Release of histamine and tryptase during continuous and interrupted cutaneous challenge with allergen in humans

Transcription

1 Release of histamine and tryptase during continuous and interrupted cutaneous challenge with allergen in humans Meir Shalit, MD,* Lawrence B. Schwartz, MD, Carolyn von AIImen, BS, Paul C. Atkins, MD, Robert M. Lavker, PhD, and Burton Zweiman, MD Philadelphia, Pa., and Richmond, Va. To help in understanding the patterns of in vivo mediator release in human allergic skin reactions, we have used a skin chamber model to challenge the denuded bases of skin blisters of 11 sensitive subjects with pollen antigens (Ags) and codeine (C), a mast cell degranulator. Challenges were performed either (1) continuously for 6 hours or (2) in an intermitten fashion that is, Ag or C for the first hour, buffer for the next 4 hours, and then Ag or C during the sixth hour. Fluids in the overlying chamber were assayed for levels of the mast cell components, histamine and tryptase. There was peak release of both histamine and tryptase during the first hour of Ag incubation (89 11 ng/ml and ng/ml, respectively). At continuous Ag-challenge sites, there was a plateau of histamine levels (8.0 to 9.5 ng/ml) during the next 4 hours, whereas tryptase levels decreased progressively to baseline levels. Challenge of continuous Ag-incubation sites with C, a mast cell activator, led to another peak release of both histamine and tryptase. At interrupted Ag-challenge sites, histamine levels decreased abruptly, and tryptase levels decreased progressively after the first hour. Rechallenge of such sites with Ag during the sixth hour induced a peak release of histamine but no increase in tryptase levels. Continuous challenge with C for up to 5 hours in other sites induced an initial peak histamine release without a subsequent plateau. However, such a plateau of histamine (but not tryptase) release occurred after an initial C challenge if Ag was subsequently incubated in a continuous fashion. Basophils were not observed in the chamber fluids or in biopsy specimens of the underlying dermis after the sixth hour of continuous Ag challenge but could be observed in small numbers relative to other leukocytes in touch preparations of the dermal surface. We conclude that the low plateau of histamine release during continuous Ag challenge is not due simply to mediator depletion and may be due to infiltrating basophils, although we have not yet convincingly found such cells in these sites at these time periods. ( J ALLRGY CLIN 1MMUNOL 1990;86: ) Ag challenge of the skin provokes an immediate wheal-and-flare response in allergic subjects. This early phase reaction may persist for hours to become From the Allergy and Immunology Section, the Department of Medicine, and the Department of Dermatology, University of Pennsylvania School of Medicine, Philadelphia, Pa., and the Department of Medicine, Medical College of Virginia, Richmond, Va. Supported in part by National Institutes of Health Grants R and P01-AR , and a grant from the Pennsylvania Allergy Association. Received for publication Feb. 23, Accepted for publication Sept. 6, Reprint requests: Burton Zweiman, MD, University of Pennsylvania School of Medicine, 512 Johnson Pavillion, Philadelphia, PA *Present address: Allergy Unit, Department of Medicine A, Hadassah Medical Center, Jerusalem, Israel. 1/1/21496 Abbreviations used PNU: Protein nitrogen unit LPR: Late-phase response Ag: Antigen B: Buffer C: Codiene an indurated LPR. (1) Few studies have attempted to assess the kinetics of allergic mediator release during the early and late response. With a skin chamber model, we have demonstrated that a single exposure to Ag resulted in release of large amounts of histamine at the first hour after challenge. 2 The levels of histamine then diminished, and no second rise was observed during the time corresponding with the LPR. In contrast, a late increase in histamine levels was 117

2 118 Shalit et al. J. ALLRGY CLIN. IMMUNOL. JULY CONTINUOUS ANTIGN [] INTRRUPTD ANTIGN _A 100 o~ c LU Z i I im I,o o,o,o :, 9 i CCr TIM(hours) FIG. 1. Histamine levels in chamber fluid from continuous (sofid bar) and interrupted (open bar) Ag challenge sites (see MATRIAL AND MTHODS for details). One site in each subject was rechallenged with C at the sixth hour. 9 Continuous Antigen 2000 ] [] Interrupted Antigen 1481 t,~n,~ c u~ 1000 I-- >- p CO[~IN TIM(hours) FIG. 2. Tryptase levels in chamber fluids from sites continuous (solid bar) and interrupted (open bar) Ag challenge sites, One site in each subject was rechallenged with C at the sixth hour. found in plasma after bronchial challenge and in nasal lavage fluid after nasal challenge with Ag. 3, 4 However, during natural exposure, the atopic subject is exposed to allergens during a period of hours and days. Therefore, we have recently studied the pattern of in vivo mediator release in human skin during continuous challenge with Ag, using the skin chamber technique. An early peak of histamine was observed followed by a rather constant low. but significant, level of histamine throughout a 9-hour period of Ag incubation. 5 A different pattern of tryptase (a human mast cell protease) release was found. The early peak of tryptase release was followed by gradually decreasing levels that were not different from control levels after the fifth hour. The present study was designed to explore possible mechanisms for the differing pattern of mediator release during prolonged exposure to Ag in the skin and to determine the source of the constant levels of histamine observed. MATRIAL AND MTHOD Subjects leven atopic healthy volunteers (six female and five male subjects, 19 to 40 years of age) with seasonal rhinitis, skin

3 VOLUM 86 Histamine and tryptase in cutaneous challenge 119 NUMBR 1 e O0 9O / 9 antigen-codeine site [] codeine site kg Z I-- U} _ TIM(hours) FIG. 3. Histamine levels in chamber fluids from sites of Ag and C challenges in five subjects. The solid bar represents the challenge sites that received Ag hourly for 4 hours and C at the fifth hour, The open bar represents sites receiving C hourly for 5 horus (see MATRIAL AND MTHODS for details) antigen-codeine 1076 [] codeine I-- Q. 556 Q: I TIM(hrs) FIG. 4. Tryptase levels in chamber fluids from sites of Ag and C challenges in five subjects. The solid bar represents the challenge sites that received Ag hourly for 4 hours and C at the fifth hour. The open bars represent sites that received C continuously for 5 hours (see MATRIAL AND MTHODS for details). sensitive to -< 10 PNU / ml of grass or ragweed-pollen extract (Greer Laboratories, Lenoir, N.C.) were selected for the study after their informed consent was obtained. None had received immunotherapy or had ingested any medication for at least 10 days before the study. Seven subjects were challenged with grass and four with ragweed-pollen extract based on their skin sensitivity. Skin chamber and Ag challenge Skin chamber studies were performed as previously described. 6 Briefly, four blisters of 1 cm diameter base, two on each forearm, were induced by gentle heat and suction. ach blister was unroofed aseptically, and 0.3 ml volume plastic collection chambers were taped in place over the denuded bases. ach one of the four blister bases in every subject was challenged with Ag by a different protocol to compare the effects of continuous versus interrupted Ag exposure on mediator release. Thus, each subject served as his/her own control. The four chambers were filled with 0.3 ml of 100 PNU/ml of pollen solution for the first hour. The fluid was removed and replaced with Ag or B solution according to the protocols below at hourly intervals for 6 hours. The protocol used in the first six subjects was a combi-

4 120 Shalit et al. J. ALLRGY CLIN. IMMUNOL. JULY O0 R~ 9 Codeine-antiflen-codeine sites I-1 Codeine-buffer-codeine sites 80 c Z i I TIM(hre) FIG. 5. Histamine levels in chamber fluids from sites of Ag and C challenges. The solid bar represents sites that received C in the first hour, Ag from the second to fourth hours, and C in the fifth hour (see MATRIAL AND MTHODS for details). The open bars represent sites that received C in the first hour, 8 from the second to fourth hours, and C in the fifth hour (see MATRIAL AND MTHODS for details). nation of continuous and intermittent Ag (100 PNU/ml) challenges through 5 hours, followed by Ag or C (10 mg/ml, ph 7.4) in the sixth hour, as illustrated below. The concentration used had been found previously by us to induce very pronounced histamine release during the first hour; this was then used as a stimulus of "maximal" mast cell histamine release. Hours Site Ag Ag Ag Ag Ag Ag 2 Ag Ag Ag Ag Ag C 3 Ag B B B B Ag 4 Ag B B B B C In five additional subjects, continuous and interrupted C (10 mg/ml) challenge of the skin chamber sites was compared to different combinations of C and Ag challenge during a 5-hour period, as illustrated below. Hours Site C Ag Ag Ag C 2 C B B B C 3 Ag Ag Ag Ag C 4 C C C C C At the end of each hour, fluids were centrifuged, and the supernatants were frozen at -20 ~ C for mediator assays. Mediator assays Histamine levels in chamber fluids were measured in duplicate by a radioenzymatic technique as previously described. 7 Because the presence of C interferes with our usual radioenzymatic assay, in all studies involving C challenge, the histamine levels were determined in the first six subjects by a fluorometric assay (kindly performed by Dr. Michael Kaliner, Allergic Diseases Laboratory, Bethesda, Md.) and in the last five subjects with a recently described radioimmunoassay. 8 Histamine in test specimens competes with an ~25I-labeled special histamine conjugate for binding to an antihistamine antibody immobilized on a test tube. Amount of 1251 binding to the tube is compared with that binding after incubation with a mixture of 125I-labeled conjugated and exogenous histamine (used in varying amount to establish a dose-response standard curve). C was found not to interfere with either the fluorometric or radioimmunoassay assay. Tryptase was measured by LISA with murine monoclonal IgG and goat polyclonal IgG antitryptase antibodies. 9 Histochemistry Smears of cells suspended in chamber fluids, imprints of blister bases, and buffy coat smears of peripheral venous blood of the subjects were allowed to dry. For the alcian blue stain, slides were heat fixed over an alcohol flame, stained wtih alcian blue (0.5%, ph <1) for 30 minutes, rinsed with distilled water, and then counterstained for 20 minutes with 0.1% safranin in 1% acetic acid./~ Staining the slides for tryptase was performed with mouse monoclonal IgG1 and kappa antitryptase antibody called 1-14 by an immunoperoxidase technique." Skin biopsies To determine cellular events within the dermis that might not be reflected in studies of chamber fluid imprints of the blister bases, we obtained 3 mm diameter biopsies of (1) the underlying dermis after a 6-hour chamber challenge with Ag or B in four subjects and (2) at sites 6 hours after intradermal Ag injection in four subjects. The biopsy specimens were fixed in Karnovsky's fixative at ph 7.3 and then embedded in pon 812 (Ladd, Inc.) Multiple adjacent 1 micron sections were cut from each of three full depth tissue alliquots of each biopsy specimen taken at random. These sections were stained with toluidine blue. The histologic

5 VOLUM 86 Histamine and tryptase in cutaneous challenge 121 NUMBR Codeine-antigen-codeine sites [] Codeine-buffer-codeine sites O} I,M r I- Q. >. tr I i I I ~2Go 2s2 ~" TIM(hrs) FIG. 6. Tryptase levels in chamber fluids from sites of Ag and C challenges. The solid bar represents sites that received C in the first hour, Ag from the second to fourth hours, and C in the fifth hour. The open bars represent sites that received C in the first hour, B from the second to fourth hours, and C in the fifth hour (see MATRIAL AND MTHODS for details). interpretation of each biopsy specimen was based on the examination of the total area of at least five sections of each of the three blocks of tissue (or at least 15 sections per biopsy). Ultrathin sections were processed for electron microsocopy in representative areas and where there was any question about the identity of cells with metachromatically stained granules. Statistical analysis The mediator findings in this subject group were found to be distributed normally (Kolmogorov-Smirnov two-sample test, adjusted for small sample size; Statworks program on a Macintosh II computer). They were then analyzed by the Student's t test for paired and unpaired values as appropriate and were expressed as mean SM. In histologic studies, the mean frequency of basophils per 1 Ixm tissue section was estimated. RSULTS The levels of histamine in chamber fluids overlying blister bases, exposed continuously to 100 PNU/ml of Ag, in the six subjects, are depicted in Fig. 1. Histamine levels peaked at the first hour (89 +_ 11 ng/ml) and reached a plateau from the second to sixth hour (8 to 12 ng/ml). When one continuous Ag challenge site (in each subject) was rechallenged with C at the sixth hour, a second peak of histamine release of 88 _+ 2 occurred, significantly higher (p < 0.001) than the hourly release of histamine during the second to fifth hour. A different pattern of tryptase release was observed at these continuous Ag challenge sites (Fig. 2). A peak level of _ 260 ng/ml was found at the end of the first hour of Ag incubation followed by gradually diminishing tryptase levels during the subsequent 5 hours of Ag challenge, as previously de- scribed. 6 However, rechallenge of these sties with C instead of Ag at the sixth hour resulted in a second peak of tryptase release of ng/ml, significantly higher (p < 0.001) than tryptase levels from the fifth and sixth hours of continuous Ag challenge (Fig. 2). To determine possible mechanisms involved in the plateau response of histamine without tryptase release during continuous Ag challenge, an interrupted challenge protocol was performed, as depicted in Figs. 1 and 2. During the first hour of Ag incubation, there was the expected peak of histamine release ( ng/ml). During the second to fifth hour, when the blister bases were incubated with B solution instead of Ag, histamine levels declined to very low levels (Fig. 1). However, a second peak of histamine release was observed at such sites when they were challenged in the sixth hour with either Ag (47 _+ 13) or C (51 8). These second peaks of histamine levels were significantly higher than the hourly histamine levels in chamber fluids obtained previously during the second to fifth hour (p < 0.001). A different temporal pattern of tryptase release was noted at interrupted challenge sites (Fig. 2). After the peak in the first hour ( ng/ml), the levels of tryptase gradually decreased during the subsequent hours in a pattern exactly the same as during continuous Ag challenge. No second peak of tryptaoe release was observed after rechallenge during the sixth hour with 100 PNU/ml of the same Ag. In contrast, a peak of 548 _+ 156 ng/ml of tryptase was observed after rechallenge with C instead of Ag in the sixth hour; this latter level was significantly (p < 0.001) larger than tryptase release during each of the previous 3 hours.

6 122 Shalit et al. j. ALLRGY CLIN. IMMUNOL. JULY 1990 FIG. 7. The superficial dermis of the base of the skin chamber after 5 hours of continuous Ag challenge. A, Marked degranulation of a representative mast cell is observed (arrow) in a 1 ~m section. (Original magnification 1000.) B, and in transmission electron microscopy. (Original magnification x8000.) C, Well-granulated mast cells were more common in the deeper dermis (arrow). Since C challenge of the prior Ag incubation sites could induce a subsequent release of both histamine and tryptase significantly more than Ag rechallenges, the release of histamine and tryptase was investigated after prolonged and interrupted C challenges of naive and previously Ag challenged skin sites in five subjects (Figs. 3 to 6). C (10 mg/ml) challenge of skin sites in the first hour induced the release of more than twice as much histamine as did Ag challenge (90 _ 17 versus 42 8; p < 0.005) challenge (Fig. 3). However, as illustrated in Figs. 3 and 5, adding either C or phosphate-buffered saline from the second to fourth hour did not induce any significant release of histamine above what is the normal spontaneous release at these skin chamber sites (2 to 7 ng/ml, personal observations in more than 100 such challenges). Most strikingly, when sites challenged with C during the first hour were then challenged with Ag starting at the second hour, a pleateau of histamine release pattern occurred from the second through the fourth hour, similar in degree to that observed when sites were challenged continuously with Ag (17 to 13 ng/ml, Fig. 5, versus 18 to 15 ng/ml, Fig. 3). These histamine levels were significantly higher than in the sites challenged continuously with C during the second to fourth hour (p < 0.025; Fig. 3). Furthermore, rechallenge with C at the fifth hour at the continuous Ag incubation site (Fig. 3) resulted in release of histamine (33 ng/ml) that was significantly higher (p < ) than C rechallenge of sites that had been previously, continuously, or intermittently exposed to C (Figs. 3 and 5, 2.4 and 8.2 ng/ml, respectively). In contrast to histamine release, the patterns of tryptase release were similar at all these sites with a progressive decline from the first to the fourth hour at continuous Ag challenge sites (Fig. 4, 860 to 40 ng/ml); this finding was similar to that occurring at sites of continuous C challenge (Fig. 4, 1076 to 156 ng/ml). However, when sites continuously challenged with Ag for 4 hours were then challenged with C in the fifth hour, a second peak of tryptase release (684 _+ 500 ng/ml) was induced (Figs. 4 and 5); the latter was significantly higher than the tryptase levels induced by the C challenge of all other sites at

7 VOLUM 86 Histamine and tryptase in cutaneous challenge 1:)3 NUMBR 1 FIG. 8. The deep dermis of the base of the skin chamber after 5 hours of continuous Ag challenge (transmission electron microscopy; original magnification x 3600). Neutrophils (IV) and eosinophils () are common; however, no basophils are observed. The typical eosinophils granules (arrow) are observed at higher magnification (insert original magnification x 4800). TABL I. Frequency of basophiles in cell preparations obtained from blister bases and from autologous peripheral venous blood Source of cells No. of subjects % Basophiles Sites of continous Ag challenge Cells suspended in chamber fluid Imprints of reaction sites Sites of interrupted Ag challenge Imprints of reaction sites* Autologous peripheral venous blood 5 0,0,0,0, ,0,0, , 2.0t 4 0.9, 1.5, 0.6, 1.0 *Only imprints were obtained because there were too few cells in these fluids to be studied. #Replicate imprints of these sites revealed no tryptase-positive cells. the fifth hour (Figs. 4 and 5; 80 to 186 ng/ml; p < 0.025). One of the possible explanations for the difference between temporal patterns of histamine and tryptase release is that basophils accumulating in the reaction site release histamine but not tryptase, because appreciable levels of tryptase are found only in mast cells. Therefore, we examined various preparations for the presence of basophils (Table I). There were very few basophils (0% to 2%) in either the cells suspended in the chamber fluids or in imprints obtained from blister bases after the fifth hour at both continuous and interrupted Ag challenge sites. These basophil frequencies were not more than in the peripheral blood of the same subjects. xamination of replicate imprints of reaction sites of interrupted Ag challenge demonstrated no tryptase-positive cells, indicating that the few alcian blue-positive cells detected were basophils and not mast cells. In light of recent studies by other investigators suggesting that basophils do not appear in overlying chamber fluids until the eighth to twelfth hour after Ag challenge, ~2 it is conceivable that the plateau of histamine release came from basophils that were still in the dennis (but not yet in the overlying blister base/chamber fluid) of our Ag challenge sites. Therefore, we performed 3 mm punch biopsies of the underlying dermis in the base of the skin chamber after 5 hours of continuous Ag stimulation in four subjects. Several consistent patterns were observed in light microscopic examination of 1 txm sections and in electron microscopy. (1) There was marked degranulation of mast cells in the superficial dermis (Fig. 7, A and B). Most of the mast cells found in the deep dennis appeared well granulated (Fig. 7, C). (2) There was a variable accumulation of neutrophils and eosinophils; however, no basophils were detected in any tissue section by either extensive light (toluidine blue)

8 124 Shalit et al. J. ALLRGY CLIN. IMMUNOL. JULY 1990 or electron microscopic studies (Fig. 8). We also examined biopsy specimens of gross LPR 6 hours after intradermal Ag in four of these study subjects. A similar pattern was observed, marked mast cell acti-, ation but no apparent Lasophils (data not presented). DISCUSSION The cutaneous LPR to Ag challenge in allergic subjects occurs in almost all individuals when the skin is challenged with relatively high concentrations of Ag This led us to hypothesize that persistent presence of Ag in the tissues is a prerequisite for the development of LPR. Gross LPR is unusual at sites of a 1-hour challenge with pollen Ag. However, with our skin chamber model, we have previously demonstrated that continuous exposure to pollen antigen induces a gross LPR accompanied by a reaction with local infiltration of activated polymorphonuclear leukocytes In addition, when the levels of preformed mediators, histamine and tryptase, were determined throughout a 9-hour period of prolonged Ag challenge, two different patterns of release were observed. During the first hour, a peak release of both mediators was noted. Thereafter, histamine was persistently released at constant low levels during the next 8 hours; in contrast, tryptase levels decreased progressively, becoming similar to levels at control sites after the fifth hour. 5 In contradistinction, no plateau of histamine release was observed if Ag was removed and replaced with B after the first hour. 5' 16 One possible explanation for the different patterns of histamine and tryptase release during continuous Ag challenge is that basophils recruited to the area from the circulation are stimulated to secrete histamine that appears in the chamber fluid at a relatively constant rate. There could be no concomitant release of tryptase, a component of mast cells but present in only minute amounts in basophils.17 Furthermore, it would be difficult to imagine a selective release of histamine from mast cells because histamine and tryptase reside in the same secretory granules. 11 The present study was designed to explore this and other possible reasons for our findings in the skin by applying various challenge protocols in allergic subjects. When blister-base sites were rechallenged with Ag 5 hours after the first hour challenge with intervening B incubation (interrupted protocol), a peak of histamine release occurred in this sixth hour while tryptase levels were not increased. It therefore is conceivable that local basophils were the origin of the histamine found in the chamber fluids. A similar conclusion was reported by Naclerio et al. 4 in nasal LPR to allergen in which a concomitant peak release of histamine without prostaglandin D2 led the authors to conclude that the cells responsible for the late histamine release were basophils. Recently, this group of authors reported increased numbers of basophils in nasal washings obtained 1 1 hours after challenge. 18 Basophils have been reported to emigrate to dermal sites of allergen challenge in atopic subjects. 12, ~9 In other studies, a selective accumulation of basophils at such sites was not observed. 2~ The low percentage of basophils found in imprints of reaction sites and in cells suspended in chamber fluids does not preclude the participation of these cells in the allergic response reported herein. The numbers of basophils detected may represent only a small part of the basophil population present at the area. Larger numbers of these cells infiltrating blister bases and the deeper dermis could have been constantly activated to release histamine while they were adhering to the surrounding tissues. Our inability to find basophils in biopsy specimens of blister bases at antigen challenge sites or intradermal LPR must cast some doubt as to their role in the histamine-release patterns we observed. Another possibility is that basophils at the surface of reaction sites had been degranulated sufficiently by Ag to obscure their detection. However, incubation of peripheral blood basophils from atopic donors in vitro with Ag alone for up to 4 hours resulted in histamine release without loss of staining (personal observation). However, the inflammatory milieu of the ongoing in vivo allergic response may not be duplicated by in vitro challenge of cell suspensions. Stimulation of human skin and rodent peritoneal mast ceils with C or morphine results in degranulation and mediator release However, human basophils are not activated by these nonimmunologic secretagogues. 23 The release of both histamine and tryptase at continuous and intermittent allergen skin sites after rechallenge with C (Figs. 1 and 2) indicates that local mast cells could still respond to a nonimmunologic stimulus and were not completely depleted of mediators by allergen challenge. However, when intermittent allergen sites were rechallenged with Ag, histamine, but not tryptase, was released, consistent with desensitization of the mast cell population. Specific desensitization has been previously demonstrated by us in our skin chamber model by use of challenges with two different Ags, 6 although mast cells stimulated by ragweed-pollen Ag for 5 hours could respond to grass but not ragweed-pollen Ag. Furthermore, it is unlikely that the absence of tryptase in these chamber fluids was due to local factors inhibiting its release or measurement because of the presence of tryptase at this same time period in the chamber fluids after C challenge. If basophils are playing a role in the sustained pattern of histamine release we observed after Ag rechallenge of Ag or C sites, they would have to be present at these sites as early as the second hour.

9 VOLUM 86 Histamine and tryptase in cutaneous challenge 125 NUMBR 1 Furthermore, they would be drawn to these sites after either C- or Ag-induced mast cell activation, since Ag rechallenge of C-, as well as of Ag-activated sites, produced virtually the same plateau of histamine release in the absence of tryptase. Only prolonged Ag, not C, is capable of producing a sustained release of histamine. We have no readily apparent explanation for this finding other than basophil influx, but two possibilities should be considered. First, perhaps Ag can permeate further through the skin than C and activate deeper mast cells that had not been activated by prior Ag challenge. The deep mast cells would release both histamine and tryptase. However, histamine is more free to diffuse (because of the size of these molecules or the affinity of tryptase for mast cell granular enzyme products, such as heparin). Therefore, only histamine arrives in our chamber fluid from these distant sites. However, there are no data to suggest that permeation of Ag is better than the smaller sized C, nor does it appear likely that diffusion of tryptase should be completely blocked. A second possibility is a local depletion of tryptase but not of histamine. However, C stimulation releases a pool of histamine and tryptase from mast cells that is not released after Ag activation of such cells, as well as a pool that is released by Ag (Figs, 2, 4, and 6). These C challenges convincingly demonstrate that mediator depletion alone cannot explain the plateau pattern of the release of histamine without tryptase observed at these sites. In summary, our findings of the different patterns of histamine and tryptase release most likely reflect a role for basophils in the appearance of histamine in the late cutaneous allergic-inflammatory response. Prevention of basophil influx and modulation of their reactions may be an effective approach to the control of chronic manifestations of allergic disease in the skin. RFRNCS 1. Dolovich J, Hargreave F, Chalmers R, Shier KJ, Gauldie J, Bienenstock J. Late cutaneous allergic responses in isolated Ig-dependent reactions. J ALLRGY CLIN IMMONOL 1975;52: Atkins PC, Miragliotta G, Talbot SF, Zweiman B, Kaplan AP. Activation of plasma Hageman factor and kallikrein in ongoing allergic reactions in the skin. J Immunol 1987;139: Durham SR, Lee TH, Cromwell O, et al. Immunologic studies in allergen-induced late-phase asthmatic reactions. J ALLRGY CLIN IMMUNOL 1984;74: Naclerio RM, Proud D, Togias AG, et al. Inflammatory mediators in late antigen-induced rhinitis. N ngl J Med 1985;313: Shalit M, Schwartz LB, Golzar N, et al. Release of histamine and tryptase in vivo after prolonged cutaneous challenge with allergen in humans. J Immunol 1988;141: Talbot SF, Atldns PC, Goetzl J, Zweiman B. Accumulation of LTC4 and histamine in human allergic skin reaction. J Clin Invest 1985;76: Atkins PC, Valenzano M, Zweiman B. Plasma concentrations of histamine measured by radioenzymatic assay: effects of histamine incubations. J ALLRGY CLIN IMMUNOL 1982;69: Morel AM, Delaage MA. Immunoanalysis of histamine through a novel chemical derivation. J ALLRGY CLIN IMMUNOL 1988; Wenzel S, Irani AA, Sanders JM, Bradford TR, Schwartz LB. Immunoassay of tryptase from human mast cells. J Immunol Methods 1986;86: Irani AA, Schecter NM, Craig S, DeBlois G, Schwartz LB. Two human mast cell subsets with different neutral protease compositions. Proc Natl Acad Sci USA 1986;83: Craig SS, Schechter NM, Schwartz LB. Ultrastructural analyses of human T and TC mast cells identified by immunoelectron microscopy. Lab Invest 1988;58: Charlesworth N, Kagey-Sobotka A, Norman PS, Lichtenstein LM. ffects of cetirizine on mast cell mediator release and cellular traffic during the cutaneous late phase response [Abstract]. J ALLRGY CLIN IMMUNOL 1987;80: Umemoto L, Poothvill J, Dolovich J, Hargreave F. Factors which influence late cutaneous allergic responses. J ALLRGY CLIN IMMUNOL 1976;58: Talbot SF, Atkins PC, Valenzano M, Zweiman B. Correlation of in vivo mediator release with late cutaneous allergic responses in humans. J ALLRGY CLIN IMMUNOL 1984;74: Shalit M, Campbell D, von Allmen C, Atkins PC, Douglas SD, Zweiman B. Neutrophil activation in human inflammatory skin reactions. J ALLRGY CLIN IMMUNOL 1987;80: Schwartz LB, Atkins PC, Bradfored TR, Fleekop P, Shalit M, Zweiman B. Release of tryptase together with histamine during the immediate cutaneous response to allergen. J ALLRGY CLIN IMMUNOL 1987;80: Castells MC, Irani AA, Schwartz LB. valuation of human peripheral blood leukocytes for mast cell tryptase. J Immunol 1987;138: Bascom R, Wachs M, Naclerio RM, Pipkorn U, Galli SJ, Lichtenstein LM. Basophil influx occurs after nasal allergen challenge: effects of topical corticosteroid treatment. J AL- LRGY CLIN IMMUNOL 1988;81: Solley GO, Gleich GJ, Jordon R, Schroeter AL. The late phase of the immediate wheal and flare skin reaction. Its dependence upon Ig antibodies. J Clin Invest 1976;58: deshazo RD, Levinson AI, Dvorak HF, Davis R. The latephase skin reaction: evidence for activation of the coagulation system in an Ig-dependent reaction in man. J Immunol 1979;122: Casale TB, Bowman S, Kaliner M. Induction of human cutaneous mast cell degranulation by opiates and endogenous opioid peptides: evidence for opiate and monopiate receptor participation. J ALLRGY CLIN IMMUNOL 1984;73: Benyon RC, Lowman MA, Church MK. Human skin mast cells: thier dispersion, purification, and secretory characterization. J Immunol 1987;138: Tharp MD, Kagey-Sobotka A, Fox CC, Marone G, Lichtenstein LM, Sullivan TJ, Functional heterogeneity of human mast cells from different anatomic sites: in vitro response to morphine sulfate. J ALLRGY CLIN IMMLINOL 1987;79: