Assays for, and cross-reactivities of, IgE antibodies to the muscle relaxants gallamine, decamethonium and succinylcholine (suxamethonium)

Assays for, and cross-reactivities of, IgE antibodies to the muscle relaxants gallamine, decamethonium and succinylcholine (suxamethonium)

Journal of Immunological Methods, 78 (1985) 293-305 Elsevier 293 JIM 03462 Assays for, and Cross-Reactivities of, IgE Antibodies to the Muscle Rela...

762KB Sizes 0 Downloads 8 Views

Journal of Immunological Methods, 78 (1985) 293-305 Elsevier

293

JIM 03462

Assays for, and Cross-Reactivities of, IgE Antibodies to the Muscle Relaxants Gallamine, Decamethonium and Succinylcholine (Suxamethonium) 1 David G. Harle *, Brian A. Baldo ,,2 and Malcolm M. Fisher ** * Kolling Institute of Medical Research, Royal North Shore Hospital, Sydney, and Department of Medicine, University of Sydney, Sydney, and ** Intensive Therapy Unit, Royal North Shore Hospital, St. Leonards, Sydney, N.S. W. 2065, Australia

(Received 1 October 1984, accepted 28 December 1984)

Two radioimmunoassays have been developed to detect IgE antibodies to succinylcholine, decamethonium and gallamine in the sera of patients who have experienced life-threatening anaphylactoid reactions following administration of a muscle relaxant drug. They involve the coupling of choline and its ethyl analogue, triethylcholine to activated Sepharose. A high degree of cross-reactivity was shown to occur between drug-reactive IgE antibodies and 6 muscle relaxants as well as choline and triethylcholine. Results suggest that the specificities of the IgE antibodies are directed towards quaternary or tertiary ammonium ions on the drugs that bind the antibodies. Molecular models of these compounds support the structure-activity relationships determined in the inhibition studies. Key words: radioimmunoassay - IgE antibodies- anaphylactoid - cross-reactivity - muscle relaxants quaternary and tertiary ammonium groups

Introduction An increasing number of serious, and occasionally fatal, anaphylactic-like (anaphylactoid) reactions are being reported following exposure of patients to muscle relaxant drugs during anaesthesia (Fisher, 1975; Dundee, 1976; Clarke, 1979; Vervloet et al., 1979; Lira and ChurchiU-Davidson, 1981; Moneret-Vautrin et al., 1981; Fisher and Munro, 1983; Stoetling, 1983). Since most reports have been clinical in nature, the mechanisms underlying the adverse reactions have not been resolved although results of recent Studies have suggested that the reactions have an immunological basis (Baldo and Fisher, 1983a, b, c). 1 This work was funded by the National Health and Medical Research Council of Australia and the Harry Daly Foundation of the Faculty of Anaesthetists, Royal Australasian College of Surgeons. 2 Address all correspondence to: Brian A. Baldo, Kolling Institute of Medical Research, Royal North Shore Hospital, St. Leonards, Sydney, N.S.W. 2065, Australia. 0022-1759/85/$03.30 © 1985 Elsevier Science Publishers B.V. (Biomedical Division)

294

We previously reported the detection of IgE antibodies that bind to the muscle relaxant drugs alcuronium, ( + )-tubocurarine and succinylcholine (Fig. 1) in the sera of some patients who experienced an anaphylactoid reaction following administration of these drugs (Baldo and Fisher, 1983a, b, c; Harle et al., 1984). The detection of antibodies that react with some of the muscle relaxants, in particular, succinylcholine, decamethonium and gallamine, poses problems because of the lack of suitable functional groups (Fig. 1) which may be used to link the molecules to a protein carrier or insoluble support. Succinylcholine may be thought of as 2 molecules of choline linked by a carbon chain derived from succinic acid. This is illustrated in Fig. 2 where the structural formulae of choline and succinylcholine are shown alongside the corresponding space-filling CPK models. We reasoned that coupling choline (via the hydroxyl group) to a solid support should produce a complex which structurally mimics the terminal portions of the succinylcholine molecule. Such a complex should be antigenically similar, if not identical, to a succinylcholine-Sepharose complex. Because of the structural similarity between succinylcholine and decamethonium (Fig. 2), it seemed likely that an immunoassay employing choline-solid-phase complex would detect antibodies to both drugs. Unlike succinylcholine and decamethonium, gallamine contains an aromatic ring.

/•CH

2

CH~=CH--CH~--N ~

H3CO

OH

OH

H/~H3

(+)-Tubocurarine

Alcuronium

+ + (CH3)3N--(CH2)lo--N(CH 3)3

H3C\+ ? ? ÷/CH:~ H3C'~NCH~CH2OCCH2CH2COCH2CH~N~-CH~ H3C CH3'

Decamethonium

Succinylcholine O II

HO C C H ~ " CH2CH2 " N+(C2Hs)3 • CH2CH2 " N÷JC2Hs)3 • CH2CH2 • N+(C2Hs)3 : Gallamine

H Pancuronium

Fig. 1. Structures of the muscle relaxant drugs used in the radioimmunoassay studies.

295

+

(CH3)3NCH2CH20H CHOLINE

0 0 + H II + (CH3)3NCH2CH2OCCH2CH2COCH2CH2N(CH3)3

SUCCINYLCHOLINE

+

+

(CH3)3NCH2CH2(CH2)6CH2CH2N(CH3)3 DECAMETHONU IM Fig. 2. Structures and CPK models of choline and the muscle relaxant drugs succinylcholine and decamethonium. The structurally similar parts of the molecules are underlined.

However, its side chains are structurally similar to the 2 'straight chain' muscle relaxants except for the presence of ethyl and not methyl groups in the ammonium ion (Fig. 1). We therefore synthesized triethylcholine (Fig. 3), the ethyl analogue of choline and coupled the compound directly to a solid support with the intention of detecting gaUamine-reactive IgE antibodies in patients' sera. This paper gives details of the 2 assays we have employed to detect IgE antibodies to succinylcholine, decamethonium and gallamine in the sera of patients allergic to

296

+

(CH3CH2)3NCH2CH2OH TRIETHyLCHOLINE

÷

N(CH2CH3)3 CH2 CH2 0 (CH3CH2)3NCH2CH20 QCH2CH2N(CH2CH3)3 fiALLAMINE Fig. 3. Structures and CPK models of t6ethylcho~ne and the muscle relaxant drug gallamine. The structurally similar parts of the moleculesare under,ned.

these drugs and examines the molecular basis of the observed cross-reactivities between the antibodies. Materials and Methods Chemicals Choline chloride and ethanolamine were obtained from Sigma Chemical Co., St. Louis, MO; methyl ethyl ketone from BDH Chemicals, Australia; (+)-tubocurarine

297

chloride, decamethonium bromide and succinylcholine chloride from Wellcome Australia, Rosebery, N.S.W.; alcuronium dichloride from Hoffmann - La Roche, Basle; gallamine triethiodide from May and Baker, West Footscray, Victoria; pancuronium bromide from Organon, Oss; 2-diethylaminoethanol and ethyl iodide from Fluka Buchs.

Subjects and sera Ten patients were bled 1 month after experiencing severe anaphylactoid reactions to muscle relaxants. Four of these patients were shown by intradermal testing (Fisher, 1981) to have reacted to succinylcholine, although none had received succinylchohne previously. One patient (Pe) had a previous severe anaphylactoid reaction to decamethonJum on first exposure to a muscle relaxant. Two patients (Do and Mo) reacted clinically to gallamine on first exposure to the drug and both were skin test positive to gallamine. The other 4 patients reacted to either alcuronium or (+)-tubocurarine. Control sera were obtained from 10 members of the hospital staff with no history of adverse reactions to anaesthesia and cord sera were obtained from the Department of Obstetrics, Royal North Shore Hospital.

Synthesis of triethylcholine 2-diethylaminoethanol (4.0 g, 3.4 mmol) was dissolved in dry methyl ethyl ketone (20 ml) in a round-bottom flask fitted with a water condenser and a drying tube. Ethyl iodide (5.34 g, 3.4 mmol) was dissolved in dry methyl ethyl ketone (20 ml) and added slowly to the substrate. The mixture was refluxed at 80°C for 8 h. The solvent was removed on a rotary evaporator and the residue dissolved in water (100 ml). The aqueous layer was then washed twice with ether (100 ml) and concentrated. The residue was then recrystallised (twice) from ethanol, and the crystals obtained were washed with ether, and vacuum dried yielding pure triethylcholine iodide (4.2 g, 84%), m.p. 280-281°C (Dauterman and Mehrotra (1963) reported a value of 283°C). A 100 mHz 1H-NMR spectrum (Varian XLFT-100) confirmed the identity of the product and was consistent with the literature (Dev6s and Krupka, 1979).

Solid-phase complexes The optimum pH, temperature and time were determined for the couphng of choline and triethylcholine to activated Sepharose. Choline chloride (60 mg in 6 ml of distilled water) was covalently coupled to 500 mg of Epoxy-activated Sepharose 6B (Pharmacia) by adjusting the pH to 12.5 with 2.5 M NaOH and gently shaking with the activated Sepharose at 35°C for 20 h. After washing with water, 0.1 M borate buffer, pH 8, and 0.1 M acetate buffer, pH 4, remaining free activated groups were blocked by incubation with 1 M ethanolamine, pH 9, at room temperature for 4 h. The same procedure was used for the coupling of triethylcholine to activated Sepharose except that coupling was performed at pH 12.6. Ethanolamine-Sepharose complex was prepared by using 300 mg of ethanolamine and 500 mg of activated Sepharose.

298

Detection of IgE antibodies by radioimmunoassay Serum (50 #1) was added to 6 mg of the solid-phase complex (choline-, triethylcholine-, ethanolamine-Sepharose or Sepharose alone), and left at room temperature for 3 h. Tubes were then washed and centrifuged 3 times with phosphate-buffered saline containing Tween 20 before adding 20,000-40,000 cpm of ~25I-anti-human IgE (Pharmacia). After standing overnight, tubes were washed 3 times and counted in a Packard Auto-Gamma Spectrometer. The presence of specific IgE antibodies was determined by the per cent radioactive uptake of 125I-anti-IgE, that is percent of counts added.

Inhibition assay Serum (50 /~1), appropriately diluted was incubated for 1 h with 50 #1 of a solution of the compound being tested before addition of the solid phase (6 mg in 100 #1). After 3 h, tubes were centrifuged at 3 500 rpm for 10 min and washed 3 times with PBS-Tween before the addition of 125I-anti-human IgE (20,000-40,000 cpm/tube). After overnight incubation at room temperature, tubes were washed 3 times and counted.

Results

Table I summarizes results obtained from direct binding radioimmunoassay studies with a number of patients who had experienced anaphylactoid reactions to muscle relaxant drugs. A positive value in the RAST assay was taken as greater than 2.5 times the cord/normal value (Baldo and Fisher, 1983c). Each of the 10 sera produced uptakes of 125I-anti-IgE of 4.2% or greater when tested in the choline-Sepharose radioimmunoassay. Uptakes for 3 of the sera, Pe, Wh and Hu were 17.5, 19.6 and 17.1% respectively, figures that can be regarded as strong positives. By contrast, control solid supports of ethanolamine-Sepharose and unmodified Sepharose gave low uptakes with these sera and along with the low values for cord and normal sera indicated that the assay was detecting choline- (and presumably succinylcholine- and decamethonium-)reactive IgE antibodies. Interestingly, patient Mi, who had not previously encountered succinylcholine, subsequently had a minor reaction (tachycardia and transient bronchospasm) to this relaxant. Similar results were obtained with the triethylcholine assay. All 10 sera (including Do and Mo, who had reacted clinically to gallamine) demonstrated triethylcholine(and most likely, gallamine-)reactive IgE antibodies, compared to cord and normal sera, which gave low radioactive uptakes. The results for the alcuronium and (+)-tubocurarine assays have been previously reported (Baldo and Fisher, 1983c) and are included for comparison to show the cross-reactivity that exists between IgE antibodies to the various muscle relaxant drugs. Further data to check the specificities of the IgE antibodies was obtained from inhibition studies, in which patients' sera were preincubated with each of the 6

-

Wh Do * Mo * Hu * Ko * Mi * Od *" Cord Normals

Ge Lo Pe

Serum

-

Succinylchofine Succinylcholine Succinylcholine/ decamethonium Succinylcholine Gallamine Gallamine Alcuronium Alcuronium ( + )-Tubocurarine ( + )-Tubocurarine ---

19.6( + ) 5.5( + ) 6.7( + ) 17.1( + ) 8.1( + ) 10.4( + ) 4.2(+) 0:8( - ) 0.8_+0.3(-)

14.1( + ) 7.9( + ) 17.5( + )

CholineSepharose

adverse reaction

5.5( + ) 3.6( + ) 10.9( + ) 10.1( + ) 2.9( + ) 7.5( + ) 4.4(+) 0.3( - ) 0.4_+0.2(-)

11.3( + ) 8.0( + ) 10.6( + )

TriethylcholineSepharose

5[ Uptake of 1~ I-anti-human IgE with

Drug eliciting

22.8( + ) 4.7( + ) 9.9( + ) 22.8( + ) 1.2( - ) 8.7( + ) 1.3(-) 1.0( - ) 1.15:0.3(-)

1.7( - ) 2.1( - ) 14.5( + )

AlcuroniumSepharose

5.2( + ) 9.6( + ) 10.7( + ) 3.1( + ) 5.3( + ) 12.2( + ) 6.4(+) 1.4( - ) 1.44-0.3(-)

9.8( + ) 10.3( + ) 23.4( + )

( + )-TubocurarineSepharose

0.6 0.9 3.0 1.5 0.8 0.8 0.7 ND ND

0.4 0.6 0.8

EthanolamineSepharose

0.7 0.9 1.C 1.2 0.8 0.6 0.6 ND ND

0.4 0.4 0.8

Sepharose

For experimental details see Materials and Methods section. Results for the alcuronium and (+)-tubocurarine assays have been previously reported (Baldo a n d Fisher, 1983e). A m o u n t s of IgE antibodies detected are represented by numbers which are percentage uptakes of 125I-anti-human IgE. Presence or absence of IgE antibodies denoted by ( + ) or ( - ). Ten normals were used as controls (mean + SD). Patients marked with (*) had not been exposed to succinylcholine. N D not done.

RESULTS O F TESTS F O R S E R U M IgE A N T I B O D I E S T O S U C C I N Y L C H O L I N E , D E C A M E T H O N I U M , G A L L A M I N E , A L C U R O N I U M A N D ( + ) - T U B O C U R A R I N E IN PATIENTS W I T H L I F E - T H R E A T E N I N G A N A P H Y L A C T O I D R E A C T I O N S T O M U S C L E R E L A X A N T S

TABLE I

t,O

Succinylcholine

Ge (1:4) (1 : 3) Hu (1 : 12) (1 : 3) Ko (1 : 2) Lo (1:2) (1 : 2) Mi (1:2) (1 : 2) Mo (1:2) Pe (1 : 12)

(1:2) Wh (1 : 12) Alcuronium (Undiluted)

GaUamine Suecinylcholine/ decamethonium

( + )-Tubocurarine

Alcuronium Succinylcholine

Alcuronium

Drug eliciting adverse reaction

Serum and dilution Alcuronium

Triethylcholine Choline Triethylcholine

68 3.3 50

8.0 9.7 74

1.3 0.64 4.3 25 9.0 1.8 5.9 5.0 28 40 0.72

(+)-Tubocurarine

6.7 9.8 240

17 15 10 9.4 2.9 4.7 4.9 4.0 5.0 54 7.7

Pancuronium

12 11 18

4.7 6.2 6.1 12 21 6.0 21 13 19 55 9.8

Succinylcholine

4.5 0.77 3.0

0.50 0.35 2.0 4,3 1.6 0.89 3.2 1.1 2.9 32 1.9

Decamethonium

5.6 9.6 5.6

1.6 13 7.8 14 8,7 6,8 5,9 11 6.0 37 8.4

Gallamine

38 22 58

12 10 20 34 30 16 130 51 70 500 23

Choline

Amount (nmol) of compound needed for 50% inhibition of the uptake of 125I-anti-human IgE

Choline 88 Triethylcholine 89 Choline 15 Triethylcholine 15 Choline 30 Choline 37 Triethyleholine 35 Choline 23 Triethyleholine 28 Triethylcholine 750 Choline 58

Sepharose coupled to

For experimental details see Materials and Methods Section.

23 44 12

16 9.5 25 22 110 6.8 20 37 22 450 20

Triethylcholine

CROSS-REACTIVITY BETWEEN MUSCLE RELAXANTS; INHIBITION STUDIES WITH SERA C O N T A I N I N G lgE ANTIBODIES TO SUCCINYLCHOLINE, D E C A M E T H O N I U M OR G A L L A M I N E

TABLE II

301 100. 90. 80.

70. Z 0 Z

60.

50. 40. 30" 20. 10. 00.1

i

10

100

COMPOUND ADDED (nmol)

Fig. 4. Inhibition by 6 muscle relaxant drugs, choline and triethylcholine of IgE binding in serum G-e to choline-Sepharose complex. Serum Ge was used at a dilution of 1:4. (O) alcuronium, (e) (+)-tubocurarine, ([2) succinylcholine, (11) decamethonium, (A) gaUamine, (,,) pancuronium, (O) triethylcholine, ( 0 ) choline.

muscle relaxants (alcuronium, (+)-tubocurarine, pancuronium, succinylcholine, decamethonium, gallamine) and with choline and triethylcholine using choline- and triethylcholine-Sepharose as solid supports. The quantifies (nmol) of these compounds required to produce 50% inhibition of the reaction of the IgE antibodies with choline- and triethylcholine-Sepharose are given in Table II. Figs. 4-7 show inhibition results obtained with sera Ge and Wh with both solid supports. It is clear from this data that there is great diversity in the affinity of the 6 muscle relaxants, choline and triethylcholine for the choline-reactive and triethylcholine-reactive antibodies. 100. 90. 80.

70. Z 0

_.E m

z

60.

,50. 40. 30. 20' 1.0. 0~).1

'i

10

1()0

COMPOUND ADDED (nmol)

Fig. 5. Inhibition by 6 muscle relaxant drugs, choline and triethylcholine of IgE binding in serum Wh to choline-Sepharose complex. Serum Wh was used at a dilution of 1 : 12. (O) aicuronium, (e) ( + )tubocurarine, (n) succinylcholine, (11) decamethonium, (z~) gallamine, (A) pancuronium, ( ~ ) triethylcholine, ( ~ ) choline.

302 lOO 9o 80 70, z 0

60.

~- 50" "r 40' 30' 20 10 0.1

1

10

100

COMPOUND ADDED (nmol)

Fig. 6. Inhibition by 6 muscle relaxant drugs, choline and triethylcholine of IgE binding in serum Ge to triethylcholine-Sepharose complex. Serum Ge was used at a dilution of 1:3. (©) alcuronium, (O) (+)tubocurarine, (t3) succinylcholine, (I) decamethonium, (zx) gallamine, (A) pancuronium, ( ~ ) triethylcholine, ( 0 ) choline.

With both solid supports, decamethonium proved to be the most potent inhibitor of all the compounds tested. Fifty per cent inhibitory values ranged from as little as 0.35 nmol of the drug required for Ge with triethylcholine-Sepharose to 32 nmol for Mo with the same solid-phase complex. On the other hand, alcuronium was clearly the least potent inhibitor, showing great variation, ranging from 3.3 nmol for Wh with choline-Sepharose to 750 nmol for Mo with triethylcholine-Sepharose.

100, 90i 80 70

/

60 Z

o

5o~

--~ 40i -'r z.

S/

30, 20~ 10~ ~.1

'I

lO

160

COMPOUND ADDED (nmol)

Fig. 7. Inhibition by 6 muscle relaxant drugs, choline and triethylcholine of IgE binding in serum Wh to triethylcholine-Separose complex. Serum Wh was used undiluted. (O) alcuronium, (O) (+)-tubocurarine, (1:3) succinyicholine, (I) decamethonium, (zx) gallamine, (A) pancuronium, ( e ) triethylcholine, ( 0 ) choline.

303 Succinylcholine and gallamine gave consistently good inhibition with both supports. In particular, succinylcholine varied from 6.0 nmol for Lo with choline-Sepharose to 55 nmol for Mo with triethylcholine-Sepharose. Gallamine showed little variation in its potency as an inhibitor with values ranging from 1.6 nmol with Ge for choline-Sepharose to 37 nmol for Mo with triethylcholine-Sepharose. Considerable variation in the potency of ( + )-tubocurarine and pancuronium was observed. For example, (+)-tubocurarine ranged from as little as 0.64 nmol for Ge with triethylcholine-Sepharose to 74 nmol for Wh with triethylcholine-Sepharose. Pancuronium varied from 2.9 nmol for Ko with choline-Sepharose to 240 nmol for Wh with triethylcholine-Sepharose. Although choline and triethylcholine might be expected to be the best inhibitors in the homologous system, i.e., choline inhibiting choline-Sepharose and triethylcholine inhibiting triethylcholine-Sepharose, this was not observed. They both exhibited moderate potency as inhibitors, but in no case were they the most potent inhibitors.

Discussion

Sera from all 10 patients who had experienced an anaphylactoid reaction to a muscle relaxant (including sera Ge, Lo, Pe and Wh who reacted clinically to succinylcholine or decamethonium) gave a positive reaction when used in an assay to detect IgE antibodies that react with choline. Likewise, IgE antibodies that reacted with triethylcholine were detected in the same 10 patients, including patients Do and Mo who reacted clinically to gallamine. From the direct binding and inhibition studies it is evident that sera from patients who reacted adversely to a muscle relaxant drug, contained IgE antibodies that cross-reacted with other muscle relaxants and with choline and triethylcholine, 2 compounds that lack muscle relaxant activity. Thus, even though a patient may have reacted clinically to 1 relaxant drug and had in their serum, antibodies that were capable of binding to that relaxant, the antibodies are also able to recognize and bind to other muscle relaxants and some other structurally related compounds. This conclusion was suggested by results obtained with radioimmunoassays that detect IgE antibodies to the muscle relaxants alcuronium and (+)-tubocurarine (Baldo and Fisher, 1983b, c). In the present study, all 10 patients demonstrated IgE antibodies which bound (+)-tubocurarine. Six of these patients also had serum IgE antibodies that reacted with alcuronium. The only structural feature common to all of the muscle relaxant drugs is the quaternary ammonium group and we have suggested (Baldo and Fisher, 1983a, b) that it is this group which is complementary to the drug-reactive IgE combining sites. These studies also showed that a number of other quaternary and tertiary ammonium-containing compounds including promethazine, neostigmine, pentolineum, trimethaphan, chlorpromazine and morphine significantly inhibited the binding of serum IgE to the muscle relaxant drug bound to Sepharose. The above mentioned drugs do not, however appear to provoke anaphylactic reactions in the patients sensitive to the muscle relaxants. This, we

304 believe, is because they generally contain only one substituted ammonium ion, unlike the muscle relaxants which contain at least 2 and which are therefore capable of cross-linking cell-bound IgE, the initiating step in mediator release from mast cells and basophils (Siraganian et al., 1975). The results summarized in this paper reveal consistent and potent inhibition of IgE binding by succinylcholine, decamethonium and gallamine. These findings suggest that the choline-Sepharose and triethylcholine-Sepharose complexes are appropriate solid phases for use in assays designed to detect antibodies to succinylcholine, decamethonium and gallamine. Decamethonium was clearly the most potent inhibitor with both solid supports while alcuronium was generally the least potent inhibitor. (+)-Tubocurarine and pancuronium showed intermediate inhibitory potency. Consideration of our previous results with alcuronium and (+)-tubocurarine (Baldo and Fisher 1983b, c), the results presented here, and the molecular structures of succinylcholine, decamethonium and gallamine, leaves no doubt about the molecular basis of the binding of muscle relaxant drugs to IgE antibodies. Both direct binding and inhibition experiments clearly implicate quaternary ammonium and tertiary ammonium groups as the allergenic (that is, the IgE-binding) determinants on the drugs. Succinylcholine and decamethonium on the one hand and gallamine on the other have quaternary ammonium groups with 3 methyl and 3 ethyl groups around the N atom respectively. With each of these drugs the terminal locations of the quaternary ammonium groups and hence their accessibility can readily be seen by examining CPK models of the molecules (Figs. 2 and 3). In addition, the structural similarity between choline and each end of the linear structures succinylcholine and decamethonium can clearly be seen in Fig. 2. Likewise, triethylcholine minus the H atom from the hydroxyl group is identical to each of the side chains projecting from the aromatic ring of gallamine. Easy accessibility of antibody to these side chains is likely since the most probable configuration of the gallamine molecule is an equilateral triangle with the quaternary ammonium groups occupying each corner of the triangle (Bovet, 1972) (Fig. 3). Neither choline nor triethylcholine were the most potent inhibitors when used with the corresponding solid support. This may be related to that the fact that both of the compounds contain only 1 quaternary ammonium group whereas each of the muscle relaxants contains 2 or more quaternary or tertiary ammonium groups. Further examination of structure-activity relationships summarized in the data contained in Table II, shows that the inhibitory potencies of the 2 rigid structures alcuronium and (+)-tubocurarine are markedly different. The latter compound proved to be a far more potent inhibitor than alcuronium. Again, the explanation appears to reside in the structure of the respective quaternary ammonium ions. Alcuronium contains 2 quaternary bridgehead N atoms, each linked to an allylic substituent. (+)-Tubocurarine contains 2 ring N atoms, I present as a quaternary ammonium ion with 2 attached methyl groups and the other as a tertiary ammonium ion with 1 methyl substituent. Thus, the quaternary ammonium ions in the choline and triethylcholine molecules show greater structural similarity with the substituted ammonium ions of (+)-tubocurarine than with the quaternary ammonium ions of

305

alcuronium. This explains the relatively poor inhibitory activity of alcuronium in the choline-Sepharose and triethylcholine-Sepharose assays. We are currently attempting to correlate patient reactions to the neuromuscular blocking drugs with data from skin tests and the radioimmunoassays and although this analysis is still incomplete, some important points are beginning to emerge. For example, the immunoassay for drug-specific IgE antibodies does not appear to be as good as skin testing for detecting sensitivity to a muscle relaxant or for distinguishing the degree of sensitivity to the 6 relaxants tested. The negative predictive value of the immunoassay seems to be better than its positive predictive value. Drug-specific IgE antibodies have been found only in patients who experienced an anaphylactic reaction to a muscle relaxant. Such antibodies have not been detected in a large number of subjects who have never reacted to these drugs. However, not all patients who had both an anaphylactic reaction and a positive skin test to a muscle relaxant reacted positively in the immunoassay. Correlations between skin tests and the presence of drug-specific IgE antibodies appear to vary from approximately 90% for succinylcholine to approximately 60% for alcuronium and (+)-tubocurarine (M.M. Fisher, B.A. Baldo and D.G. Harle, unpublished). It has proved difficult to compare both methods for the screening of alternative drugs since conclusions on the safety or otherwise of a particular muscle relaxant can only be made on the basis of challenge tests which are too dangerous to be of practical use. Hence, for the present we are inclined to use both methods but to place a greater emphasis on the skin test findings.

References Baldo, B.A. and M.M. Fisher, 1983a, Nature (London) 306, 262. Baldo, B.A. and M.M. Fisher, 1983b, Mol. ImmunoL 20, 1393. Baldo, B.A. and M.M. Fisher, 1983c, Anacsth. Intensive Care 11, 194. Bovet, D., 1972, in: Neuromuscular Blocking and Stimulating Agents, Vol. 1, International Encyclopedia of Pharmacology and Therapeutics, Section 14, ed. J. Cheymol (Pergamon Press, Oxford) pp. 243-294. Clarke, R.S.J., 1979, Intravenous Anaesthetic Agents, ed. J. Dundee (Edward Arnold, London) pp. 87-118. Dauterman, W.C. and K.M. Mehrotra, 1963, J. Neurochem. 10, 113. Dev6s, R. and R.M. Krupka, 1979, Biochim. Biophys. Acta 557, 469. Dundee, J.W., 1976, Br. J. Anaesth. 48, 57. Fisher, M.M., 1975, Anaesth. Intensive Care 3, 180. Fisher, M.M., 1981, Anaesth. Intensive Care 9, 242. Fisher, M.M. and I. Munro, 1983, Anaesth. Analg. 62, 559. Harle, D.G., B.A. Baldo and M.M. Fisher, 1984, Lancet i, 930. Lim, M. and H.C. ChurchiU-Davidson, 1981, in: Adverse Reactions to Anaesthetic Drugs, Monographs in Anaesthesiology, No. 8, ed. J.W. Thornton (Elsevier, Oxford) pp. 65-136. Moneret-Vautrin, D.A., M.C. Laxenaire and R. MoeUer, 1981, Clin. Allergy 11, 175. Siraganian, R.P., W.A. Hook and B.B. Levine, 1975, Immunochemistry 12, 149. Stoetling, R.K., 1983, Anaesth. Analg. 62, 341. Vervloet, D., A. Amaud, P. Vellieux, S. Kaplanski and J. Charpin, 1979, J. Allergy Clin. Immunol. 63, 348.