Experimental Pulmonary Eosinophilia in Mice
by Ascaris suum Extract1 •2
MAKOTO NOGAMI, MATSUNOBU SUKO, HIROKAZU OKUDAIRA, TERUMASA MIYAMOTO, JUNJI SHIGA, MAMORU ITO, and SHIRO KASUYA
Introduction
Eosinophils accumulate in the alveoli of the lung in various diseases, including pulmonary infiltration with eosinophilia (PIE) syndrome (1). In vitro, eosinophils are cytotoxic for lung parenchymal cells, and eosinophil granule-derived major basic protein is cytotoxic for tracheal cells (2). In order to clarify the role of eosinophils in PIE syndrome and in other lung disorders with eosinophilic infiltration, the development of an animal model in which eosinophils accumulate in the lung to a great extent is necessary. Crofton and coworkers (1) classified PIE syndrome into five groups. Among them, simple pulmonary eosinophilia and tropical pulmonary eosinophilia are known to be caused by parasites. Transnasal administration has been reported to place material in the lower respiratory tract (3). Therefore, it is supposed that transnasal administration of extract from parasites could cause eosinophilia in the respiratory tract. The purpose of this study was to establish ~, model of eosinophilic infiltration in the lung using an extract of the parasite Ascaris suum, and, for the elucidation of the mechanism of eosinophilia, to see if T-cells, mast cells, and IgE are involved in this model. For this purpose, cyclosporin treatment (4), athymic nude mice, genetically mast-cell-deficient (WB X C57BL/6)FI-W/Wv (W/WV) mice (5), IgE high-responder Balb-C mice, and IgE low-responderSJL mice (6)wereused. Methods Animals C57BLl6, Balb-C, and SJL mice were purchased from Charles River Japan, Atsugi, Japan. C57BLl6-nu/nu (C57BLl6 nude) mice were obtained from the Central Institute for Experimental Animals, Kawasaki, Japan. W/Wv mice were purchased from the Shizuoka Agriculture Cooperative Association for Laboratory Animals, Hamamatsu, Japan. The mice used in this study were male, 7 to 10wk of age, and bred in a specific-pathogen-
SUMMARY The transnasal administration of an extract of the parasite Ascaris suum (Asc) to C57BU6 mice for 3 wk produced marked eosinophilia in the bronchoalveolar lavage (BAL) fluid. This pulmonary eosinophilia was not accompanied by blood eosinophilia. The oral administration of cyclosporin, 50 mg/kg body weight, every other day significantly suppressed the pulmonary eosinophilia. Athymic C57BL/6-nu/nu mice failed to develop pUlmonary eosinophilia. These data indicate that pUlmonary eosinophilia caused by this parasite extract is T-cell-dependent. Genetically mast-celldeficient (WB X C57BU6)F1-W/Wv (W/Wv) mice developed marked eosinophilia in the BAL, which shows that mast cells are not necessary in the formation of eosinophilia in BAL In this model. C57BU6, W/Wv, and Balb-C mice that developed eosinophilia in the BAL also showed elevated totallgE levels and IgE titers against Asc in the sera. On the other hand, C57BU6-nu/nu mice and IgE low-responder SJL mice, which developed little eosinophilia, did not show elevated total IgE levels and IgE titers against Asc in the sera. However, oral administration of cyclosporln, 50 mg/kg, which inhibited eosinophilia in the BAL in C57BU6 mice, did not significantly Inhibit the elevation of totallgE or IgE against Asc in the sera. This indicates that serum IgE production is not required for the formation AM REV RESPIR DIS 1990; 141:1289-1295 of eosinophilia.
free environment (free from infection with Corynebacterium kutscheri, Bacillus piliformis, Salmonel/a, Pseudomonas aeruginosa, Pasturel/a pneumotropica, Bordetel/a branchiseptica,Mycoplasma pulmonis, Sendai virus, or mouse hepatitis virus). Male Sprague-Dawley rats weighing more than 400 g were purchased from CLEA Japan, Tokyo, Japan.
Parasite Extract The Ascaris suum extract (Asc) was obtained in the following manner (7). Adult Ascaris suum worms were obtained from a slaughterhouse. The worms were freeze-dried, defatted with ethyl ether, extracted with veronalbuffered saline (VBS), and dialyzed against VBS. The extracts were dissolved in saline at a concentration of 5 mg/ml. Administration of Asc, Saline, or Ovalbumin to C57BL/6 Mice Each C57BLl6 mouse was anesthetized by ethyl ether and given 40 ul/day of the above Asc solution by trans nasal drops 2 consecutive days a week for I, 2, or 3 wk; 48 h after the last administration, blood was collected from retroocular veins using a glass microsyringe, and bronchoalveolar lavage(BAL) fluid was obtained as described below. The mice not used for BAL were used for histologic studies as described below. Some C57BLl6 mice were administered saline or 5 mg/ml saline solution of ovalbumin (OA), crystallized five times (Seikagaku Kogyo, Tokyo, Japan),
for 3 wk, and the BAL fluid was collected in the same manner as the Asc.
Administration of Asc to Several Strains of Mice Balb-C, SJL, C57BLl6 nude, and W/W V mice were anesthetized by ethyl ether and administered Asc 2 consecutive days a week for 3 wk as described above; 48 h after the last administration, blood and BAL fluid werecollected 3S described below. Cyc/osporin Treatment Cyclosporin (Cs) was kindly supplied by Sandoz (Tokyo, Japan). Cs was dissolved in olive oil at a concentration of 2 mg/ml just before use. Some C57BLl6 mice, which had been administered Asc for 3 wk in the same manner as described above, were orally adminis-
(Received in original form February 7, 1989 and in revised form September ll, 1989) 1 From the Department of Medicine and Physical Therapy and the Department of Pathology, Faculty of Medicine, University of Tokyo, Tokyo; the Central Institute of Experimental Animals, Kawasaki; and the Department of Parasitology, Gifu University School of Medicine, Gifu, Japan. 2 Correspondence and requests for reprints should be addressed to Makoto Nogarni, Department of Medicine and Physical Therapy, Faculty of Medicine, University of Tokyo, 7-3-1, Hongo, Bunkyoku, Tokyo, Japan 113.
1289
1290
tered the Cs solution through a gastric tube at 50 mg/kg body weighteveryother day from the first day of Asc administration to the day the mice were killed. For the control animals, olive oil alone was administered in the same manner. Bronchoalveolar Lavage The mice were killed by ethyl ether and resected by median incision. The trachea was exposed in the neck, and a 21-gauge Teflon'" tube (Terumo, Tokyo, Japan) was inserted into the trachea. Saline, I ml, was infused and collected through the tube repeatedly to gain 4 ml of BAL fluid for each mouse; 5 ml of saline were required to gain 4 ml of BAL fluid. The cells in the fluid werethen washed with RPMI 1640supplemented with L-glutamine(GIBCO Laboratories, Grand Island, NY), centrifuged three times, and suspended in 400 Itl of RPMI 1640. The cells of BAL fluid were counted with a hemacytometer by trypan blue exclusion. The cells in the BAL fluid were sprayed onto a glass slide by Cytospin II (Shandon Southern Products, Cheshire, England) and stained with Diff-Quik'" (Midorijuji, Kobe, Japan), and the cell differentials were counted. At least 500 cells were counted for cell differentials (3). Numbers of eosinophils in the BAL were calculated by multiplying total cell numbers by percentages of eosinophils. Part of the cellsin the BAL fluid were fixed in 2.5070 solution of glutaraldehyde and postfixed in 2% osmium tetroxide solution for 2 h. After the dehydration in graded ethanol solutions, the cells were embedded in Epon 512. Ultrathin sections were cut, doublestained with uranyl acetate and lead citrate, and then examined with a Nippon Denshi JM 100C electron microscope (Nippon Denshi, Tokyo, Japan). Histologic Study The lungs of C57BL/6 mice not used for the BAL study were fixed in 10% formalin, embedded in paraffin, sectioned, and stained with hematoxylin-eosin. Blood Leukocyte Counts and Cell Differentials The blood samples were studied for leukocyte counts and cell differentials. For leukocyte counts, the blood was stained with Turk's solution, and the cells were counted in a hemacytometer. Glass slides were streaked with blood and stained with May-Grunwald and Giemsa solutions, and leukocyte differentials were counted microscopically. Numbers of eosinophils were calculated by multiplying total leukocyte counts by percentages of eosinophils. Serum IgE Levels Serum specimens wereprepared from part of the blood collected. Serum IgE was determined by enzyme-linked immunosorbent assay (ELISA) as previously described by Hirano and coworkers (8). Briefly, 96-well EIA plates (Coster, Cambridge, MA) were
NOGAMI, SUKO, OKUDAIRA, MIYAMOTO, SHIGA, ITO, AND KASUYA
used. PBS containing 0.05% Tween" 20 was used to dilute the samples and for washing. For blocking nonspecific binding sites, the same buffer containing 1% BSA was used. The volume pipetted in this assay was 50 ul, except for blocking, where it was 100 ul. The wells were coated with monoclonal rat antimouse IgE antibody (YamasaShoyu, Tokyo, Japan) at 4 C overnight. After blocking, 1:10 or I :100 dilutions of sera or standard twofold serial dilutions were applied. Monoclonal mouse IgE antidinitrophenyl antibody (SPE-7; Miles Laboratories, Naperville, IL) was used as standard. The plate was kept for I h at room temperature, and then biotinylated monoclonal rat antimouse IgE (Yamasa Shoyu, Tokyo, Japan), which recognizes the different epitope on the Fc fragment of IgE from that used for the primary coating, was added. After I h at room temperature, the wells were washed, and peroxidase-labeled avidin D (Vector Laboratories, Burlingame, CAl was added and kept for I h at room temperature. The substrate to develop the reaction was the solution containing 0.04% H 20 2 and 0.004% o-phenylene-diamine (Sigma Chemical Co., St. Louis, MO) in 0.2 M sodium phosphate, 0.1 M sodium citrate buffer (pH, 5.0). Thirty minutes after addition of the substrate, reaction was stopped with 4 N H 2S04 and was read at 490 nm with an ELISA reader (Corona Electric Co., Tokyo, Japan). Serum IgE titers against Asc were determined by passive cutaneous anaphylaxis (PCA) reaction by the modified method of Okudaira and coworkers (9). Briefly, 50 ul of serial dilutions of sera (e.g., I:5, I:10, 1:20) were injected intracutaneously into shaved
dorsal skin of a Sprague-Dawley rat. After 18 h, the rat was given an intravenous injection of I ml of I% Evans blue (Merck, Darmstadt, Germany), which was immediately followed by an intracutaneous injection of 50 ul of Asc (10 ug/rnl) to each site where the sera were injected. Twenty minutes later, the rat was killed by carbon dioxide gas, and the reaction was evaluated by examining the inside of the skin. The PCA titer was expressed by the reciprocal of the final dilution givinga positive bluing reaction 5 mm or more in diameter.
0
Statistics Student's t test was used to compare the values between two groups. Bonferroni's test was used to compare the values among more than two groups (10). A p value below 0.05 was regarded as statistically significant.
Results HAL Fluid
Asc was given transnasally to see if eosinophilia in the BAL was induced. The numbers of eosinophils in the BAL fluids of several strains of mice are shown in figure 1. C57BL/6 mice given Asc for 3 wk developed significant eosinophilia as compared with C57BL/6 mice given saline for 3 wk (p < 0.01). Eosinophils in the BAL fluid of C57BL/6 mice given Asc for 3 wk are shown in figure 2A, whereas figure 2B shows macrophages and lymphocytes in the BAL fluid of C57BL/6 mice given saline for 3 wk. The
Cell number
DI
(x 104)
total cell number in SAL
eosinophil number in SAL
50
WBB6Fl -W/Wv
BALB/c
S-.JL
C57BL/6
C57BL/6
Asc
Asc
Saline
Asc
n=10
n=9
-nu/nu Asc
n= 5
n=10
n= 5
Fig. 1. Eosinophil numbers in the bronchoalveolar lavage (BAL) fluid of several strains of mice given Ascarissuum extract or saline transnasally for 3 wk. Bars show mean values ± standard deviations. Asc = Ascaris suum extract; * = statistically significant (p < 0.01) as compared with C57BU6 mice with Asc administration.
1291
EXPERIMENTAL PULMONARY EOSINOPHILIA IN MICE BY ASCARIS SUUM
B
c Fig. 2. Cells in the bronchoalveolar lavage (BAL) fluid. A. Cells in C57BU6 mice given Ascaris suum extract (Asc) transnasally for 3 weeks. Light microscopy. Diff-Quik stain. Original magnification, x 400. B. Cells in C57BU6 mice given saline transnasally for 3 weeks. Light microscopy. Diff-Quik stain. Original magnification, x 400. C. An eosinophil containing specific granules in C57BU6 mice given Asc transnasally for 3 wk. Electron microscopy. Original magnification, x 5,000. D. A degranulated eosinophil in C57BU6 mice given Asc transnasally for 3 wk. Original magnification, x 5,000.
electron microscopy of the BAL fluid cells of C57BL/6 mice given Asc for 3 wk shows the eosinophils with characteristic granules in the cytoplasm (figure 2C). Some eosinophils from three of 10 mice showed degranulation in the electron microscopy although less than 50/0 of the total eosinophils in number (fig-
ure 2D). W/Wv and Balb-C mice given Asc for 3 wk also showed eosinophilia in the BAL fluid (figure 1). Because C57BL/6 mice produced the most eosinophils in the BAL, further studies were carried out mainly on C57BL/6 mice. The time required for the development of eosinophilia in C57BL/6 mice was
studied. The cell numbers and fractions in the BAL fluid of C57BL/6 mice given Asc for 1 to 3 wk are shown in table 1. The mice given Asc for 1 or 2 wk showed less eosinophilia than did those given Asc for 3 wk (p < 0.01); 3 wk were necessary to cause marked eosinophilia. In order to see if eosinophilia in the
1292
NOGAMI, SUKO, OKUDAIRA, MIYAMOTO, SHIGA, ITO, AND KASUYA
TABLE 1 CELL NUMBERS AND CELL DIFFERENTIALS IN THE BRONCHOALVEOLAR LAVAGE (BAL) FLUID OF C57BLl6 MICE
Material Asc
Weeks
Cell Numbers (x 104 )
E
N
M
L
Number of Eosinophils (x 104 )
1 2 3
10 ± 4 15 ± 4 60 ± 10
5 ± 5 20 ± 10 51 ± 8
30 ± 5 14 ± 8 9 ± 3
64 ± 5 65 ± 13 35 ± 6
1 ± 1 1 ± 1 5 ± 3
0.5 ± 0.8' 3.3 ± 2.1' 31.0 ± 7.0
Saline
3
8±2
Ovalbumin
3
8 ± 2
Cell Differentials (%)
o± o±
0
o±
0
97 ± 2
3 ± 1
0.0 ± 0.0'
0
O±O
97 ± 1
3 ± 1
0.0 ± 0.0'
Definition of abbreviations: Asc = Ascaris suum extract; S = saline; OA = ovalbumin; E = eosinophil; N = neutrophil; M = macrophage; L = lymphocyte; • = statistically significant (p < 0.01) as compared with mice given Asc for 3 wk. Figures show mean values ± standard deviations. The number of mice tested was 10 for each group except ovalbumin, where it was 5. Weeks means those for which the materials were transnasally administered.
en Asc transnasally for 3 wk, which showed significantly fewer eosinophils in the BAL fluid than in C57BLl6 mice given Asc (p < 0.01) (figure 3A). Cs treatment significantly inhibited the development of eosinophilia in the BAL fluid of C57BL/6 mice given Asc for 3 wk as compared with olive oil (p < 0.01) (figure 3B). In order to see if mast cells are involved in this eosinophilia, W/Wv mice were given Asc for 3 wk. W/Wv mice developed a statistically significant increase in eosinophils as compared with W/Wv mice given saline (p < 0.01) (figure 4).
Histologic Study Cell number
(xl0') Cell number
( x 10')
50
DI
50
DI
total cellnumber in BAl
total cellnumber in BAl
eosinophil number in BAl
o C57BL/6 +Asc
C57BL/6
n=10
+Asc
eosinophil number in BAl
Blood Leukocyte Counts and Cell Differentials C57BL/6 +Asc
C57BL/6 +Asc
+
+
cyclosporin no, Oliveoil p.o, 50mg/kg everyother day everyother day n= 5 n=10
-nu/nu n=5
The lungs of C57BLl6 mice given Asc for 3 wk developed an accumulation of eosinophils and lymphocytes around vessels and alveolar septa (figure 5A and B). The lungs of C57BL/6 mice given saline for 3 wk or of nontreated mice showed a slight infiltration of neutrophils and lymphocytes into the alveolar septa, but few eosinophils were found (figure 5C and D).
Fig. 3. Eosinophil numbers in the bronchoalveolar lavage (BAL) fluid of C57BU6 nude mice and cyclosporintreated C57BU6 mice given Ascaris suum extract (Asc). All mice were given Asc transnasally for 3 wk. Bars show mean values ± standard deviations. A. Eosinophil numbers in C57BU6 nude mice.• = statistically significant (p < 0.01) as compared with the eosinophil number in C57BU6 with Asc administration. B. Eosinophil numbers in cyclosporin treatment (50 mgikg body weight every other day). ' = statistically significant (p < 0.01) as compared with the eosinophil number in C57BU6 with Asc and olive oil administration.
The peripheral blood leukocyte count was carried out to see if pulmonary eosinophilia in C57BL/6 mice was accompanied by blood eosinophilia. The peripheral blood leukocyte count of C57BL/6 mice at the third week is shown in table 2. No significant increase in blood eosinophils was found in C57BL/6 mice given Asc compared with C57BL/6 mice given saline.
Serum IgE Levels Cell number
(X 10 4)
DI total
WBB6FI-W/WV
eosinophil
cell number
number
In SAL
in SAL
Fig. 4. Eosinophil numbers in the bronchoalveolar lavage (BAL) fluid of (WB X C57BU6)Fl-W/wv mice given Ascaris suum extract transnasally for 3 wk. Bars show mean values ± standard deviations. WBB6Fl-W/wv = (WB X C57BU6)Fl-W/wv mice; Asc = Ascaris suum extract; , = statistically significant (p < 0.01) as compared with the eosinophil number in mice with saline administration.
WBB6FI-W/Wv
Ase
Saline
n=5
n=5
BAL fluid of C57BLl6 mice can be caused by other materials, OA was given transnasally for 3 wk. OA failed to show eosinophilia in the BAL fluid in C57BLl6
mice as compared with Asc (p < 0.01) (table 1). To see if T-cells are involved in eosinophilia, C57BLl6 nude mice were giv-
The following study was carried out to see if serum IgE levels were correlated with eosinophilia in the BAL. The serum IgE levels are shown in figure 6. In terms of serum IgE levels, there was no statistically significant difference between C57BLl6 and W/Wv or C57BLl6 and Balb-C mice given Asc for 3 wk, whereas SJL or C57BL/6 nude mice given Asc for 3 wk and C57BLl6 mice given OA for 3 wk showed significantly lower serum IgE levels than did C57BLl6 mice given Asc for 3 wk (p < 0.01). C57BLl6 mice given Asc for 3 wk showed statistically significant elevated IgE as compared with C57BL/6 mice given saline for 3 wk (p < 0.01). IgE antibody titers against Asc by PCA reaction are shown in figure 7. There was no statistically significant difference in titers between C57BLl6 and W/Wv or C57BLl6 and Balb-C mice given Asc for 3 wk, where-
1293
EXPERIMENTAL PULMONARY EOSINOPHI!..IA IN MICE BY ASCARIS SUUM
Fig . 5. Lung histology. Light microscopy. A. C57BU6 mice given Ascaris suum extract (Asc) transnasally for 3 wk . Hematoxylin-eosin stain . Fixed in formalin and embedded in paraffin . Original magnification, x10. B. C57BU6 mice given Asc transnasally for 3 wk. Hematoxylin-eosin stain. Fixed in for malin and embedded in paraffin . Orig inal magn ification , x 330. Eosinophils are indicated by thick arrows. A macrophage is indicated by a thin arrow. C. C57BU6 mice given saline transnasally for 3 wk. Hematoxylin-eosin stain . Fixed in formalin and embedded in paraffin . Original magnification, xl0. D. Nontreated C57BU6 mice of the same age as A, B, and C. Hematoxylin-eosin stain. Fixed in formalin and embedded in paraf fin. Or iginal magnification, x10.
as SJL or C57BL/6 nude mice given Asc for 3 wk and C57BLl6 mice given OA for 3 wk showed significantly lower titers than did C57BLl6 mice given Asc for 3 wk (p < 0.05).The titers in C57BLl6 mice given Asc for 3 wk were significantly elevated compared with C57BLl6 mice given saline for 3 wk (p < 0.05). Oral administration of Cs, 50 mg/kg, which
inhibited eosinophilia in the BAL in C57BLl6 mice given Asc for 3 wk, did not prevent the production of total IgE or IgE antibody against Asc in the sera (figures 6 and 7). Discussion
The transnasal administration of Asc to C57BL/6 mice for 3 wk caused marked
eosinophilic infiltration in the lung. Simple pulmonary eosinophilia and tropical pulmonary eosinophilia in human beings are known to be caused by parasitic infection (1). Sun and coworkers (2) developed eosinophilia in the lung of guinea pigs by the aerosol administration of polymyxin B (2). Laycock and colleagues (11) developed lung eosinophilia in rats
1294
NOGAMI, SUKO, OKUDAIRA, MIYAMOTO, SHIGA, ITO, AND KASUYA
TABLE 2 BLOOD LEUKOCYTE COUNTS AND CELL DIFFERENTIALS IN C57BLl6 MICE
Mat
Leukocyte Count (x 103//1 /)
N
E
B
L
M
Number of Eosinophils (x 1(2)
Asc Saline
5.0 ± 0.5 4.5 ± 0.7
50.5 ± 5.0 49.5 ± 5.0
4.0 ± 1.0 4.0 ± 1.5
0.5 ± 0.1 0.5 ± 0.3
40.0 ± 5.0 41.0 ± 5.0
5.0 ± 1.0 5.0 ± 1.0
2.0 ± 0.5 1.8 ± 0.6 NS
Leukocyte Differentials (%)
Definition of abbreviations: Mat = material transnasally administered; E = eosinophil; N = neutrophil; B = basophil; L = lymphocyte; M = macrophage; Asc = Ascaris suum extract; S = saline; NS = statistically not significact as compared with mice given Asc. Figures show mean values ± standard deviations. The number ot mice tested tor each group was 10. Each material was transnasally administered tor 3 wk as described in text.
log IgE (ng/ml)
5
•
4
I
3
• •
f
2
•
-f* --*
C57BL/6 Asc
WBB6Fl
-w/Wv
BALB/c Asc
SJL
C57BL/6
C57BL/6
C57BL/6
C57BL/6
Asc
-nu/nu
OA
Saline
Asc
Asc
+
Asc
Cs
Fig. 6. Serum IgE levels in several strains of mice transnasally administered Ascaris suum extract, ovalbumin, or saline for 3 weeks. Bars show mean values ± standard deviations. Asc = Ascaris suum extract; OA = ovalbumin; Cs = cyclosporin treatment as described in METHODS in text; • = statistically significant (p < 0.01)as compared with C57BU6 with Asc admi nistration; NS = statistically not significant as compared with C57BU6 with Asc administration.
log (peA titers)
3 2
•
-_fN'
C57BL/6 Asc
•
- fN'
•
- - * --* - - * - - * WBB6FI
-w/Wv Asc
BALB/c Asc
SJL
C57BL/6
C57BL/6
C57BL/6
C57BL/6
Asc
-nu/nu
OA
Saline
Asc
Asc
+
Cs
Fig. 7. Serum PCA titers against Ascaris suum extract (Asc) determined by PCA reaction in several strains of mice transnasally administered Asc, ovalbumin, or saline for 3 wk. Bars show mean values ± standard deviations. Asc = Ascaris suum extract; OA = ovalbumin; Cs = cyclosporin treatment as described in METHODS in text; • = statistically significant (p < 0.05) as compared with C57BU6 mice with Asc administration; NS = statistically not significant as compared with C57BU6 with Asc administration.
by the intravenous injection of Sephadex particles. However, these models produced a relatively small percentage of eosinophils when assessed by BAL. The eosinophilia in our present model was caused by a parasite extract, which could be helpful to elucidate the mechanism of pulmonary eosinophilia. The result that 3 wk of Asc administration were necessary to develop eosinophilia in the BAL of C57BLl6 mice indicates that eosinophilia in this model is not simply or solely dependent on the direct eosinophil chemotactic activity of Asc. Some mechanisms can be hypothetically considered for the development of eosinophilia. One possibility is that Asc-specific IgE is produced, expressed on the surface of mast cells, and IgE-mediated mediator release from mast cells is caused by Asc. Mast cells are known to release various mediators, including eosinophil chemotactic factor of anaphylaxis and plateletactivating factor, which have potent chemotactic activity for eosinophils (12-14). Another possible mechanism is that Asc binds to the surface Ig receptors of T-cells and causes the releaseof cytokines, including interleukin-5, which has the ability to proliferate eosinophils (15). With regard to the IgE-mediated mast cell mediator release, W/WV mice, which are known to be geneticallymast-cell-deficient (5), developed marked eosinophilia in the BAL fluid by Asc. This shows that mast cells are not necessarily required in the formation of eosinophilia in the BAL fluid in this model. We also applied our model of pulmonary eosinophilia to several strains of mice, including IgE low-responder SJL mice (6) to see if serum IgE antibody levels are correlated with eosinophilia in the BAL. The result was that C57BLl6, W/Wv, and Balb-C mice, which developed marked eosinophilia in the BAL, also showed elevated total IgE levels and IgE titers against Ase, whereas SJL and C57BLl6 nude mice, which developed significantly fewer eosinophils, did not. However, oral administration of Cs, 50 mg/kg, which inhibited eosinophilia in the BAL in C57BLl6 mice, did not prevent the production of total IgE or IgE against Asc in the sera. This indicates that serum IgE production seems to accompany marked eosinophilia in this model, but it is not required for the development of eosinophilia. Watanabe and coworkers (16)demonstrated that blood eosinophilia by Nippostrongylus brasiliensis
EXPERIMENTAL PULMONARY EOSINOPHILIA IN MICE BY ASCARIS SUUM
and Trichinella spiralis was produced in SJA/9 mice without elevating serum 19B. They speculated that IgE-independent mechanism, including cytokine release from T-cells, might be involved in murine blood eosinophilia. Whether the slight eosinophilia in the BAL in SJL mice in our model is correlated with IgE-independent mechanism or not needs further study. With regard to T-cell involvement, we used Cs treatment and nude mice. The eosinophilic infiltration in our model was inhibited by Cs, and C57BLl6 nude mice failed to develop eosinophilia as assessed by BAL. These results indicate that T-cells may be involved in the formation of eosinophilia in our model of PIE syndrome. Basten and Beeson (17) demonstrated that blood eosinophilia in rats caused by the intravenous injection of Trichinella spiralis larvae is inhibited by neonatal thymectomy or antilymphocyte sera. They concluded that blood eosinophilia by Trichinella spiralis is related to lymphocytes. Ponzio and Speirs (18) showed by murine cell transfer that eosinophilia in the peritoneal exudate caused by tetanus toxoid is dependent on T-cells.Kano and coworkers (19)showed that eosinophilia, by the injection of Ascarissuum eggs into the peritoneal cavity, is T-cell-dependent. Thomson and colleagues (20) showed that Cs inhibits blood, lymph node, and liver eosinophilia in mice induced by cyclophosphamide administration and immunization with OA in complete Freund adjuvant. Their model failed to produce eosinophilia in the lung. Our data indicate that pulmonary eosinophilia caused by a parasite extract applied transnasally is also T-cell-dependent. Further study is necessary to see if interleukin-5 and other cytokines are involved in this pulmo-
nary eosinophilia. Because blood eosinophilia was not observed in this model, cytokines might be produced and acting in the lung. In conclusion, we report a murine model of pulmonary eosinophilia caused by the transnasal administration of a parasite extract, which is dependent on T-cells but not dependent on IgE or mast cells.
1295
ies I. Characterization of rat monoclonal anti-IgE antibodies and the use of these antibodies for determinations of serum IgE levelsand for anaphylactic reactions. Int Arch Allergy Appl Immunol 1988; 85:47-54. 9. Okudaira H, Suzuki T, Ogita T. A study of the rat passive cutaneous anaphylaxis (PCA) reaction for the assay of mouse IgE antibody. J Immunol Methods 1980; 33:369-77. 10. Wallenstein S, Zucker CL, Fleiss J. Some statistical methods useful in circulation research. Circ Res 1980; 47:1-9. 11. Laycock SM, Smith H, Spicer BA. Airway hyper-reactivity and blood, lung and airway eoAcknowledgment sinophilia in rats treated with Sephadex particles. The writers thank Sandoz, Tokyo, Japan, for Int Arch Allergy Appl Immunol 1986; 81:363-7. kindly supplying cyclosporin. 12. KayAB, Austen KF.The IgE-mediated release of an eosinophil leukocyte chemotactic factor from human lung. J Immunol 1971; 107:899-902. References 13. Wasserman SI, Goetzl EJ, Austen KF. Per1. Crofton JW, LivingstoneJL, Aswarld NC, Robformed eosinophil chemotactic factor of anaphyerts ATV. Pulmonary eosinophilia. Thorax 1952; laxis (ECF-A). J Immunol 1974; 112:351-8. 7:1-35. 14. Wardlaw AJ, Kay AB. PAF-acether is a po2. Sun X, Davis WB, Fukuda Y,Ferrans VJ, Crystent chemotactic factor for human eosinophils. J tal RG. Experimental polymyxin B-induced inter- Allergy Clin Immunol 1986; 77:236. stitiallung disease characterized by an accumula- 15. Yamaguchi Y, Suda T, Suda J, et al. Purified tion of cytotoxic eosinophils in the alveolar strucinterleukin 5 supports the terminal differentiation tures. Am Rev Respir Dis 1985; 131:103-8. and proliferation of murine eosinophilic precur3. TakizawaH, Suko M, KobayashiN, et al. Expersors. J Exp Med 1988; 167:43-56. imental hypersensitivity pneumonitis in the mouse: 16. Watanabe N, Katakura K, Kobayashi A, histologicand immunologic features and their mod- Okumura K, Ovary Z. Protective immunity and ulation with cyclosporin A. J AllergyClin Immunol eosinophilia in IgE-deficient SJA/9 mice infected 1988; 81:391-400. with Nippostrongylus brasiliensis and Trichinella 4. Hess AD, Esa AH, Colombani PM. Mecha- spiralis. Proc Nat! Acad Sci USA 1988;85:4460-2. nism of action of cyclosporin: effect on cells of 17. Basten A, Beeson PB. Mechanism of eothe immune system and on subcellular events in sinophilia II. Role of the lymphocyte. J Exp Med T cell activation. Transplant Proc 1988;20(Suppl: 1970; 131:1288-305. 29-40). 18. Ponzio NM, Speirs RS. Lymphoid cell depen5. Kitamura Y, Go S, Hatanaka K. Decrease of dence of eosinophil response to antigen. VI. The v mast cellsin W/W mice and their increase by bone effect of selective removal of Tor B lymphocytes marrow transplantation. Blood 1978; 52:447-52. on the capacity of primed spleen cells to adoptive6. Okudaira H, TeradaE, YokoyamaM, Miyamo- ly transferred immunity to tetanus toxoid. Immuto T. IgE-isotype-specific suppressor cells in the nology 1975; 28:243-51. mouse: characterization using tetraparental chimera 19. Kano S, Kasuya S, Ohtomo H. Eosinophilia mice of high (DBA/2) and low (SJL) IgE respond- in mouse regulated by multiple factors following er embryos. Int Arch Allergy Appl Immunol1988; repeated stimulations with Ascaris eggs into the 85:462-6. peritoneal cavity. Int Arch Allergy Appl Immunol 7. Kasuya S, Otsuka S, Kano S, Ohtomo H. Inhib- 1989; 89:54-9. itory immunization using complete Freund's adju- 20. ThomsonAW, BrownPAJ, SimpsonJG, Whitvant on ongoing eosinophilia induced by repeated ing PH, Davidson RJL. Inhibition by cyclosporin antigen-stimulations in guinea pigs. Int Arch Al- A of IgE production and cyclophosphamidelergy Appl Immunol 1987; 83:303-9. induced eosinophilia in rats immunized with non8. Hirano T, Miyajima H, Kitagawa H, et al. parasite antigen. Immunol Lett 1985; 9:93-7. Studies on murine IgE with monoclonal antibod-
by Ascaris suum Extract1 •2
MAKOTO NOGAMI, MATSUNOBU SUKO, HIROKAZU OKUDAIRA, TERUMASA MIYAMOTO, JUNJI SHIGA, MAMORU ITO, and SHIRO KASUYA
Introduction
Eosinophils accumulate in the alveoli of the lung in various diseases, including pulmonary infiltration with eosinophilia (PIE) syndrome (1). In vitro, eosinophils are cytotoxic for lung parenchymal cells, and eosinophil granule-derived major basic protein is cytotoxic for tracheal cells (2). In order to clarify the role of eosinophils in PIE syndrome and in other lung disorders with eosinophilic infiltration, the development of an animal model in which eosinophils accumulate in the lung to a great extent is necessary. Crofton and coworkers (1) classified PIE syndrome into five groups. Among them, simple pulmonary eosinophilia and tropical pulmonary eosinophilia are known to be caused by parasites. Transnasal administration has been reported to place material in the lower respiratory tract (3). Therefore, it is supposed that transnasal administration of extract from parasites could cause eosinophilia in the respiratory tract. The purpose of this study was to establish ~, model of eosinophilic infiltration in the lung using an extract of the parasite Ascaris suum, and, for the elucidation of the mechanism of eosinophilia, to see if T-cells, mast cells, and IgE are involved in this model. For this purpose, cyclosporin treatment (4), athymic nude mice, genetically mast-cell-deficient (WB X C57BL/6)FI-W/Wv (W/WV) mice (5), IgE high-responder Balb-C mice, and IgE low-responderSJL mice (6)wereused. Methods Animals C57BLl6, Balb-C, and SJL mice were purchased from Charles River Japan, Atsugi, Japan. C57BLl6-nu/nu (C57BLl6 nude) mice were obtained from the Central Institute for Experimental Animals, Kawasaki, Japan. W/Wv mice were purchased from the Shizuoka Agriculture Cooperative Association for Laboratory Animals, Hamamatsu, Japan. The mice used in this study were male, 7 to 10wk of age, and bred in a specific-pathogen-
SUMMARY The transnasal administration of an extract of the parasite Ascaris suum (Asc) to C57BU6 mice for 3 wk produced marked eosinophilia in the bronchoalveolar lavage (BAL) fluid. This pulmonary eosinophilia was not accompanied by blood eosinophilia. The oral administration of cyclosporin, 50 mg/kg body weight, every other day significantly suppressed the pulmonary eosinophilia. Athymic C57BL/6-nu/nu mice failed to develop pUlmonary eosinophilia. These data indicate that pUlmonary eosinophilia caused by this parasite extract is T-cell-dependent. Genetically mast-celldeficient (WB X C57BU6)F1-W/Wv (W/Wv) mice developed marked eosinophilia in the BAL, which shows that mast cells are not necessary in the formation of eosinophilia in BAL In this model. C57BU6, W/Wv, and Balb-C mice that developed eosinophilia in the BAL also showed elevated totallgE levels and IgE titers against Asc in the sera. On the other hand, C57BU6-nu/nu mice and IgE low-responder SJL mice, which developed little eosinophilia, did not show elevated total IgE levels and IgE titers against Asc in the sera. However, oral administration of cyclosporln, 50 mg/kg, which inhibited eosinophilia in the BAL in C57BU6 mice, did not significantly Inhibit the elevation of totallgE or IgE against Asc in the sera. This indicates that serum IgE production is not required for the formation AM REV RESPIR DIS 1990; 141:1289-1295 of eosinophilia.
free environment (free from infection with Corynebacterium kutscheri, Bacillus piliformis, Salmonel/a, Pseudomonas aeruginosa, Pasturel/a pneumotropica, Bordetel/a branchiseptica,Mycoplasma pulmonis, Sendai virus, or mouse hepatitis virus). Male Sprague-Dawley rats weighing more than 400 g were purchased from CLEA Japan, Tokyo, Japan.
Parasite Extract The Ascaris suum extract (Asc) was obtained in the following manner (7). Adult Ascaris suum worms were obtained from a slaughterhouse. The worms were freeze-dried, defatted with ethyl ether, extracted with veronalbuffered saline (VBS), and dialyzed against VBS. The extracts were dissolved in saline at a concentration of 5 mg/ml. Administration of Asc, Saline, or Ovalbumin to C57BL/6 Mice Each C57BLl6 mouse was anesthetized by ethyl ether and given 40 ul/day of the above Asc solution by trans nasal drops 2 consecutive days a week for I, 2, or 3 wk; 48 h after the last administration, blood was collected from retroocular veins using a glass microsyringe, and bronchoalveolar lavage(BAL) fluid was obtained as described below. The mice not used for BAL were used for histologic studies as described below. Some C57BLl6 mice were administered saline or 5 mg/ml saline solution of ovalbumin (OA), crystallized five times (Seikagaku Kogyo, Tokyo, Japan),
for 3 wk, and the BAL fluid was collected in the same manner as the Asc.
Administration of Asc to Several Strains of Mice Balb-C, SJL, C57BLl6 nude, and W/W V mice were anesthetized by ethyl ether and administered Asc 2 consecutive days a week for 3 wk as described above; 48 h after the last administration, blood and BAL fluid werecollected 3S described below. Cyc/osporin Treatment Cyclosporin (Cs) was kindly supplied by Sandoz (Tokyo, Japan). Cs was dissolved in olive oil at a concentration of 2 mg/ml just before use. Some C57BLl6 mice, which had been administered Asc for 3 wk in the same manner as described above, were orally adminis-
(Received in original form February 7, 1989 and in revised form September ll, 1989) 1 From the Department of Medicine and Physical Therapy and the Department of Pathology, Faculty of Medicine, University of Tokyo, Tokyo; the Central Institute of Experimental Animals, Kawasaki; and the Department of Parasitology, Gifu University School of Medicine, Gifu, Japan. 2 Correspondence and requests for reprints should be addressed to Makoto Nogarni, Department of Medicine and Physical Therapy, Faculty of Medicine, University of Tokyo, 7-3-1, Hongo, Bunkyoku, Tokyo, Japan 113.
1289
1290
tered the Cs solution through a gastric tube at 50 mg/kg body weighteveryother day from the first day of Asc administration to the day the mice were killed. For the control animals, olive oil alone was administered in the same manner. Bronchoalveolar Lavage The mice were killed by ethyl ether and resected by median incision. The trachea was exposed in the neck, and a 21-gauge Teflon'" tube (Terumo, Tokyo, Japan) was inserted into the trachea. Saline, I ml, was infused and collected through the tube repeatedly to gain 4 ml of BAL fluid for each mouse; 5 ml of saline were required to gain 4 ml of BAL fluid. The cells in the fluid werethen washed with RPMI 1640supplemented with L-glutamine(GIBCO Laboratories, Grand Island, NY), centrifuged three times, and suspended in 400 Itl of RPMI 1640. The cells of BAL fluid were counted with a hemacytometer by trypan blue exclusion. The cells in the BAL fluid were sprayed onto a glass slide by Cytospin II (Shandon Southern Products, Cheshire, England) and stained with Diff-Quik'" (Midorijuji, Kobe, Japan), and the cell differentials were counted. At least 500 cells were counted for cell differentials (3). Numbers of eosinophils in the BAL were calculated by multiplying total cell numbers by percentages of eosinophils. Part of the cellsin the BAL fluid were fixed in 2.5070 solution of glutaraldehyde and postfixed in 2% osmium tetroxide solution for 2 h. After the dehydration in graded ethanol solutions, the cells were embedded in Epon 512. Ultrathin sections were cut, doublestained with uranyl acetate and lead citrate, and then examined with a Nippon Denshi JM 100C electron microscope (Nippon Denshi, Tokyo, Japan). Histologic Study The lungs of C57BL/6 mice not used for the BAL study were fixed in 10% formalin, embedded in paraffin, sectioned, and stained with hematoxylin-eosin. Blood Leukocyte Counts and Cell Differentials The blood samples were studied for leukocyte counts and cell differentials. For leukocyte counts, the blood was stained with Turk's solution, and the cells were counted in a hemacytometer. Glass slides were streaked with blood and stained with May-Grunwald and Giemsa solutions, and leukocyte differentials were counted microscopically. Numbers of eosinophils were calculated by multiplying total leukocyte counts by percentages of eosinophils. Serum IgE Levels Serum specimens wereprepared from part of the blood collected. Serum IgE was determined by enzyme-linked immunosorbent assay (ELISA) as previously described by Hirano and coworkers (8). Briefly, 96-well EIA plates (Coster, Cambridge, MA) were
NOGAMI, SUKO, OKUDAIRA, MIYAMOTO, SHIGA, ITO, AND KASUYA
used. PBS containing 0.05% Tween" 20 was used to dilute the samples and for washing. For blocking nonspecific binding sites, the same buffer containing 1% BSA was used. The volume pipetted in this assay was 50 ul, except for blocking, where it was 100 ul. The wells were coated with monoclonal rat antimouse IgE antibody (YamasaShoyu, Tokyo, Japan) at 4 C overnight. After blocking, 1:10 or I :100 dilutions of sera or standard twofold serial dilutions were applied. Monoclonal mouse IgE antidinitrophenyl antibody (SPE-7; Miles Laboratories, Naperville, IL) was used as standard. The plate was kept for I h at room temperature, and then biotinylated monoclonal rat antimouse IgE (Yamasa Shoyu, Tokyo, Japan), which recognizes the different epitope on the Fc fragment of IgE from that used for the primary coating, was added. After I h at room temperature, the wells were washed, and peroxidase-labeled avidin D (Vector Laboratories, Burlingame, CAl was added and kept for I h at room temperature. The substrate to develop the reaction was the solution containing 0.04% H 20 2 and 0.004% o-phenylene-diamine (Sigma Chemical Co., St. Louis, MO) in 0.2 M sodium phosphate, 0.1 M sodium citrate buffer (pH, 5.0). Thirty minutes after addition of the substrate, reaction was stopped with 4 N H 2S04 and was read at 490 nm with an ELISA reader (Corona Electric Co., Tokyo, Japan). Serum IgE titers against Asc were determined by passive cutaneous anaphylaxis (PCA) reaction by the modified method of Okudaira and coworkers (9). Briefly, 50 ul of serial dilutions of sera (e.g., I:5, I:10, 1:20) were injected intracutaneously into shaved
dorsal skin of a Sprague-Dawley rat. After 18 h, the rat was given an intravenous injection of I ml of I% Evans blue (Merck, Darmstadt, Germany), which was immediately followed by an intracutaneous injection of 50 ul of Asc (10 ug/rnl) to each site where the sera were injected. Twenty minutes later, the rat was killed by carbon dioxide gas, and the reaction was evaluated by examining the inside of the skin. The PCA titer was expressed by the reciprocal of the final dilution givinga positive bluing reaction 5 mm or more in diameter.
0
Statistics Student's t test was used to compare the values between two groups. Bonferroni's test was used to compare the values among more than two groups (10). A p value below 0.05 was regarded as statistically significant.
Results HAL Fluid
Asc was given transnasally to see if eosinophilia in the BAL was induced. The numbers of eosinophils in the BAL fluids of several strains of mice are shown in figure 1. C57BL/6 mice given Asc for 3 wk developed significant eosinophilia as compared with C57BL/6 mice given saline for 3 wk (p < 0.01). Eosinophils in the BAL fluid of C57BL/6 mice given Asc for 3 wk are shown in figure 2A, whereas figure 2B shows macrophages and lymphocytes in the BAL fluid of C57BL/6 mice given saline for 3 wk. The
Cell number
DI
(x 104)
total cell number in SAL
eosinophil number in SAL
50
WBB6Fl -W/Wv
BALB/c
S-.JL
C57BL/6
C57BL/6
Asc
Asc
Saline
Asc
n=10
n=9
-nu/nu Asc
n= 5
n=10
n= 5
Fig. 1. Eosinophil numbers in the bronchoalveolar lavage (BAL) fluid of several strains of mice given Ascarissuum extract or saline transnasally for 3 wk. Bars show mean values ± standard deviations. Asc = Ascaris suum extract; * = statistically significant (p < 0.01) as compared with C57BU6 mice with Asc administration.
1291
EXPERIMENTAL PULMONARY EOSINOPHILIA IN MICE BY ASCARIS SUUM
B
c Fig. 2. Cells in the bronchoalveolar lavage (BAL) fluid. A. Cells in C57BU6 mice given Ascaris suum extract (Asc) transnasally for 3 weeks. Light microscopy. Diff-Quik stain. Original magnification, x 400. B. Cells in C57BU6 mice given saline transnasally for 3 weeks. Light microscopy. Diff-Quik stain. Original magnification, x 400. C. An eosinophil containing specific granules in C57BU6 mice given Asc transnasally for 3 wk. Electron microscopy. Original magnification, x 5,000. D. A degranulated eosinophil in C57BU6 mice given Asc transnasally for 3 wk. Original magnification, x 5,000.
electron microscopy of the BAL fluid cells of C57BL/6 mice given Asc for 3 wk shows the eosinophils with characteristic granules in the cytoplasm (figure 2C). Some eosinophils from three of 10 mice showed degranulation in the electron microscopy although less than 50/0 of the total eosinophils in number (fig-
ure 2D). W/Wv and Balb-C mice given Asc for 3 wk also showed eosinophilia in the BAL fluid (figure 1). Because C57BL/6 mice produced the most eosinophils in the BAL, further studies were carried out mainly on C57BL/6 mice. The time required for the development of eosinophilia in C57BL/6 mice was
studied. The cell numbers and fractions in the BAL fluid of C57BL/6 mice given Asc for 1 to 3 wk are shown in table 1. The mice given Asc for 1 or 2 wk showed less eosinophilia than did those given Asc for 3 wk (p < 0.01); 3 wk were necessary to cause marked eosinophilia. In order to see if eosinophilia in the
1292
NOGAMI, SUKO, OKUDAIRA, MIYAMOTO, SHIGA, ITO, AND KASUYA
TABLE 1 CELL NUMBERS AND CELL DIFFERENTIALS IN THE BRONCHOALVEOLAR LAVAGE (BAL) FLUID OF C57BLl6 MICE
Material Asc
Weeks
Cell Numbers (x 104 )
E
N
M
L
Number of Eosinophils (x 104 )
1 2 3
10 ± 4 15 ± 4 60 ± 10
5 ± 5 20 ± 10 51 ± 8
30 ± 5 14 ± 8 9 ± 3
64 ± 5 65 ± 13 35 ± 6
1 ± 1 1 ± 1 5 ± 3
0.5 ± 0.8' 3.3 ± 2.1' 31.0 ± 7.0
Saline
3
8±2
Ovalbumin
3
8 ± 2
Cell Differentials (%)
o± o±
0
o±
0
97 ± 2
3 ± 1
0.0 ± 0.0'
0
O±O
97 ± 1
3 ± 1
0.0 ± 0.0'
Definition of abbreviations: Asc = Ascaris suum extract; S = saline; OA = ovalbumin; E = eosinophil; N = neutrophil; M = macrophage; L = lymphocyte; • = statistically significant (p < 0.01) as compared with mice given Asc for 3 wk. Figures show mean values ± standard deviations. The number of mice tested was 10 for each group except ovalbumin, where it was 5. Weeks means those for which the materials were transnasally administered.
en Asc transnasally for 3 wk, which showed significantly fewer eosinophils in the BAL fluid than in C57BLl6 mice given Asc (p < 0.01) (figure 3A). Cs treatment significantly inhibited the development of eosinophilia in the BAL fluid of C57BL/6 mice given Asc for 3 wk as compared with olive oil (p < 0.01) (figure 3B). In order to see if mast cells are involved in this eosinophilia, W/Wv mice were given Asc for 3 wk. W/Wv mice developed a statistically significant increase in eosinophils as compared with W/Wv mice given saline (p < 0.01) (figure 4).
Histologic Study Cell number
(xl0') Cell number
( x 10')
50
DI
50
DI
total cellnumber in BAl
total cellnumber in BAl
eosinophil number in BAl
o C57BL/6 +Asc
C57BL/6
n=10
+Asc
eosinophil number in BAl
Blood Leukocyte Counts and Cell Differentials C57BL/6 +Asc
C57BL/6 +Asc
+
+
cyclosporin no, Oliveoil p.o, 50mg/kg everyother day everyother day n= 5 n=10
-nu/nu n=5
The lungs of C57BLl6 mice given Asc for 3 wk developed an accumulation of eosinophils and lymphocytes around vessels and alveolar septa (figure 5A and B). The lungs of C57BL/6 mice given saline for 3 wk or of nontreated mice showed a slight infiltration of neutrophils and lymphocytes into the alveolar septa, but few eosinophils were found (figure 5C and D).
Fig. 3. Eosinophil numbers in the bronchoalveolar lavage (BAL) fluid of C57BU6 nude mice and cyclosporintreated C57BU6 mice given Ascaris suum extract (Asc). All mice were given Asc transnasally for 3 wk. Bars show mean values ± standard deviations. A. Eosinophil numbers in C57BU6 nude mice.• = statistically significant (p < 0.01) as compared with the eosinophil number in C57BU6 with Asc administration. B. Eosinophil numbers in cyclosporin treatment (50 mgikg body weight every other day). ' = statistically significant (p < 0.01) as compared with the eosinophil number in C57BU6 with Asc and olive oil administration.
The peripheral blood leukocyte count was carried out to see if pulmonary eosinophilia in C57BL/6 mice was accompanied by blood eosinophilia. The peripheral blood leukocyte count of C57BL/6 mice at the third week is shown in table 2. No significant increase in blood eosinophils was found in C57BL/6 mice given Asc compared with C57BL/6 mice given saline.
Serum IgE Levels Cell number
(X 10 4)
DI total
WBB6FI-W/WV
eosinophil
cell number
number
In SAL
in SAL
Fig. 4. Eosinophil numbers in the bronchoalveolar lavage (BAL) fluid of (WB X C57BU6)Fl-W/wv mice given Ascaris suum extract transnasally for 3 wk. Bars show mean values ± standard deviations. WBB6Fl-W/wv = (WB X C57BU6)Fl-W/wv mice; Asc = Ascaris suum extract; , = statistically significant (p < 0.01) as compared with the eosinophil number in mice with saline administration.
WBB6FI-W/Wv
Ase
Saline
n=5
n=5
BAL fluid of C57BLl6 mice can be caused by other materials, OA was given transnasally for 3 wk. OA failed to show eosinophilia in the BAL fluid in C57BLl6
mice as compared with Asc (p < 0.01) (table 1). To see if T-cells are involved in eosinophilia, C57BLl6 nude mice were giv-
The following study was carried out to see if serum IgE levels were correlated with eosinophilia in the BAL. The serum IgE levels are shown in figure 6. In terms of serum IgE levels, there was no statistically significant difference between C57BLl6 and W/Wv or C57BLl6 and Balb-C mice given Asc for 3 wk, whereas SJL or C57BL/6 nude mice given Asc for 3 wk and C57BLl6 mice given OA for 3 wk showed significantly lower serum IgE levels than did C57BLl6 mice given Asc for 3 wk (p < 0.01). C57BLl6 mice given Asc for 3 wk showed statistically significant elevated IgE as compared with C57BL/6 mice given saline for 3 wk (p < 0.01). IgE antibody titers against Asc by PCA reaction are shown in figure 7. There was no statistically significant difference in titers between C57BLl6 and W/Wv or C57BLl6 and Balb-C mice given Asc for 3 wk, where-
1293
EXPERIMENTAL PULMONARY EOSINOPHI!..IA IN MICE BY ASCARIS SUUM
Fig . 5. Lung histology. Light microscopy. A. C57BU6 mice given Ascaris suum extract (Asc) transnasally for 3 wk . Hematoxylin-eosin stain . Fixed in formalin and embedded in paraffin . Original magnification, x10. B. C57BU6 mice given Asc transnasally for 3 wk. Hematoxylin-eosin stain. Fixed in for malin and embedded in paraffin . Orig inal magn ification , x 330. Eosinophils are indicated by thick arrows. A macrophage is indicated by a thin arrow. C. C57BU6 mice given saline transnasally for 3 wk. Hematoxylin-eosin stain . Fixed in formalin and embedded in paraffin . Original magnification, xl0. D. Nontreated C57BU6 mice of the same age as A, B, and C. Hematoxylin-eosin stain. Fixed in formalin and embedded in paraf fin. Or iginal magnification, x10.
as SJL or C57BL/6 nude mice given Asc for 3 wk and C57BLl6 mice given OA for 3 wk showed significantly lower titers than did C57BLl6 mice given Asc for 3 wk (p < 0.05).The titers in C57BLl6 mice given Asc for 3 wk were significantly elevated compared with C57BLl6 mice given saline for 3 wk (p < 0.05). Oral administration of Cs, 50 mg/kg, which
inhibited eosinophilia in the BAL in C57BLl6 mice given Asc for 3 wk, did not prevent the production of total IgE or IgE antibody against Asc in the sera (figures 6 and 7). Discussion
The transnasal administration of Asc to C57BL/6 mice for 3 wk caused marked
eosinophilic infiltration in the lung. Simple pulmonary eosinophilia and tropical pulmonary eosinophilia in human beings are known to be caused by parasitic infection (1). Sun and coworkers (2) developed eosinophilia in the lung of guinea pigs by the aerosol administration of polymyxin B (2). Laycock and colleagues (11) developed lung eosinophilia in rats
1294
NOGAMI, SUKO, OKUDAIRA, MIYAMOTO, SHIGA, ITO, AND KASUYA
TABLE 2 BLOOD LEUKOCYTE COUNTS AND CELL DIFFERENTIALS IN C57BLl6 MICE
Mat
Leukocyte Count (x 103//1 /)
N
E
B
L
M
Number of Eosinophils (x 1(2)
Asc Saline
5.0 ± 0.5 4.5 ± 0.7
50.5 ± 5.0 49.5 ± 5.0
4.0 ± 1.0 4.0 ± 1.5
0.5 ± 0.1 0.5 ± 0.3
40.0 ± 5.0 41.0 ± 5.0
5.0 ± 1.0 5.0 ± 1.0
2.0 ± 0.5 1.8 ± 0.6 NS
Leukocyte Differentials (%)
Definition of abbreviations: Mat = material transnasally administered; E = eosinophil; N = neutrophil; B = basophil; L = lymphocyte; M = macrophage; Asc = Ascaris suum extract; S = saline; NS = statistically not significact as compared with mice given Asc. Figures show mean values ± standard deviations. The number ot mice tested tor each group was 10. Each material was transnasally administered tor 3 wk as described in text.
log IgE (ng/ml)
5
•
4
I
3
• •
f
2
•
-f* --*
C57BL/6 Asc
WBB6Fl
-w/Wv
BALB/c Asc
SJL
C57BL/6
C57BL/6
C57BL/6
C57BL/6
Asc
-nu/nu
OA
Saline
Asc
Asc
+
Asc
Cs
Fig. 6. Serum IgE levels in several strains of mice transnasally administered Ascaris suum extract, ovalbumin, or saline for 3 weeks. Bars show mean values ± standard deviations. Asc = Ascaris suum extract; OA = ovalbumin; Cs = cyclosporin treatment as described in METHODS in text; • = statistically significant (p < 0.01)as compared with C57BU6 with Asc admi nistration; NS = statistically not significant as compared with C57BU6 with Asc administration.
log (peA titers)
3 2
•
-_fN'
C57BL/6 Asc
•
- fN'
•
- - * --* - - * - - * WBB6FI
-w/Wv Asc
BALB/c Asc
SJL
C57BL/6
C57BL/6
C57BL/6
C57BL/6
Asc
-nu/nu
OA
Saline
Asc
Asc
+
Cs
Fig. 7. Serum PCA titers against Ascaris suum extract (Asc) determined by PCA reaction in several strains of mice transnasally administered Asc, ovalbumin, or saline for 3 wk. Bars show mean values ± standard deviations. Asc = Ascaris suum extract; OA = ovalbumin; Cs = cyclosporin treatment as described in METHODS in text; • = statistically significant (p < 0.05) as compared with C57BU6 mice with Asc administration; NS = statistically not significant as compared with C57BU6 with Asc administration.
by the intravenous injection of Sephadex particles. However, these models produced a relatively small percentage of eosinophils when assessed by BAL. The eosinophilia in our present model was caused by a parasite extract, which could be helpful to elucidate the mechanism of pulmonary eosinophilia. The result that 3 wk of Asc administration were necessary to develop eosinophilia in the BAL of C57BLl6 mice indicates that eosinophilia in this model is not simply or solely dependent on the direct eosinophil chemotactic activity of Asc. Some mechanisms can be hypothetically considered for the development of eosinophilia. One possibility is that Asc-specific IgE is produced, expressed on the surface of mast cells, and IgE-mediated mediator release from mast cells is caused by Asc. Mast cells are known to release various mediators, including eosinophil chemotactic factor of anaphylaxis and plateletactivating factor, which have potent chemotactic activity for eosinophils (12-14). Another possible mechanism is that Asc binds to the surface Ig receptors of T-cells and causes the releaseof cytokines, including interleukin-5, which has the ability to proliferate eosinophils (15). With regard to the IgE-mediated mast cell mediator release, W/WV mice, which are known to be geneticallymast-cell-deficient (5), developed marked eosinophilia in the BAL fluid by Asc. This shows that mast cells are not necessarily required in the formation of eosinophilia in the BAL fluid in this model. We also applied our model of pulmonary eosinophilia to several strains of mice, including IgE low-responder SJL mice (6) to see if serum IgE antibody levels are correlated with eosinophilia in the BAL. The result was that C57BLl6, W/Wv, and Balb-C mice, which developed marked eosinophilia in the BAL, also showed elevated total IgE levels and IgE titers against Ase, whereas SJL and C57BLl6 nude mice, which developed significantly fewer eosinophils, did not. However, oral administration of Cs, 50 mg/kg, which inhibited eosinophilia in the BAL in C57BLl6 mice, did not prevent the production of total IgE or IgE against Asc in the sera. This indicates that serum IgE production seems to accompany marked eosinophilia in this model, but it is not required for the development of eosinophilia. Watanabe and coworkers (16)demonstrated that blood eosinophilia by Nippostrongylus brasiliensis
EXPERIMENTAL PULMONARY EOSINOPHILIA IN MICE BY ASCARIS SUUM
and Trichinella spiralis was produced in SJA/9 mice without elevating serum 19B. They speculated that IgE-independent mechanism, including cytokine release from T-cells, might be involved in murine blood eosinophilia. Whether the slight eosinophilia in the BAL in SJL mice in our model is correlated with IgE-independent mechanism or not needs further study. With regard to T-cell involvement, we used Cs treatment and nude mice. The eosinophilic infiltration in our model was inhibited by Cs, and C57BLl6 nude mice failed to develop eosinophilia as assessed by BAL. These results indicate that T-cells may be involved in the formation of eosinophilia in our model of PIE syndrome. Basten and Beeson (17) demonstrated that blood eosinophilia in rats caused by the intravenous injection of Trichinella spiralis larvae is inhibited by neonatal thymectomy or antilymphocyte sera. They concluded that blood eosinophilia by Trichinella spiralis is related to lymphocytes. Ponzio and Speirs (18) showed by murine cell transfer that eosinophilia in the peritoneal exudate caused by tetanus toxoid is dependent on T-cells.Kano and coworkers (19)showed that eosinophilia, by the injection of Ascarissuum eggs into the peritoneal cavity, is T-cell-dependent. Thomson and colleagues (20) showed that Cs inhibits blood, lymph node, and liver eosinophilia in mice induced by cyclophosphamide administration and immunization with OA in complete Freund adjuvant. Their model failed to produce eosinophilia in the lung. Our data indicate that pulmonary eosinophilia caused by a parasite extract applied transnasally is also T-cell-dependent. Further study is necessary to see if interleukin-5 and other cytokines are involved in this pulmo-
nary eosinophilia. Because blood eosinophilia was not observed in this model, cytokines might be produced and acting in the lung. In conclusion, we report a murine model of pulmonary eosinophilia caused by the transnasal administration of a parasite extract, which is dependent on T-cells but not dependent on IgE or mast cells.
1295
ies I. Characterization of rat monoclonal anti-IgE antibodies and the use of these antibodies for determinations of serum IgE levelsand for anaphylactic reactions. Int Arch Allergy Appl Immunol 1988; 85:47-54. 9. Okudaira H, Suzuki T, Ogita T. A study of the rat passive cutaneous anaphylaxis (PCA) reaction for the assay of mouse IgE antibody. J Immunol Methods 1980; 33:369-77. 10. Wallenstein S, Zucker CL, Fleiss J. Some statistical methods useful in circulation research. Circ Res 1980; 47:1-9. 11. Laycock SM, Smith H, Spicer BA. Airway hyper-reactivity and blood, lung and airway eoAcknowledgment sinophilia in rats treated with Sephadex particles. The writers thank Sandoz, Tokyo, Japan, for Int Arch Allergy Appl Immunol 1986; 81:363-7. kindly supplying cyclosporin. 12. KayAB, Austen KF.The IgE-mediated release of an eosinophil leukocyte chemotactic factor from human lung. J Immunol 1971; 107:899-902. References 13. Wasserman SI, Goetzl EJ, Austen KF. Per1. Crofton JW, LivingstoneJL, Aswarld NC, Robformed eosinophil chemotactic factor of anaphyerts ATV. Pulmonary eosinophilia. Thorax 1952; laxis (ECF-A). J Immunol 1974; 112:351-8. 7:1-35. 14. Wardlaw AJ, Kay AB. PAF-acether is a po2. Sun X, Davis WB, Fukuda Y,Ferrans VJ, Crystent chemotactic factor for human eosinophils. J tal RG. Experimental polymyxin B-induced inter- Allergy Clin Immunol 1986; 77:236. stitiallung disease characterized by an accumula- 15. Yamaguchi Y, Suda T, Suda J, et al. Purified tion of cytotoxic eosinophils in the alveolar strucinterleukin 5 supports the terminal differentiation tures. Am Rev Respir Dis 1985; 131:103-8. and proliferation of murine eosinophilic precur3. TakizawaH, Suko M, KobayashiN, et al. Expersors. J Exp Med 1988; 167:43-56. imental hypersensitivity pneumonitis in the mouse: 16. Watanabe N, Katakura K, Kobayashi A, histologicand immunologic features and their mod- Okumura K, Ovary Z. Protective immunity and ulation with cyclosporin A. J AllergyClin Immunol eosinophilia in IgE-deficient SJA/9 mice infected 1988; 81:391-400. with Nippostrongylus brasiliensis and Trichinella 4. Hess AD, Esa AH, Colombani PM. Mecha- spiralis. Proc Nat! Acad Sci USA 1988;85:4460-2. nism of action of cyclosporin: effect on cells of 17. Basten A, Beeson PB. Mechanism of eothe immune system and on subcellular events in sinophilia II. Role of the lymphocyte. J Exp Med T cell activation. Transplant Proc 1988;20(Suppl: 1970; 131:1288-305. 29-40). 18. Ponzio NM, Speirs RS. Lymphoid cell depen5. Kitamura Y, Go S, Hatanaka K. Decrease of dence of eosinophil response to antigen. VI. The v mast cellsin W/W mice and their increase by bone effect of selective removal of Tor B lymphocytes marrow transplantation. Blood 1978; 52:447-52. on the capacity of primed spleen cells to adoptive6. Okudaira H, TeradaE, YokoyamaM, Miyamo- ly transferred immunity to tetanus toxoid. Immuto T. IgE-isotype-specific suppressor cells in the nology 1975; 28:243-51. mouse: characterization using tetraparental chimera 19. Kano S, Kasuya S, Ohtomo H. Eosinophilia mice of high (DBA/2) and low (SJL) IgE respond- in mouse regulated by multiple factors following er embryos. Int Arch Allergy Appl Immunol1988; repeated stimulations with Ascaris eggs into the 85:462-6. peritoneal cavity. Int Arch Allergy Appl Immunol 7. Kasuya S, Otsuka S, Kano S, Ohtomo H. Inhib- 1989; 89:54-9. itory immunization using complete Freund's adju- 20. ThomsonAW, BrownPAJ, SimpsonJG, Whitvant on ongoing eosinophilia induced by repeated ing PH, Davidson RJL. Inhibition by cyclosporin antigen-stimulations in guinea pigs. Int Arch Al- A of IgE production and cyclophosphamidelergy Appl Immunol 1987; 83:303-9. induced eosinophilia in rats immunized with non8. Hirano T, Miyajima H, Kitagawa H, et al. parasite antigen. Immunol Lett 1985; 9:93-7. Studies on murine IgE with monoclonal antibod-
