Leukotriene C4 production by normal-density and low-density eosinophils of atopic individuals and other patients with eosinophilia

Leuk iene C4 produ ion by normaland low-density eosinophils of atopic individuals and other patients eosinophilia H. F. Kauffman, Ph.D., B. van der Belt, J. G. R. de Mon9 H. Boelens, G. H. Ko~ter, M.D.,* and K. de Vries, M.D.

ity

M.D.,

Groningen, The Netherlands

With the use of a percoU gradient separation procedure, eosinophils of individuals" with asttvna and with allergy could be separated into normal- and low-density cell fractions. The presence of low-density eosinophils possibly reflects an ongoing process of activation of these cells induced by the allergic reaction. Ca-ionophore-induced leukotriene (LT) C4 production, in the absence of added substrates, demonstrated a decreased potency for LT generation by low-density eosinophils compared with the LT generation of normal-density cells (57 -+- 33 ng and 103 +- 44 ng per 106 cells, respectively). In contrast with the Ca-ionophore-induced LT formation, incubations with serum-treated zymosan in the presence of glutathione demonstrated higher productions of LTC~ with the low-density eosinophilic subpopulation compared with normal-density cells. This is compatible with a possibly higher expression of complement C3b receptors' on the low-density eosinophils. Total arylsulfatase contents demonstrated that low-density eosinophils are not degranulated with respect to their small granules. Although release of the large granules by low-density eosinophils cannot be excluded, electron.microscopy studies indicated that degranulation is not the only (or major)factor that determines' the density of the various eosinophilic subpopulations. (J ALLERGYCLIN IMMUNOL1987:79:611-9

The role of eosinophils in asthma and other atopic diseases has attracted much attention for several years. There is a fundamental change from the eosinophil being regarded as a cell with beneficial effects regulating mast cell degranulation 1' 2 to being regarded as a cell with potential harmful effects in atopic diseases leading to bronchial asthma. The possible harmful involvement of eosinophils is supported by several observations: (1) The cytotoxic substance major basic protein from eosinophils is f o u , d in sputum of individuals with asthma, demonstrating an inverse relationship with lung function. 3 (2) Release of the eosinophil cationic protein into the circulation was observed after allergen challenge during the early obstructive reaction. 4 (3) Data from the literature indicate that the eosinophil is the main in-

From the Departments of Altergology and *Pulmonology, Groningen, The Netherlands. Supported by The Netherlands Asthma Fonds Grant No. 82.25. Received for publication Dec. 30, 1985. Accepted for publication Sept. 20, 1986. Reprint requests: H. F. Kauffman, Ph.D., Department of Atlergology, Clinic for Internal Medicine, State University Hospital, Oostersingel 59, 9713 EZ Groningen, The Netherlands.

Abbreviations used LT: Leukotriene LTC~: Leukotriene C~ LTB4: Leukotriene B, AS: Arylsulfatase BSA: Bovine serum albumin Tyrode-ACM: Tyrode buffer supplied with Ca 2 Mg 2~. and BSA STZ: Serum-treated zymosan EM: Eleetron-miemscopy(ic) PMN: Polymorphonuclear leukocyte

filtrating cell in asthma s and was demonstrated to be present in elevated numbers in bronchoalveolar lavages of subjects with asthma. 6s From recent studies evidence has been obtained that eosinophils play an important role in the development of the late obstructive reaction after allergen inhalation. It was demonstrated that the eosinoplail is the main infiltrating cell during the first phase o f the late obstructive reaction, whereas neutrophil numbers were not different from control values. ~ This observation was supported by a recent study of late reac-

611

612

Kauffman et al.

J. ALLERGYCLIN. tMMUNOL. APRIL 1987

flecting different states of activity. 12' ,3 Low-density eosinophils are also characterized by an increased expression of Fc gamma '3 and Fc epsilon m4membrane receptors and produce more LTC4 after incubation with IgG-coated particles. 15 The present study investigates (1) if there is a heterogeneity in eosinophils of individuals with asthma and with eosinophilia and (2) if this heterogeneity is reflected in differences in AS content, EM characteristics, and the potency of LT formation between normal- and low-density eosinophils.

.0zJ #

1oo 8O R om

6O

t,0 20

I

I

I

I

100 80

9

+HI o

>= o

I

x

60

o

9

o~ 60 20 4)v-~ia

u+

I

I

I

g

I

lgl

I

I

FIG. 1. Purity and recovery of eosinophilic leukocytes by Percoll gradient centrifugation of allergic individuals with asthma. Symbols represent eosinophils of individual patients described in Table I. Connected symbols are individuals of whom most cells were in low-density state;/, II, and/11 are cells recovered from the interphase, as described under Material and Methods; / and II are lowdensity eosinophils; III is normal-density eosinophils; Illl is Percoll layer with intermediate density of 1,102.

tions in red cedar asthma demonstrating similar infiltrations of eosinophils during the late reaction, whereas infiltration of neutrophils was observed 25 hours after allergen challenge when airway inflammation was decreasing. 9 Recently, eosinophils have also been demonstrated to produce the spasmogenic LTC4, whereas neutrophils produced the chemotactic factor LTB4. ~~ 11 Therefore, it was proposed that eosinophils are attracted by chemotactic substances during mast cell degranulation in early asthmatic reactions. This is possibly followed by an active liberation of the spasmogenic LTs by eosinophils during the late obstructive p h a s e ) Studies with purified eosinophils from individuals with eosinophilia have demonstrated that eosinophils can occur as cells of varying densities re-

MATERIAL AND METHODS Purification of e o s i n o p h i l s

Thirty to forty milliliters of venous blood was drawn from either healthy volunteers, allergic patients with obstructive reactions, or patients with eosinophilia of other etiology. Sedimentation of erythrocytes in the EDTA blood samples was performed by addition of 12 ml of gelatine (3% in 0!. 15 mol/L NaC1). The huffy coat cells, collected after 60 minutes, were centrifuged, resuspended in 2 • 6 ml of RPMI medium, and separated on Ficoll-Paque (density 1.077 gm/ml, Pharmacia Fine Chemicals, Uppsala, Sweden). The PMN cell pellet was washed twice in RPMI at 4 ~ C and reconstituted in 4 ml of autologous plasma that was made slightly hypertonic by the addition of 200 ~,1 of sodium chloride (0.75 mol/L). Two milliliter portions of this cellplasma suspension were layered on Percoll gradients ~at were prepared as described earlier '6 with densities of 1.085, 1.094, 1.110, and 1.115 gm/ml (3 ml volumes of each layer). Cell numbers applied to each gradient tube did not exceed 4.107 cells. After centrifugation (20 minutes at 1000 x g, at 20 ~ C) the cell interfaces were carefully collected, washed in Tyrode medium (two times), and reconstituted in Tyrode supplied with 1.8 mmol/L CaCI2, 0.5 mmol/L MgC12, and 0.1% BSA (Tyrode-ACM). The cell populations sampled at the densities 1.085, 1.094, and 1.110 (Fig. 1, bands 1, H, and III, respectively) consisted of purified neutrophils, neutrophils and low-density eosinophils, and purified normal-density eosinophils, respectively. Eosinophils of healthy individuals purified by this procedure are mainly recovered in band III, demonstrating minor numbers in band II (recovery ~< 10%), and no eosinophils in band I. Cell c o u n t i n g a n d v i a b i l i t y

Cytocentrifuge smears were prepared from each fraction by mixing 10 ixl of cell suspension with 5 Ixl of autologous plasma, and differential counts were performed after staining with May-Grtinwald/Giemsa. Total cell count was estimated in a Coulter (Coulter Electronics, Inc., Hialeah, Fla.) counter. Viability tests with trypan blue exclusion were excellent and exceeded 96%. F u n c t i o n tests

Production of LTC, by the different cell populations (band I, II, and III) was tested with 1.106 cells per milliliter

VOLUME79

Leukotriene C4 production by eosmophiis

NUMBER 4

613

TABLE I. Percentage of whole blood eosinophils and total IgE of a group of allergic individuais* Patient No.

Symbol in Fig. I t

% Eosinophils

Total IgE (IU/ml)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

9 C) 9 A 9 [] 9 -] + / \ x (1) @ [~

16.0 0.5 6.7 8.0 6.0 6.0 7.0 6.0 1.3 17.0 8.0 7.0 4.0 7.0 7.5 16.3

500 7. 32 NP NP 1500 4~ NP 680 640 NP 54 110t) 270 135 39{}

NP = not performed. *All individuals demonstrated manifestations of asthma and/or rhinitis and were positive in skin tests with common allergens (mainly house dust mites). No steroid therapy was used except patient 2 who used 2.5 mg of prednisolone daily +Symbols correspond with the separation profiles of Fig. 1.

of q'yrode-ACM. Cells were preincubated for 15 minutes at 37 ~ C, and then 0.5 rnl portions were added to tubes containing Ca ionophore (A23187, final concentration 20 v.mol/L, Sigma Chemical Co., St. Louis, Mo.). The cells were incubated for an additional 15 minutes at 370 C. Complement receptor (C3b-dq~-~ndent) LTC, generation was tested with STZ as a ~ l a n t after the incubation conditions as des c r i ~ d by Bruynzeel et al. t7 In short, 1.1@ cells per milliliter of calcium and magnesium containing Tyrode buffer without lISA together with 5 mmol/L of glutathione were incubated with I mg/ml of STZ for 30 minutes at 37 ~ C. The S T was prepared according to the method of Busse et al.'~ Liberation of mediators was terminated by placing the cells on ice and by c e n t r i ~ i n g for 10 minutes at 600 x g at 4 ~C. The supematants were directly frozen in liquid nitrogen and stored at - 80 ~ C until they were analyzed. The supematants were tested for LTC4 by use of the radioimmunoassay kit (New England Nuclear, Boston, Mass,) following the instructions of the manufacturers. When incubations were performed in buffer supplied with BSA (0.1%), additional calibration curves were run with 100 I~l of stand~ds diluted in Tyrode-ACM. The ~ s e n e e of BSA (0.03% end concentration) caused a s l i ~ t shift of the curve to the left (10%), compared with the curve with the use of standard assay conditions. A1t h o u # this radioinmmnoassay for LTC, demonstrated crossreactivity with LTD,, this did not influence the quantity of LTC4 ~ u c t i o n , since eosinophils were demonstrated to produce exclusively LTC,, and neutrophils produce LTB, only." LTC4 production by the eosinophils was verified for some cell supematants in cooperation with J. Verhagen and P. L. B. Bruynzeel at the Department of Bio-Organic Chem-

istry, State University of Utrecht, with the reversed-phase high-performance liquid chromatography procedures as described previously. TM ,7 Total AS was tested by freezing and thawing the cell suspensions three times and rrmastwing the AS activity, of the supernat~ described by low-density the AS contents of the low-density eosinophi|s were compared with the AS content of normal-density cells. In addition, EM studies were performed with support of the Department of Hematology.* In this latter study, counting of electron-dense and electron-lucern granules was: performed for both low- and normal-density eosinophils of cell isolates from the same patient.

RESULTS In Fig. 1, the purity and recovery of eosinophils o f 16 separations are iUttstrated for individuals with allergic reactions. In Table I, the information with respect to percentage o f eosinophits in whole blood, total IgE, and clinical manifestations are illustrated. In five cases, the eosinophils demonstrated a normal density pattern similar to that o f healthy individuals with < 10% o f low-density cells in hand II. Eleven individuals demortstratexl varying leve|s of low-density eosinophils ( > 10%) from which in five cases the low-density cells were p r e ~ n t as most of

*Head, Prof. Dr. R. Halle and help by Mrs. N Btom.

614

K a u f f m a n et al.

J. ALLERGY CLIN. IMMUNOL. APRIL 1987

160

A

f

100

t--

L.J

J

%0

50

120

100

80

I

I

I

10

20

I

I

/,0

60

fime (rain) ------

60

~0

.)t.

20

!

I[

FIG. 2. Production of LTC+ by purified neutrophils (I); lowdensity eosinophils in the presence of neutrophils (11); purified normal-density eosinophils (111) by the Ca-ionophore A23187 (20 p,mol/L). Patients with asthma and allergy (e); patient with asthma and allergy during acute phase (&); patients with intrinsic asthma and eosinophilia, no allergy (*); patients with eosinophilia with other etiology, like eczema, hypereosinophilic syndrome, or unknown etiology (e). The difference of LTC4 production between hypodense and normal-density eosinophils is significant, according to the Student's t test (p < 0.005).

the total cell count (connected symbols, Fig. 1). In one case almost all cells were in the very low-density region of band I (Fig. 1 and Table I, patient 1, 16% eosinophils in whole blood). Formation of LTC+by both low-density and normaldensity eosinophils from individuals with eosinophilia of different etiology was studied after addition of Ca ionophore (A23187, 20 i~mol/L) in a Tyrode medium supplied with only Ca and Mg and 0.1% BSA without addition of either arachidonic acid or glutathione. The latter substances were omitted in order to test the intrinsic capacity of the cells for LT formation and to avoid masking of possible exhaustion of internal sources by previous activation. The LT formation measured for nine individuals with various numbers of eosinophils in the low-density region is demonstrated in Fig. 2. It is demonstrated that the low-density eosinophils in band II generally produced only half of the leukotrienes (57 _ 33 ng/106 eosinophils) com-

FIG. 3. Generation of LTC+ by Ca ionophore (20 p,moi/L) with normal-density (111) and low-density (11) eosinophils and the corresponding PMN fraction after FicolI-Paque separation (C). Normal-density eosinophils (e) (61.5%), low-density eosinophils (o) (4%), and PMN fraction (A) (7.5% eosinophils). Black bar at 20 minutes incubation is the spontaneous LTC+ production of the cell types in the absence of Ca ionophore.

pared with the normal-density eosinophils (103 ___ 44 ng/106 eosinophils; p < 0.005). Patients with asthma and allergy either in a resting phase (Fig. 2, e), or in an acute phase (A), or intrinsic asthma (cold, *) demonstrated the same difference in eosinophils when eosinophils were compared with eosinophils of other etiology (eczema or unknown). Only one patient with hypereosinophilic syndrome did not demonstrate the remarkable difference in LT synthesis between low- and normal-density eosinophils (Fig. 2, bandH, 72.6, and bandlll, 66.2 ng/106 eosinophils). With cells of one patient, LT generation was studied in time for normal-density (III) and low-density (II) eosinophils and the PMN fraction (obtained after the Ficoll-Paque separation) containing both low- and normal-density eosinophils. Fig. 3 illustrates that the higher LT generation by normal-density cells is mainly due to a higher rate of production during the first 30 minutes of incubation, which is followed by stabilization or a decline in LTC4 after prolonged incubation. The PMN cell fraction containing both normal- and low-density cells demonstrates an intermediate LT production, which indicates that the differences in LT generation as described above are not artifacts induced by the separation procedure. The lower production of LTC4 by low-density eosinophils could not be ascribed to the interaction of neutrophils present in these cell preparations. In accordance with similar observations by Kajita et al.,~~ mixing of purified eosinophils with increasing quantities of r m u ~ i l s did not result in

VOLUME 79 NUMBER 4

Leukotriene C,, production by eosi~ophils

615

TABLE II. Influence of neutrophilic leukocytes (99.6% neutrophils) on the Ca-ionophore-induc~.~d LTC. generation by purified eosinophils (98.9% eosinophils) goginophils (%*) LTC./I 06 eosinophils (ng)

98.9 18.0

79.1 31.3

59.3

39.6

19. ~

0

33.6

33.6

3 !. 3

ND

ND = not detectable. *Eosinophils (1.10 ~ cells per milliliter) were mixed with increasing quantities of neutrophils (1.10 ~ cells per milliliter) with volume ratios of 5/0.4,1, 3/2, 2/3, and 1/4. The mixtures were incubated with 10 ixmol/L of Ca ionophore for 30 minutes at 37 ~ C accordins to the procedure described under Material and Methods.

a suppression of the LTC4 generation (Table II). This result indicates that there is no cell-cell interaction in the production of LTC4 from eosinophils nor increased degradation of LTC4 by oxygen-derived products from the neutrophilic cells. Receptor-mediated LT formation was studied with STZ as stimulant. In contrast with the Ca-ionophore-mediated LT formation, STZ treatment induced a higher LTC4 generation with lowdensity eosinophils compared with the normal-density cells (Table III). The latter result is compatible with either a higher complement-receptor expression on low-density eosinophils and/or the higher state of activity of the subpopulation of eosinophils. Total AS contents of 12 cell preparations of purified normal-density eosinophils and neutrophils demonstrate that the eosinophil AS content ( 6 8 4 _ 71 U/10 b cells) is about 20 times higher than that of neutrophils (39 -+ 8 U/106 cells). This large difference in AS content makes it possible to calculate AS contents of low-density eosinophils present in either band II or band I, even when the neutrophils make up most of the cells. In Table IV the AS contents of both normal- and low-density eosinophils are demonstrated for ceil layers containing at least 10% eosinophits, corrected for the contribution of the neutrophil AS. Generally, the low-density eosinophils (bands I and II) demonstrate similar AS contents comparable with the normal-density (band III) eosinophils. In one case (Table IV, patient 5) the AS values were clearly below the mean value found for normaldense eosinophils. However, this lower concentration of AS was found for all eosinophils from this patient gathered at the different densities. These latter observations suggest that degranulation of the AScontaining granules is not involved in the appearance of low-density eosinophils. EM studies revealed differences in electron-dense granules between normal- and low-density eosinophils from one patient with highly elevated numbers of eosinophils (Table V). Cells from this patient demonstrated a more activated morphology characterized by many electron-lucent granules or vacuoles, dyssym-

TABLE i11. LTC, generation by low-density and normal-density eosinophits (nanogram per 10' cells) induced by STZ (1 mg/ml) Patient No.

1 2 3 4

L o w density

3.8 65.6 26.6 71.7

(70. t)* (4.2) (46.5) (15.0)

I~rmal density

0. I (98.5) 29.7 (42.2) t ~.9 (98.9) It.i (82.4}

*Number in parentheses is percent of eosinophils

merry m distribution of granules, which were generally gathered at one site of the cell and excluded from the nuclear hof, and many extrusions from the outer membrane (not presented). However, this pattern Ifound when eosinophilia is >15%) was not characteristic for individuals with smaller numbers of eosinophils that are generally found in patients with asthma and allergy (Table I). Often. no clear morphological differences between eosinophils sampled at the different densities could be observed. This is illustrated (Fig. 4. EM pictures l and llI) in eosinophils demonstrating a similar and normal morphology, although they were ~mpled at the densities 1,085 (I) and l, 1 l0 (III), respectively. Similar morphology was also found for eosinophils demonstrating a highly activated morphology with vacuoles (Fig. 4, O and H) and sampled at the densities 1.078 (O) and 1.094 (IlL Cells from these two bands isolated from patient 5. as presented in Table IV (the very light density cells, band O at density 1.078. is not listed in Table IV), are cells with low concentrations of AS that indieate prior degranulation. It is important to note that these highly activated and degranulated cells clearly demonstrating vacuolation can be isolated at higher densities ( Fig. 4, 1I, density 1.094) than cells demonstrating normal morphology (Fig. 4, I. density 1.085). Therefore. from the EM studies (as demonstrated in Fig. 4), it ts concluded that degranulation and/or vacuolation of eosinophils is not the only or main parameter that determines the location of the cells at the different densities.

616

Kauffman et al.

J. ALLERGYCLIN.IMMUNOL. APRIL 1987

T A B L E IV. AS contents (in units per 108 eosinophils) of normal-density (band III) and hypodense

(band I or II) eosinophils after separation on Percoll gradients Band

Purity (% eos)

Patient No.

I

II

III

I

II

III

1 2 3 4 5 6 7 8 9 10 11 Mean SD

--550 811 388 620 -----592 175

750 697 705 -599 718 542 931 532 700 758 693 117

606 826 777 -476 -749 896 479 717 616 681 147

0.7 0 5.5 48 25 23 0 0 2 0 0

63 84 81 0 84 55 11 8.5 90 4 63

97 98 96 0 93 64 88 99 100 97 95

Diagnosis

IA AA IA Eczema Eczema OE Eczema AA AA OE Eczema

eos = Eosinophils; AA = allergy plus asthma; IA = intrinsic asthma; OE = other etiology.

T A B L E V. Electron-dense granules in low-density (band II) and normal-density (band III) e o s i n o p h i l s m e a s u r e d by m e a n s of EM Band II

Band III

+ EDG*

- EDGt

+ E D G / - EDG

+ EDG

- EDG

+ E D G / - EDG

24.6:1:

10.5~:

2.3

34:1:

6.0~

5.7

*Electron-dense granules. tGranules without electron-dense material. :~Ten EM pictures of both band II and band III were compared, and the mean values were calculated.

DISCUSSION

This study was performed to investigate first, if allergic individuals with asthma have subpopulations of eosinophils with lower densities compared with eosinophils of normal individuals. Second, functional differences between eosinophil subpopulations were studied with respect to Ca-ionophore- and STZ-induced liberation of LTC4, AS content, and EM morphology. It was demonstrated that by use of a standard procedure for Percoll gradient centrifugation, eosinophils of patients with asthma and with eosinophilia could be separated into a fraction of normal-density eosinophils sampled at a density of 1.110 (band II1) and low-density cells found at the densities 1.094 and 1.085 gm/ml (bands H and I). Our results are in accordance with a recent article of Fukuda et al. ,2~ describing the density distributions of eosinophils in individuals with and without asthma under isotonic conditions. We also found similar eosinophilic subpopulations in individuals with asthma and with allergy.

In our study, the difference between normal-density and low-density eosinophils was found at a higher density to that reported by Fukuda et al. This is likely due to the difference in isolation procedure. In our study, hypertonic conditions were used in order to obtain normal-density eosinophils with a high degree of purity, 22 which enables the study of metabolic parameters in the absence of contaminating neutrophils. The presence of low-density eosinophils in most patients studied (11 of 16, Fig. 1) is important because in other studies these cells were demonstrated to be activated with respect to receptor expression, cytotoxic properties, and oxygen metabolism. ~2"14 It is important to note that elevations in receptor expression for both eosinophils and neutrophils were observed in the circulation during both early and late obstructive reactions after allergen inhalation. 23 This suggests that activation of circulating PMN cells is induced by mediators liberated from the tung tissue during the obstructive reactions. It is not clear if these mediators liberated after allergen challenge are re-

VOLUME79

Leukotriene C4 production by eosinophits 617

NUMBER 4

FIG. 4. EM studies of eosinophils gathered at different densities. I and III, eosinophils purified from two patients demonstrating normal morphology isolated at the densities 1.085 (I) and 1,1 t0 (111). Original magnification x 10,000. O and II, eosinophils from one patient with eczema (patient 5, Table IV). Cells were isolated at the densities 1.078 (0) and 1.094 (11), respectively. Original magnification x 10,000.

sponsible for the appearance of low-density blood eosinophils. Another possible explanation for the appearance of low-density blood eosinophils is their return to the circulation from the lung after the latephase reaction. Infiltration of the lung by eosinophils during the late-phase reactions. 9 is generally followed by a strong elevation in circulating eosinophils 24 hours after allergen challenge. 24 2s These recirculating eosinophils may be low-density eosinophils as a result of their previous activities. We, therefore, also measured the abilities of the cells for LTC4 production without addition of any substrates, such as arachi-

donic acid, glutathione (as is often done in other studies),~0, ~,. ,7. ,9.26 in order to test the intrinsic capacity of the cell for LT generation. A clear difference was observed between the eosinophil fracti~s. Hypodense eosinophils produced only half the amounts of LTC4 compared with normal-density eosinophils (Fig. 2). In agreement with the results of Kajita et al., 2~this difference in production is not due to t h e interaction of neutrophils, either by production of LTB4 or production of large quantities of oXygen radicals, since separate experiments mixing purified eosinophils with increasing numbers of neutrophils in-

618

Kauffman et al.

fluenced LTC4 production only marginally (Table II). However, it cannot be excluded that low-density eosinophils have an altered arachidonic acid metabolism leading to a diminished production of LTC4 formation but increased productions of other metabolites, which has been demonstrated for eosinophils in the presence of arachidonic acid.26' 27 Our observations on the LTC4 generation by lowdensity eosinophils are opposite to the findings of Kajita et al.,2~ who demonstrated greater amounts of LT generated by the low-density cells. It is possible that differences in the isolation procedure (the use of fetal calf serum 2~ or different incubation conditions are responsible for the observed contrasting findings. However, there is no obvious explanation from the differences in the procedures as described. The lower production of LTC4 by low-density eosinophils might be evidence that low-density eosinophils have been attracted to and infiltrated in the inflamed tissue and returned to the circulation with loss of substrates necessary for LT generation. This latter interpretation is in accordance with the observation that eosinophils are greatly increased in the circulation 24 hours after allergen inhalation when this challenge was followed by early and/or late obstructive reactions. 24' 25 When STZ was used as a C3b receptor-mediated stimulus for LT production, both normal-density and low-density eosinophils produced smaller quantities of LTC4 compared with productions with Ca-ionophore as stimulus, which is in accordance with previous observations by Bruynzeel et al.17 However, in contrast with Ca-ionophore, STZ induced greater amounts of LTC4 with low-density eosinophils compared with the normal-density cells. This increased production varied for the different cell isolates from two to seven times that of the normal-density cells. Similar increased cell activity by low-density eosinophils was also described for LTC4 generation with IgG-dependent stimulation 15 and oxygen consumption with STZ as a stimulant 13 and suggest a higher receptor expression and/or greater stimulus-response coupling in these activated eosinophil subpopulations. When total AS contents of both low- and normaldensity eosinophils were studied, no clear evidence was found for loss of AS by degranulation in hypodense eosinophils. A decrease in AS content was observed in only a few cases (Table IV, Nos. 2, 3, and 7). AS contents are, however, the parameter for degranulation of only the small granules. 2s Therefore, it cannot be excluded that the low-density eosinophils are ceils that have exclusively released a part of their large granules containing substances like major basic protein, eosinophil peroxidase, or eosinophil cationic protein. 28 Also our

J. ALLERGY CLIN. IMMUNOL. APRIL 1987

EM studies supported only in part the concept that low-density eosinophils appear as a result of degranulation. In Fig. 4, studies are presented of cells demonstrating similar morphology, although cells were isolated at different densities. In addition, cells with a highly activated morphological appearance demonstrating much vacuolation and degranulation (Fig. 4, H), could be found at higher densities than eosinophils found at lower densities demonstrating normal morphology (Fig. 4). On the other hand, in one isolate, normal-density eosinophils demonstrated more electron-dense granules compared with the hypodense eosinophils. Our results are in accordance with similar observations of Tai et al. 29 who demonstrated that light-density cells were morphologically normal and who suggested that alterations in cell density are due to changes in cell volume with swelling of the granule material and not simply granule loss. Further study of the nature of the low-density eosinophils, in particular quantification of enzymic parameters of the granules, is needed before a definitive conclusion can be reached. In addition, it is interesting to note that in Table IV, four patients were included demonstrating eczema and peripheral eosinophilia. All four individuals demonstrated hypodense eosinophils, three of four with the most cells in the low-density state. Recently, extensive deposition of toxic eosinophil proteins in skin tissue of patients with atopic dermatitis was demonstrated. 30. 31 These observations indicate the skin as another site of infiltration by eosinophils with active liberation of toxic proteins and (possibly) LTs. We thank the Departments of Hematology and Pulmonary Diseases of the State University Hospital of Groningen for their help, J. Verhagen and P. L. B. Bruynzeel for performing determinations of LTC4by means of reversed-phase high-performance liquid chromatography, Mrs. B. Schilizzi for correction of the manuscript, and Mrs. J. Kruitbosch and Mrs. I. Lambeck for their expert secretarial help. REFERENCES

1. Kay AB. Functionsof the eosinophilleukocyte.Br J Haematol 1976;33:313. 2. SchatsM, WassermanS, PattersonR. The eosinophiland the lung. Arch Intern Med 1982;142:1515. 3. FrigasE, LoegeringDA, SolleyGO, FarrowGM, GleichGJ. Elevated levels of the eosinophil granule major basic protein in the sputum of patients with bronchial asthma. Mayo Clin Proc 1981;56:345. 4. DahlR, VengeP, OlisonI. Variationsof bloodeosinophilsand eosinophilcationicprotein in serumin patients with bronchial asthma: studies during inhalation challenge test. Allergy 1978;33:211. 5. Ffigas E, GleichGJ. Tile eosinophilarid the pathophysiology of asthma. J ALLERGYCLrNIMMtmOL1986;77:527.

VOLUME79 NUMBER4

6. Godard Ph, Chantreuil J, Damon M, Coupe M, Flandre D, Crastet de Paulet A, Michel FB. Functional assessment of alveolar macrophages: comparison of cells from asthmatics and normal subjects. J ALLERGYCLINIMMUNOL1982;70:88. 7. Diaz P, Gallequillos FR, Gonzales MC, Pantin CFA, Kay AB. Bronchoalveolar lavage in asthma: the effect of disodium cromoglycate on leukocyte counts, immunoglobulins, and complement. J ALLERGYCLtN IMMUNOL1984;74:41. 8. Monchy de JGR, Kauffman HF, Venge P, Ko~ter GH, Jansen HM. Sluiter HJ, de Vries K. Bronchoalveolar eosinophilia during allergen-induced late asthmatic reactions. Am Rev Respit Dis 1985;131:373. 9. Lain S, Chan-Yeung M, Lericbe J, Kyek K, Philips D. Cellular changes in bronchial lavage fluids following late asthmatic reactions in patients with red cedar asthma. Am Rev Respir Dis 1985;129:A42. 10. Weller PF, Lee CW, Foster DW, Corey EJ, Austen KF, Lewis RA. Generation and metabolism of 5-1ipoxygenase pathway leukotrienes by human eosinophils: predominant production of leukotriene C4. Proc Natl Acad Sci USA 1983;80:7626. 11. Verhagen J, Bruynzeel PLB, Koedam JA, Wassink GA, de Boer M, Terpstra GK, Kreukniet J, Veldink GA, Vliegenthart JFG. Specific leukotriene formation by purified human eosinophils and neutrophils. FEBS Lett 1984;168:23. 12. Prin L, Capron J, Capron M, Gosset PH, Taelman H, Tonnel AB, Capron A. Heterogeneity of human eosinophils. II. Variability of respiratory burst activity related to cell density. Clin Exp lmmunol 1984;57:735. 13. Winqvist I, OIofsson I, Olsson I, Persson AM, Hallberg T. Altered density, metabolism, and surface receptors of eosinophils in eosinophilia. Immunology 1982;47:531. 14. Capron M, Spiegelberg HL, Prin L, Bennich H, Butterworth AE, Pierce RJ, Atiouaissi M, Capron A. Role of IgE-receptors in effect or function of human eosinophils. J Immunol 1984; 132:462. 15. Shaw RJ, Walsh GM, Cromwell D, Mogbel R, Spry CJF, Kay AB. Activated human eosinophils generate SRS-A leucotrienes following IgG-dependent stimulation. Nature 1985;316:150. 16. Kauffman HF, Levering PR, de "Cries K. A single centrifugation step method for the simultaneous separation of different leukocytes with special reference to basophilic leukocytes. J lmmunol Methods 1983;57:1. 17. Bruynzeel PLB, Kok PTM, Hamelink ML, Kijne AM, Verhagen J. Exclusive leucotriene C4 synthesis by purified human eosinophils induced by opsonized zymosan. FEBS Lett 1985;189:350.

Leukotriene C4 production by eosinophils

619

18. Busse WW. Decreased granulocyte response to isoprotenerol in asthma during upper respiratory ~ract inlections. Am Rev Respir Dis 1977;115:738. 19. Wasserman SI, Austen KF. Arylsulfatase B of human lung: isolation, characterization, and interaction with slow-reacting substances of anaphylaxis. Ctin Invest 1976;5"7:738. 20. Kajita T, Yui Y, Mita H, Taniguchi N. Sait~) H, Mishima T, Shida T. Release of leucotriene C~ from human eo,qnophils and its relation to the cell density, lnt Arcbs Allergy Appl Immunol 1985;78:406. 21. Fukuda T, Dunnette SL. Reed CE, Ackerman SJ. Peters MS, Gleich GJ. Increased numbers of hypodensc eosinophils in the blood of patients with asthma. Am Rev Respir Dis 1985; 132:981. 22. Vadas M, David J, Butterworth AE, Pisani NT, Siongok TA. A new method for the purification of human eosinophils and neutrophils, and a comparison of the ability of these cells to damage schistosomula of Schistosoma mansoni. J hnmunol 1979;122:1228. 23. Durham SR, Caroll M, Walsh GM, Kay AB. l,eukocyte activation in allergen-induced late-phase asthmatic reaction. N Engl J Meal 1984;311:1398. 24. Booy-Noord H, de Vales K, Sluiter ILl, Orie NGM. Late bronchial obstructive reaction to experimental inhalation of house dust extract. Clin Allergy 1972;2:63. 25. Durham SR, Kay AB. Eosinophils, bronchial hyperreactivity, and late-phase asthmatic reactions. Clin Allergy 1985; 15:411. 26. Henderson WR, Harley JB, Fauci AS. Arachidonic acid metabolism in normal and hypereosinophilic syndrome human eosinophils: generation of leucotrienes B4, C~, D~. and 15lipoxygenase products. Immunology 1984:51:679. 27. Bruynzeel PLB, Kok PTM. Vietor RJ. Verhagen J On the optimal conditions of LTC, formation by human eosinophils in vitro. Prostaglandins Leukotrienes Med 1985:20:11. 28. Weller PF. Eosinophilia. J ALLERGYCLtNIMMU~)L1984;73:1. 29. Tai PC, Bakes DM, Barkans JR. Spry CJF. Plasma membrane antigens on light density and activated human blood eosinophils. Clin Exp Immunol 1985:60:427 30. Leiferman KM, Ackerman SJ. Sampson HA, Haugen BA. Venencie PY, Gleich GJ. Dermal deposition of eosinophil granule major basic protein in atopic dermatitis: comparison with onchocercciasis. N Engl J Med 1985:313:282. 31. Spry CJF, Tai PC, Barkans J. Tissue localization of human eosinophil cationic proteins in allergic diseases. Int Archs Allergy Appl lmmunol 1985;77:252