Int. Archs Allergy appl. Immun. 55: 178-188 (1977)
Enhancement by a Serum Factor of Immunoglobulin-Mediated Histamine Release from Human Leukocytes Bengt-Âke Petersson and Gunnemar Stdlenheim Department of Medical and Physiological Chemistry, University of Uppsala, The Biomedical Center, Uppsala
Introduction Antigen-induced histamine release from the leukocytes of allergic donors is an exten sively used in vitro method to study cellular mechanisms involved in allergic reactions [7, 8, 13]. Antibodies against human immu noglobulin E (anti-IgE) and immunoglobu lin G (anti-IgG) can also initiate the secre tion of histamine from the leukocytes of both allergic and normal donors [4, 6, 9]. It is currently agreed that the mechanisms for antigen and anti-IgE-triggered release are identical [9]. The Staphylococcus aureus cell wall constituent, protein A, is another substance used during recent years to study mechanisms involved in allergic histamine release from human leukocytes [14-16]. The usefulness of protein A depends on its
capacity to combine with the Fc part of hu man IgG but not with IgE [3, 17]. Repeated washings of the leukocytes decreased their capacity to secrete histamine when treated with protein A, while the release induced by optimum amounts of anti-IgE was not sig nificantly decreased. However, when re peatedly washed leukocytes were incubated in autologous serum, their capacity to re spond to protein A stimulation was re stored. When serum was separated on a Se phadex G-200 column, material with capac ity to enhance protein A-induced release was eluted in the peaks containing IgG and albumin (peaks II and III, respectively). Further fractionations revealed that the re sensitizing activity was mediated by material not related to immunoglobulins [16]. In this report the capacity of serum and serum frac-
Downloaded by: Univ. of California Santa Barbara 128.111.121.42 - 3/7/2018 3:20:29 AM
Abstract. When serum was fractionated on Sephadex G-200, the material eluted in the second and third major peaks had a very pronounced capacity to enhance IgE- as well as IgG-mediated histamine release from the leukocytes of both normal and allergic donors. Unseparated serum on the other hand had a low capacity to stimulate anti-IgE-induced histamine release. Besides resulting in a higher histamine release, pretreatment with serum fractions also increased the rate of histamine release. Further purification revealed that the stimulating activity of the material in the second peak was mediated by a trypsin-sensitive component, probably active in low concentration.
179
Histamine Release from Human Leukocytes
tions to enhance IgE-mediated histamine re lease has been studied.
Peak
II
Peak
in
Materials and Methods Antibodies against Human IgE Anti-IgE sera were produced in rabbits as pre viously described [15]. IgG from these anti-IgE sera was isolated by chromatography on a DEAEcellulose column followed by further purification on protein A linked to Sepharose 6-B (Pharmacia Fine Chemicals, Uppsala). To obtain specific an ti-IgE, the rabbit IgG preparation was passed through a column containing Sepharose-linked hu man IgG. The unabsorbed material was collected, and will be referred to as anti-IgE throughout this report.
Antigens Allergen extracts were purchased from Vitrum (Stockholm). Extracts diluted 1:100 were dialyzed against Tris-buffered saline containing KC1 for 48 h at 4 °C. Before use in histamine-release ex periments, human serum albumin (0.3 mg/ml), MgCl2 (IX 10-3 m ), and CaCl2 (0.6X10-3M) were added. Quantitative Determination of Immuno globulin E A radioimmunologic method was used to de termine the IgE content of serum fractions used in the histamine release experiments [1], The deter minations were kindly performed by Dr. S. G. O. Johansson. Separations of Serum on Sephadex G-200 Human serum (25 ml) was run on a Sephadex G-200 column (3.5X160 cm) in Tris-buffered sa line containing KC1. The column was eluted at a
Fig. 1. Separation of human serum (25 ml) on a Sephadex G-200 column (3.5X160 cm) in Tris-buffered saline containing KC1. The column was eluted at a rate of 16 ml/h and 3 fractions/h. The fractions were diluted 1:10 before reading the absorbance at 280 nm. Peak II and III material was pooled as shown in the upper part of the fig ure and concentrated to 25 ml.
rate of 16 ml/h, three fractions being collected per hour. Figure 1 shows the separation profile for one of the runs. Peak II and peak III material were pooled as shown in the upper part of the fig ure and concentrated to 25 ml. The protein con centration, determined according to Lowry et al. [12], was 13.3 mg/ml in peak II and 26.1 mg/ml in peak III. Peak II contained less than 5 ng of TgE/ml and peak III less than 1 ng of IgE/ml. Separation of Peak II Material on a QAE-Sephadex Column The material (25 ml) from peak II of the Sephadex G-200 column was dialyzed against 0.0125 M Tris-HCl (pH 7.0) containing 0.05 M NaCl, then applied to a QAE-Sephadex (Pharma cia) column (40 ml) in the same buffer. After ma terial not bound to the column had been eluted with the starting buffer, the NaCl concentration was changed stepwise to 0.1, 0.15, and finally 0.2 M. The material eluted in each step was
Downloaded by: Univ. of California Santa Barbara 128.111.121.42 - 3/7/2018 3:20:29 AM
Antibodies against Human IgG An IgG preparation from rabbit antiserum against human IgG (heavy chain specific) was pur chased from Dakopatts, Copenhagen. This anti-IgG was from the same preparation as that used in a previous report [16]. Before use, a 1:5 dilu tion of anti-IgG was dialyzed against Tris-buffered saline containing KC1 (Tris 0.025 M, NaCl 0.120 M and KC1 0.005 M, pH 7.6) for 72 h at 4 °C.
Petersson/St&lenheim
180
Trypsin Digestion Peak 11-3 material. 200 ng protein/ml, in Trisbuffered saline containing KC1, was treated with 8 /(g trypsin (Worthington Biochemical Corp.) per milliliter for 4 h at 37 °C. The reaction was stopped by adding 16 ug of soybean trypsin inhib itor (Sigma Chemical Company) per ml and the solution was incubated for an additional 30 min. Before use in histamine release experiments, albu min was added to a concentration of 0.3 mg/ml. Control samples contained trypsin and trypsin-in hibitor in Tris-A. Isolation of Leukocytes The method introduced by Lichtenstein and Osier was used to isolate leukocytes from the blood of normal or allergic persons [10]. Briefly, the blood was mixed with dextran (Pharmacia) and ethylenediaminetetraacetic acid (EDTA) and was allowed to stand at room temperature for 4560 min. The upper leukocyte-rich fraction was washed either once or three times with Tris-A (Tris-buffered saline containing KC1 and 0.3 mg albumin/ml). In the latter procedure the samples were incubated for 20 min at 37 °C after the sec ond washing. Leukocytes washed three times will be called repeatedly-washed leukocytes [14]. The allergic donors participating in this investigation were sensitive to one or several of the usual aller gens. This was shown both in cutaneous tests and in RAST (radioallergosorbent test). Histamine Release The histamine-release experiments were per formed on 2.0 ml samples containing 0.6-1.0X107 leukocytes/ml. The effect of washing on IgE-
mediated release was tested by incubating oncewashed or repeatedly-washed cells with anti-IgE in Tris-ACM (Tris-A containing calcium and magne sium) at 37 °C for 40 min. A two-step procedure was used to study the capacity of serum and serum fractions to enhance the release of histamine, triggered by anti-IgE, an tigen and anti-IgG. Firstly, repeatedly-washed cell samples were incubated at 37 °C with 1.0-ml por tions of Tris-A, serum or serum fractions for spe cific time intervals as indicated. After washing the leukocytes once in Tris-A at 4 °C, step 2, which consisted of incubating the cells with the histami ne-releasing agent in Tris-ACM, was performed. Then the amount of histamine liberated in step 2 was determined. Total histamine content was de termined in two samples boiled for 3 min. Histamine Assay Histamine was biologically assayed using seg ments of guinea pig ileum as published [2, 15]. Human IgG, in excess, was added to anti-IgG-containing fractions (Cohn fraction II, Kabi AB, Stockholm).
Results Effect of Washings on Anti-IgE-Induced Histamine Release A total of 28 experiments were per formed to study histamine release from on ce-washed and repeatedly-washed cells, as induced by optimum amounts of anti-IgE in Tris-ACM. The decreased release as an effect of repeated washings, was calculated in proportion to the release obtained after one washing. In 12 of the experi ments a decrease of between 12 and 70°/o, with an average of 31°/o, was obtained. A decrease of less than ten per cent was de tected in the remaining experiments. In 16 of the experiments the release after one washing exceeded 50% of the total histam ine content and in this group a marked de crease (more than 10% of the release after
Downloaded by: Univ. of California Santa Barbara 128.111.121.42 - 3/7/2018 3:20:29 AM
pooled and concentrated to 25 ml. After dialysis against Tris-buffered saline containing KC1, the materials were used in histamine-release experi ments. Preliminary experiments indicated that the ma terial eluted with 0.15 M NaCl was most effective in enhancing anti-IgE-initiated release. This mate rial, designated peak II-3, was used further. The 25 ml pool had a protein concentration of 0.76 mg/ml and contained less than 1 ng of IgE/mg protein. When peak II-3 material was run on polyacrylamide gel electrophoresis, one faint band and four strong bands were seen on the gels.
one washing) was only obtained in 4 experi ments. In the other group, giving a release of less than 50% after one washing, repeat ed washings resulted in a striking decrease (more than 10%) in 8 out of 12 experi ments. If leukocyte preparations from the same donor were tested on different occa sions, the susceptibility to washings could vary. With a few exceptions, the cell donors participating in the following experiments were selected with a view to obtaining leu kocyte preparations releasing less than 50% of their histamine on treatment with opti mum amounts of anti-IgE. Optimum Pretreatment Conditions for the Enhancement of Histamine Release The capacity of peak II and peak III ma terial to enhance histamine release from the leukocytes was measured in the two-step procedure already described. The first incu bation step was carried out on cell samples in buffer (control) or different dilutions of peak II or peak III material, for various time intervals. The second incubation was for 40 min at 37 °C with optimum amounts of anti-IgE. The pretreatment with peak II and peak III material resulted in a marked potentiation of the histamine release. In most of the experiments, maximum en hancement was obtained after 20 min pre treatment with peak II material diluted 1:2 or with peak III material diluted 1:4. In a few experiments, 40 min incubation was needed to get maximum enhancement. For the remaining experiments in this report, step 1 treatment of samples was carried out for 40 min at 37 °C. In all but a few experiments, incubation of repeatedly washed leukocytes with TrisA for an additional 40 min at 37 °C, did not further decrease the capacity of anti-IgE to
181
induce histamine secretion. Thus, the great er histamine release from leukocytes pre treated with serum fraction, as compared with buffer-pretreated cells, was a stimula tion effect, and not dependent on a decrease in vitality of the control cells. Furthermore, no histamine release was detected when leukocytes pretreated with serum, peak II (1:2) or peak III (1:4) mate rial, were incubated for 40 min at 37 °C with Tris-ACM buffer, instead of anti-IgE, in step 2. All peak II and peak III material used in this investigation was prepared from the serum of a normal blood donor. Testing four additional sera showed that all of them contained material with capacity to stimu late anti-IgE-triggered release. Study of the Effect of Peak II and Peak III Pretreatment on Anti-IgE-Induced Histamine Release First the effect of pretreatment of the cells with a 1:4 dilution of peak III material was tested. The two-step procedure pre viously outlined was used. Cells pretreated with Tris-A buffer only were used as control samples. The second incubation, with antiIgE in Tris-ACM, was performed on all samples. After various time intervals, the reaction was stopped by addition of 1 ml ice-cold Tris-A, and the samples were cen trifuged in the cold. Figure 2 shows the re sults of three of these experiments. It can be clearly seen that peak Ill-treated cells re leased greater amounts of histamine than did the buffer-treated cells. This was true when optimum (1.5 ,ug/ml) or suboptimum (0.5 jug/ml) amounts of anti-IgE were used. Pretreatment with peak III material also re duced the lag period. These two effects of peak III treatment were observed in all the
Downloaded by: Univ. of California Santa Barbara 128.111.121.42 - 3/7/2018 3:20:29 AM
Histamine Release from Human Leukocytes
Petersson/Stâlenheim
182
b
Fig. 2. Time course of anti-IgE(1.5 ug/ml in a, b, and 0.5 «g/ml in c)-induced histamine release from buffer (O) and peak III ( # ) pretreated leua
c
kocytes. a, b, and c represent experiments with leukocytes from different donors.
b
Anti-lgE, mg pnotein/ml
Fig. 3. Histamine release induced with varying amounts of anti-IgE from buffer and serum fraction pretreated leukocytes. O = Release from buffer-pretreated leukocytes; • = release from peak III (a) or peak II (b) pretreated leukocytes.
experiments performed. In the experiment shown in figure 2c maximum level was reached more rapidly with peak III than with buffer-treated cells. This effect was less pronounced in other experiments. In another set of experiments the ability of different concentrations of anti-IgE to re lease histamine from leukocytes pretreated with peak II and peak III material was in
vestigated. The first-step incubation was with Tris-A (control), peak II material or peak III material. The second-step incuba tion was with varying amounts of anti-IgE in Tris-A CM, for 40 min at 37 °C. The re sults are shown in figure 3 from which it can be seen that, at all anti-IgE concentra tions used, pretreatment with peak II or peak III material resulted in a marked in crease in histamine release, compared with the control set of samples. In experiments where otpimum concentrations of anti-IgE released high amounts of histamine (70-100°/o) from buffer pretreated leuko cytes, the enhancing effect of peak II or peak III pretreatment was more pronounced when sub- or supraoptimum amounts of anti-IgE were used (fig. 3b). In the experiments shown in table I, leu kocytes were incubated in buffer, autolo gous serum, peak II material or peak III material for 40 min at 37 °C in the first step. In the second step the samples were treated with optimum amounts of anti-IgE for 40 min at 37 °C. The cell donors for these experiments were selected with a view to obtaining less than 50°/o histamine release
Downloaded by: Univ. of California Santa Barbara 128.111.121.42 - 3/7/2018 3:20:29 AM
a
183
Histamine Release from Human Leukocytes
Experi- Anti-IgE (1.5 /tg/ml)-induccd histamine ment release (in %) from leucocytes pretreated at No. 37°C for 40 min with: Tris-A serum 1:2 peak 111:2 peak 111 1:4 1 2 3 4 5 6 7 8 9 10 11
16 18 23 24 24 24 30 39 47 48 48
24 25 28 30 not tested 32 40 not tested 64 59 67
55 47 50 38 67 58 81 82 87 90 not tested
52 52 65 51 65 66 80 81 86 90 95
from buffer-pretreated cells on treatment with optimum amounts of anti-lgE. Serum diluted 1:2 was used since dilutions between 1:1 and 1:32 gave essentially the same de gree of enhancement. Dilutions 1:64 to 1:512 had smaller effects. Another reason for choosing this serum concentration was that serum diluted 1:2 gave maximum resensilizalion of protein A-induced histamine release [16]. Pretreatment of the leukocytes with peak II or peak III material resulted in a very pronounced enhancement of the anti-IgEinduced release. If the increase was calcu lated as percent of the release induced from buffer-treated cells, the enhancement ob tained with peak II varied between 58 and 244% with an average of 135. For peak III the corresponding figures were 83 and 225% with an average of 145. Serum, on the other hand, had a very low capacity to stimulate anti-IgE-induced release. The enhancement was between 22 and 50% with an average of 33%.
Study of the Effect of Peak II ami Peak III Pretreatment on the Antigen-Induced Release Firstly, leukocytes from subjects suffer ing from an allergy were treated with buffer (control) or peak III material. Secondly, they were incubated with varying amounts of antigen diluted in Tris-ACM (for 40 min at 37 °C). The antigens used were pollen extracts of birch (diluted to 10'7-10"3) and timothy (diluted to 10‘14-1 0 '3). The results are shown in figure 4 from which it can be seen that pretreatment with peak III material markedly enhanced the antigen-induced his tamine secretion from the leukocytes of these donors. Figure 4b shows an experi ment where optimum amounts of antigen released high amounts of histamine (79%) from buffer pretreated cells also. However, at sub- or supraoptimum concentrations of antigen, the stimulating effect of pretreat ment with peak III material was more pron ounced. Antigen-triggered release was enhanced as an effect of peak II pretreatment (experia
Dilution o f birchpollen extract
b
Dilution o f tim othypollen extract
Fig. 4. Histamine release induced with birch pollen antigen (a) or timothy pollen antigen (b) from buffer (O) or peak III (# ) pretreated leuko cytes.
Downloaded by: Univ. of California Santa Barbara 128.111.121.42 - 3/7/2018 3:20:29 AM
Table I. Anti-IgE-induced histamine release from buffer, serum, peak II and peak lll-pretrcated leuko cytes
Petersson/Stâlenheim
184
Experiment No.
1 2 3 4
Anti-IgG (1:200)-induced histamine release (in %) from leukocytes pretreated at 37 °C for 40 min with: Tris-A
peak III 1:4
8 23 37 40
51 55 75 87
Table III. Anti-IgE-induced histamine release from buffer, peak II, peak II-3 and trypsinated peak II-3 pretreated leukocytes Experi- Additive in step 1 ment (pretreatment at 37 °C No. for 40 min)
Histamine release in step 2 % (test with antiIgE, 1.5 MS/m\)
1
buffer 24 peak II 1:2 67 peak II-3 200 MS protein/ml 71 peak II-3 100 fig protein/ml 67
2
buffer peak II 1:2 peak II-3 100 MS protein/ml peak II-3 10 MS protein/ml peak II-3 1 MS protein/ml
ments not shown), in the same way as antiIgE-initiated release.
21 40 38 29 23
3
Study of the Effect of Peak 111 Pretreatment on Anti-IgG-induced Histamine Release Using the experimental procedure al ready described the effect of peak III pre treatment on anti-IgG-triggered release was investigated. In the second step, a 1:200 di lution of anti-IgG was used [16]. The ex periments in table II show that antiIgG-triggered release was also markedly stimulated. If the increase was calculated as percent of the release induced from buffertreated cells, the enhancement obtained var ied between 102 and 538%.
buffer 28 peak II-3 200 MS protein/ml 52 peak IT-3 100 MS protein/ml 51 peak II-3 10 MS protein/ml 33 peak II-3 1 m s protein/ml 26 trypsin-digested peak II-3 35 200 MS protein/ml buffer containing trypsin and trypsin inhibitor 26
4
buffer 31 peak II-3 200 MS protein/ml 62 trypsin-digested peak II-3 200 MS protein/ml 36 buffer containing trypsin and trypsin inhibitor 33
Capacity of Native, and TrypsinDigested, Peak 11-3 Material to Enhance Anti-1gE-Induced Release In a series of experiments, leukocytes were preincubated with peak II-3 material (from QAE-Sephadex) before the treat ment with anti-IgE. Table III shows that peak II-3 material with protein concentra tions of 200 or 100 pg per ml had a high ca pacity, even in comparison with unseparat-
ed peak II material, to enhance anti-IgE-induced release. If peak II-3 material was di luted to 10 pg protein/ml it still had some capacity to stimulate the release, while 1 pg protein/ml was inactive. On the other hand, trypsin-digested peak II-3 material (200 pg protein/ml) had a very low capacity, compared with undigested ma terial to enhance anti-IgE induced release (experiments 3 and 4, table III).
Downloaded by: Univ. of California Santa Barbara 128.111.121.42 - 3/7/2018 3:20:29 AM
Table II. Anti-IgG-induced histamine release from buffer and peak 111-pretreated leukocytes
Discussion The previously observed decrease in protein A reactivity, as a result of washings, suggested that some factor necessary for his tamine release was eluted from the leuko cytes [14, 16]. In 31 out of 43 experiments, repeated washings completely removed pro tein A reactivity and in the remaining ex periments caused a greater than 50% de crease. On the other hand, anti-IgE-induced release was rather resistant to repeated washings; although it had been found earlier [7] that a factor with capacity to enhance antigen-triggered release was eluted from whole viable leukocytes during 30 min in cubation at 15-20 °C. Repeated washings caused a greater than 50% decrease in anti-IgE-induced release in only 2 out of 28 experiments. A decrease of less than 10% was detected in 16 of the experiments. Fur thermore, it has been recently reported that an 85-100% decrease in protein A reactivi ty could be obtained without any decrease in release triggered by anti-IgE or anti-IgG [14, 16]. Histamine secretion from repeat edly washed leukocytes (i.e. lacking reactiv ity to protein A alone) could also be in duced by protein A in complex with IgG [14]. The differences in susceptibility to washings indicated that anti-IgE, anti-IgG, protein A and protein A-IgG complexes have different requirements for the factors which are capable of modulating the degree of histamine release. Since some leukocyte preparations from donors suffering from an allergy released low amounts of histamine (below 30-50%), even when treated with optimum antigen concentrations, Lichtenstein and Osier [10] discussed in their original report the possi ble existence of some nonimmunoglobulin
185
factor necessary for antigen-triggered his tamine secretion. In the middle of the 1960s, the same authors reported that in corporation of 10% human serum in the an tigen solution enhanced the histamine re lease induced by suboptimum amounts of antigen. At optimum antigen concentrations no effect was seen [11]. They managed to isolate from serum a lipoprotein and a low molecular weight factor with capacity to po tentiate antigen-induced release [7]. Ishizaka et al. [6] showed that anti-IgE-triggered histamine release from the leukocytes of both allergic and normal donors was en hanced by some serum component not relat ed to the complement system. This result was confirmed later by Grant and Lichten stein [5]. In addition to the stimulation of IgE-mediated release, serum had the capaci ty to effectively resensitize repeatedlywashed leukocytes thus restoring their ca pacity to liberate histamine on protein A treatment [16]. The active material was eluted both in protein peaks II and III when serum was fractionated on Sephadex G-200. This result, together with the fact that di alysis did not decrease the capacity of serum to resensitize leukocytes, indicated that the factors were not identical with the lipopro tein or the low molecular weight factor stud ied by Lichtenstein and coworkers [7, 16]. Material with resensitizing capacity was not bound to a protein A-Sepharose column, and washings of resensitized leukocytes again decreased their capacity to release his tamine on protein A treatment [16]. These results indicated that the active components must exert their effect after binding to the leukocytes. The possibility was considered that the factors could have a stimulating ef fect on immunologically induced histamine secretion in general. Thus, the capacity of
Downloaded by: Univ. of California Santa Barbara 128.111.121.42 - 3/7/2018 3:20:29 AM
Histamine Release from Human Leukocytes
peak II and III material to enhance IgE-mediated release has been investigated. Other investigators have routinely been including 1-10% serum in their antigen or anti-IgE solutions when studying the poten tiating effect of serum on the histamine re lease induced with these agents [5, 6, 11]. Since the active components in peak II and III material were bound to the leukocytes, it was possible in this, and in a recent report, to study the stimulating effect of serum and scrum fractions in a two-step procedure. The advantage of such a method is that the leukocytes both in control and test samples are treated with the triggering agent under similar conditions, for example, in the pres ence of protein in the same concentration. Use of the two-step procedure therefore rules out stimulation by an unspecific prote in effect; a possibility discussed in an earlier work [11]. Whether washing resulted in a decrease in IgE-mediated histamine release or not, pretreatment of repeatedly washed leuko cytes with peak II and III material actually had a striking effect on histamine release in duced by both anti-IgE and antigen (fig. 3, 4). Enhancement was obtained at all antigen and anti-IgE concentrations tested (fig. 3, 4). If optimum amounts of the releasing agents released high quantities of histamine from buffer-pretreated cells, the effect of peak II or peak III pretreatment could not, of course, have been so pronounced. In this case the stimulating effect was better visual ized with sub- or supraoptimum amounts (fig. 3, 4). Furthermore, the time-course ex periments revealed that peak III pretreat ment, besides resulting in a higher histamine release, also reduced the length of the lag period (fig. 2). Autologous serum enhanced anti-IgE-
Petersson/Stálenheim
triggered release but the effect was small compared with the effect of peak II or peak III pretreatment. Table I shows a series of II experiments, where the stimulating effect obtained as a result of pretreatment with peak II material, peak III material or autol ogous serum was studied. The blood donors were selected with a view to obtaining leu kocyte preparations releasing less than 50% of their histamine when buffer-pretreated cells were incubated with optimum amounts of anti-IgE. Peak II and peak III material gave, with rather few exceptions, compara ble enhancement of the anti-IgE-induced re lease. The average stimulation by peak II was 135% and by peak III, 145%. The cor responding value for serum was only 33%. In these experiments, serum diluted 1:2 was used. Undiluted serum or serum diluted more than 1:2 (down to 1:512) did not re sult in better stimulation. The ineffective ness of serum in producing an enhancement comparable with that of peak II and peak III material, indicates the possible existence in whole serum, of histamine release inhibi tors or inactivators of the enhancing factors. This has in fact been discussed earlier in connection with the lipoprotein, which had the capacity to increase antigen-induc ed histamine release from human leukocy tes [7], Furthermore, a few experiments showed that peak III-pretreated leukocytes released considerably higher amounts of histamine than did buffer-preincubated cells, during treatment with optimum amounts of antiIgG (table II). When peak II material was further frac tionated on a QAE-Sephadex column, ac tive material was obtained in a pool with low protein content. 100 /
Enhancement by a Serum Factor of Immunoglobulin-Mediated Histamine Release from Human Leukocytes Bengt-Âke Petersson and Gunnemar Stdlenheim Department of Medical and Physiological Chemistry, University of Uppsala, The Biomedical Center, Uppsala
Introduction Antigen-induced histamine release from the leukocytes of allergic donors is an exten sively used in vitro method to study cellular mechanisms involved in allergic reactions [7, 8, 13]. Antibodies against human immu noglobulin E (anti-IgE) and immunoglobu lin G (anti-IgG) can also initiate the secre tion of histamine from the leukocytes of both allergic and normal donors [4, 6, 9]. It is currently agreed that the mechanisms for antigen and anti-IgE-triggered release are identical [9]. The Staphylococcus aureus cell wall constituent, protein A, is another substance used during recent years to study mechanisms involved in allergic histamine release from human leukocytes [14-16]. The usefulness of protein A depends on its
capacity to combine with the Fc part of hu man IgG but not with IgE [3, 17]. Repeated washings of the leukocytes decreased their capacity to secrete histamine when treated with protein A, while the release induced by optimum amounts of anti-IgE was not sig nificantly decreased. However, when re peatedly washed leukocytes were incubated in autologous serum, their capacity to re spond to protein A stimulation was re stored. When serum was separated on a Se phadex G-200 column, material with capac ity to enhance protein A-induced release was eluted in the peaks containing IgG and albumin (peaks II and III, respectively). Further fractionations revealed that the re sensitizing activity was mediated by material not related to immunoglobulins [16]. In this report the capacity of serum and serum frac-
Downloaded by: Univ. of California Santa Barbara 128.111.121.42 - 3/7/2018 3:20:29 AM
Abstract. When serum was fractionated on Sephadex G-200, the material eluted in the second and third major peaks had a very pronounced capacity to enhance IgE- as well as IgG-mediated histamine release from the leukocytes of both normal and allergic donors. Unseparated serum on the other hand had a low capacity to stimulate anti-IgE-induced histamine release. Besides resulting in a higher histamine release, pretreatment with serum fractions also increased the rate of histamine release. Further purification revealed that the stimulating activity of the material in the second peak was mediated by a trypsin-sensitive component, probably active in low concentration.
179
Histamine Release from Human Leukocytes
tions to enhance IgE-mediated histamine re lease has been studied.
Peak
II
Peak
in
Materials and Methods Antibodies against Human IgE Anti-IgE sera were produced in rabbits as pre viously described [15]. IgG from these anti-IgE sera was isolated by chromatography on a DEAEcellulose column followed by further purification on protein A linked to Sepharose 6-B (Pharmacia Fine Chemicals, Uppsala). To obtain specific an ti-IgE, the rabbit IgG preparation was passed through a column containing Sepharose-linked hu man IgG. The unabsorbed material was collected, and will be referred to as anti-IgE throughout this report.
Antigens Allergen extracts were purchased from Vitrum (Stockholm). Extracts diluted 1:100 were dialyzed against Tris-buffered saline containing KC1 for 48 h at 4 °C. Before use in histamine-release ex periments, human serum albumin (0.3 mg/ml), MgCl2 (IX 10-3 m ), and CaCl2 (0.6X10-3M) were added. Quantitative Determination of Immuno globulin E A radioimmunologic method was used to de termine the IgE content of serum fractions used in the histamine release experiments [1], The deter minations were kindly performed by Dr. S. G. O. Johansson. Separations of Serum on Sephadex G-200 Human serum (25 ml) was run on a Sephadex G-200 column (3.5X160 cm) in Tris-buffered sa line containing KC1. The column was eluted at a
Fig. 1. Separation of human serum (25 ml) on a Sephadex G-200 column (3.5X160 cm) in Tris-buffered saline containing KC1. The column was eluted at a rate of 16 ml/h and 3 fractions/h. The fractions were diluted 1:10 before reading the absorbance at 280 nm. Peak II and III material was pooled as shown in the upper part of the fig ure and concentrated to 25 ml.
rate of 16 ml/h, three fractions being collected per hour. Figure 1 shows the separation profile for one of the runs. Peak II and peak III material were pooled as shown in the upper part of the fig ure and concentrated to 25 ml. The protein con centration, determined according to Lowry et al. [12], was 13.3 mg/ml in peak II and 26.1 mg/ml in peak III. Peak II contained less than 5 ng of TgE/ml and peak III less than 1 ng of IgE/ml. Separation of Peak II Material on a QAE-Sephadex Column The material (25 ml) from peak II of the Sephadex G-200 column was dialyzed against 0.0125 M Tris-HCl (pH 7.0) containing 0.05 M NaCl, then applied to a QAE-Sephadex (Pharma cia) column (40 ml) in the same buffer. After ma terial not bound to the column had been eluted with the starting buffer, the NaCl concentration was changed stepwise to 0.1, 0.15, and finally 0.2 M. The material eluted in each step was
Downloaded by: Univ. of California Santa Barbara 128.111.121.42 - 3/7/2018 3:20:29 AM
Antibodies against Human IgG An IgG preparation from rabbit antiserum against human IgG (heavy chain specific) was pur chased from Dakopatts, Copenhagen. This anti-IgG was from the same preparation as that used in a previous report [16]. Before use, a 1:5 dilu tion of anti-IgG was dialyzed against Tris-buffered saline containing KC1 (Tris 0.025 M, NaCl 0.120 M and KC1 0.005 M, pH 7.6) for 72 h at 4 °C.
Petersson/St&lenheim
180
Trypsin Digestion Peak 11-3 material. 200 ng protein/ml, in Trisbuffered saline containing KC1, was treated with 8 /(g trypsin (Worthington Biochemical Corp.) per milliliter for 4 h at 37 °C. The reaction was stopped by adding 16 ug of soybean trypsin inhib itor (Sigma Chemical Company) per ml and the solution was incubated for an additional 30 min. Before use in histamine release experiments, albu min was added to a concentration of 0.3 mg/ml. Control samples contained trypsin and trypsin-in hibitor in Tris-A. Isolation of Leukocytes The method introduced by Lichtenstein and Osier was used to isolate leukocytes from the blood of normal or allergic persons [10]. Briefly, the blood was mixed with dextran (Pharmacia) and ethylenediaminetetraacetic acid (EDTA) and was allowed to stand at room temperature for 4560 min. The upper leukocyte-rich fraction was washed either once or three times with Tris-A (Tris-buffered saline containing KC1 and 0.3 mg albumin/ml). In the latter procedure the samples were incubated for 20 min at 37 °C after the sec ond washing. Leukocytes washed three times will be called repeatedly-washed leukocytes [14]. The allergic donors participating in this investigation were sensitive to one or several of the usual aller gens. This was shown both in cutaneous tests and in RAST (radioallergosorbent test). Histamine Release The histamine-release experiments were per formed on 2.0 ml samples containing 0.6-1.0X107 leukocytes/ml. The effect of washing on IgE-
mediated release was tested by incubating oncewashed or repeatedly-washed cells with anti-IgE in Tris-ACM (Tris-A containing calcium and magne sium) at 37 °C for 40 min. A two-step procedure was used to study the capacity of serum and serum fractions to enhance the release of histamine, triggered by anti-IgE, an tigen and anti-IgG. Firstly, repeatedly-washed cell samples were incubated at 37 °C with 1.0-ml por tions of Tris-A, serum or serum fractions for spe cific time intervals as indicated. After washing the leukocytes once in Tris-A at 4 °C, step 2, which consisted of incubating the cells with the histami ne-releasing agent in Tris-ACM, was performed. Then the amount of histamine liberated in step 2 was determined. Total histamine content was de termined in two samples boiled for 3 min. Histamine Assay Histamine was biologically assayed using seg ments of guinea pig ileum as published [2, 15]. Human IgG, in excess, was added to anti-IgG-containing fractions (Cohn fraction II, Kabi AB, Stockholm).
Results Effect of Washings on Anti-IgE-Induced Histamine Release A total of 28 experiments were per formed to study histamine release from on ce-washed and repeatedly-washed cells, as induced by optimum amounts of anti-IgE in Tris-ACM. The decreased release as an effect of repeated washings, was calculated in proportion to the release obtained after one washing. In 12 of the experi ments a decrease of between 12 and 70°/o, with an average of 31°/o, was obtained. A decrease of less than ten per cent was de tected in the remaining experiments. In 16 of the experiments the release after one washing exceeded 50% of the total histam ine content and in this group a marked de crease (more than 10% of the release after
Downloaded by: Univ. of California Santa Barbara 128.111.121.42 - 3/7/2018 3:20:29 AM
pooled and concentrated to 25 ml. After dialysis against Tris-buffered saline containing KC1, the materials were used in histamine-release experi ments. Preliminary experiments indicated that the ma terial eluted with 0.15 M NaCl was most effective in enhancing anti-IgE-initiated release. This mate rial, designated peak II-3, was used further. The 25 ml pool had a protein concentration of 0.76 mg/ml and contained less than 1 ng of IgE/mg protein. When peak II-3 material was run on polyacrylamide gel electrophoresis, one faint band and four strong bands were seen on the gels.
one washing) was only obtained in 4 experi ments. In the other group, giving a release of less than 50% after one washing, repeat ed washings resulted in a striking decrease (more than 10%) in 8 out of 12 experi ments. If leukocyte preparations from the same donor were tested on different occa sions, the susceptibility to washings could vary. With a few exceptions, the cell donors participating in the following experiments were selected with a view to obtaining leu kocyte preparations releasing less than 50% of their histamine on treatment with opti mum amounts of anti-IgE. Optimum Pretreatment Conditions for the Enhancement of Histamine Release The capacity of peak II and peak III ma terial to enhance histamine release from the leukocytes was measured in the two-step procedure already described. The first incu bation step was carried out on cell samples in buffer (control) or different dilutions of peak II or peak III material, for various time intervals. The second incubation was for 40 min at 37 °C with optimum amounts of anti-IgE. The pretreatment with peak II and peak III material resulted in a marked potentiation of the histamine release. In most of the experiments, maximum en hancement was obtained after 20 min pre treatment with peak II material diluted 1:2 or with peak III material diluted 1:4. In a few experiments, 40 min incubation was needed to get maximum enhancement. For the remaining experiments in this report, step 1 treatment of samples was carried out for 40 min at 37 °C. In all but a few experiments, incubation of repeatedly washed leukocytes with TrisA for an additional 40 min at 37 °C, did not further decrease the capacity of anti-IgE to
181
induce histamine secretion. Thus, the great er histamine release from leukocytes pre treated with serum fraction, as compared with buffer-pretreated cells, was a stimula tion effect, and not dependent on a decrease in vitality of the control cells. Furthermore, no histamine release was detected when leukocytes pretreated with serum, peak II (1:2) or peak III (1:4) mate rial, were incubated for 40 min at 37 °C with Tris-ACM buffer, instead of anti-IgE, in step 2. All peak II and peak III material used in this investigation was prepared from the serum of a normal blood donor. Testing four additional sera showed that all of them contained material with capacity to stimu late anti-IgE-triggered release. Study of the Effect of Peak II and Peak III Pretreatment on Anti-IgE-Induced Histamine Release First the effect of pretreatment of the cells with a 1:4 dilution of peak III material was tested. The two-step procedure pre viously outlined was used. Cells pretreated with Tris-A buffer only were used as control samples. The second incubation, with antiIgE in Tris-ACM, was performed on all samples. After various time intervals, the reaction was stopped by addition of 1 ml ice-cold Tris-A, and the samples were cen trifuged in the cold. Figure 2 shows the re sults of three of these experiments. It can be clearly seen that peak Ill-treated cells re leased greater amounts of histamine than did the buffer-treated cells. This was true when optimum (1.5 ,ug/ml) or suboptimum (0.5 jug/ml) amounts of anti-IgE were used. Pretreatment with peak III material also re duced the lag period. These two effects of peak III treatment were observed in all the
Downloaded by: Univ. of California Santa Barbara 128.111.121.42 - 3/7/2018 3:20:29 AM
Histamine Release from Human Leukocytes
Petersson/Stâlenheim
182
b
Fig. 2. Time course of anti-IgE(1.5 ug/ml in a, b, and 0.5 «g/ml in c)-induced histamine release from buffer (O) and peak III ( # ) pretreated leua
c
kocytes. a, b, and c represent experiments with leukocytes from different donors.
b
Anti-lgE, mg pnotein/ml
Fig. 3. Histamine release induced with varying amounts of anti-IgE from buffer and serum fraction pretreated leukocytes. O = Release from buffer-pretreated leukocytes; • = release from peak III (a) or peak II (b) pretreated leukocytes.
experiments performed. In the experiment shown in figure 2c maximum level was reached more rapidly with peak III than with buffer-treated cells. This effect was less pronounced in other experiments. In another set of experiments the ability of different concentrations of anti-IgE to re lease histamine from leukocytes pretreated with peak II and peak III material was in
vestigated. The first-step incubation was with Tris-A (control), peak II material or peak III material. The second-step incuba tion was with varying amounts of anti-IgE in Tris-A CM, for 40 min at 37 °C. The re sults are shown in figure 3 from which it can be seen that, at all anti-IgE concentra tions used, pretreatment with peak II or peak III material resulted in a marked in crease in histamine release, compared with the control set of samples. In experiments where otpimum concentrations of anti-IgE released high amounts of histamine (70-100°/o) from buffer pretreated leuko cytes, the enhancing effect of peak II or peak III pretreatment was more pronounced when sub- or supraoptimum amounts of anti-IgE were used (fig. 3b). In the experiments shown in table I, leu kocytes were incubated in buffer, autolo gous serum, peak II material or peak III material for 40 min at 37 °C in the first step. In the second step the samples were treated with optimum amounts of anti-IgE for 40 min at 37 °C. The cell donors for these experiments were selected with a view to obtaining less than 50°/o histamine release
Downloaded by: Univ. of California Santa Barbara 128.111.121.42 - 3/7/2018 3:20:29 AM
a
183
Histamine Release from Human Leukocytes
Experi- Anti-IgE (1.5 /tg/ml)-induccd histamine ment release (in %) from leucocytes pretreated at No. 37°C for 40 min with: Tris-A serum 1:2 peak 111:2 peak 111 1:4 1 2 3 4 5 6 7 8 9 10 11
16 18 23 24 24 24 30 39 47 48 48
24 25 28 30 not tested 32 40 not tested 64 59 67
55 47 50 38 67 58 81 82 87 90 not tested
52 52 65 51 65 66 80 81 86 90 95
from buffer-pretreated cells on treatment with optimum amounts of anti-lgE. Serum diluted 1:2 was used since dilutions between 1:1 and 1:32 gave essentially the same de gree of enhancement. Dilutions 1:64 to 1:512 had smaller effects. Another reason for choosing this serum concentration was that serum diluted 1:2 gave maximum resensilizalion of protein A-induced histamine release [16]. Pretreatment of the leukocytes with peak II or peak III material resulted in a very pronounced enhancement of the anti-IgEinduced release. If the increase was calcu lated as percent of the release induced from buffer-treated cells, the enhancement ob tained with peak II varied between 58 and 244% with an average of 135. For peak III the corresponding figures were 83 and 225% with an average of 145. Serum, on the other hand, had a very low capacity to stimulate anti-IgE-induced release. The enhancement was between 22 and 50% with an average of 33%.
Study of the Effect of Peak II ami Peak III Pretreatment on the Antigen-Induced Release Firstly, leukocytes from subjects suffer ing from an allergy were treated with buffer (control) or peak III material. Secondly, they were incubated with varying amounts of antigen diluted in Tris-ACM (for 40 min at 37 °C). The antigens used were pollen extracts of birch (diluted to 10'7-10"3) and timothy (diluted to 10‘14-1 0 '3). The results are shown in figure 4 from which it can be seen that pretreatment with peak III material markedly enhanced the antigen-induced his tamine secretion from the leukocytes of these donors. Figure 4b shows an experi ment where optimum amounts of antigen released high amounts of histamine (79%) from buffer pretreated cells also. However, at sub- or supraoptimum concentrations of antigen, the stimulating effect of pretreat ment with peak III material was more pron ounced. Antigen-triggered release was enhanced as an effect of peak II pretreatment (experia
Dilution o f birchpollen extract
b
Dilution o f tim othypollen extract
Fig. 4. Histamine release induced with birch pollen antigen (a) or timothy pollen antigen (b) from buffer (O) or peak III (# ) pretreated leuko cytes.
Downloaded by: Univ. of California Santa Barbara 128.111.121.42 - 3/7/2018 3:20:29 AM
Table I. Anti-IgE-induced histamine release from buffer, serum, peak II and peak lll-pretrcated leuko cytes
Petersson/Stâlenheim
184
Experiment No.
1 2 3 4
Anti-IgG (1:200)-induced histamine release (in %) from leukocytes pretreated at 37 °C for 40 min with: Tris-A
peak III 1:4
8 23 37 40
51 55 75 87
Table III. Anti-IgE-induced histamine release from buffer, peak II, peak II-3 and trypsinated peak II-3 pretreated leukocytes Experi- Additive in step 1 ment (pretreatment at 37 °C No. for 40 min)
Histamine release in step 2 % (test with antiIgE, 1.5 MS/m\)
1
buffer 24 peak II 1:2 67 peak II-3 200 MS protein/ml 71 peak II-3 100 fig protein/ml 67
2
buffer peak II 1:2 peak II-3 100 MS protein/ml peak II-3 10 MS protein/ml peak II-3 1 MS protein/ml
ments not shown), in the same way as antiIgE-initiated release.
21 40 38 29 23
3
Study of the Effect of Peak 111 Pretreatment on Anti-IgG-induced Histamine Release Using the experimental procedure al ready described the effect of peak III pre treatment on anti-IgG-triggered release was investigated. In the second step, a 1:200 di lution of anti-IgG was used [16]. The ex periments in table II show that antiIgG-triggered release was also markedly stimulated. If the increase was calculated as percent of the release induced from buffertreated cells, the enhancement obtained var ied between 102 and 538%.
buffer 28 peak II-3 200 MS protein/ml 52 peak IT-3 100 MS protein/ml 51 peak II-3 10 MS protein/ml 33 peak II-3 1 m s protein/ml 26 trypsin-digested peak II-3 35 200 MS protein/ml buffer containing trypsin and trypsin inhibitor 26
4
buffer 31 peak II-3 200 MS protein/ml 62 trypsin-digested peak II-3 200 MS protein/ml 36 buffer containing trypsin and trypsin inhibitor 33
Capacity of Native, and TrypsinDigested, Peak 11-3 Material to Enhance Anti-1gE-Induced Release In a series of experiments, leukocytes were preincubated with peak II-3 material (from QAE-Sephadex) before the treat ment with anti-IgE. Table III shows that peak II-3 material with protein concentra tions of 200 or 100 pg per ml had a high ca pacity, even in comparison with unseparat-
ed peak II material, to enhance anti-IgE-induced release. If peak II-3 material was di luted to 10 pg protein/ml it still had some capacity to stimulate the release, while 1 pg protein/ml was inactive. On the other hand, trypsin-digested peak II-3 material (200 pg protein/ml) had a very low capacity, compared with undigested ma terial to enhance anti-IgE induced release (experiments 3 and 4, table III).
Downloaded by: Univ. of California Santa Barbara 128.111.121.42 - 3/7/2018 3:20:29 AM
Table II. Anti-IgG-induced histamine release from buffer and peak 111-pretreated leukocytes
Discussion The previously observed decrease in protein A reactivity, as a result of washings, suggested that some factor necessary for his tamine release was eluted from the leuko cytes [14, 16]. In 31 out of 43 experiments, repeated washings completely removed pro tein A reactivity and in the remaining ex periments caused a greater than 50% de crease. On the other hand, anti-IgE-induced release was rather resistant to repeated washings; although it had been found earlier [7] that a factor with capacity to enhance antigen-triggered release was eluted from whole viable leukocytes during 30 min in cubation at 15-20 °C. Repeated washings caused a greater than 50% decrease in anti-IgE-induced release in only 2 out of 28 experiments. A decrease of less than 10% was detected in 16 of the experiments. Fur thermore, it has been recently reported that an 85-100% decrease in protein A reactivi ty could be obtained without any decrease in release triggered by anti-IgE or anti-IgG [14, 16]. Histamine secretion from repeat edly washed leukocytes (i.e. lacking reactiv ity to protein A alone) could also be in duced by protein A in complex with IgG [14]. The differences in susceptibility to washings indicated that anti-IgE, anti-IgG, protein A and protein A-IgG complexes have different requirements for the factors which are capable of modulating the degree of histamine release. Since some leukocyte preparations from donors suffering from an allergy released low amounts of histamine (below 30-50%), even when treated with optimum antigen concentrations, Lichtenstein and Osier [10] discussed in their original report the possi ble existence of some nonimmunoglobulin
185
factor necessary for antigen-triggered his tamine secretion. In the middle of the 1960s, the same authors reported that in corporation of 10% human serum in the an tigen solution enhanced the histamine re lease induced by suboptimum amounts of antigen. At optimum antigen concentrations no effect was seen [11]. They managed to isolate from serum a lipoprotein and a low molecular weight factor with capacity to po tentiate antigen-induced release [7]. Ishizaka et al. [6] showed that anti-IgE-triggered histamine release from the leukocytes of both allergic and normal donors was en hanced by some serum component not relat ed to the complement system. This result was confirmed later by Grant and Lichten stein [5]. In addition to the stimulation of IgE-mediated release, serum had the capaci ty to effectively resensitize repeatedlywashed leukocytes thus restoring their ca pacity to liberate histamine on protein A treatment [16]. The active material was eluted both in protein peaks II and III when serum was fractionated on Sephadex G-200. This result, together with the fact that di alysis did not decrease the capacity of serum to resensitize leukocytes, indicated that the factors were not identical with the lipopro tein or the low molecular weight factor stud ied by Lichtenstein and coworkers [7, 16]. Material with resensitizing capacity was not bound to a protein A-Sepharose column, and washings of resensitized leukocytes again decreased their capacity to release his tamine on protein A treatment [16]. These results indicated that the active components must exert their effect after binding to the leukocytes. The possibility was considered that the factors could have a stimulating ef fect on immunologically induced histamine secretion in general. Thus, the capacity of
Downloaded by: Univ. of California Santa Barbara 128.111.121.42 - 3/7/2018 3:20:29 AM
Histamine Release from Human Leukocytes
peak II and III material to enhance IgE-mediated release has been investigated. Other investigators have routinely been including 1-10% serum in their antigen or anti-IgE solutions when studying the poten tiating effect of serum on the histamine re lease induced with these agents [5, 6, 11]. Since the active components in peak II and III material were bound to the leukocytes, it was possible in this, and in a recent report, to study the stimulating effect of serum and scrum fractions in a two-step procedure. The advantage of such a method is that the leukocytes both in control and test samples are treated with the triggering agent under similar conditions, for example, in the pres ence of protein in the same concentration. Use of the two-step procedure therefore rules out stimulation by an unspecific prote in effect; a possibility discussed in an earlier work [11]. Whether washing resulted in a decrease in IgE-mediated histamine release or not, pretreatment of repeatedly washed leuko cytes with peak II and III material actually had a striking effect on histamine release in duced by both anti-IgE and antigen (fig. 3, 4). Enhancement was obtained at all antigen and anti-IgE concentrations tested (fig. 3, 4). If optimum amounts of the releasing agents released high quantities of histamine from buffer-pretreated cells, the effect of peak II or peak III pretreatment could not, of course, have been so pronounced. In this case the stimulating effect was better visual ized with sub- or supraoptimum amounts (fig. 3, 4). Furthermore, the time-course ex periments revealed that peak III pretreat ment, besides resulting in a higher histamine release, also reduced the length of the lag period (fig. 2). Autologous serum enhanced anti-IgE-
Petersson/Stálenheim
triggered release but the effect was small compared with the effect of peak II or peak III pretreatment. Table I shows a series of II experiments, where the stimulating effect obtained as a result of pretreatment with peak II material, peak III material or autol ogous serum was studied. The blood donors were selected with a view to obtaining leu kocyte preparations releasing less than 50% of their histamine when buffer-pretreated cells were incubated with optimum amounts of anti-IgE. Peak II and peak III material gave, with rather few exceptions, compara ble enhancement of the anti-IgE-induced re lease. The average stimulation by peak II was 135% and by peak III, 145%. The cor responding value for serum was only 33%. In these experiments, serum diluted 1:2 was used. Undiluted serum or serum diluted more than 1:2 (down to 1:512) did not re sult in better stimulation. The ineffective ness of serum in producing an enhancement comparable with that of peak II and peak III material, indicates the possible existence in whole serum, of histamine release inhibi tors or inactivators of the enhancing factors. This has in fact been discussed earlier in connection with the lipoprotein, which had the capacity to increase antigen-induc ed histamine release from human leukocy tes [7], Furthermore, a few experiments showed that peak III-pretreated leukocytes released considerably higher amounts of histamine than did buffer-preincubated cells, during treatment with optimum amounts of antiIgG (table II). When peak II material was further frac tionated on a QAE-Sephadex column, ac tive material was obtained in a pool with low protein content. 100 /
