2O, 9.9 mm glucose nd 25 mm NHO:,) ws used; in some experiments, which re indicted in the text, the clcium concentrtion ws reduced to.5 mm. All Krebs medi were equilibrted with 95% 3 nd 5% O2 in order to mintin their ph between 7.3 nd 7.4 t 37#{176}. Rt phrenic nerve-diphrgm preprtions. Sprgue-Dwley rts (15-25 g) were stunned nd decpitted, nd the diphrgm together with the thorcic contents, djoining ribs nd liver were removed nd plced in Krebs solution (21#{176}). In order to study the response of the diphrgm muscle to Ah injection nd to nerve stimultion, the right phrenic nerve nd thorcic inferior ven cv were dissected free nd the bdominl inferior ven cv nd veins drining the left hemidiphrgm were tied nd cut s described by others (Burgen et l., 1949; Pterson, 1965). A centrl wedge of the right diphrgm round the right phrenic vein ws cut out, nd the thorcic inferior ven cv ws cnnulted with polythene cnnul. The muscle ws suspended (1 g oftension) in 15-ml bth t 37#{176} nd could be stimulted to contrct by stimulting the phrenic nerve or by injecting Ah into the phrenic vein nd muscle through the cnnul in the thorcic inferior ven cv. The nerve ws stimulted with rectngulr pulses (suprmximl voltge,.1 Hz,.3 msec) through n electrode similr to tht described by Mogey et l. (1949). The mount of Ah pplied to the muscle ws vried by chnging the volume injected (.6-. 15 ml) or by chnging the concentrtion of Ah injected (5-2 x 1 M). The rte ofinjection of Ah ws kept constnt by using Hmilton (R 7-2) syringe operted by push button. Injections of Ah were mde very 1 minutes nd the bth solution ws chnged fter three or four injections to prevent chnges in end-plte sensitivity produced by Ah diffusing into the bth fluid (Pterson, 1965). Muscle contrctions were mesured by force displcement trnsducer nd recorded on polygrph recorder. When Ah relese from the phrenic nerve diphrgm ws to be mesured, the right or the left hemidiphrgm together with its phrenic nerve ws suspended in smll rectngulr bth (37#{176}) tht contined 1 ml of Krebs medium. In these expenments the Krebs medium lwys contined physostigmine sulfte (3 x 1 M) nd the muscle ws bthed in this medium for 6 minutes before n experiment ws strted. In ech experiment, medium ws collected for mesurement of Ah relese during consecutive 15-minute periods of rest or stimultion; nerve stimultion ws t 2 Hz for the 1st 13 minutes of the 15-minute collection period. The Ah in liquots (.4 ml) of the fluid collected ws extrcted into tetrphenylboron-heptnone solution within 3 minutes of collection. Perfused superior cervicl gngli of the ct. ts were nesthetized with ether nd then with chlorlose (8 mg/kg iv.). The right superior cervicl gnglion ws isolted nd perfused by the

578 WRIGHT AND OLLIER Vol. 2 method of Kibjkow (1933) s described by ollier nd Lng (1969); perfusion ws with Krebs solution contining choline (5 x 1 M). In some experiments the effect of neomycin B on gnghionic trnsmission nd on the response to gnglionic stimultion by nicotine ws ssessed by mesuring the contrction ofthe ipsilterl nictitting membrne with trnsducer nd polygrph recorder. The cut pregnghionic nerve ws stimulted with rectngulr pulses (suprmximl voltge,.3 msec; 5 Hz for 3 seconds); nd, using Hmilton syringe, nicotine trtrte (4-12 nmol in. 1 ml) ws injected t constnt rte into the rteril cnnuh close to the gnghion. Other experiments tested the effect ofneomycin B on Ah relese from gngli. Perfusion ws done with Krebs solution contining choline nd physostigmine (1.5 x 1 M), nd Ah relese ws incresed by pregnglionic nerve stimultion (2 Hz) for 2 minutes; ech period of stimultion ws followed by 13 minutes of rest. Isolted superior cervicl gngli of the rt. Rts (25-3 g) were nesthetized with urethne (1.- 1.2 g/kg i.p.). The right nd left superior cervicl gngli were removed together with pproximtely 2 cm of pregnglionic trunk nd 2 to 3 mm of the internl nd externl crotid nerves. The gngli were deshethed with the id of dissecting microscope nd plced in 5-ml bth t 37#{176} contining low Krebs solution (,.5 mm) supplemented with choline (1 x 1- M). The pregnghionic trunk ws drwn into smll stimulting electrode; the postgnglionic internl or externl crotid nerve ws drwn into miniture suction recording electrode (Lrrbee et l., 1963) from which postgnglionic ction potentils could be mesured. Both gnghi were tested for excitbility. To mesure ccumultion ofclcium, 1 i of4 (New Englnd Nucler nd, Dorvl, Quebec, specific ctivity.62 i/mmole) ws dded to the bth contining the gngli nd 5 minutes lter the test gnglion ws stimulted (1 Hz for 2 minutes); the control gnghion ws not stimulted. At the end ofthis period of stimultion, both gngli were wshed with 2 ml of Krebs solution (4#{176}) six times t 2-minute intervls. Pre- nd postgnghionic nerves were removed nd the gnghi were blotted, cut up nd plced in vil contining 1 pi of distilled wter nd 1 ml of tissue solubilizer (NS; Amershm/Serle, Des Plines, Ill.). The contents of the vils were incubted t 5#{176} for 2 to 3 hours until the tissue ws completely dissolved, 3.il of glcil cetic ws dded to neutrlize the bse nd thereby reduce quenching nd 15 ml of toluene (contining 2,5- diphenyloxzole, 4 g/l nd p-bisl2-(5-phenyloxzolyl)jbenzene, 5 mg/l) ws dded so tht rdioctivity could be determined by liquid scintilltion spectrometer. Ten microliters of the incubting bth solution ws prepred for determintion of rdioctivity in n identicl mnner to the gnghi. Rt nococcygeus preprtion. Sprgue-Dwley rts (2-3 g) were stunned nd decpitted nd the pired nococcygeus muscles were removed s described by Gillespie (1972). Individul muscles were suspended (1 g of resting tension) in 5 ml of Krebs solution (contining EDTA, 4 x 1 M, nd scorbic cid, 1 x 1 M) t 37#{176}; the tissue ws rrnged in n inverted V s described by Lnger (197) so tht it could be stimulted through two pltinum wires. Muscle contrctions to field stimultion (3-4 V,.4-1 msec, 5-2 Hz) or to drugs dded to the bth were recorded on polygrph recorder. In order to mesure nordrenhine relese, the tissue ws exposed to 3 j.i of DL-nordrenhne bitrtrte (3H7) (New Englnd Nucler nd, specific ctivity 1. i/mmoh) for 3 minutes during which the preprtion ws stimulted (5 Hz for 1- minute trins every 3 minutes) for the 1st 15 mmutes. At the end of this incubtion period, preprtions were wshed with Krebs solution every mmute for the 1st 5 minutes, nd then every 5 minutes for the durtion of the experiment. The 5-minute wshes were collected nd n hiquot (.5 ml) ws plced in 15 ml of solvent system described by ollier nd Lng (1969) for determintion of rdioctivity. Ionized clcium. Ionized clcium in Krebs solution nd Krebs solution contining neomycin B ws mesured using the Orion clcium ion electrode (model 92-2) in conjunction with n Orion digitl ph/mv meter (model 71). lcium stndrds (.5, 1 nd 2 mm) were prepred with l2 stndrd solution (.1 M) in Nl solution (154 mm). A 2- fold increse of 12 in stndrd solutions gve millivolt increse of 5.5 to 7.5; when the millivolt increse for 2-fold increse of l2 fell below 5.5, the electrode membrne ws chnged. The procedure for mesurement of ionized clcium in smples ws tht of Httner et t. (197) in which clcium stndrds nd unknown smples were lternted until stble millivolt differences were obtined. It ws necessry to equilibrte the Krebs solutions with 95% 2 nd 5% O2 throughout the mesurement procedure s loss of the O2 ffected the ionized clcium. When Krebs solution ws prepred in the usul mnner with nhydrous 12 powder to give concentrtion of 2.5 mm, the mesured ionized clcium ws 1.61 ±.26 mm. In experiments mesuring ionized clcium, 12 stndrd solution ws used to prepre the Krebs solution (l2 2.5 mm) nd the mesured ionized clcium ws 1.87 ±.22 mm, suggesting tht the powdered l2 ws not completely nhydrous. Ah ssy. The Ah content of smples collected from the rt diphrgm or the ct gnghion prepr-

1977 NEOMYIN AND TRANSMITTER RELEASE 579 tion ws mesured by rdioenzymic ssy described by Goldberg nd Mcmn (1973). The procedure (see Kto et t., 1975 for detils) involved extrcting Ah from.4 ml of Krebs solution into.4 ml of heptnone contining tetrphenylboron (1 mg/ml), then extrcting the Ah from.3 ml of the orgnic phse into. 15 ml of 1 N H1;.1 ml of this cid ws then trnsferred to microtubes, nd the H1 ws removed by lyophihiztion in vcuum centrifuge. The residue from lyophiliztion ws dissolved in 3 il of rection medium contining denosine 5 -triphosphte (ATP) (.8 mm); Mgl2, (12.5 mm), glycylglycine (83 mm), nd choline kinse (3 g of protein, specific ctivity 33 mol/g of protein per mm, prepred from yest); this rection mixture ws incubted t 3#{176} for 15 minutes. During this stge the choline in the smple, but not the Ah, ws converted to phosphorylcholine by choline kinse. The second stge of the ssy involved dding 5 pi of solution contining cetylcholinesterse (1 zg, specific ctivity 1.4 unith/tg) nd 1y 2PIATP (New Englnd Nucler nd; bout.4 j.ti, specific ctivity 2-4 i! mmoh) to ech smple nd incubting t 3#{176} for 1 minutes. During this stge the choline formed by the ction of cetylcholinesterse on the Ah in the smple ws converted to t 2Plphosphorylcholine. This [32Plphosphorylchohine ws seprted from unchnged [32P}ATP by pplying the smple to n ion exchnge column (5 x 5 mm, Dowex AG 1X8, formte form) nd eluting with 1.8 ml ofnh4ooh (75 mm); the elute ws collected into Aqusol (New Englnd Nucler nd) for determintion of rdioctivity. In ech ssy pproprite blnks were processed in the sme wy s were the smples to be ssyed; the mount of rdioctivity generted from blnks depends to some extent upon their choline content (Kto et l., 1975). When Ah relese ws mesured from the ct gnglion, the perfusion medium of Krebs solution contining choline nd physostigmine ws used s the blnk becuse perfusion of this medium through the gnglion did not lter its choline content. When Ah relese ws mesured from the rt diphrgm preprtion, choline ws not dded to the Krebs medium nd incubtion with tht tissue incresed the choline content of the medium nd incresed the blnk determintion; thus, in these experiments the blnk ws determined by hydrolyzing the Ah in representtive.4-ml liquots before extrcting nd ssying. This procedure ws done by dding 2.il of 1 N NOH to.4 ml of Krebs solution (ph> 1), incubting t 8#{176} for 15 minutes, nd then dding 2.il of 1 N H1 to neutrlize the bse (ph 6-8). Neomycin B nd streptomycin in concentrtions greter thn tht chieved in the experiments hd no effect on the enzyme ssy. Duplicte stndrds (1 nd 2 pmol ofah in.4 ml ofkrebs solution with physostigmine) were prepred t the time of collection of the test smples nd were extrcted nd ssyed in mnner identicl with the test smples. The rdioctivity generted from these Ah stndrds ws used to clculte the mount of Ah in the test smples. Drugs used. The following drugs were used in the study; d-tubocurrine chloride, nordrenhine bitrtrte (lbiochem, Los Angeles, lif.), cetylcholine chloride (Lemtte-Boinot, Pris, Frnce), choline chloride (BDH hemicls Ltd., Poole, Englnd), physostigmine sulfte (Nutritionl Biochemicls orportion, levelnd, Ohio), nicotine trtrte (K nd K lbortories, Plinview, N.Y.), phentolmine mesylte (ib ompny Ltd., Dorvl, Quebec) nd tetrodotoxin (Sigm hemicl ompny, St. Louis, Mo.). Before ech experiment neomycin B sulfte nd streptomycin sulfte were dissolved in distilled wter to produce. 1 M solutions, which were djusted to ph 7.4 by dding 1 N NOH solution. Sttistics. Sttisticl differences were nlyzed by the unpired nd pired Student s t tests (twotiled). Results re expressed s men ± SE. Results Site of neuromusculr block produced by neomycin or streptomycin. In these experiments the rt phrenic nerve-diphrgm preprtion ws set up such tht muscle contrction could be evoked by nerve stimultion or by injecting Ah. The result of prt of typicl experiment is shown in figure 1. A smll rtifct ws produced when norml sline ws injected into the phrenic vein, but injection of Ah redily produced much greter response. Ah injection produced twitch with mximl height of pproximtely twice the height of the twitch produced by single suprmximl stimulus pplied to the phrenic nerve. In ll experiments, n mount of Ah ws used which produced twitch of similr height to tht induced by single nerve impulse; the mount of Ah required to produce this response rnged from 5 to 4 nmol (mode 15 nmol). When control responses to Ah nd to nerve stimultion were constnt, the neuromusculr blocking drugs were dded to the bth nd ech concentrtion of drug ws llowed to ct for 1 minutes, t which time stedy-stte blockde ws pproched. oncentrtions were incresed every 1 minutes until the response to nerve stimultion ws inhibited

58 WRIGHT AND OLLIER Vol. 2 2g 2mm i/ I / 4N 6N8N I I!7ItL FIG. 1. A representtive trcing to demonstrte the contrctile response of rt diphrgm to phrenic nerve stimultion nd to injected Ah: s, injection of.1 ml of norml sline;, injection of 1 nmol of Ah in.1 ml of norml sline; 4N, 6N nd 8N represent the responses fter 1 minutes in the presence of4, 6 nd 8 x 1 M neomycin B, respectively; the lst segment is the response 2 minutes fter wshing out neomycin. Frequency ofnerve stimultion ws.1 Hz; Ah ws injected every 1 minutes. U ft N.omycinl i,io (N) by 9 to 1%. The preprtion ws then wshed until the response to Ah injection nd to nerve stimultion hd returned to norml (see fig. 1 for n exmple). In ech preprtion, it ws possible to test the effects of d-tubocurrme, neomycin B nd streptomycin on the response to Ah injection nd the response to nerve stimultion. By plotting the percentge of the initil response remining in the presence of drug ginst hog concentrtion of drug, log dose-response plots were obtined, which could be fitted to stright lines by regression nlysis (fig. 2, A nd B). All three drugs, tubocurrine, neomycin nd streptomycin, blocked muscle responses to injected Ah nd to nerve stimultion. The reltive pre- nd postsynptic ctions of the two ntibiotics were ssessed by clculting the inhibition of the response to injected Ah induced by concentrtion of drug tht blocked the response to nerve stimultion by 5% (tble 1). At equieffective neuromusculr blocking concentrtions, d-tubocurrine, which is presumed to ct entirely postsynpticlly (Koelle, 1975), reduced the response to injected Ah significntly (P <.5) more thn did streptomycin or neomycin; streptomycin ws lso significntly (P <.1) more effective s n nti-ah gent thn w neomycin. These results suggest tht neomycin B nd streptomycin lso hve presynptic blocking effect t the neuromusculr junction nd tht this presynptic ction is more predominnt for neomycin B thn for streptomycm. This supposition ws tested by mesuring ft U It d-tvb.cer.rin. FIG. 2. Effect of incresing concentrtions of drugs on the contrctile response of rt diphrgm to phrenic nerve stimultion nd to Ah injection. Ech point represents the muscle response (s percentge of the control response) fter 1 minutes of exposure to the concentrtion ofdrug indicted. The stright log dose-response lines were clculted by regression nlysis. A. A typicl experiment is iilustrted compring the reltionship of the log doseresponse lines to Ah injection (Ah; open symbols) with the log dose-response lines to nerve stimultion (nerve; filled symbols) for neomycin B (Neo B) nd d-tubocurrine (d-tubo). B. A typicl experiment is shown compring streptomycin (Sm) with d-tubocurrine in the sme wy s demonstrted in A. the effect of the ntibiotics on Ah relese t the neuromusculr junction. Effect of neomycin nd streptomycin on Ah relese from hemidiphrgms. The concentrtions of neomycin B nd streptomycin which inhibited by 5% the response of the diphrgm to phrenic nerve stimultion in the previous experiments (tble 1) were used to test the effect of the drugs on Ah relese. These experiments mesured both the mount of Ah relesed spontneously by the phrenic nerve- (N)

1977 NEOMYIN AND TRANSMITTER RELEASE 581 TABLE 1 omprison of the bility of drugs to inhibit neuromusculr trnsmission with the bility to inhibit the response to injected cetyicholine d-tubocurrine Streptomycin Neomycin B oncentrtion of drug (M) tht inhibited 8.7 ± 1.2 x 1#{176} 5.4 ±.5 x 1 by 5% the response to Ah injection oncentrtion of drug (M) tht inhibited 6.5 ±.4 x 1-i 1.2 ±.1 x 1- by 5% the response to nerve stimultion Residul response to Ah injection in the 14.8 ± 2.4 27.1 ± 4.3 presence of concentrtion of drug tht blocked response to nerve stimultion by 5% 8.8 ± 1.4 x 1 6.2 ±.5 x 1-- 54.8 ± 2.Gr Mesurements re the men ±S.E. from five experiments. b P <.5 s compred to the sme mesurement for neomycin B. P <.1 s compred to the sme mesuement for streptomycin. P <.5 s compred to the sme mesurements for neomycin B nd streptomycin. diphrgm preprtion nd the mount relesed during nerve stimultion (2 Hz for 13 minutes). In ll experiments, Ah relese ws mesured first in the bsence of drug, then in the presence of drug, nd then fter wshing out the drug. A typicl experiment is illustrted by figure 3 which shows tht neomycin B clerly reduced the evoked Ah relese, but hd little effect upon resting relese; the effect of neomycin B ws redily reversed by wshing. In 12 mesurements, spontneous relese in drug-free medium ws 2.7 ±.2 pmol/min nd this ws not significntly (P >.5) ltered in the presence of neomycin B or streptomycin; spontneous relese in the presence of neomycin B nd streptomycin ws 2.5 ±.3 nd 2.6 ±.2 pmol/min, respectively. Phrenic nerve stimultion incresed Ah relese to 3 to 4 times the spontneous relese (fig. 3). In the present experiments, evoked relese ws clculted by subtrcting the resting relese in the period before the stimultion period from the mount of Ah relesed during the stimultion period. Neomycin B or streptomycin reversibly decresed the evoked relese of Ah, nd neomycm B ws somewht more effective thn ws streptomycin t equieffective blocking concentrtions (fig. 4), lthough this difference ws not sttisticlly significnt (P =.8). Effect of neomycin on ionized clcium. One striking property of the neuromusculr block produced by the minoglycosides is the reversl of the block with clcium. This led orrdo Tim. mm N.o B6X14M FIG.3. A representtive experiment demonstrting the effect of neomycin B on spontneous nd evoked relese of Ah from the rt phrenic nervediphrgm preprtion. The filled brs represent the mount of Ah relesed during suprmximl nerve stimultion (2 Hz for the first 13 minutes of the 15-minute collection period) nd the open brs represent Ah relese t rest. Neomycin B, 6 x 1 M, ws present in the Krebs medium during the period indicted. (1963) nd Sobek et l. (1963) to suggest tht these drugs might be cting by decresing ionized clcium. Elmqvist nd Josefsson (1962) nd Pittinger (197) hve shown tht blocking concentrtions of neomycin nd streptomycin, respectively, do not significntly ffect ionized

582 WRIGHT AND OLLIER Vol. 2. U. V N.oB 6X14M SM i2z1o3m FIG. 4. Effect of neomycin B (6 x 1 M) nd streptomycin (1.2 x 1 M) on the evoked relese of Ah from the rt phrenic nerve-diphrgm preprtion. Fifteen-minute periods of stimultion in the presence of the ntibiotics (htched nd lined brs) were brcketed by 15-minute periods of stimultion in the bsence of drugs (dotted brs) (see fig. 3). Evoked relese ws clculted by subtrcting the Ah collected during the 15-minute rest period before stimultion from the Ah collected during the 15-minute stimultion period nd expressed s percentge of the relese during the initil control stimultion period. Vlues represent men ±S.E. from six experiments., P <.1 compred with brcketed controls. clcium, nd the present experiments confirmed this for neomycin. In Krebs solution (1,, 2.5 mm), prepred with l2 stndrd solution, the mesured ionized clcium ws 1.87 ±.22 mm, nd neomycin B (up to 2 mm) hd no significnt effect (ionized clcium 1.84 ±.25 mm); with higher concentrtions of neomycin B (4 nd 8 mm), mesured ionized clcium ws depressed to 1.75 ±.37 nd 1.52 ±.33 mm, respectively. As neuromusculr nd gnghionic blockde is esily demonstrble with concentrtions of2 mm nd less of neomycm B, it is unlikely tht clcium cheltion by neomycin is contributing to the presynptic blocking effect demonstrted in the present experiments, nd reduced clcium influx into presynptic nerve terminls is more likely explntion for the reduced Ah relese. Site of the block of gnglionic trnsmission produced by neomycin. To test the hypothesis tht neomycin B ws competing t the nerve terminls for the clcium necessry for Ah relese, the neuromusculr junction ws not suitble becuse of the lrge mount of clcium uptke by muscle. Blustein (1971) demonstrted tht rt gngli cn be used to mesure clcium uptke into pregnglionic nerve terminls with stimultion. This finding, together with the informtion tht neomycin blocks gnghionic trnsmission by n ction ntgonized by clcium (orrdo nd Rmos, 1958), suggested wy to test the clcium competition hypothesis, but first the site of the block of gnglionic trnsmission by neomycin hd to be elucidted. This site ws determined by experiments using the perfused superior cervicl gnglion of the ct. Initil experiments compred the effect of neomycin B on the response of the nictitting membrne to pregnghionic nerve stimultion, with the effect of neomycin on the response to nicotine injected close to the gnglion. In four experiments, the response of the nictitting membrne to pregnglionic nerve stimultion ws blocked by 5 to 1% when the gnglion ws perfused with Krebs medium contining neomycin B (2 x 1o-: M), but this concentrtion of neomycin B hd no effect on the nictitting membrne response to submximl concentrtion of nicotine injected close to the gnglion (fig. 5 shows prt of the record from typicl experiment). This indictes tht neomycin is blocking gnghionic trnsmission by predominntly presynptic ction. To confirm this presynptic blocking ction, the effect of neomycin on Ah relese into the venous effluent ws mesured in six experiments in which gngli were perfused with Krebs medi contining physostigmine nd choline. Spontneous Ah relese ws below the sensitivity of the Ah ssy (5 pmol), but 2g ii N.o B tthymy i4 FIG. 5. Representtive experiment demonstrting the effect of neomycin B perfused through the isolted ct superior cervicl gnghion on the response of the nictitting membrne to pregnghionic nerve stimultion nd to nicotine injected into the gnghion. At S, the pregnglionic nerve ws stimulted (5 Hz for 3 seconds). At 6n nd 8n, 6 or 8 nmol of nicotine trtrte in.1 ml of norml sline ws injected into the gnghion. i, before perfusion with neomycin; ii, 1 minutes fter perfusion with neomycin B (2 x 1 M); iii, 15 minutes fter switching to neomycin-free medi.

977 NEOMYIN AND TRANSMITTER RELEASE 583 Ah relese during pregnglionic nerve stimultion (2 Hz for 2 minutes) ws esily mesured. In ech experiment, evoked Ah relese ws mesured before neomycin, 13 minutes fter switching to perfusion medium contining neomycin B (2 x 1 M), nd 13 minutes fter switching bck to neomycin-free medium. The drug reduced evoked relese of Ah to 61 ± 4% of control vlues (fig. 6A) nd the drug effect ws redily reversed by wshing. In five other experiments (fig. 6B), the effect of neomycin upon the relese of Ah ws mesured during perfusion with low (.5 mm) medium. Under these conditions the relese of Ah by pregnglionic nerve stimultion ws bout 36% of tht mesured in the norml clcium mechum, nd this ws expected from the results of others (Hrvey nd McIntosh, 194; Dougls et t., 1961). In the low clcium medium, neo- mycin B reduced evoked Ah relese to undetectble levels (<1% of control) nd these conditions were used to test the effect of neomycin upon 4 uptke in rt gngli. 4 uptke in rt gngli. These experiments mesured the ccumultion of 4 by pired isolted superior cervicl gngli, bthed in low Krebs solution. One gnglion of the pir ws stimulted (1 Hz for 2 mmutes) nd the uptke (picomoles per gnglion) in ech gnghion ws mesured. By subtrcting the uptke in the nonstimulted from the uptke in the stimulted gnghion, mesure of the uptke due to stimultion ws obtined. Four experiments confirmed the results of Blustein (1971) tht pregnghionic stimultion incresed 4 ccumultion by 5 to 1% (fig. 7). When these experiments were repeted in the presence of concentrtion of d- E ft E V 3 E U, 3 ft V Sf, NOB 2x13M NB 2x13M FIG. 6. Effect of neomycin B on the evoked relese of Ah from the ct superior cervicl gnglion (A) during perfusion with norml Krebs medium (l2, 2.5 mm) nd (B) during perfusion with low clcium Krebs medium (l,,.5 mm). The dotted brs represent control evoked relese (2 Hz for 2 minutes) before perfusion with neomycin B nd 13 minutes fter perfusion with neomycin-free medi, nd the htched brs represent evoked relese 13 minutes fter perfusion with neomycin B (2 x 1 M). The relese of Ah into low clcium medium contining neomycin ws undetectble; the br in tht instnce represents the limit of detection of the Ah ssy (5 pmol). Vlues represent men ±S.E. from five experiments. < (.5) compred with brcketed controls. ontrol d-tubo N.o B 2x1O4M 2x13M FIG. 7. The effect ofd-tubocurrine nd neomycin B on the increse in 4 uptke by rt superior cervicl gngli induced by pregnglionic nerve stimultion (1 Hz for 2 minutes). Uptke ws clculted s the difference between the ccumultion by stimulted nd nonstimulted gngli. Ech br plus verticl line represents the men ±S.E. of the stimulted 4 uptke in picomoles from four experiments under the conditions indicted. All experiments were done with low Krebs medium (l2.5 mm) nd the drugs tht were used blocked gnglionic trnsmission completely. *P <.1 compred to 4 uptke under control conditions nd in the presence of d-tubocurrine.

584 WRIGHT AND OLLIER Vol. 2 tubocurrine (2 x 1 M) which completely blocked synptic trnsmission, s mesured by the postgnghionic ction potentil, the incresed 4 uptke in the stimulted gnglion remined; this result suggests tht this uptke ws occurring in the pregnghionic nerve. In the presence of concentrtion of neomycin (2 x 1 M), which in low Krebs medium completely blocks gnglionic trnsmission, the incresed ccumultion of 4 induced by pregnglionic nerve stimultion ws bolished (fig. 7). Effect of neomycin B on nordrenergic preprtion. Initil experiments tested the effect of neomycin B upon the mechnicl response of the rt isolted nococcygeus preprtion to field stimultion or to dded nordrenline (1 x b_s to 5 x 1 M). The response to field stimultion ws inhibited by neomycin B (1-4 x 1 M) nd this block ws prtilly reversed by n increse in the clcium concentrtion by 2 mm (fig. 8A). However, the re- 2g[ 2mm 29L(j A Neo B mm.s ss -.s-s 2mM B FIG. 8. The effect of neomycin B nd neomycin B plus clcium on the contrctile response of the isohted rt nococcyeus muscle to field stimultion nd to nordrenline. A. ontrctile responses to field stimultion re shown (suprmximl voltge,.4 msec; 2 Hz for 1 seconds). B. Ech trcing represents the response to nordrenhine in cumultive concentrtions: 1 x 1, 2 x 1_8, 4 x 1 nd 1 x 1 M. i, control; ii, dditionl 12, 2 mm; iii, neomycin B, 1 mm; iv, neomycin B, 1 mm nd dditionl l2, 2 mm. sponse of the muscle to nordrenline ws lso inhibited by these sme concentrtions of neomycin B (fig. 8B) nd this block ws lso reversed by clcium, confirming the results of Goodmn et t. (1974) on ortic smooth muscle. Thus, in order to study effects on sympthetic nerve endings, it ws necessry to mesure nordrenline relese. Muscles were incubted with [3Hlnordrenline, wshed nd then stimulted by field stimultion (2 Hz for 2 minutes) to relese 3H-nordrenline; stimultion significntly incresed the mount of rdioctivity relesed into the bth for two nd sometimes three periods of stimultion. This increse in the efflux of rdioctivity during field stimultion ws due to stimultion of nerves: first, contrctile response to Ah (1 x 1 M) produced no incresed efflux of rdioctivity; necond, the contrctile response to field stimultion ws bolished by phentolmine (1 x 1 M), but the incresed efflux ofrdioctivity ws not bolished; nd third, both the contrctile response nd the incresed efflux of rclioctivity during field stimultion were bolished by tetrodotoxin (1 x 1 M). To test the effect of neomycin B on the relese of rdioctivity during field stimultion, pired preprtions were loded with [3H]nordrenline s described in Methods ; fter 7 to 8 minutes of wshing, neomycin B (4 x 1 M) ws dded to one of the preprtions nd 2 minutes lter both preprtions were stimulted (2 Hz) for the initil 2 minutes of 5- minute collection period. The neomycin B ws then wshed out, nd fter 25 minutes of wshing, both preprtions were gin stimulted for the initil 2 minutes of the 5-minute collection period. Figure 9 shows the results of these experiments. The percentge increse in relese of rdioctivity bove n verge bseline efflux during the first stimultion period in the presence of neomycin B is compred with the percent increse during the first stimultion period without neomycin B; neomycin B significntly inhibited this increse in relese. In ddition, this blockde of relese is reversible becuse when the neomycin B ws wshed out relese during the second stimultion period ws incresed, wheres in the control preprtion relese ws decresed during the second stimultion period (fig. 9). Therefore, neomycin B ppers to reversibly block nordrenline relese s well s Ah relese.

1977 NEOMYIN AND TRANSMITTER RELEASE 585 18 16 g14 3 E12 U, #{149}. i ::: E.8 ;: ::: : ::..: #{149}.: 4 2: POoB- - Si S2 + - FIG. 9. The effect of neomycin B on the relese of nordrenline from the rt nococcygeus muscle induced by field stimultion. The pired nococcygeus preprtions hd been previously incubted with [3H]nordrenhine. The increse in the efflux of rdioctivity during stimultion ws expressed s percentge of the bse-line efflux. Relese into norml Krebs medium from one nococcygeus muscle during two stimultion periods (2 Hz for 2 minutes) 25 minutes prt (51, 52) ws compred with relese from the opposite nococcygeus muscle under identicl conditions except tht neomycin B, crosshtched br (4 x 1 M), ws present during the first stimultion period (Sb). Vlues represent the men ±S.E. from five experiments. **P <.5 compred with control Si. *1) <.2 compred with neomycin B 52. Discussion The minoglycosides re group of ntibiotics with common toxic effects: neuromusculr blockde, ototoxicity nd nephrotoxicity. Neomycin is the most toxic of the minoglycosides nd is now no longer recommended for use prenterlly in mn becuse of this toxicity. There hs been some difference of opinion s to the site of ction of the neuromusculr prlysis by the minoglycosides (see Introduction). This controversy hs probbly risen becuse different investigtors hve used different minoglycoside ntibiotics nd different concentrtions of ionized clcium. We hve shown tht t given ionized clcium concentrtion (1.6 mm), the neuromusculr block produced by neomycin hs greter presynptic component thn the neuromusculr block produced by streptomycm; neomycin is t lest twice s potent s streptomycin in inhibiting Ah relese from the neuromusculr junction (see fig. 4), but is less potent thn streptomycin in inhibiting the response to injected Ah (see tble 1). This fct would explin why Elmqvist nd Josefsson (1962) nd Vitl Brzil nd Prdo-Frnceschi (1969), who studied neomycin, were most impressed with the presynptic blocking effect, wheres Dretchen et t. (1973) nd Dunkley et l. (1973), who studied streptomycin, were most impressed with the postsynptic blocking effects. Ionized clcium concentrtion will lso ply mjor role s to whether the pre- or postsynptic effect is predominnt. Incresing ionized clcium concentrtion will decrese the presynptic component of the block nd it will hve no effect or it could potentilly increse the postsynptic block (Elmqvist nd Josefsson, 1962; Nstuk nd Liu, 1966). On the other hnd, decresing ionized clcium will mrkedly potentite the presynptic block (see fig. 7) nd it would not be expected to ffect the postsynptic block. In mn, where the norml plsm ionized clcium concentrtion is.94 to 1.33 mm (Moore, 197), the presynptic block my be more predominnt nd bnormlly low ionized clcium concentrtions could be very importnt in potentiting the neuromusculr blocking effect of the minoglycosides. The incresed uptke ofclcium into gngli ssocited with stimultion is not blocked by postsynptic blocking gents (fig. 9; Blustein, 1971), but it ws blocked by neomycin under conditions where neomycin inhibited Ah relese. This is suggestive evidence tht neomycin, like mgnesium (Blustein, 1971), cts to block the clcium influx tht is necessry for Ah relese. It hs been suggested tht mgnesium my lso ct by competing with clcium t n intrcellulr site (ooke et!., 1973 ; Miledi, 1973) in ddition to competing for uptke t the externl membrne. It is unlikely tht neomycin cts intrcellulrly s it is lrge chrged polyctionic orgnic bse nd its effects re rpid in onset nd rpidly reversible. If neomycin blocks Ah relese by blocking the influx of clcium tht triggers neurotrnsmitter relese, the drug might be expected to block the clcium-dependent relese of other neurotrnsmitters nd hormones (Rubin, 197). This ws shown for nordrenline relese from

586 WRIGHT AND OLLIER Vol. 2 drenergic nerve terminls. Other processes tht depend upon uptke lso pper to be sensitive to neomycin nd other minoglycosides; thus, in rbbit ortic strips nd vrious cnine rteries, neomycin inhibited the contrctile responses which re dependent upon utiliztion of clcium from superficil sites (Goodmn et t., 1974; Adms nd Goodmn, 1975); gentmicin produced negtive inotropic effect on rt myocrdium which ws ntgonized in competitive mnner by clcium (Adms, 1975); nd neomycin nd streptomycin competitively inhibited 4 uptke into frgmented srcoplsmic reticulum (Firhurst nd Mcri, 1975). Therefore the minoghycosides my be useful phrmcologicl tool for studying the processes involved in clcium movement cross membrnes in vriety of systems. Acknowledgments. We wish to thnk Jn Elliot nd Lorett Bonet for their ssistnce in these experiments. We thnk Roussel Ltd. (nd) of Montrel, Quebec for providing neomycin B sulfte (lso clled frmycetin sulfte) nd Pfizer ompny Ltd. of Montrel, Quebec for providing streptomycin sulfte. References ADAMS, H. R. : Direct myocrdil depressnt effects of gentmicin. Eur. J. Phrmcol. 3: 272-279, 1975. ADAMS, H. R. AND GOODMAN, F. R.: Differentil inhibitory effect of neomycin on contrctile responses of vrious cnine rteries. J. Phrmcol. Exp. Ther. 193: 393-42, 1975. BLAUSTEIN, M. P.: Pregnglionic stimultion increses clcium uptke by sympthetic gnghi. Science (Wshington) 173: 391-393, 1971. BURGEN, A. S. V., DIKENS, F. AND ZATMAN, L. J.: The ction ofbotulinum toxin on the neuromusculrjunction. J. Physiol. (London) 19: 1-24, 1949. OLLIER, B. AND LANG,.: The metbolism of choline by sympthetic gnghion. n. J. Physiol. Phrmcol. 47: 119-126, 1969. OOKE, J. D., OMoio, K. AND QUASTEL, D. M. J.: The role of clcium in depolriztion secretion coupling t the motor nerve terminl. J. Physiol. (London) 228: 459-497, 1973. ORRADO, A. P. : Respirtory depression due to ntibiotics. lcium in tretment. Anesth. Anlg. 42: 1-5, 1963. ORRADO, A. P. AND RAM5, A..: Neomycin: Its curriform nd gnglioplegic ction. Rev. Brsil. Biol. 18: 81-85, 1958. DOUGLAS, W. W., LywooD, D. W. AND STRAUB, R. W.: The stimulnt effect of brium on the relese of cetylcholine from the superior cervicl gnglion. J. Physiol. (London) 156: 515-522, 1961. DRETHEN, K. L., GERGIS, S. D., SOKOL, M.D. AND LONG, J. P.: Effect of vrious ntibiotics on neuromusculr trnsmission. Eur. J. Phrmcol. 18: 21-23, 1972. DRETHEN, K. L., SOKOLL, M.D., GERGIS, S. D. AN LONG, J. P.: Reltive effects of streptomycin o motor nerve terminl nd endplte. Eur. J. Phr mcol. 22: 1-16, 1973. DUNKLEY, B., SGHvI, I. AND GOLDSTEIN, G.: hr cteriztion of neuromusculr block produced b streptomycin. Arch. Int. Phrmcodyn. Ther. 21 213-223, 1973. ELMQVIST, D. AND JOSEFSSON, J.. : The nture o neuromusculr block produced by neomycin Act. Physiol. Scnd. 54: 15-11, 1962. FAIRHURST, A. S. AND MARI, J.: Aminoglycosid interctions in skeletl muscle preprtions. Life Sci. 16: 1321-133, 1975. GILLESPIE, J. S.: The rt nococcygeus muscle n its response to nerve stimultion nd to som drugs. Brit. J. Phrmcol. 45: 44-416, 1972. GOLDBERG, A. M. AND MAMAN, R. E. : The deter mintion of picomole mounts of cetylcholine i mmmlin brin. J. Neurochem. 2:1-8,1973. GOODMAN, F. R., WEIss, G. D. AND ADAMS, H. R.: Altertions by neomycin of 4 movements n contrctile responses in vsculr smooth muscle. J. Phrmcol. Exp. Ther. 188: 472-48, 1974. HARVEY, A. M. AND MAINTOSH, F.. : lcium nd synptic trnsmission in sympthetic gnglion. J. Physiol. (London) 97: 48-416, 194. HATTNER, R. S., JOHNSON, J. W., BERNSTEIN, D. S., WAcHMAN, A. AND BRAHMAN, J.: Electrochemicl determintion of pprent ionized serum clcium using clcium-selective electrode: the method nd vlues in norml humns nd comprison to totl serum clcium. lin. him. Act 28: 67-75, 197. KATO, A.., OLLIER, B., hson, D. AND WRIGHT, J. M.: The effect of tropine upon cetylchohine relese from ct superior cervicl gngli nd rt corticl slices: Mesurement by rdio-enzymic method. n. J. Physiol. Phrmcol. 53: 15-157, 1975. KIBJAKOW, A. W. : Uber humorle Ubertrgung der Erregung von einem Neuron uf ds ndere. Pfluegers Arch. Gesmte Physiol. Menschen Tiere 232: 432-443, 1933. KOELLE, G. B.: Neuromusculr blocking gents. In The Phrmcologicl Bsis of Therpeutics, ed. by L. S. Goodmn nd A. Gilmn, p. 578, The Mcmilln ompny, New York, 1975. LANGER, S. Z. : The metbolism of [3H}nordrenhine relesed by electricl stimultion from the isolted nictitting membrne of the ct nd from the vs deferens of the rt. J. Physiol. (London) 28: 515-546, 197. LARRABEE, M. G., KLINGMAN, J. D. AND LEIHT, W. S.: Effects oftemperture, clcium nd ctivity on phospholipid metbolism in sympthetic gnglion. J. Neurochem. 2: 549-57, 1963. MILEDI, R.: Trnsmitter relese induced by injection of clcium ions into nerve terminls. Proc. Roy. Soc. London B Biol. Sci. 183: 421-425, 1973. MOGEY, G. A., TREVAN, J. W. AND YOUNG, P. A.: The ssy of d-tubocurrine chloride on the isolted rt-diphrgm: A sttisticl exmintion. Anlyst 74: 577-582, 1949. MOORE, E. W.: Ionized clcium in norml serum, ultrfiltrtes, nd whole blood determined by ionexchnge electrodes. J. lin. Invest. 49: 318-334, 197. NASTUK, W. L. AND LIU, J. H.: Muscle postjunctionl membrne chnges in chemosensitivity produced by clcium. Science (Wshington) 15: 266-267, 1966. PATERSON, G.: Twitch responses with cetylcholine

977 NEOMYIN AND TRANSMITTER RELEASE 587 in the isolted innervted nd chroniclly denervted rt diphrgms nd their modifiction by neuromusculr blocking gents. J. Phrm. Phrmcoh. 17: 281-294, 1965. PITTINGER,. B.: ph nd streptomycin influences upon ionic clcium in serum. Anesth. Anlg. 49: 54-545, 197. PITTINGER,. B. AND ADAMSON, R.: Antibiotic blockde of neuromusculr function. Annu. Rev. Phrmcol. 12: 169-184, 1972. PITTINGER,. B., ERYASA, Y. AND ADAMSON, R.: Antibiotic-induced prlysis. Anesth. Anlg. 49: 487-51, 197. PITTINGER,. B., LONG, J. P. AND MILLER, J. R.: The neuromusculr blocking ction of neomycin: A concern of the nesthesiologist. Anesth. Anlg. 37: 276-282, 1958. PRIDGEN, J. E.: Respirtory rrest thought to be due to intrperitonel neomycin. Surgery 4: 571-574, 1956. RUBIN, R. P. The role of clcium in the relese of neurotrnsmitter substnces nd hormones. Phrmcol. Rev. 22: 389-428, 197. SOBEK, V., HAVA, M., MIKULAKOVA, J. AND WAITzovA, D.: Uber den Mechnismus der toxischen Wirkungen von Neomycin und die Moglichkeiten ihrer Beeinflussung. Arzneimittel-Forschung 13: 391-397, 1963. TIMMERMAN, J.., LONG, J. P. AND PITTINGER,. B.: Neuromusculr blocking properties of vrious ntibiotic gents. Toxicol. Appl. Phrmcol. 1: 299-34, 1959. VITAL BRAZIL,. AND ORRADO, A. P.: The curriform ction of streptomycin. J. Phrmcol. Exp. Ther. 12: 452-459, 1957. VITAL BRAZIL,. AND PRADO-FRANESHI, J.: The nture of neuromusculr block produced by neomycin nd gentmicin. Arch. mt. Phrmcodyn. Ther. 179: 78-85, 1969.