Antigen-induced activity in man



Correlation with bronchospasm disodium cromoglycate

and inhibition

Paul C. Atkins, M.D., Michael F.A.C.P. Philadelphia, Pa.

E. Norman,

M.D., and Burton


Zweiman, M.D.,

We have previously reported increased neutrophil chemotactic activity in sera obtained after positive antigen inhalation responses in atopic subjects. This report describes the kinetics of appearance of this serum activity and the effects of antigen dose and disodium cromoglycate pretreatment on the response in IO ragweed-sensitive subjects. Signijicantly increased chemotactic activity was present as early as I min, peaked at 10 min, and persisted through 24 hr after inhalation of antigen. The increased chemotactic activio correlated with the degree of bronchospasm induced by antigen inhalation and the amount of antigen administered. The increased chemotactic activity and bronchospasm were blocked by administration of disodium cromoglycate prior to antigen challenge. These findings are consistent with a postulated antigen-induced anaphylactic release of chemotactic activity. The correlation of this activity with the degree of bronchospasm and its appearance after administration of even small doses of antigen suggest that this activity may be important in antigen-mediated bronchospasm.

The pathophysiology of bronchial asthma is thought to include spasm of the smooth muscle lining the tracheobronchial tree as well as edema and marked cellular infiltration of the bronchial wall.’ While there have been numerous studies regarding the nature and degree of smooth muscle spasm in asthma, the mechanisms of the cellular accumulation have not been well defined. Previous work by us2 as well as others3 has documented the presence of neutrophils in the skin early in the course of immediate hypersensitivity reactions in humans. In attempting to link these observations with bronchial asthma, we have described the appearance of increased serum neutrophi1 chemotactic activity occurring during the induction of IgE-mediated bronchospasm in humans.4 The present report characterizes this chemotactic activity

From the Allergy and Immunology Section and the Department of Pediatrics, University of Pennsylvania School of Medicine. Supportedby Grants NIH 5 MOlRROO40, ROI AI 14332,and Al 07031 from the National Institutes of Health and Biomedical ResearchInstitutional Grant No. .5-SO7-054I5- IS. Received for publication March 13, 1978. Accepted for publication May 19, 1978. Reprint requeststo: Paul C. Atkins, M.D., Allergy and Immunology Section, Department of Medicine, University of Pennsylvania, 510 JohnsonPavilion, 36th & Hamilton Walk, Philadelphia, Pa. 19174. 0091-6749/7810362-0149$00.70/0

0 1978 The C. V. Mosby


in relation to: ( 1) the kinetics of its appearance; (2) the dosage of antigen administered; (3) the degree of bronchospasm produced; and (4) the effects of prior administration of disodium cromoglycate.



and inhalation


Ten subjects were selected after informed consent was obtained. All were skin test-reactive to ragweed extract* (I ,000 PNU/ml) and had demonstrated positive bronchial challenges after inhalation of ragweed extract* (from IOO10,000 PNU/ml), defined as 20% decrease of I-set forced expiratory volume (FEV,). None of the subjects were smokers, receiving medication or immunotherapy, or symptomatic from asthma for at least 48 hr from the time of study. In 6 of the IO subjects with positive challenges, ragweed inhalation was repeated after administration of a total of 100 mg disodium cmmoglycate,? inhaled in five divided doses, 24 hr before and 30 min immediately preceding the second inhalation challenge. All inhalation challenges were performed using a dosimeter,$ with FEV, measured by spirometer (Survey spirometer, Warren E. Collins, Inc.. Braintree, Mass.), as previously described.“ *Greer Laboratories, Lenoir, N. C. tInta Rx, Fisons Corporation, Bedford, Mass. *Provided by Richard Rosenthal.M.D., JohnsHopkins University School of Medicine. Vol. 62, No. 3, pp. 149-155





and Zweiman









AFTER (minutes






CHALLENGE or hours)

FIG. 1. Serum neutrophil chemotactic activity (mean and standard error of the mean) from I,$ min to 48 hr after inhalation of a dose of ragweed pollen causing a greaterthan 20% decrease in FEV, in 10 ragweed-sensitive subjects. *p < 0.01 compared to time 0.




Blood was collected immediately before and IO min after inhalation of the challenge dose (unless otherwise noted). Serum was harvested and frozen immediately (-70” C). The chemotactic activity for neutruphilic leukocytes was assayed in semm specimens previously heated at 56” C as described by US,~using a modified Boyden chamber assay and 3-p Millipore filters (Millipore Corporation, Bedford, Mass.). In brief, 2.5 x IO6neutmphils suspended in 0.5 ml buffer (medium - 199*) with 10% heat-inactivated fetal calf serum? were layered on top of the filter and the bottom compartment was charged with 0.8 ml of solution to be tested. ?he chambers were then incubated for 3 hr at 37” C, filters removed, stained with hematoxylin and eosin, coded, and counted as reported below. A minor modification of the counting method was made after pilot studies showed that it lead to more sensitive and reproducible readings. The distance of furthest cell penetration into the filter (a), estimated by an ocular micrometer, and the density of cells per high-power field at that distant layer (b), were multiplied to determine a chemotactic index (axb). As shown in Table I, this chemotactic index (axb) was compared with distance of cell penetration into the filter (a) and density of cells/HPF (b). It is clear that although both distance and cell density increased with increasing *Microbiological Associates, Bethesda, Md. Kirand Island Biological Company, Grand Island, N. Y.

concentration of stimuli, the rates of increases were different. The product, chemotactic index, correlated more significantly with the dose than either component individually. In addition, comparison of the chemotactic index to density of cells/HPF in Table II showed similar reproducibility for buffer, but the index was more reproducible for normal sera and bacterial filtrate than the density of cells alone. For these reasons, the chemotactic index of 5 highpower fields in each of two coded, duplicate filters was averaged and used as the expression of chemotactic response in our studies. All experiments included: (I) paired patient sera obtained before and after each inhalation challenge; (2) bacterial filtrate-a positive chemotactic control consisting of broth filtrate of Eschen’chia coli*$ and (3) buffer (medium-199 with 10% heat-inactivated fetal calf sera).



All analyses were performed using the Monroe calculator.? The mean chemotactic indices were compared using Student’s t test.5 Values before and after disodium cromoglycate were compared using Student’s t test for paired values.5 Correlation coefficients were determined by linear regression analysis.5 *Harry Morton, Ph.D., Professorof Bacteriology, William Pepper Laboratory, Hospital University of Pennsylvania. tMonroe Calculator Model No. 1860, Litton Co., Haverford, Pa.



62 3







LOG DOSE OF ANTIGEN RESULTING IN A POSITIVE CHALLENGE FIG. 2. Serum neutrophil chemotactic activity (mean and standard error of the mean) is plotted (solid line) with % decrease in FEV, (dashed line], mean, and standard error of the mean on the ordinates. The dose of ragweed inhaled is plotted on the abscissa as logarithmic decrements from that dose, eliciting a positive challenge (loo) in 10 ragweed-sensitive subjects.* p < 0.01 compared to baseline values (before antigen administration) + positive challenge defined by >20% decrease in FEV,.

RESULTS Kinetics of appearance neutrophil chemotactic



of increased activity


The chemotactic activity in serum specimens obtained from 10 subjects prior to positive ragweed inhalation challenge, from ‘/z to 60 min after challenge, and 24 and 48 hr after challenge is shown in Fig. 1. Significantly increased neutrophil chemotactic activity compared to chemotactic activity before challenge (I ,529 t 167) was present as early as I min (3,252 t 323). peaked at 10 min (6,445 -t- 1,169), persisted for at least 24 hr (3,100 r+ 447), and returned to baseline by 48 hr after inhalation of ragweed antigen (I,5 11 ? 153). The increase in chemotactic activity at 24 hr was observed in 8 of our IO subjects. Relationship of serum neutrophil chemotactic activity and bronchospasm the dose of antigen inhaled


Serial measurements of FEVr and serum chemotactic activity were assessed 10 min after inhalation of each of a series of increasing doses of ragweed antigen. The relationship between the amount of antigen administered, degree of bronchospasm, and appearance of increased neutrophil chemotactic activity in these IO ragweed-reactive subjects is shown in Fig. 2. The minimal antigen dose inducing a decrease in



FtG. 3. Correlation in 10 subjects of the mean percent increase in serum neutrophil chemotactic activity (ordinate) with mean percent decrease in FEV, (abscissa) after each of two concentrations of antigens inhaled (100 and 10-l).

152 Atkins, Norman, and Zweiman





TABLE I. Correlation doses of chemotactic

of quantitation stimuli*

of chemotaxis

by three different




with increasing

Dose of stimulus

Bacterial filtrate





Chemotactic index?

























Cells /HPF Distance?penetration Normal sera

Chemotactic index?


Cells /HPF







100% 2916












*Meanof 13experiments. t(Mean numberof cells/HPF) X (distancepenetrationinto the filter in microns) for 10HPFin 2 duplicatefilters. Meannumberof cells/ HPFfor 10HPFin 2 duplicatefilters. $Meanmicron penetration from cell surface layer for 10 HPFin 2 duplicatefilters. TABLE II. Quantitation different methods*

of chemotaxis

Chemotactic stimulus


Chemotactic index

21.7 r 3.0t

2,173 !Z 112


(5%) 1,441 + 106 (7%)

Bacterial filtrate Normal sera

9.0 k 1.3 (14.4%)


3.6 + 0.3


by two

422 5 44 (10%)

*Twentyexperiments. t Meannumberof cells/HPF f standarderrorof themean.(Mean numberof cells/HPF) x (meandistancepenetrationin microns)t standarderrorof the mean. $% variationin parentheses. FEVl of 220% is denoted as IO0 concentration. At this dose the mean neutrophil chemotactic index (solid line) was 3,597 t 330 and the mean percent fall in FEVl (dashed line) was 35 2 5, both significantly different from baseline values (p < 0.00 1). At concentrations of one-tenth the positive challenge dose (lo-‘), both the chemotactic index (2,313 + 223) and the percent fall in FEVl (10 ? I .6) were also significantly different from baseline values (p < 0.05). It should be emphasized that this dose induced no respiratory symptoms and that the mean percent decrease in FEV, observed was less than the 20% reduction commonly accepted as a meaningful change after inhalation challenge.6 In addition, serum chemotactic activity returned to baseline values within 20 min after the 10-l antigen dose unlike the sustained increase observed after the IO0 antigen dose

(Fig. 1). Finally, at concentrations of one-hundreth the positive challenge ( 1OU2),neither the chemotactic index (1,453 & 191) nor the percent decrease in FEVl (2.7 & 1.2) was significently different from baseline values. To further define the relationship between changes in airway resistance and serum chemotactic activity after antigen inhalation, the relative changes in FEV, and chemotactic activity were compared at 10 min for each antigen dose inhalation in the 10 subjects, by linear regression analysis (Fig. 3). A significant correlation between mean percent decrease FEVl and mean percent increase chemotactic activity at 10 min after challenge was observed (r = 0.79, p < 0.001). Effect of prior administration of disodium cromoglycate upon antigen-induced neutrophil chemotactic activity Neutrophil chemotactic activity before and after the administration of disodium cromoglycate was measured in an attempt to further define the relationship between antigen administration and the appearance of increased neutrophil chemotactic activity. Six of the 10 subjects underwent a second challenge at least I wk following the initial challenge and after the administration of disodium cromoglycate. A total of 100 mg disodium cromoglycate (in 5 divided doses) was inhaled over 24 hr. The last dose was administered 30 min prior to the inhalation of antigen. As shown in Fig. 4, A, there was no significant increase in the neutrophil chemotactic activity after inhalation of 10-l antigen dose compared to baseline values when disodium cromoglycate was administered (solid line,


Neutrophil chemotactic activity

62 3

Correlation coefficient

P value











p value









1,990 ? 103 vs 1,757 + 114). In contrast, there was significantly increased chemotactic activity compared to baseline values when the same subjects inhaled the same amount of antigen ( 10-l) without prior disodium cromoglycate administration (dashed line, 2,689 -+ 216 vs 1,731 + 195, p < 0.05). Furthermore, the serum chemotactic activity after inhalation of 10-i antigen dose was significantly lower after disodium cromoglycate administration than in the absence of this pretreatment ( 1,990 + 103 vs 2,689 * 2 16, p < 0.05). At IO0 antigen dose, with and without disodium cromoglycate administration, significantly increased chemotactic activity was observed compared to baseline values (p < 0.01). Although the absolute chemotactic activities were not significantly different from each other (4,004 + 375 vs 2,689 ? 216, p > 0.05), they were lower after drug administration. As shown in Fig. 4, B, a significant mean percent decrease in FEVr occurred after administration of both IO-’ (9.3 + 4.6) and IO0 (32 rt 6) antigen doses in these 6 subjects without any pretreatment compared to baseline values (dashed line, p > 0.01). However, after pretreatment with disodium cromoglycate (solid line), no significant decrease in FEV, was observed at either 10-i (5.6 + 2.9) or IO0 (12.4 2 5) antigen doses in these same subjects. DISCUSSION We have investigated the changes in serum neutrophi1 chemotactic activity after antigen inhalation challenge. We had previously shown that this increased activity occurred only after antigen-induced bronchospasm, was non-complement-derived, and contained in serum fractions greater than 50,000 daltons.4 The results reported here further strengthen the association


of this activity with antigen-induced anaphylactic mediator release and bronchospasm. The timing of appearance of increased chemotactic activity, its dependence upon the amount of antigen inhaled, and its decrease after disodium cromoglycate administration are consistent with antigen-induced, anaphylactic origin of this activity. Studies in humans have demonstrated a rise in plasma histamine after positive inhalation challenge occurring within 5 min, peaking at 15 min, and disappearing in 30 min after antigen administration. Other studies after cold challenge of subjects with cold urticaria demonstrated release of histamine and eosinophil and neutrophil chemotactic factors in venous blood draining the sites of these reactions.** g All three of these mediators were detectable at % to 2 min, peaked at 5 min, and were no longer present 30 min after the challenge. The differences in timing of appearance of neutrophil chemotactic activities between cold- and antigenchallenged subjects are slight, but may reflect the difference between collection locally (regional arm vein) versus systemically (venous blood after pulmonary inhalation) or the difference between physical (cold) vs antigen-induced anaphylactic challenge. Despite these differences, the appearance of chemotactic activity within I min after inhalation of antigen correlates well with these previous studies. Perhaps the most striking observations reported here are in regard to the significant differences in chemotactic activity occurring before and after pretreatment with disodium cromoglycate administration. While the exact mode of action of this drug is unknown, studies in humans suggest that it interferes with the degradation of intracellular cyclic 3’,5’adenosine monophosphate (AMP) in the target cell (mast cell or circulating basophil) by inhibition of the action of a specific phosphodiesterase. lo Elevated levels of cyclic 3’,5’-AMP are known to inhibit mediator release.” Studies in animal models suggest that it may interfere with mediator release by impeding calcium influx into the cell.‘* Very little disodium cromoglycate is systemically absorbed, less than 10%. I2 Therefore, the reduction of antigen-induced chemotactic activity after pretreatment with this drug is likely due to inhibition of activation of mast cells lining the tracheobronchial tree, and provides the strongest evidence to date of the anaphylactic origin of this chemotactic activity. There are several factors that could explain the failure of disodium cromoglycate to significantly block chemotactic activity at IO0 antigen dose. Of necessity, the drug was administered about I hr before inhalation of 10’ antigen compared to 30 min




and Zweiman



4500 r









+ ,







2o A&
















FIG. 4. Serum neutrophil chemotactic activity (mean and standard error of the mean) (A) and % decrease in FEV, (mean and standard error of the mean) (BJ after inhalation of increasing doses of ragweed in 6 subjects exhibiting positive challenges at loo antigen dose, before (dashed/he) and after (solid line) disodium cromoglycate administration. *, p < 0.05, difference from loo antigen dose. t, p < 0.05, difference between, before, and after disodium cromoglycate.

prior to the 10-l antigen dose. In addition, a 12% decrease of FEVi occurred after IO0 antigen despite disodium cromoglycate administration. Thus, it appears that the drug decreased the antigen sensitivity about 1 log since the values for both FEVi and chemotactic activity at loo were similar to responses seen in these same subjects at 10-l antigen dose without drug administration. In addition to an early appearance of neutrophil chemotactic activity, we have detected its presence for up to 24 hr after the inhalation of ragweed antigen. We cannot determine from our studies whether this results from the continued release of chemotactic activity, its persistence in the circulation, or recruitment of other mediators by an ongoing immediate hypersensitivity reaction. Further elucidation of this phenomenon awaits further characterization of this neutrophil chemotactic activity. However, the persistence of this chemotactic activity should be viewed in the context of recent descriptions of late-onset bronchospastic and skin reactions after ragweed administration. These late-onset asthmatic responses occur in the more highly sensitive subjects13 while the late-phase skin responses14occur with administration of large antigen doses. These responses usually begin at 4 hr, peak at 8 to 12 hr, but

have been noted to persist for 24 to 48 hr in some individuals. Interestingly, in the late-phase skin test responses, over 20% of the infiltrating cells appear to be neutrophilic leukocytes. l5 These observations in the skin may correlate with the persistence of increased neutrophil chemotactic activity for 24 hr that we have observed only at the highest antigen doses ( IO0 not IO-‘). However, we could make no correlation with late-onset asthmatic responses since only 2 of our IO subjects complained of pulmonary symptoms persisting for 12 to 24 hr and none had any decrease in FEVr at 24 hr. We must also comment upon the relationship of these phenomena to seasonal asthma. First, it must be noted that increased chemotactic activity appeared at doses involving small decreases ( 10%) in FEV,, when subjects were asymptomatic. We have recently demonstrated” that seasonal asthma may be a subclinical phenomenon which could be contributed to in part by increased chemotactic activity occurring in association with subtle decreases in FEV,. Second, the direct correlation of this activity with degree of bronchospasm also encourages further observations during seasonal asthma. Finally, the amount of antigen delivered to the lungs in our studies based upon the calculations of Wilson16 and Rosenthal17 and their



62 3

colleagues would appear to be in the range of 0. IO to 1 10 protein nitrogen units (PNU) of ragweed extract for the IO0 antigen dose in our 10 subjects. The release of chemotactic activity occurring after deposition of between 0.01 to I1 PNU of ragweed antigen (IO-’ antigen dose) is clearly within the realm of in vivo seasonal exposure. In conclusion, we have observed an early appearance and persistence of serum neutrophil chemotactic activity which was related to the amount of antigen inhaled. The amount of chemotactic activity could be reduced by prior disodium cromoglycate administration, suggesting an anaphylactic origin. The correlation of this activity with the amount of bronchospasm, and the appearance of this activity after the inhalation of subthreshold doses of antigen, suggest that investigations of this activity during in vivo seasonal exposure would be fruitful. The presence and persistence of this activity in the serum after antigen inhalation suggest possible pathogenetic roles particularly with reference to the inflammatory cell component of the observed bronchial responses. REFERENCES I. Terr, A. I.: Bronchial asthma, in Baum, G. L., editor: Textbook of pulmonary diseases, ed. 2, Boston, 1974, Little, Brown & Co., p. 423. 2. Atkins, P. C., Green, G. R., and Zweiman, B.: Histologic studies of human skin test responses to ragweed, compound 48180 and histamine, J. ALLERGY CLIN. IMMUNOL. 51:263, 1973. 3. Kline, B. S., Cohen, M. B., and Rudolph, S. A.: Histologic changes in allergic and nonallergic wheals, J. ALLERGY 3:53 I, 1932. 4. Atkins, P. C., Norman M., Weiner, H., et al.: Release of neutrophil chemotactic activity during immediate hypersensitivity reactions in humans, Ann. Intern. Med. 86~415, 1977. 5. Kilpatrick, S. J.: Statistical principles in health care informa-






I I.

12. 13.








tion, Baltimore, 1973, University Park Press, p, I55- 158; pp. 164-167;~~. 168-172. Chai, H., Fari-, R. S., Yroehhch, L. A., et al.: Standardization of bronchial inhalation challenge procedures, J. ALLERGY CLIN. IMMUNOL. 56323, 1975. Bhat, K. N., Arroyave, C. M., Mamey, S. R., et al.: Plasma histamine changes during provoked bronchospasm in asthmatic patients, J. ALLERGY CLIN. IMMUNOL. S&647, 1976. Sorer, N. A., Wasserman, S. I., and Austen, K. F.: Cold urticaria: Release into the circulation of histamine and eosinophil chemotactic factor of anaphylaxis during cold challenge, N. Engl. J. Med. 294:687, 1976. Wasserman, S. I., Soter, N. A., Center, D. M., et al.: Cold urticaria: Recognition and characterization of a neutrophil chemotactic factor which appears in serum during experimental cold challenge, J. Chn. Invest. 60: 189, 1977. Lavin. N., Rachelefsky, G. S., and Kaplan, S. A.: An action of disodium cromoglycate: Inhibition of cyclic 3’,5’-AMP phosphodiesterase, J. ALLERGY CLIN. IMMUNOL. 57~80, 1976. Kaliner, M., and Austen, K. F.: Immunologic release of chemical mediators from human tissues, Ann. Rev. Pharmacol. 15:177-189, 1975. Intal (cromolyn sodium). A monograph, Bedford, Mass., 1973, Fisons Corporation, p. 42; p, 54. Robertson, D. G., Kerigan, A. T., Hargreave, F. E., et al.: Late asthmatic responses induced by ragweed pollen allergen, J. ALLERGY CLIN. IMMUNOL. 54:244, 1974. Solley, G. O., Gleich, G. J., Jordan, R. E., et al.: The late phase of the immediate wheal and flare skin reactions: Its dependence upon IgE antibodies, J. Clin. Invest. 58:408, 1976. Weiner, H., Atkins, P. C., Altose, M., et al.: Ragweed rhinitis and asthma defined by history, pulmonary function tests, and inhalation challenge, J. ALLERGY CLIN. IMMUNOL. 57x219, 1976. (Abst.) Wilson, A. F., Novey, H. S., Berke, R. A., et al.: Deposition of inhaled pollen and pollen extract in human airways, N. Engl. J. Med. 288: 1056, 1973. Rosenthal, R. R., Bruce, C. A., Lichtenstein, L. M., et al.: The role of inhalation challenge, Int. Arch. Allergy Appl. Immunol. 49:89. 1975.

Antigen-induced neutrophil chemotactic activity in man. Correlation with bronchospasm and inhibition by disodium cromoglycate.

Antigen-induced activity in man neutrophil chemotactic Correlation with bronchospasm disodium cromoglycate and inhibition Paul C. Atkins, M.D., M...
609KB Sizes 0 Downloads 0 Views