Agents Actions, 35 (1992)
0065-4299/92/040170-09 $1.50+ 0.20/0 9 1992 Birkh/iuser Verlag, Basel
Hyperosmolarity selectively enhances IgE-receptor-mediated histamine release from human basophils B.W. Nielsen, T. Bjerke, T. M. E. Damsgaard, T. Herlin, K. Thestrup-Pedersen 1 and P.O. Schiotz Department of Pediatrics and 1Dermato-Venerology,UniversityHospital of Aarhus, DK-8000 Aarhus C, Denmark
Abstract
Increased osmotic pressure has been reported to cause non-cytotoxic histamine release (HR) from human basophils, as well as a potentation of H R induced by anti-IgE. In this study, the effects of hyperosmolar Na-K-acetate (300-600 mOsm/kg H 2 0 ) on H R was studied in washed human blood cells from newborns, adult volunteers and patients with severe atopic dermatitis. These three patient groups represesented 3 very distinct populations with respect to total plasma IgE content, medians were < 0.2 IU/ml, 20.5 IU/ml and 2508 IU/ml, respectively. Increasing osmolarity to 500 mOsm/kg H 2 0 caused little H R in the absence of other stimuli, whereas at 600 mOsm/kg H 2 0 a significant increase in spontaneous H R was seen. The H R induced by anti-IgE and Concanavalin A, acting through the IgE-receptor, was increased approximately twofold at 500 mOsm/kg H20. Responses were highly correlated to results at 300 mOsm/kg H20. The use of 600 mOsm/kg H 2 0 buffers caused a further increase in most, but not all blood samples. The potentiation of IgE-receptor-mediated H R when using hyperosmolar media was clearly independent of plasma IgE contents, and did not change the concentration-response to anti-IgE. In contrast, H R induced by the IgE-receptor-independent stimuli, Formyl-met-leu-phe and calcium ionophore A 23187, were not enhanced at all by incrased osmotic pressure. We conclude, that hyperosomolar media selectively enhance IgE-receptor-mediated HR. The use of hyperosmolar media may therefore be beneficial in a diagnostic application of washed blood H R assays used in allergydiagnosis.
Introduction
degranulation occurred without other stimuli [3,
It was reported several years ago by H o o k & Siraganian, that the use of a Na- or K-acetate buffer could increase the responsiveness of human basophils to anti-IgE [1]. Furthermore, a dose-dependent non-cytotoxic enhancement of IgE-mediated histamine release from human basophils was reported, when anti-IgE was added to the cells in a hyperosmolar environment [1, 2]. If basophils were subjected to osmolarities exceeding 500600 mOsm/kg H20, a significant non-cytotoxic
Based on these findings, we have employed a hyperosmolar PIPES-buffered Na-K-acetate buffer in the microfibre-based histamine release assay. This assay has been used for the study of basophil histamine release (HR) in washed blood preparations from different patient categories [5-7]. In all these studies, H R could be measured in more than 90% of the blood samples examined. The present study examines the effects of hyperosmolar Na-K-acetate buffer on H R induced by IgE-
4].
Agents Actions, 35 (1992) mediated as well as non-IgE-dependent stimuli. Furthermore, the effects are studied in blood samples from three distinct populations, i.e. cord blood, healthy adults and patients with severe atopic dermatitis.
Materials and methods
Reagents PIPES-acetate buffer. Four different PIPES-acetate buffers were used, differing only in Na-acetate concentration and hence in osmolarity. The buffers were termed, according to their osmolarity, as follows: PA300 PA400 PA500 and PA600
(140 m M Na-acetate, 305 mOsm/kg H20), (186 m M Na-acetate, 401 mOsm/kg H20), (232 m M Na-acetate, 497 mOsm/kg H20) (278 m M Na-acetate, 593 mOsm/kg HzO ).
These buffers all contained PIPES (PiperazineN,N'-bis[2-ethanesulfonic acid]) 10mM, K-acetate 5 mM, as well as CaC1 0.6 m M and MgC1 1.1 mM, heparin 7.5 IE/ml, glucose 2.8 m M and human serum albumin 0.16 g/l, and were adjusted to pH 7.40 with Tris[hydroxymethyl]aminomethane.
Preparation of washed blood samples. Three separate series of experiments were conducted: Adult volunteers: Blood samples, containing 20 ml of blood anticoagulated with 3.8 m M of EDTA, were obtained by venipuncture from 22 adult volunteers, none of whom were taking any form of medication at the time of blood sampling. Plasma was removed after high-speed centrifugation (2755 g), and the blood was reconstituted to original volume with PA300. Each sample was then split equally into 4 ten-milliliter tubes. The cells in each tube were resuspended and washed four times in one of the four different PA-buffers. The same PA-buffer was employed for the H R determinations as well as for diluting the secretagogues employed. Final osmolarity was measured in the supernatant from the last washing step, using an Advanced Instruments Wide-Range osmometer, and was found to be identical to that of the PA-buffer employed.
171 Cord blood: Cord blood samples, containing 10 ml of blood anticoagulated with EDTA (10mM), were obtained from 50 consecutive newborns. The washed blood was prepared as above, using only the PA300 and PA500 due to the limited amount of blood available. Atopic dermatitis patients: A smaller group (n = 10) of blood samples from patients with severe atopic dermatitis were also included, as basophil granulocytes from such patients have been reported to show high spontaneous H R [8]. Two 10ml-blood samples, drawn from the same venipuncture, were obtained. One sample was anticoagulated with EDTA 5 m M preventing Ca + +dependent HR. In the other sample, the nonchelating agent Li-Heparin (15 IU/ml), was employed. Washed blood cells were prepared as above, employing only the PA300 and PA500 buffers.
Secretagogues. Rabbit-anti-human IgE (anti-IgE; Behring-Werke, Marburg, Germany) in final concentrations of 0.04, 0.4, 4, 40 and 400 IU/ml, Concanavalin A (ConA; Pharmacia, Uppsala, Sweden) 10.4, 20.8, 41.6 and 83.2 Ixg/ml, N-formyl-methionyl-leucyl-phenylalanine (FMLP; Sigma, St. Louis, MO) 10 -8, 10 - 6 and 10 - 4 M, and the calcium ionophore A23187 (Sigma, St. Louis, MO) at 0.25, 0.5, 1.0 and 2.0 ~tM were employed. Histamine release assay. Histamine release was determined by the microfibre-based assay as previously described [6, 9, 10]. All histamine release determinations were performed in triplicate. All H R values are expressed as the difference (washed blood cells+secretagogue) - (washed blood cells + PA-buffer). Estimation of background histamine release was performed only in the blood samples from adult volunteers, calculated as the difference between opthaldialdehyde fluorescence in wells containing washed blood cells + PA-buffer, and wells containing only the PA-buffer. The intra-assay variation was estimated on the basis of the numerical difference between 5 separate triplet-groups of wells containing unstimulated blood, evenly distributed across the microfibrecoated 96-well plate. IgE-measurements. Total plasma IgE content was quantitated by a sandwich-ELISA-method as currently used in clinical practice.
172
Statistics. Non-parametric statistical methods were employed for comparison of group data, including Kruskal-Wallis' analysis of variance by ranks and Wilcoxon's test for paired comparisons. When calculating the correlations between H R with different PA-buffers in each of the three series of experiments, a simple regression analysis, based on the actual values, was employed. Values of p < 0.05 were considered significant.
Agents Actions, 35 (1992) H i s t a m i n e Release ( n g / m l Blood) A
6o -
A
45 AAA
a0 A
A
A A
A
A ^A^A
AAA Results
The three experimental series each represented a distinct population with respect to the total plasma IgE content. In the 22 blood samples from adult volunteers, the median total IgE content was 20.5 IU/ml (range: < 5-414 IU/ml), whereas in cord blood plasma the median was
0065-4299/92/040170-09 $1.50+ 0.20/0 9 1992 Birkh/iuser Verlag, Basel
Hyperosmolarity selectively enhances IgE-receptor-mediated histamine release from human basophils B.W. Nielsen, T. Bjerke, T. M. E. Damsgaard, T. Herlin, K. Thestrup-Pedersen 1 and P.O. Schiotz Department of Pediatrics and 1Dermato-Venerology,UniversityHospital of Aarhus, DK-8000 Aarhus C, Denmark
Abstract
Increased osmotic pressure has been reported to cause non-cytotoxic histamine release (HR) from human basophils, as well as a potentation of H R induced by anti-IgE. In this study, the effects of hyperosmolar Na-K-acetate (300-600 mOsm/kg H 2 0 ) on H R was studied in washed human blood cells from newborns, adult volunteers and patients with severe atopic dermatitis. These three patient groups represesented 3 very distinct populations with respect to total plasma IgE content, medians were < 0.2 IU/ml, 20.5 IU/ml and 2508 IU/ml, respectively. Increasing osmolarity to 500 mOsm/kg H 2 0 caused little H R in the absence of other stimuli, whereas at 600 mOsm/kg H 2 0 a significant increase in spontaneous H R was seen. The H R induced by anti-IgE and Concanavalin A, acting through the IgE-receptor, was increased approximately twofold at 500 mOsm/kg H20. Responses were highly correlated to results at 300 mOsm/kg H20. The use of 600 mOsm/kg H 2 0 buffers caused a further increase in most, but not all blood samples. The potentiation of IgE-receptor-mediated H R when using hyperosmolar media was clearly independent of plasma IgE contents, and did not change the concentration-response to anti-IgE. In contrast, H R induced by the IgE-receptor-independent stimuli, Formyl-met-leu-phe and calcium ionophore A 23187, were not enhanced at all by incrased osmotic pressure. We conclude, that hyperosomolar media selectively enhance IgE-receptor-mediated HR. The use of hyperosmolar media may therefore be beneficial in a diagnostic application of washed blood H R assays used in allergydiagnosis.
Introduction
degranulation occurred without other stimuli [3,
It was reported several years ago by H o o k & Siraganian, that the use of a Na- or K-acetate buffer could increase the responsiveness of human basophils to anti-IgE [1]. Furthermore, a dose-dependent non-cytotoxic enhancement of IgE-mediated histamine release from human basophils was reported, when anti-IgE was added to the cells in a hyperosmolar environment [1, 2]. If basophils were subjected to osmolarities exceeding 500600 mOsm/kg H20, a significant non-cytotoxic
Based on these findings, we have employed a hyperosmolar PIPES-buffered Na-K-acetate buffer in the microfibre-based histamine release assay. This assay has been used for the study of basophil histamine release (HR) in washed blood preparations from different patient categories [5-7]. In all these studies, H R could be measured in more than 90% of the blood samples examined. The present study examines the effects of hyperosmolar Na-K-acetate buffer on H R induced by IgE-
4].
Agents Actions, 35 (1992) mediated as well as non-IgE-dependent stimuli. Furthermore, the effects are studied in blood samples from three distinct populations, i.e. cord blood, healthy adults and patients with severe atopic dermatitis.
Materials and methods
Reagents PIPES-acetate buffer. Four different PIPES-acetate buffers were used, differing only in Na-acetate concentration and hence in osmolarity. The buffers were termed, according to their osmolarity, as follows: PA300 PA400 PA500 and PA600
(140 m M Na-acetate, 305 mOsm/kg H20), (186 m M Na-acetate, 401 mOsm/kg H20), (232 m M Na-acetate, 497 mOsm/kg H20) (278 m M Na-acetate, 593 mOsm/kg HzO ).
These buffers all contained PIPES (PiperazineN,N'-bis[2-ethanesulfonic acid]) 10mM, K-acetate 5 mM, as well as CaC1 0.6 m M and MgC1 1.1 mM, heparin 7.5 IE/ml, glucose 2.8 m M and human serum albumin 0.16 g/l, and were adjusted to pH 7.40 with Tris[hydroxymethyl]aminomethane.
Preparation of washed blood samples. Three separate series of experiments were conducted: Adult volunteers: Blood samples, containing 20 ml of blood anticoagulated with 3.8 m M of EDTA, were obtained by venipuncture from 22 adult volunteers, none of whom were taking any form of medication at the time of blood sampling. Plasma was removed after high-speed centrifugation (2755 g), and the blood was reconstituted to original volume with PA300. Each sample was then split equally into 4 ten-milliliter tubes. The cells in each tube were resuspended and washed four times in one of the four different PA-buffers. The same PA-buffer was employed for the H R determinations as well as for diluting the secretagogues employed. Final osmolarity was measured in the supernatant from the last washing step, using an Advanced Instruments Wide-Range osmometer, and was found to be identical to that of the PA-buffer employed.
171 Cord blood: Cord blood samples, containing 10 ml of blood anticoagulated with EDTA (10mM), were obtained from 50 consecutive newborns. The washed blood was prepared as above, using only the PA300 and PA500 due to the limited amount of blood available. Atopic dermatitis patients: A smaller group (n = 10) of blood samples from patients with severe atopic dermatitis were also included, as basophil granulocytes from such patients have been reported to show high spontaneous H R [8]. Two 10ml-blood samples, drawn from the same venipuncture, were obtained. One sample was anticoagulated with EDTA 5 m M preventing Ca + +dependent HR. In the other sample, the nonchelating agent Li-Heparin (15 IU/ml), was employed. Washed blood cells were prepared as above, employing only the PA300 and PA500 buffers.
Secretagogues. Rabbit-anti-human IgE (anti-IgE; Behring-Werke, Marburg, Germany) in final concentrations of 0.04, 0.4, 4, 40 and 400 IU/ml, Concanavalin A (ConA; Pharmacia, Uppsala, Sweden) 10.4, 20.8, 41.6 and 83.2 Ixg/ml, N-formyl-methionyl-leucyl-phenylalanine (FMLP; Sigma, St. Louis, MO) 10 -8, 10 - 6 and 10 - 4 M, and the calcium ionophore A23187 (Sigma, St. Louis, MO) at 0.25, 0.5, 1.0 and 2.0 ~tM were employed. Histamine release assay. Histamine release was determined by the microfibre-based assay as previously described [6, 9, 10]. All histamine release determinations were performed in triplicate. All H R values are expressed as the difference (washed blood cells+secretagogue) - (washed blood cells + PA-buffer). Estimation of background histamine release was performed only in the blood samples from adult volunteers, calculated as the difference between opthaldialdehyde fluorescence in wells containing washed blood cells + PA-buffer, and wells containing only the PA-buffer. The intra-assay variation was estimated on the basis of the numerical difference between 5 separate triplet-groups of wells containing unstimulated blood, evenly distributed across the microfibrecoated 96-well plate. IgE-measurements. Total plasma IgE content was quantitated by a sandwich-ELISA-method as currently used in clinical practice.
172
Statistics. Non-parametric statistical methods were employed for comparison of group data, including Kruskal-Wallis' analysis of variance by ranks and Wilcoxon's test for paired comparisons. When calculating the correlations between H R with different PA-buffers in each of the three series of experiments, a simple regression analysis, based on the actual values, was employed. Values of p < 0.05 were considered significant.
Agents Actions, 35 (1992) H i s t a m i n e Release ( n g / m l Blood) A
6o -
A
45 AAA
a0 A
A
A A
A
A ^A^A
AAA Results
The three experimental series each represented a distinct population with respect to the total plasma IgE content. In the 22 blood samples from adult volunteers, the median total IgE content was 20.5 IU/ml (range: < 5-414 IU/ml), whereas in cord blood plasma the median was
