- Email: [email protected]
Late allergic bronchial reactions and the effect of allergen provocation on the complement system
Late allergic bronchial reactions and the effect of allergen provocation on the complement system Gunnemar St~lenheim, M.D., and Lavinia Machado, M.D. Uppsa/a, Sweden
Forty-nine bronchial provocations were performed with different allergens in allergic asthmatic patients. Immediate bronchial reactions occurred in 33 instances, and late reactions occurred in nine instances. The material was divided into three groups according to patients with no reactions, patients with early reactions alone, and patients with late reactions. Titers of whole complement and factors C2 and C3 were determined hemolytically, and the serum protein contents of factors C3, C4, and Cl q were measured immunochemically. Decreases in complement and complement components were observed on a number of occasions after bronchial challenge. When the groups were compared with respect to mean change in complement titers, the three groups were indistinguishable. No change was noted in the ratio between the complement contents in venous and arterial blood as a result of provocation, nor did the specific hemolytic activity of factor C3 change. No C3 split products could be traced. It is likely that there will be a higher proportion of patients with C2 consumption among those with late bronchial reactions. No special feature could be assigned to these individuals, however. It is concluded that no major complement activation is regularly found in allergic asthmatic patients as a result of exposure to allergens. However, groups of patients in whom such activation takes place might exist. (J ALLERGYCLIN IMMUNOL 75:508-12, 1985.)
The significance of late bronchial reactions to allergen provocation has been discussed for a number of years. 1.2 Clinical asthma is more like the late reaction than the early one, and the occurrence of a late reaction is better correlated to the severity of asthma than is that of an early reaction. 2' 3 Neither the late reaction nor the asthmatic attack is simply the result of histamine action, but the responsible mediators have not yet been well-defined. There are reasons to believe that the late allergic reactions, in the bronchi as well in the skin, result from the effects of a concert of mediators, most probably including SRS-A leukotrienes as important members. Another effector system, the complement system, has been incriminated From the Department of Lung Medicine, Uppsala University Hospital, and Department of Medical and PhysiologicalChemistry, Uppsala University, Uppsala, Sweden. Supported by grants from the Swedish Medical Research Council, the Swedish National Association against Heart and Chest Diseases, the Research Fund of the Swedish Medical Society, and the Swedish Association against Asthma and Allergy. Reprint requests: Gunnemar St~lenheim, Dr., Dept. of Lung Medicine, Uppsala University Hospital, Fack S-751 85 Uppsala, Sweden. Received for publication March 2, 1984. Accepted for publication Aug. 28, 1984.
508
Abbreviations used
PEFR: CH~o: SRID: IEP:
Peak expiratory flow rate 50% Hemolysis complement unit Single radial immunodiffusion Immunoelectrophoresis
by some authors, 4' 5 whereas other authors have found no support for this idea. 68 This argument thus appears to be still unsettled. We have observed, however, that products from an activated complement system may elicit rise to late allergic skin reactions, 9 and we are therefore trying to elucidate the possible role of the complement system in asthma and particularly in late bronchial reactions.
MATERIAL AND METHODS Patients and provocation The patients, 47 in all, were referred to the allergy section for investigation because of suspected allergic asthma. The provocations were performed as part of this investigation. Criteria for inclusion in the study were a positive RAST and a positive skin test. Forty-nine provocations were done. The mean age of the patients was 33.9 yr (range 17 to 56 yr). At the time of testing, the patients had no respiratory
VOLUME75
Late bronchial reactions and complement 509
NUMBER 4
60
O0 ;4 40
0
20
,_ 9 20
o 40
q. 60
80
15 M i n FIG. 1. Decrements in PEFR in percent of initial values 15 min and 6 hr after allergen challenge. TABLE I. Early and late reactions after bronchial p r o v o c a t i o n s w i t h different allergens
No. Positive early reactions Positive late reactions
Mites
Dust
Animal
Pollen
17 15 4
18 11 3
9 3 1
5 3 1
symptoms, and they had not used any asthma or allergy medication for 12 hr and no cromoglycate or corticosteroids for 1 wk. Allergens used in the routine tests of the clinic were used in the provocations. Seventeen tests were performed with Dermatophagoides pteronyssinus (house dust mite, ALK; Allergologisk Laboratorium, Copenhagen, Denmark), 18 with house dust (Dome Laboratories, Slough, England), four with cat dander (ALK), four with dog dander (ALK), two with birch pollen (Spectralgen, Pharmacia Diagnostics AB, Uppsala, Sweden), one with rat dander (ALK), and three with timothy pollen (Spectralgen) extracts. If the patient had positive RAST or skin tests for more than one allergen, the allergen most suspected of causing asthmatic symptoms, as judged from the history, was used for the provocation. Two patients were tested with two different allergens. In these cases more than 1 wk was allowed to elapse between the tests. For the provocation a Pari Inhaler Boy (Paul Ritzan Pari Werkka Gmbh, Starnberg am See, West Germany) was used. The patient breathed in the allergen extract in a mask continuously until 1 ml was consumed. The test started with a blank inhalation of Coca's solution and was followed by inhalations of increasing concentrations of the allergen. The initial dose chosen was the dose judged as safe from the clinical standpoint. There was
an interval of 15 min between the different doses, and the test was stopped if the PEFR had decreased by 20% of the initial value, if the patient had definite breathing discomfort, or if wheezing was heard on lung auscultation. At most three (or in some cases four) doses of allergen were tested. In all tests PEFR was measured before the first inhalation, immediately and 15 min after each inhalation, and 6 hr after the last one. In 25 cases an extensive lung function test was undertaken,before the first inhalation and 15 min and 6 hr after the last inhalation.'~ Venous blood was drawn before and 15 min and 6 hr after the inhalation, and in 23 cases blood samples were also drawn from the femoral artery on these occasions. The blood was centrifuged, and the serum was removed within 1 hr and stored at - 7 0 ~ C until it was tested. The study was approved by the Committee for Ethics of the medical faculty, Uppsala, and informed consent was obtained from each patient before inclusion to the study.
Complement determinations CH~o,'1 C2 titer, ~ and C3 titer '3 were measured, and IEP 14 was performed according to established methods. C3 and C4 were measured by SRID by use of Partigen (Behringwerke AG, Marburg-Lahn, West Germany) and with the use of the procedure suggested by the manufacturers. C l q
510 St~llenheim and Machado
J. ALLERGYCLIN.IMMUNOL. APRIL 1985
TABLE II. Complement levels 15 minutes after allergen provocations in percent of prechallenge values • SD
No reactions Only early reactions Late reactions
No.
CHso
C2
C3
16 24 9
105.1 + 17.1 107.7 • 16.7 113.5 --- 30.4
108.3 • 27.2 101.4 • 30.2 93.1 • 40.4
100.1 • 29.0 102.9 • 20.0 108.0 • 29.4
TABLE III. Complement levels 6 hours after allergen provocations in percent of prechallenge values ___SD
No reactions Only early reactions Late reactions
No.
CHso
C2
C3
16 24 9
104.4 - 19.6 106.4 • 11.2 120.2 • 35.7
101.9 • 26.6 107.9 • 20.8 108.4 • 61.3
105.3 • 23.1 98.9 • 16.1 101.1 • 20.2
protein was measured by SRID according to the method described by Mancini et al.'5 SD The within-the-day method error ( x 100) was 1% mean for the CH~0, 3% for the C2, 1.9% for the C3 titrations, 2.2% for the C3, and 2.6% for the C4 protein determinations. All samples from the same provocation were always run simultaneously. RESULTS
Thirty-two of the forty-nine provocations resulted in a positive immediate reaction with a decrease in PEFR of 15% or more. Eight of the patients also had a late bronchial reaction (Table I). In one case a late reaction occurred without being preceded by a detectable early reaction. The mean decrement in PEFR was 52% (_+23.7) for the early reactions that were followed by late ones and was 41.4% ( _+ 16.2) for the isolated early reactions. The coefficient of correlation between changes in PEFR in early and late reactions was 0.22 (Fig. 1). The mean value for CHso was unchanged both at 15 min and at 6 hr in the groups with early reactions alone and late reactions as well as in the group without any bronchial reactions (Tables 1I and II1). Of the patients who reacted with complement changes, an approximately equal number demonstrated increases as well as decreases in this activity. The same pattern was observed for C2 and C3 titers as well as for protein concentrations of C3, C4, and C l q that was measured with SRID. To check if degradation products were present in the blood after provocation, two analyses were performed. First, the blood was assayed for "converted" C3 by means of IEP, but no such conversion was found in any of the cases. Second, the ratio between hemolytically active C3 and C3 protein was calculated. Measured as a mean value, this ratio did not change in any patient group as a
result of provocation, but in a few individual patients the ratio changed after challenge. A decrease exceeding 15% after 15 min or 6 hr was observed in four (25%) patients without bronchial reactions, in eight (33%) patients with only an early reaction, and in five (56%) patients with a late reaction. An attempt was made to determine if complement was consumed during the passage of blood through the lungs. Thus the complement titers and complement protein content were measured simultaneously in venous and arterial blood. The ratios between venous and arterial blood before, 15 min after, and 6 hr after provocation were calculated. The mean changes in all of the components measured (Tables IV and V) were small in comparison with the variation as expressed by the standard deviation. Three of the four patients without bronchial reactions, seven of the 14 with only an early reaction, and none of the five with a late reaction demonstrated decreases in this arteriovenous ratio for CHs0 of more than 15%. The question arises if it is the patients with the strongest bronchial reactions that get complement alterations. Among the six patients with a decrease in PEFR exceeding 60% after 15 min, only one had a CHs0, C2, or C3 value that was more than 15% lower after the provocation than before. In none of these six patients was the ratio of hemolytic C3 to C3 protein below 85% of the initial value after the provocation. Five patients had more than 25% lower PEFR 6 hr after the provocation than before. None had lower CHso values 15 min or 6 hr after the provocation than before. Three had C2 values that were more than 25% lower than the initial value both after 15 min and after 6 hr. Changes exceeding 15% were most often observed for C2. Of 16 patients in whom no bronchial obstruction developed, three (19%) demonstrated a decrease in C2 after 15 min and two (13%) after 6
VOLUME75
Late bronchial reactions and complement 511
NUMBER 4
TABLE IV. The ratio between complement titers in arterial and venous blood drawn 15 minutes after provocation expressed in percent of the prechallenge ratio • SD
No reactions Only early reactions Late reactions
No.
CH~,
C2
C3
4 14 5
82.3 -+ 37.6 94.0 -+ 15.8 97.3 • 7.6
82.5 • 18.3 97.1 • 17.6 112.8 • 9.8
122.0 _+ 27.3 108.3 • 37.1 105.8 • 25.1
TABLE V. The ratio between complement titers in arterial and venous blood drawn 6 hours after
provocation expressed in percent of the prechallenge ratio • SD
No reactions Only early reactions Late reactions
No.
CH~
C2
C3
4 14 5
103.0 _+ 37.6 104.0 • 22.0 93.0 • 5.6
94.8 • 36.8 103.2 • 22.0 116.5 • 5.1
91.3 +_ 43.9 108.1 • 17.9 91.3 • 22.2
hr. Of the 24 patients with isolated early bronchial reactions, seven (29%) had low values of C2 at 15 min and one (4%) at 6 hr. In five (56%) of the nine patients reacting with late bronchial obstruction, there was a decrease in C2 of more than 15% after 15 min, and in four (44%) the C2 values were low after 6 hr. A correlation coefficient of 0.20 was found between decrements in C2 after 15 min and in PEFR after 6 hr when the values of all 49 patients were considered. Other correlations between complement changes and bronchial reactions were lower. DISCUSSION Late allergic reactions were originally considered to take place if the allergic patient had IgG antibodies against the allergen, ~ but it has since been demonstrated beyond doubt that IgE can mediate late reactions that can be elicited by any ordinary allergen provided a sufficiently large dose is administered. J6-1~ The late bronchial reaction resembles the asthmatic attack more closely than does the early one with respect to clinical features, physiologic characteristics, and effects of pharmaceutical agents.'-' 3 Histamine does not produce late reactions or states of the same kind as asthmatic attacks. The late reaction possesses more the characteristics of an inflammatory reaction in the airways~9 than bronchoconstriction produced only by smooth muscle. Mediators to be looked for should thus have the potential to produce inflammation, and probably a number of mediators act together in the production of the late reaction. The complement system is a system with an ability to produce many of the components in the inflammatory process. On activation of the complement, anaphylatoxins are generated. These recruit leukocytes to the site of activation and trigger these cells to release their granular content, including histamine.
The anaphylatoxins also induce vasodilatation, increased vascular permeability, and contraction of smooth muscles. Purified anaphylatoxin injected into the skin is very efficient in eliciting an urticarial reaction.Z~ Only about 20 ng of C3a is needed to produce a wheal-and-flare reaction. Some years ago we were able to demonstrate that anaphylatoxin could also elicit late cutaneous reactions. 9 Therefore, if complement is activated at a bronchial provocation, it is also possible that anaphylatoxin will take part in the development of the bronchial reaction. Arroyave et al. 4 titrated complement before and after provocation of allergic asthmatic patients and patients with hypersensitivity to acetylsalicylic acid. They observed signs of complement consumption in three of seven atopic individuals and in five of seven aspirin-intolerant persons. In the former group this consumption took place within 30 min and in the second group between one and 4 hr after the challenge. Hsieh 5 studied 20 asthmatic children, three of whom demonstrated reductions in serum C4 and/or factor B after allergen provocation. Two of these patients had precipitating antibodies, and one had a late bronchial reaction. Hutchcroft and G u z , 6 o n the other hand, found no complement decrease in their patients during allergeninduced asthma. They also studied arterial blood samples from two patients without finding any signs of complement activation. Baur et al. 8 found a complement decrease in two of their 16 patients during the late reaction. Pryjma et al. 7 noted a reproducible fall in serum complement after allergen provocation in some children, but there was no correlation between this reduction and the appearance of late bronchial reactions. Among our 32 patients with early bronchial reactions, eight (25%) had late reactions. Late reactions occurred with all four groups of allergens used (Table
512
St,~lenheim and Machado
J. ALLERGYCLIN. IMMUNOL. APRIL 1985
I). Changes in complement titers were observed in a few patients, but there was no difference between the patients who did not have any bronchial reaction, those who developed only an early reaction, and those with a late reaction. The mean postchallenge titers of whole complement and complement components, in relation to the prechallenge ones, were very close to 100% both 15 min and 6 hr after the test in the three groups of patients. The variation was fairly wide. Since complement components react as acute phase proteins and elevations of the titers were frequently observed after the provocations, an attempt was made to determine whether complement consumption or activation took place in the pulmonary circulation. For this purpose the change in the arteriovenous ratio for whole complement components was measured, but no significant changes in these ratios were observed. Another possible way of ascertaining whether a component may be present or not in the circulation in the inactivated form is to calculate the ratio between the hemolytic activity and the protein content of the component. This was done for C3, but that ratio did not change as a result of provocation in any of the groups of patients. It is concluded that bronchial provocation with allergens does not result in significant complement inactivation, not even if the acute phase character of complement is taken into consideration. This does not necessarily rule out the possibility, however, that complement may be involved in the production of bronchial reactions, early or late. There might be subgroups of patients with allergic asthma in whom complement consumption takes place as a result o f exposure to an allergen. Thus, it appears as if significant consumption is somewhat more common in the group o f patients with late bronchial reactions. It does not appear as if this is dependent on the type of allergen used. As yet there is no clue as to what distinguishes patients with complement consumption from other patients. The very high biologic potency of anaphylatoxins has to be kept in m i n d ? ~ It may be estimated very roughly that activation of much less than 1% o f the total amount of circulating C3 and C5 would be sufficient to produce enough anaphylatoxins to cause significant bronchial obstruction. This extent of complement activation is far too small to be detected by the methods used in this study. Methods are being developed to approach this problem from another angle. We thank Ms. ,~sa B~ickstrrm and Ms. Monica Ferm for excellent technical assistance and registered nurses Barbro Ekblad and Eva Persson for assistance with provocation tests.
REFERENCES
1. Pepys J: Hypersensitivity diseases of the lungs due to fungi and organic dusts. Monogr Allergy 4:19, 1969 2. Warner JO: Significance of late reactions after bronchial challenge with house dust mite. Arch Dis Child 51:905, 1976 3. Price JF, Hey EN, Soothill JF: Provocation to the skin, nose, and lung in children with asthma: immediate and dual hypersensitivity reactions. Clin Exp Immunol 47:587, 1982 4. Arroyave CM, Stevenson DD, Vaughan JH, Tan EM: Plasma complement changes during bronchospasm provoked in asthmatic patients. Clin Allergy 7:173, 1977 5. Hsieh K-H: Changes of serum complement and eosinophil count in antigen-induced bronchospasm. Ann Allergy 41:182, 1978 6. Hutchcroft BJ, Guz A: Levels of complement components during allergen-induced asthma. Clin Allergy 8:59, 1978 7. Pryjma JR, MiklaszewskaJ, Haluszka J, Scislicki A: Decrease of complement hemolytic activity after an allergen-housedustbronchial provocation test. J ALLERGYCLINIMMUNOL70:306, 1982 8. Baur X, Dorsch W, Becher T: Levels of complement factors in human serum during immediate and late asthmatic reactions and during acute hypersensitivitypneumonitis. Allergy 35:383, 1980 9. SthlenheimG, Zetterstrrm O: Late cutaneous allergic reactions without the participation of IgE. Monogr Allergy 14:264, 1979 10. Machado L, SthlenheimG, Malmberg P: Early and late allergic bronchial reactions. Physiologicalcharacteristics. Clin Allergy (accepted for publication) 11. Mayer MM: Complement and complement fixation. In Kabat EA, Mayer MM, editors: Experimental Immunochemistry. Springfield, I11., 1967, Charles C Thomas, Publisher, p 133 12. Cooper NR, Polley MJ, Miiller-EberhardHJ: The second component of human complement (C2): quantitative molecular analysis of its reactions in immune hemolysis. Immunochemistry 7:341, 1970 13. Miiller-Eberhard HJ, Nilsson UR, Dalmasso AP, Polley MJ, Calcott MA: A molecular concept of immune cytolysis. Arch Pathol 82:205, 1966 14. St~lenheim G, Castensson S: Protein A from Staphylococcus aureus. Conversion of complement factor C3 by aggregates between IgG and protein A. FEBS Len 14:79, 1971 15. Mancini G, Carbonara AO, Heremans JF: Immunochemical quantitation of antigens by single radial immunodiffusion, lmmunochemistry 2:235, 1965 16. Solley GO, Gleich GJ, Jordan RE, Schroeter AL: The late phase of the immediate wheal-and-flare skin reaction. Its dependence upon IgE antibodies. J Clin Invest 58:408, 1976 17. Umemoto L, Poothullil J, Dolovich J, Hargreave FE: Factors which influence late cutaneous allergic responses. J ALLERGY CLIN IMMUNOL58:60, 1976
18. Hargreave FE, Dolovich J, Robertson DG, Kerigan AT: The late asthmatic responses. Can Med Assoc J 110:415, 1974 19. Cartier A, Thomson NC, Frith PA, Roberts R, Tech M, Hargreave FE: Allergen-induced increase in bronchial responsiveness to histamine: relationship to the late asthmatic response and change in airway caliber. J ALLERGYCLIN IMMUNOL 70:170, 1982 20. Wuepper KD, Bokisch VA, Miiller-Eberhard HJ, Stoughton RB: Cutaneous responses to human C3 anaphylatoxin in man. Clin Exp Immunol 11:13, 1972
Forty-nine bronchial provocations were performed with different allergens in allergic asthmatic patients. Immediate bronchial reactions occurred in 33 instances, and late reactions occurred in nine instances. The material was divided into three groups according to patients with no reactions, patients with early reactions alone, and patients with late reactions. Titers of whole complement and factors C2 and C3 were determined hemolytically, and the serum protein contents of factors C3, C4, and Cl q were measured immunochemically. Decreases in complement and complement components were observed on a number of occasions after bronchial challenge. When the groups were compared with respect to mean change in complement titers, the three groups were indistinguishable. No change was noted in the ratio between the complement contents in venous and arterial blood as a result of provocation, nor did the specific hemolytic activity of factor C3 change. No C3 split products could be traced. It is likely that there will be a higher proportion of patients with C2 consumption among those with late bronchial reactions. No special feature could be assigned to these individuals, however. It is concluded that no major complement activation is regularly found in allergic asthmatic patients as a result of exposure to allergens. However, groups of patients in whom such activation takes place might exist. (J ALLERGYCLIN IMMUNOL 75:508-12, 1985.)
The significance of late bronchial reactions to allergen provocation has been discussed for a number of years. 1.2 Clinical asthma is more like the late reaction than the early one, and the occurrence of a late reaction is better correlated to the severity of asthma than is that of an early reaction. 2' 3 Neither the late reaction nor the asthmatic attack is simply the result of histamine action, but the responsible mediators have not yet been well-defined. There are reasons to believe that the late allergic reactions, in the bronchi as well in the skin, result from the effects of a concert of mediators, most probably including SRS-A leukotrienes as important members. Another effector system, the complement system, has been incriminated From the Department of Lung Medicine, Uppsala University Hospital, and Department of Medical and PhysiologicalChemistry, Uppsala University, Uppsala, Sweden. Supported by grants from the Swedish Medical Research Council, the Swedish National Association against Heart and Chest Diseases, the Research Fund of the Swedish Medical Society, and the Swedish Association against Asthma and Allergy. Reprint requests: Gunnemar St~lenheim, Dr., Dept. of Lung Medicine, Uppsala University Hospital, Fack S-751 85 Uppsala, Sweden. Received for publication March 2, 1984. Accepted for publication Aug. 28, 1984.
508
Abbreviations used
PEFR: CH~o: SRID: IEP:
Peak expiratory flow rate 50% Hemolysis complement unit Single radial immunodiffusion Immunoelectrophoresis
by some authors, 4' 5 whereas other authors have found no support for this idea. 68 This argument thus appears to be still unsettled. We have observed, however, that products from an activated complement system may elicit rise to late allergic skin reactions, 9 and we are therefore trying to elucidate the possible role of the complement system in asthma and particularly in late bronchial reactions.
MATERIAL AND METHODS Patients and provocation The patients, 47 in all, were referred to the allergy section for investigation because of suspected allergic asthma. The provocations were performed as part of this investigation. Criteria for inclusion in the study were a positive RAST and a positive skin test. Forty-nine provocations were done. The mean age of the patients was 33.9 yr (range 17 to 56 yr). At the time of testing, the patients had no respiratory
VOLUME75
Late bronchial reactions and complement 509
NUMBER 4
60
O0 ;4 40
0
20
,_ 9 20
o 40
q. 60
80
15 M i n FIG. 1. Decrements in PEFR in percent of initial values 15 min and 6 hr after allergen challenge. TABLE I. Early and late reactions after bronchial p r o v o c a t i o n s w i t h different allergens
No. Positive early reactions Positive late reactions
Mites
Dust
Animal
Pollen
17 15 4
18 11 3
9 3 1
5 3 1
symptoms, and they had not used any asthma or allergy medication for 12 hr and no cromoglycate or corticosteroids for 1 wk. Allergens used in the routine tests of the clinic were used in the provocations. Seventeen tests were performed with Dermatophagoides pteronyssinus (house dust mite, ALK; Allergologisk Laboratorium, Copenhagen, Denmark), 18 with house dust (Dome Laboratories, Slough, England), four with cat dander (ALK), four with dog dander (ALK), two with birch pollen (Spectralgen, Pharmacia Diagnostics AB, Uppsala, Sweden), one with rat dander (ALK), and three with timothy pollen (Spectralgen) extracts. If the patient had positive RAST or skin tests for more than one allergen, the allergen most suspected of causing asthmatic symptoms, as judged from the history, was used for the provocation. Two patients were tested with two different allergens. In these cases more than 1 wk was allowed to elapse between the tests. For the provocation a Pari Inhaler Boy (Paul Ritzan Pari Werkka Gmbh, Starnberg am See, West Germany) was used. The patient breathed in the allergen extract in a mask continuously until 1 ml was consumed. The test started with a blank inhalation of Coca's solution and was followed by inhalations of increasing concentrations of the allergen. The initial dose chosen was the dose judged as safe from the clinical standpoint. There was
an interval of 15 min between the different doses, and the test was stopped if the PEFR had decreased by 20% of the initial value, if the patient had definite breathing discomfort, or if wheezing was heard on lung auscultation. At most three (or in some cases four) doses of allergen were tested. In all tests PEFR was measured before the first inhalation, immediately and 15 min after each inhalation, and 6 hr after the last one. In 25 cases an extensive lung function test was undertaken,before the first inhalation and 15 min and 6 hr after the last inhalation.'~ Venous blood was drawn before and 15 min and 6 hr after the inhalation, and in 23 cases blood samples were also drawn from the femoral artery on these occasions. The blood was centrifuged, and the serum was removed within 1 hr and stored at - 7 0 ~ C until it was tested. The study was approved by the Committee for Ethics of the medical faculty, Uppsala, and informed consent was obtained from each patient before inclusion to the study.
Complement determinations CH~o,'1 C2 titer, ~ and C3 titer '3 were measured, and IEP 14 was performed according to established methods. C3 and C4 were measured by SRID by use of Partigen (Behringwerke AG, Marburg-Lahn, West Germany) and with the use of the procedure suggested by the manufacturers. C l q
510 St~llenheim and Machado
J. ALLERGYCLIN.IMMUNOL. APRIL 1985
TABLE II. Complement levels 15 minutes after allergen provocations in percent of prechallenge values • SD
No reactions Only early reactions Late reactions
No.
CHso
C2
C3
16 24 9
105.1 + 17.1 107.7 • 16.7 113.5 --- 30.4
108.3 • 27.2 101.4 • 30.2 93.1 • 40.4
100.1 • 29.0 102.9 • 20.0 108.0 • 29.4
TABLE III. Complement levels 6 hours after allergen provocations in percent of prechallenge values ___SD
No reactions Only early reactions Late reactions
No.
CHso
C2
C3
16 24 9
104.4 - 19.6 106.4 • 11.2 120.2 • 35.7
101.9 • 26.6 107.9 • 20.8 108.4 • 61.3
105.3 • 23.1 98.9 • 16.1 101.1 • 20.2
protein was measured by SRID according to the method described by Mancini et al.'5 SD The within-the-day method error ( x 100) was 1% mean for the CH~0, 3% for the C2, 1.9% for the C3 titrations, 2.2% for the C3, and 2.6% for the C4 protein determinations. All samples from the same provocation were always run simultaneously. RESULTS
Thirty-two of the forty-nine provocations resulted in a positive immediate reaction with a decrease in PEFR of 15% or more. Eight of the patients also had a late bronchial reaction (Table I). In one case a late reaction occurred without being preceded by a detectable early reaction. The mean decrement in PEFR was 52% (_+23.7) for the early reactions that were followed by late ones and was 41.4% ( _+ 16.2) for the isolated early reactions. The coefficient of correlation between changes in PEFR in early and late reactions was 0.22 (Fig. 1). The mean value for CHso was unchanged both at 15 min and at 6 hr in the groups with early reactions alone and late reactions as well as in the group without any bronchial reactions (Tables 1I and II1). Of the patients who reacted with complement changes, an approximately equal number demonstrated increases as well as decreases in this activity. The same pattern was observed for C2 and C3 titers as well as for protein concentrations of C3, C4, and C l q that was measured with SRID. To check if degradation products were present in the blood after provocation, two analyses were performed. First, the blood was assayed for "converted" C3 by means of IEP, but no such conversion was found in any of the cases. Second, the ratio between hemolytically active C3 and C3 protein was calculated. Measured as a mean value, this ratio did not change in any patient group as a
result of provocation, but in a few individual patients the ratio changed after challenge. A decrease exceeding 15% after 15 min or 6 hr was observed in four (25%) patients without bronchial reactions, in eight (33%) patients with only an early reaction, and in five (56%) patients with a late reaction. An attempt was made to determine if complement was consumed during the passage of blood through the lungs. Thus the complement titers and complement protein content were measured simultaneously in venous and arterial blood. The ratios between venous and arterial blood before, 15 min after, and 6 hr after provocation were calculated. The mean changes in all of the components measured (Tables IV and V) were small in comparison with the variation as expressed by the standard deviation. Three of the four patients without bronchial reactions, seven of the 14 with only an early reaction, and none of the five with a late reaction demonstrated decreases in this arteriovenous ratio for CHs0 of more than 15%. The question arises if it is the patients with the strongest bronchial reactions that get complement alterations. Among the six patients with a decrease in PEFR exceeding 60% after 15 min, only one had a CHs0, C2, or C3 value that was more than 15% lower after the provocation than before. In none of these six patients was the ratio of hemolytic C3 to C3 protein below 85% of the initial value after the provocation. Five patients had more than 25% lower PEFR 6 hr after the provocation than before. None had lower CHso values 15 min or 6 hr after the provocation than before. Three had C2 values that were more than 25% lower than the initial value both after 15 min and after 6 hr. Changes exceeding 15% were most often observed for C2. Of 16 patients in whom no bronchial obstruction developed, three (19%) demonstrated a decrease in C2 after 15 min and two (13%) after 6
VOLUME75
Late bronchial reactions and complement 511
NUMBER 4
TABLE IV. The ratio between complement titers in arterial and venous blood drawn 15 minutes after provocation expressed in percent of the prechallenge ratio • SD
No reactions Only early reactions Late reactions
No.
CH~,
C2
C3
4 14 5
82.3 -+ 37.6 94.0 -+ 15.8 97.3 • 7.6
82.5 • 18.3 97.1 • 17.6 112.8 • 9.8
122.0 _+ 27.3 108.3 • 37.1 105.8 • 25.1
TABLE V. The ratio between complement titers in arterial and venous blood drawn 6 hours after
provocation expressed in percent of the prechallenge ratio • SD
No reactions Only early reactions Late reactions
No.
CH~
C2
C3
4 14 5
103.0 _+ 37.6 104.0 • 22.0 93.0 • 5.6
94.8 • 36.8 103.2 • 22.0 116.5 • 5.1
91.3 +_ 43.9 108.1 • 17.9 91.3 • 22.2
hr. Of the 24 patients with isolated early bronchial reactions, seven (29%) had low values of C2 at 15 min and one (4%) at 6 hr. In five (56%) of the nine patients reacting with late bronchial obstruction, there was a decrease in C2 of more than 15% after 15 min, and in four (44%) the C2 values were low after 6 hr. A correlation coefficient of 0.20 was found between decrements in C2 after 15 min and in PEFR after 6 hr when the values of all 49 patients were considered. Other correlations between complement changes and bronchial reactions were lower. DISCUSSION Late allergic reactions were originally considered to take place if the allergic patient had IgG antibodies against the allergen, ~ but it has since been demonstrated beyond doubt that IgE can mediate late reactions that can be elicited by any ordinary allergen provided a sufficiently large dose is administered. J6-1~ The late bronchial reaction resembles the asthmatic attack more closely than does the early one with respect to clinical features, physiologic characteristics, and effects of pharmaceutical agents.'-' 3 Histamine does not produce late reactions or states of the same kind as asthmatic attacks. The late reaction possesses more the characteristics of an inflammatory reaction in the airways~9 than bronchoconstriction produced only by smooth muscle. Mediators to be looked for should thus have the potential to produce inflammation, and probably a number of mediators act together in the production of the late reaction. The complement system is a system with an ability to produce many of the components in the inflammatory process. On activation of the complement, anaphylatoxins are generated. These recruit leukocytes to the site of activation and trigger these cells to release their granular content, including histamine.
The anaphylatoxins also induce vasodilatation, increased vascular permeability, and contraction of smooth muscles. Purified anaphylatoxin injected into the skin is very efficient in eliciting an urticarial reaction.Z~ Only about 20 ng of C3a is needed to produce a wheal-and-flare reaction. Some years ago we were able to demonstrate that anaphylatoxin could also elicit late cutaneous reactions. 9 Therefore, if complement is activated at a bronchial provocation, it is also possible that anaphylatoxin will take part in the development of the bronchial reaction. Arroyave et al. 4 titrated complement before and after provocation of allergic asthmatic patients and patients with hypersensitivity to acetylsalicylic acid. They observed signs of complement consumption in three of seven atopic individuals and in five of seven aspirin-intolerant persons. In the former group this consumption took place within 30 min and in the second group between one and 4 hr after the challenge. Hsieh 5 studied 20 asthmatic children, three of whom demonstrated reductions in serum C4 and/or factor B after allergen provocation. Two of these patients had precipitating antibodies, and one had a late bronchial reaction. Hutchcroft and G u z , 6 o n the other hand, found no complement decrease in their patients during allergeninduced asthma. They also studied arterial blood samples from two patients without finding any signs of complement activation. Baur et al. 8 found a complement decrease in two of their 16 patients during the late reaction. Pryjma et al. 7 noted a reproducible fall in serum complement after allergen provocation in some children, but there was no correlation between this reduction and the appearance of late bronchial reactions. Among our 32 patients with early bronchial reactions, eight (25%) had late reactions. Late reactions occurred with all four groups of allergens used (Table
512
St,~lenheim and Machado
J. ALLERGYCLIN. IMMUNOL. APRIL 1985
I). Changes in complement titers were observed in a few patients, but there was no difference between the patients who did not have any bronchial reaction, those who developed only an early reaction, and those with a late reaction. The mean postchallenge titers of whole complement and complement components, in relation to the prechallenge ones, were very close to 100% both 15 min and 6 hr after the test in the three groups of patients. The variation was fairly wide. Since complement components react as acute phase proteins and elevations of the titers were frequently observed after the provocations, an attempt was made to determine whether complement consumption or activation took place in the pulmonary circulation. For this purpose the change in the arteriovenous ratio for whole complement components was measured, but no significant changes in these ratios were observed. Another possible way of ascertaining whether a component may be present or not in the circulation in the inactivated form is to calculate the ratio between the hemolytic activity and the protein content of the component. This was done for C3, but that ratio did not change as a result of provocation in any of the groups of patients. It is concluded that bronchial provocation with allergens does not result in significant complement inactivation, not even if the acute phase character of complement is taken into consideration. This does not necessarily rule out the possibility, however, that complement may be involved in the production of bronchial reactions, early or late. There might be subgroups of patients with allergic asthma in whom complement consumption takes place as a result o f exposure to an allergen. Thus, it appears as if significant consumption is somewhat more common in the group o f patients with late bronchial reactions. It does not appear as if this is dependent on the type of allergen used. As yet there is no clue as to what distinguishes patients with complement consumption from other patients. The very high biologic potency of anaphylatoxins has to be kept in m i n d ? ~ It may be estimated very roughly that activation of much less than 1% o f the total amount of circulating C3 and C5 would be sufficient to produce enough anaphylatoxins to cause significant bronchial obstruction. This extent of complement activation is far too small to be detected by the methods used in this study. Methods are being developed to approach this problem from another angle. We thank Ms. ,~sa B~ickstrrm and Ms. Monica Ferm for excellent technical assistance and registered nurses Barbro Ekblad and Eva Persson for assistance with provocation tests.
REFERENCES
1. Pepys J: Hypersensitivity diseases of the lungs due to fungi and organic dusts. Monogr Allergy 4:19, 1969 2. Warner JO: Significance of late reactions after bronchial challenge with house dust mite. Arch Dis Child 51:905, 1976 3. Price JF, Hey EN, Soothill JF: Provocation to the skin, nose, and lung in children with asthma: immediate and dual hypersensitivity reactions. Clin Exp Immunol 47:587, 1982 4. Arroyave CM, Stevenson DD, Vaughan JH, Tan EM: Plasma complement changes during bronchospasm provoked in asthmatic patients. Clin Allergy 7:173, 1977 5. Hsieh K-H: Changes of serum complement and eosinophil count in antigen-induced bronchospasm. Ann Allergy 41:182, 1978 6. Hutchcroft BJ, Guz A: Levels of complement components during allergen-induced asthma. Clin Allergy 8:59, 1978 7. Pryjma JR, MiklaszewskaJ, Haluszka J, Scislicki A: Decrease of complement hemolytic activity after an allergen-housedustbronchial provocation test. J ALLERGYCLINIMMUNOL70:306, 1982 8. Baur X, Dorsch W, Becher T: Levels of complement factors in human serum during immediate and late asthmatic reactions and during acute hypersensitivitypneumonitis. Allergy 35:383, 1980 9. SthlenheimG, Zetterstrrm O: Late cutaneous allergic reactions without the participation of IgE. Monogr Allergy 14:264, 1979 10. Machado L, SthlenheimG, Malmberg P: Early and late allergic bronchial reactions. Physiologicalcharacteristics. Clin Allergy (accepted for publication) 11. Mayer MM: Complement and complement fixation. In Kabat EA, Mayer MM, editors: Experimental Immunochemistry. Springfield, I11., 1967, Charles C Thomas, Publisher, p 133 12. Cooper NR, Polley MJ, Miiller-EberhardHJ: The second component of human complement (C2): quantitative molecular analysis of its reactions in immune hemolysis. Immunochemistry 7:341, 1970 13. Miiller-Eberhard HJ, Nilsson UR, Dalmasso AP, Polley MJ, Calcott MA: A molecular concept of immune cytolysis. Arch Pathol 82:205, 1966 14. St~lenheim G, Castensson S: Protein A from Staphylococcus aureus. Conversion of complement factor C3 by aggregates between IgG and protein A. FEBS Len 14:79, 1971 15. Mancini G, Carbonara AO, Heremans JF: Immunochemical quantitation of antigens by single radial immunodiffusion, lmmunochemistry 2:235, 1965 16. Solley GO, Gleich GJ, Jordan RE, Schroeter AL: The late phase of the immediate wheal-and-flare skin reaction. Its dependence upon IgE antibodies. J Clin Invest 58:408, 1976 17. Umemoto L, Poothullil J, Dolovich J, Hargreave FE: Factors which influence late cutaneous allergic responses. J ALLERGY CLIN IMMUNOL58:60, 1976
18. Hargreave FE, Dolovich J, Robertson DG, Kerigan AT: The late asthmatic responses. Can Med Assoc J 110:415, 1974 19. Cartier A, Thomson NC, Frith PA, Roberts R, Tech M, Hargreave FE: Allergen-induced increase in bronchial responsiveness to histamine: relationship to the late asthmatic response and change in airway caliber. J ALLERGYCLIN IMMUNOL 70:170, 1982 20. Wuepper KD, Bokisch VA, Miiller-Eberhard HJ, Stoughton RB: Cutaneous responses to human C3 anaphylatoxin in man. Clin Exp Immunol 11:13, 1972