Cutaneous immediate hypersensitivity in man: Effects of systemically administered adrenergic drugs Jerrold Howard Sax

A. Kram, M.D., Henry R. Bourne, M.D.,* I. Maibach, M.D., and Kenneth 1. Melmon,

Francisco,

M.D.

Calif.

in vitro models of immediate hypersensitivity, beta adrenergic drugs inhibit the antigen-induced release of histamine and other mediators from mast cells and basophils. Epinephrine, an agent with both beta and alpha adrenergic properties, is clinically useful for treating immediate hypersensitivity reactions. We examined the eflects of intravenody administered adrenergic drugs on cutaneous rvheal-and-flare reactions to antigens and histamine in 7 normal men. Both epinephrine and isoproterenol, a betn agonist, inhibited shin reactions produced by antigen or histamine. Phenylephrine, an alpha adrenergic agonist, produced no efect. These results suggest that epinephrine’s inhibition of cutaneous immediate hypersensitivity in vivo is mediated by beta, rather than alpha, receptors. This inhibitory effect in viva is probably not mediated solely by preventing allergic release of inflammatory mediators since the skin reactions to histamine were inhibited as well. In

Epinephrine is widely used in the acute treatment of immediate hypersensitivity reactions. However, epinephrine’s mechanism of action in these disorders is not known. Recent in vitro experiments, using antigen-sensitized lung fragmentslm3 and leukocytes from allergic human subjects,4s5 have shown that several adrenergic drugs, including epinephrine, can prevent the IgE-mediated release of histamine and other mediators of hypersensitivity. The effects of epinephrine and its chemical congeners on allergic mediator release can be separated pharmacologically into 2 classes:beta adrenergic agents, such as isoproterenol, inhibit mediator release; alpha adrenergic agents, such as phenylephrine, have no effect on histamine release from basophils” and, in fact, increase antigen-triggered release of mediators from lung fragments.” In the antigen-sensitized lung, as in many other systems, epinephrine itself can exert both beta and alpha effects. Epinephrine alone, like isoproterenol, inhibits release of mediators ; when propranolol is present, however, epinephrine augments release.le3 The beta adrenergic effects of epinephrine and its congeners are mediated by inFrom the Departments of Medicine and Dermatology, the Division of Clinical Pharmacology and the Cardiovascular Research Institute, University of California, San Francisco. Supported by Grants Nos. GM 16496, HL 15851, GM 00001, GM 01791, and FDA Contract No. 74-46. Received for publication Aug. 26, 1974. Reprint requests to: Henry R. Bourne, M.D., Division of Clinical Pharmacology, 1089-M University of California Medical Center, San Francisco, Calif. 94143. “Established Investigator of the American Heart Association. Vol.

56, No.

5, pp.

387-398

388

Kram

et al.

J, ALLERGY

CLIN. IMMUNOI. NOVEMBER 1975

increased production of cyclic adenosinc 3’,5’-monop2losphate I cAMI ) in l(w. Itocytes” and in mast cell~‘-~~; alpha adrenergic &e&s may bc rclatetl to a tii.crease in cellular cABII’.i-~’ Because of the potential clinical implications of these in vitro experiments, KC’ ha\-e investigated the effects of 3 clinically used aclrenergic agents, administeiw! intravenously, on cutaneous immediate hypersensitivity in man. \Ye ask(d whether epinephrine can reproducibly suppress the wheal-and-flare responses caused by intradermal administration of specific antigens and a putative allergic mediator such as histamine, and whether a predominantly alpha adrcnergic agent, phenylephrine, and a predominantly beta active agent, isoproterenol, would prw duce the same effects as cpinephrine, which possesses “mixed” alpha and beta adrenergic activity. METHODS The subjects used in this study (see below) had previously documented cutaneous whealand-flare responses to intradermal antigen and histamine. They were placed in the prone position. Electrocardiographic leads and an arm blood pressure cuff were attached; a slow intrii. venous infusion of 0.9% saline was started. When pulse rate and blood pressure were stable (usually after 15 min), the subjects received the first of 6 pairs of intradermal injections of antigen and histamine, at separate sites on the back. Injection sites and drug administration were determined according to the following protocol: The back, from the first thoracic vertebra to the third lumbar vertebra, was divided into were placed to the left of the midthirds. The first injections, made during saline infusion, line in the upper third, antigen medially (about 5 cm from the midline) and histamine laterally (8-9 cm inside the posterior axillary line). The injection sites were photographed (see below I 6 min after antigen and histamine injection. Ten nun later the first drug was begun intravenously; rate of infusion was regulated over a lo-15 min period to produce a stable change in heart rate and blood pressure (which were measured every 2 min). The antigen and histamine injections were repeated on the right side of the upper third of the back, symmetrically placed with regard to the first injections (antigen medial, histamine lateral). Drug infusion r,ontinued for an additional 6 min, at which time the injection sites were photographed and the drug infusion stopped. After pulse and l~lood pressure had returned to pretreatment values and were stable for 10 min (usually 43 min after stopping the drug), the same sequence was repeated on the middle third of the back, again with a repeat saline infusion and intradermal injections on the left side, followed by infusion of the second drug and intradermal injections on t,he right. Finally, the sequence was repeated a third time on the lower back; saline and the third drug were used. The duration of the study for each subject was about 5 hr. The order of drug infusions in each subject was randomized, and intradermal injections during saline infusion were repeated before each drug, because of a previous observations that cutaneous reactivity to intradermal antigen and histamine may vary according to the level of injection on the back. In addition, the randomization procedure should minimize possible “ carryover” effects of previous injections and drug infusions on subsequent cutaneous responses.

Subjects Seven healthy men, between the ages of 21 and 35 yr, n’ere chosen because each demonstrated a. n-heal-and-flare reaction to intradermal injection of Trichophyton or mixed grass antigen, as well as histamine. The subjects had no hypertension, asthma, or other cardiorespiratory disease as determined by history, physical examination, and electrocardiogram, and were taking no medications at the time of the study. All subjects gave written, informed consent to the experimental protocol, which was approved by the Committee on Human Experimentation of the Academic Senate of the University of California, San Francisco.

VOLUME NUMBER

56 5

TABLE

I. Cardiovascular

Cutaneous

responses

Drua Infusion rate (&min) Pulse change (bpm) Blood pressure change (mm Hg) Systolic Diastolic *All values indicated

Drug

to adrenergic I

immediate

hypersensitivity

389

drugs*

Iao~roterenol

I

EWmohrine

1

Phenvleohrine

2.03 f 0.20 f31 * 3.1

12.1 f 1.0 +15 i 1.5

124 f. 26 -29 f 2.6

+25 -17

f38 f3

f 3.4 h1.9

+26 f +19 f

in

An increase

3~ 6.9 f 6.3

are expressed as the mean + SEM by “+,” and a decrease by “-.”

of observations

7 subjects.

5.4 3.6 is

infusions

Drugs were diluted in 0.9% saline at the following concentrations (expressed as the salt) : epinephrine HCl (Parke, Davis), 20 pg/ml; isoproterenol HCl (Winthrop), 4 pg/ml; phenylephrine HCl (Winthrop), 400 fig/ml. Drug infusions were regulated by a Harvard infusion pump at a rate aiming, on the basis of a previous study, 7 to produce stable rises of 40 mm Hg in systolic blood pressure. This blood pressure was not always achieved, however, because we arbitrarily limited the infusion rates to prevent tachycardia greater than 125 bpm or bradycardia of less than 35 bpm.

lntradermal

iniections

and assessment

of wheal-and-flare

reaction

Histamine phosphate, 8.25 pg, Trichophyton antigen (kindly provided by Dr. Nigel Cruickshank), 3 gg, and mixed grasses antigen (1: 10,000 dilution of antigen, University of California, San Francisco, Pediatric Allergy Clinic) were injected intradermally in a volume of 30 pl, with the use of a calibrated Agla microsyringe (Burroughs Wellcome). Five subjects received Trichophyton and 2 received mixed grasses antigen. The resulting cutaneous wheal-and-flare reaction was photographed on Kodachrome II film with a fixed-distance camera. Coded slides were projected and areas of wheal and flare were measured separately with a compensating planimeter (K & E Model No. 4236). These measurements were performed by a technician who did not know the code.

RESULTS

The 3 adrenergic agents produced distinctly different patterns of cardiovascular response (Table I), in accordance with their relative beta and alpha adrenergic activities: isoproterenol caused an increase in heart rate and a widened pulse pressure, while phenylephrine stimulated a rise in both systolic and diastolic blood pressure and lowered the pulse rate ; epinephrine produced a rise in systolic pressure and pulse rate. These effects, and the drug infusion rates required to produce them, were quantitatively similar to those observed in a previous study of the same drugs in healthy young men.7 As expected from previous studies,6 the magnitude of cutaneous wheal-andflare reactions to intradermal antigen and histamine varied considerably among the subjects. In the face of this variability, we assessedthe effects of drug infusions (Fig. 1) by plotting the size of skin reactions during infusion of a particular drug, measured on the y axis, against the reaction sizes during the corresponding saline infusion, measured on the x axis. Thus, clustering of points on either side of the line of identity (dashed lines in Fig, 1) would indicate a drug-induced change in skin reactivity: An inhibitory effect would be reflected by a shift of points below and to the right of the line, and a shift in the opposite direction would result from an enhancing action of the drug.

390

Kram

J. ALLERGY

et al.

PHENYLEPHRINE

ISOPROTERENOL 0.4

0.8

CLIN. IMMUNOL. NOVEMBER 1975

1.2

1.6

2.0

0.4

0.8

1.2

1.6

0.4

2.0

0.8

1 2

1.6

2.0

D /‘O p < (0) / = 0.125 (01 . I 0.“‘” .‘o . a40 0.031

1’

IO

,

/

,’

08

0 l

.

0

/

10

20

30

50

10

REACTION Scale

FIG. 1. The effect injection of antigen during infusion of and phenyiephrine responding value either side of the reactivity. P values above and below

20

DURING

km?) :+--l-b

40

30

SALINE

50

INFUSION

2

WHEAL

.

50

FLARE

0

of drugs on the size of the wheal-and-flare reaction after intradermai (upper panels) and histamine (lower panels). The size of reaction drugs (isoproterenol in panels A and D, epinephrine in panels 8 and E, in panels C and F), measured on the y axis, is plotted against the corduring saline infusion, measured on the x axis. Clustering of points on line of identity (dashed line] indicates a drug-induced change in skin were calculated from the sign test comparing the number of values the line of identity.ls

The overall pattern of effects on skin reactions, plotted in this way, was generally consistent, although changes did not achieve statistical significance in all cases (see legend, Fig. 1). The two agents with beta adrenergic activity, isoproterenol and epinephrine, reduced the size of skin reactions (Fig. 1, 11, B, D, and E). Inhibition of the flare reaction appeared greater with epinephrine than with isoproterenol. Phenylephrine, a predominantly alpha adrenergic amine, had no inhibitory effect on the size of skin reactions (Fig. 1, C and P), and may even have produced a modest increase in the size of the cutaneous flare response to histamine. Most strikingly, each of the drugs altered cutaneous responses to both antigen and histamine in the same direction and to comparable extents (compare Fig. 1, A-C with Fig. 1, respectively). DISCUSSION

D-F,

Beta adrenergic drugs stimulate synthesis of cAMP and inhibit the IgEmediated release of allergic mediators from human leukocyte@ 5 and lung fragments1-3; alpha-adrenergic agents may produce the opposite effect.le3 Can these

VOLUME NUMBER

56 5

Cutaneous

immediate

hypersensitivity

391

in vitro observations be used to reliably predict and explain the effects of the samedrugs on in vivo cutaneous immediate hypersensitivity reactions, which are also mediated by IgE and involve release of vasoactive substancesfrom cutaneous mast cells? This hypothesis has not been extensively studied. Yamamoto, Greaves, and Fairley8 made the important observation that epinephrine inhibits the IgEmediated release of histamine from sensitized human skin slices incubated in vitro. The inhibitory effect of epinephrine was blocked by the beta adrenergic antagonist, propranolol, and potentiated by theophylline, a drug that blocks degradation of intracellular CAMP in many tissues. Inhibition of histamine release from the skin slices by a synthetic derivative of CAMP also supported the suggestions that mediator release from mast cells in human skin is inhibited by beta adrenergic amines working through intracellular CAMP. A few in vivo studies have suggested that beta adrenergic agents can reduce cutaneous immediate hypersensitivity reactions (wheal and flare). Pretreatment of forearm skin by iontophoretic application of isoproterenol reduced the diameter of wheal responses to allergens; similar pretreat,ment with propranolol increased the size of allergen-induced wheals.” In 1951, sublingually administered isoproterenol was reportedlo to inhibit cutaneous reactivity to antigens in man. Orally administered salbutamol, a recently developed beta adrenergic drug, inhibited the immediate skin reaction produced by Ascaris antigen in monkeys.ll Recently, several drugs related to the CAMP system, including isoproterenol, have been shown to decreaseallergic reactions in the rat.iz Our results confirm and extend these observations, with regard to the wheal and flare induced by antigens (Fig. 1). At doses that produced cardiovascular responses,both agents with beta adrenergic activity, isoproterenol and epinephrine, inhibited skin reactivity, while phenylephrine, the alpha adrenergic agent, did not. Thus, both adrenergic inhibition of histamine release in vitro and skin reactivity to antgen in vivo can be roughly classified as beta adrenergic responses.These results are consistent with the hypothesis that prevention of mediator release is at least in part responsible for the inhibition of skin reactivity by drugs like epinephrine. If adrenergic drugs acted only by increasing CAMP in cutaneous mast cells, thus preventing mediator release, they should not also affect skin reactivity to histamine, a putative mediator of immediate hypersensitivity reactions. In our experiments, however, the beta adrenergic drugs reduced histamine-induced skin reactions in addition to those induced by antigens, and phenylephrine did not (Fig. 1). In previous experiments, isoproterenol (sublingually) lo and salbutamol (orally)11 did not affect skin reactivity to histamine. A possible reason for the different results of our experiments is that intravenous administration of drugs may have produced greater adrenergic effects; direct comparison is difficult, since the earlier studies did not report cardiovascular responses. In any case, intravenously administered isoproterenol and epinephrine, at dosesthat produce prominent cardiovascular effects, probably affect skin reactivity by one or more mechanisms in addition to inhibition of mediator release from mast cells; e.g., by altering vascular tone.

392

Kram

et al.

J. ALLERGY

CLIN. IMMUNOL. NOVEMBER 1975

In summary, we have found that in vitro observations’-‘~ h of the efYects 01 :i adrenergic drugs on IgE-mediated histamine release reliably predicted the offcct~ of the same drugs in viva on cutaneous immediate hyperscnsiti~ity reactions. Thv mechanism observed in vitro, howcvcr, vannot, entirely explain the inhibition 01 skin reactivity, because the same drugs also inhibitccl skin lesions incluccd by histamine. The results suggest that opinephrine, a drug clinically usetl to trout cutaneous and generalized immediate hypersensitivity reactioq wts 1)~ rnort* than one mechanism. Every beta adrenergic response so far tested is mediated by intracellular ~AILYW.~” Beta-, but not alpha-, adrenergic drugs reduw skin I’(‘act,ivity, which suggests that epinephriuc’s therapeutic effectiveness iu vutaneow allergic reactions is largely mediated by CAMP, although not exclusively by pwventing mediator release from mast cells. We wish to acknowledge California Medical Facility out, the studies.

the invaluable assistance given by Mr. George in Vacaville, California, in screening the patients

HubaId, of thi, and in (harrying

REFERENCES 1 Orange, R. P., Kaliner, M. A., La Xaia, I’. ant? slow reacting substance of anaphylaxis from human lung. I. Modulation by agents influenc ing cellular levels of cyclic 3’,5’-adenosine monophosphate, J. Exp. Med. 134 (Suppl.1 : 1367, 1971. 3 Kaliner, M., Orange, R. P., and Austen, K. F.: Immunological rclcase of histamine anti slow reaating substance of anaphylaxis from human lung. IV. Enhancement by cholinergic and alpha adrenergic stimulation, J. Exp. Med. 136: 556, 1972. 4 Liehtenstein, L. M., and Margolis, S.: Histamine release in citro: Inhibition by catecholamines and methylxanthines, Science 161: 902, 1968. 5 Bourne, H. R., Lichtenstein, L. M., and Melmon, K. L.: Pharmacologic control of allergic histamine release in wit+o: Evidence for an inhibitory role of 3’,5’ adenosine monophosphatr in human leukocytes, J. Immunol. 108: 695, 1972. 6 Galant, S. P., and Maibach, H. I.: Reproducibility of allergy epicutaneous test technique?, J. ALLERGI' CLI~~\‘.~MIIMUNOI..51: 245, 1973. 7 Mueller, P. S., and Horwitz, D.: Plasma free fatty acid and blood glucose responses to analogues of norepinephrine in man, J. Lipid Res. 3: 257, 1962. 8 Yamamoto, S., Greaves, M. W., and Fairley, V. M.: Cyclic AMP-induced inhibition of TgEmediated hypersensitivity in human skin, Immunology 24: 77, 1973. 9 Rhereff’, R. H., Harwell, W., Lieberman, P., Rosenberg, E. W., and Robinson, H.: Effect of beta atlrenergic stimulation and blockade on immediate hypersensitivity skin test reactions, J. ALLERGY CLIN. TMIMUNOL. 52: 325, 1973. 10 Sheldon, .J. M., Husted, J. R., a.nd Lovell, R. G.: Effects of inuprrl on antigen-:tntil)od?and histamine skin reaction, Ann. Allergy 9: 45, 1951. 11 Perprr, R. J., Sanda, M., and Lichtenstein, 1~. M.: The relationship of in t,itro and in L%LV~ allergic histamine release: Tnhibition in primates try cAMP active agents, rut. Arch. Allergy 43: 837, 1972. 12 Taylor, W. A., Frances, D. H., Sheldon, I),, and Roth, I. M.: The anti-anaphylactic actions of disodium cromoglycate, theophylline, isoprenaline and prostaglandins, Int. Arch. Allergy Appl. Immunol. 46: 104, 1974. 13 Robison, G. A., Butcher, R. W., and Sutherland, E. W.: Cyclic AMP, New York, 1971, Academic Press, Inc., pp. 224-226. 14 Noether, G.: Introduction to statistics-a first approach, Boston, 1971, Houghton Mifflin co., p. 113.

Cutaneous immediate hypersensitivity in man: effects of systemically administered adrenergic drugs.

Cutaneous immediate hypersensitivity in man: Effects of systemically administered adrenergic drugs Jerrold Howard Sax A. Kram, M.D., Henry R. Bourne,...
503KB Sizes 0 Downloads 0 Views