Histochemistry52, 273-279 (1977)
Histochemistry 9 by Springer-Verlag 1977
Localization of Eosinophils in the Thymus by the Peroxidase Reaction Edith Mfiller Institute of Anatomy, Universityof Ziirich (Director: Prof. Dr. G. T6ndury) GloriastraBe 19, CH-8006 Ziirich, Switzerland
Summary. Thymuses of human fetuses and infants and of young mice were investigated histochemically for peroxidase. Eosinophils were shown to be the only peroxidase-positive cells in the thymus. In human thymuses the eosinophilic cells were predominantly localized in medullar areas, with concentration of cell clusters at the cortico-medullar junction, around or inside Hassall's bodies and occasionally in high numbers in the intraseptal vessels of the cortex. In the normal mouse the eosinophils were evenly distributed throughout the medulla. Treatment with corticosteroids or X-rays produced a severe involution of the thymus with concommitant change in cellular pattern. The central areas of the thymus residue contained lymphocytes while the peripheral regions consisted of reticuloepithelia, macrophages and numerous eosinophils. Azathioprine did not change the morphology of the thymus. The numbers of eosinophils were slightly reduced, the distribution pattern remaining unchanged.
Introduction It is well known that the thymus contains numerous enzymes with activities that fluctuate as an expression of the functional state of the organ. Peroxidase(EC 1.11.1.7) activity, however, is not measurable (Arvy, 1973). Experiments with intraperitoneally injected horse-radish peroxidase showed to some extent the fate of the exogenously applied enzyme in the thymus (Gervin and Holtzman, 1972). The authors found that, within 30 min of injection, the peroxidase was localised in macrophages both in the blood vessels and adjacent connective tissue. In saline-injected controls, peroxidase activity was found in erythrocytes and occasionally in cells resembling eosinophil leucocytes.
274
E. Mtiller
However, it has been demonstrated that eosinophils contain large quantities of peroxidase (Ryt6maa, 1960; Archer, 1963; Bainton and Farquhar, 1970; Okun et al., 1970) which has different properties from myeloperoxidase (Archer, 1965). The aim of the present study was to analyse the normal distribution of peroxidase-active cells and the fluctuation of their numbers in the thymus after treatment with corticosteroids and X-rays, both of which influence the migration of eosinophils (Essellier etal., 1960; Lee and Domm, 1967; Henning etal., 1973; Blau, 1967; Ghossein et al., 1975). Materials and Methods The tissue specimens of h u m a n thymuses were obtained from stillborn fetuses (22, 25 and 27 weeks of gestation); from four children still-born at term; and from five children aged 1, 2~/2, 4, 5, and 12 m o n t h s who had died suddenly without previous manifestation of illness. For the experiments with immunesuppressive agents, thymuses from I C R mice, all male 9 weeks old, were used. They were divided into 3 groups of ten and each group was treated as follows: group a) sodium cortisoneacetate (1 rag) was injected i.p. daily for two days and the animals killed 24 h after the second treatment; group b) the entire bodies of the animals were irradiated for 16 min with 850 r (180 kv, 0.5 m m Cu-filter) and the mice were killed 24, 48 and 72 h after this single X-ray application; group c) azathioprine sodium (6 mg/kg, Imuran, Burroughs Wellcome & Co) was injected i.p. daily for 5 days and the animals killed three days later after the final injection. The control animals were killed at the same time intervals. Small pieces of thymuses were frozen at - 7 0 ~ with solid CO2 and cut at 8 g m in the cryostat. The sections were dried at 4 ~ C before incubation for enzy/ne demonstration or staining by the Giemsa-May-Grfinwald method. In light microscopy sections, peroxidase was demonstrated by the benzidine method of Van Duijn, 1955 (Pearse, 1972) with and without addition of L-DOPA. In electronmicroscopy studies the enzyme was visualized by the diaminobenzidine reaction of Grah a m and Karnovsky (Okun et al., 1970). Tyrosine was demonstrated by the method of Rappaport (Pearse, 1972). Control sections were incubated in media lacking HzO2, benzidine or L - D O P A respectively.
Results
The distribution of peroxidase-active cells in the human thymuses studied was variable. Usually they were diffusely spread in the medulla (Fig. 1) and were occasionally concentrated as cell clusters on the cortico-medullary border or around the Hassall bodies (Fig. 2). Intact and disrupted peroxidase-positive cells were also demonstrated within the Hassall bodies (Fig. 3). In the cortex, only single isolated peroxidase-active cells could be observed. Occasionally the intraseptal blood vessels of the cortex were surrounded by peroxidase-active cells, whereas none or few were found in the medulla. Fig. 1. T h y m u s from a 2-month-old child post mortem, incubated for peroxidase. Dark dots are cells with peroxidase-reaction product. Enzyme-positive cells spread mostly in the medullar parts, or as cell clusters on the cortico-medullar (cm) border, or along the septal vessels (sv). ca. x 28 Fig. 2. Medullar part from a thymus of a newborn child. Note the numerous peroxidase-active cells around one and inside of another Hassall's body (h). ca. x 93
Fig. 3. Detail of a medullar part from a t h y m u s of a newborn child. Two Hassall's bodies (h) loaded with peroxidase-active cells, x 375 Fig. 4. Eosinophil granulocyt in the mouse t h y m u s incubated for peroxidase. Heavy reaction-deposit in the granules (g). The cell nucleus (n) is visible in two parts, x 14,000
Fig. 5. Normal mouse t h y m u s incubated for peroxidase. Note the n u m e r o u s active cells spread almost regularly t h o u r o u g h the medulla. Only single cells with peroxidase-reaction product are in the cortex, x 56
Fig. 6. Corticosteroid-treated mouse thymus incubated for peroxidase. No cortico-medullary structure in the atrophied thymus. The peroxidase-active cells now mostly in the border and less in the middle of the organ, x 56
Eosinophils in the Thymus
277
After treatment of mice with corticosteroids and X-ray irradiation, the Tyrosinase (EC 1.10.3.1) had the same distribution as peroxidase (EC 1.11.1.7) in cells in the thymus. In m o u s e thymuses, the peroxidase-active cells were also confined to the medulla where they were regularly dispersed (Fig. 5). The EM-investigations of the thymuses incubated for peroxidase, showed enzyme-active granules characteristic of eosinophils (Fig. 4). After treatment of mice with corticosteroids and X-ray irradiation, the thymus involuted to a small residue, the central region of which contained densly packed lymphocytes and single peroxidase-active cells. The periphery showed a loose arrangement of reticular epithelia, numerous macrophages and a high number of peroxidase-positive cells (Figs. 6 and 7). In azathioprine-treated animals, the thymus retained normal structure but the peroxidase-active cells were less numerous than in controls. Most of the peroxidase-active cells were located as clusters on the cortico-medullary border or in the intraseptal vessels (Fig. 8).
Discussion
A stable histochemical method is a prerequisite for the study of peroxidaseactivity in the thymus. By adding L-DOPA (Okun et al., 1970), the quick-fading reaction product of the benzidine-peroxidase method (van Duijn, 1955) is stabilized. This allows an easy and reliable light microscopical study of the distribution of peroxidase-active cells in the tissue. At the EM-level, this method provides strong evidence that tile eosinophil contain peroxidase (Okun et al., 1970; Bainton and Farquhar, 1970). Recently, studies of c o r t i c o s t e r o i d influence on eosinophils (Thorn and Forsham, 1948; Essellier etal., 1954) have undergone a revival. Lee and Domm (1967) described, that within six hours after administration of 1.5 mg of hydrocortisone to rats, eosinophils become prominent either at the cortico-medullary border, or as perivascular clusters. Henning et al. (1973) found an extreme involution of rabbit thymus and a 2--3 fold increase of eosinophils at its borders 12 h after hydrocortisone treatment (10 mg/kg). In the present study the thymus of hydrocortisone-treated mice, displayed within 24 h a similar extreme involution and distribution pattern of eosinophils, to that observed by Henning et al. (1973). Eosinophils disappear from the blood stream after corticosteroid treatment and accumulate predominantly in the lymphoid organs (Essellier et al., 1954; Ryt6maa, 1960). This behaviour may be attributed to some eosinophilotactic substance mediated by lymphocytes, as reported by Litt (1964) and by Basten and Beeson (1970). Studies on the thymus-dysgenetic nude mouse Fig. 7. X-ray treated mouse thymus. Note here the same structure and peroxidase-active cell-pattern as in the corticosteroid-treated mouse Figure 6. x 56 Fig. 8, lmuran-treated mouse thymus incubated for peroxidase. Arrows indicate enzyme-positive ceil clusters (cO on the cortico-medullar border. In the medullar p a r t - t e s s peroxidase-active cells are present compared with the control mouse t h y m u s Figure 5. • 56
278
E. Mfiller
(Miller et al., 1976; Hsu and Hsu, 1976) and on thymic-depleted mice (Ponzio and Speirs, 1973) have indicated that the presence of T-lymphocytes is necessary for eosinophil migration into the affected tissue sites. As Blau (1967) has described, 48 h after X-ray-treatment the guinea-pig thymus undergoes an involution. This involution is paralleled by a concentration of X-ray-resistant lymphocytes in central parts and a disappearance of lymphocytes from the peripheral parts, which leads to the inverted appearance. Four days after the X-ray exposure the eosinophils become more prominent in Hassall's bodies. This observation of the involution of the thymus is confirmed by the results described here, with the exception that eosinophils become more prominent at the border of the thymus residue (Fig. 7). In conclusion, both corticosteroid- and X-ray-treatment lead to a decrease in the number of lymphocytes in the thymus, as a result of cell death (La Pushin and de Harven, 1971). The increased numbers of eosinophils in the thymus residue might be due to some eosinophilotactic substance, released in connection with the destruction of T-lymphocytes, which seems to influence migration of eosinophils (Turnbull and Kay, 1976; Basten and Beeson, 1970). Azathioprine, did not change the morphology of the thymus or the distribution pattern of the eosinophils, which were only slightly reduced in number. The action of this substance may be similar to that of methotrexate and cyclophosphamide, which both suppresse the eosinophil response in the animal on inoculation with parasites (Boyer et al., 1970). In the h u m a n thymuses studied, the distribution of eosinophils was variable (Figs. 1-3). This variability could be attributed to different illnesses and causes of death in children. In this case a stress situation leading to high cortisone blood levels before death might be assumed. In such circumstances the cortisone would lead to the eosinophil distribution pattern similar to that found shortly after cortisone administration in the animal (Blau, 1967). Thus, only in some cases could the distribution pattern of eosinophils in the thymus of children be correlated with pattern observed in the experimental animal studies.
Acknowledgments. I am grateful to PD Dr. G. Kistler, Institute of Anatomy, University Ztirich, for providing the electronmicrograph Figure 4, to Miss Walther, Dermatologic Clinic, Kantonspital Ziirich, for carrying out the X-ray radiation and to Miss Tunesi for technical assistance.
References Archer, G.T., Air, G., Jackas, M., Morell, D.B.: Studies on rat eosinophil peroxidase. Biochim. biophys. Acta (Amst.) 99, 96-101 (1965) Archer, G.T., Hirsch, J.G.: Isolation of granules from eosinophil leucocytes and study of their enzyme content. J. exp. Med. 118, 277~85 (1963) Arvy, L. : Enzymologyof the thymus. In: Handbuch der Histochemie,Vol. VIII, Part l, Supplements. Stuttgart: Gustav Fischer 1973 Bainton, D.F., Farquhar, M.G.: Segregation and packaging of granule enzymes in eosinophilic leukocytes. J. Ceil Biol. 45, 54-73 (1970) Basten, A., Beeson, P.B.: Mechanism of eosinophilia. II. Role of the lymphocyte. J. exp. Med. 131, 1288 1305 (1970) Blau, J.N. : The dynamic behaviour of Hassall's corpuscles and the transport of particulate matter in the thymus of the guinea-pig. Immunology 13, 281-292 (1967)
EosinophiIs in the Thymus
279
Boyer, M.H., Basten, A., Beeson, P.B. : Mechanism of eosinophilia. III. Suppression of eosinophilia by agents known to modify immune responses. Blood 36, 458-469 (1970) Essellier, A.F., Jeanneret, R.L., Morandi, L. : The mechanism of glucocorticoid eosinopenia. Contribution to the physiology of eosinophile granulocytes. Blood 9, 531-549 (1954) Gervin, S.W., Holtzman, E.: The fate of exogenous peroxidase in the thymus of newborn and young adult mice. J. Histochem. Cytochem. 20, 445-462 (1972) Ghossein, N.A., Bosworth, J.L., Stacey, R.T., Muggia, F.M, Krishnaswamy, V. : Radiation-related eosinophilia. Radiology 117, 413 417 (1975) Henning, W., M611mann, H., Kindler, J., Reisch, J., Alfes, H. : Einflug yon synthetischen Glucocorticoiden und von Metopiron auf ketosteroidhaltige Zellen des postnatalen Kaninchenthymus. Z. Zellforsch. 143, 37 44 (1973) Hsu, Ch.K., Hsu, S.H.: hnmunopathology of schistosomiasis in athymic mice. Nature (Lond.) 262, 397-399 (1976) La Pushin, R.W., Harven, E. de: A study of gluco-corticosteroid-induced pyknosis in the thymus and lymph node of the adrenalectomized rat. J. Cell Biol. 50, 583-597 (1971) Lee, R.E., Domm, L.V.: A histological and histochemical study on the effects of adrenal cortical steroids in the fetal and neonatal thymus. Anat. Rec. 157, 105-116 (1967) Litt, M.: Eosinophils and antigen-antibody reactions. Ann. N.Y. Acad. Sci. 116, 964-985 (1964) Miller, A.M., Colley, D.G., McGarry, M.P. : Spleen cells from schistosoma mansoni-infected mice produce diffusible stimulator of eosinophilopoiesis in vivo. Nature (Lond.) 262, 586 587 (1976) Okun, M.R., Edelstein, L.M., Or, N., Hamada, G., Donnellan, B., Lever, W.F.: Histochemical differentiation of peroxidase-mediated from tyrosinase-mediated melanin formation in mammalian tissues. Histochemie 23, 295-309 (1970) Pearse, A.G.E.: Histochemistry. Theoretical and applied. IIVd Edition, Vol. 2. Edinburgh and London: Churchill Livingstone 1972 Ponzion, N.M., Speirs, R.S.: Lymphoid cell dependence of eosinophil response to antigen. III. Comparison of the rate of appearance of two types of memory ceils in various lymphoid tissues at different times of priming. J. Imnmnol. 110, 1363-1370 (1973) Ryt6maa, T.: Organ distribution and histochemical properties of eosinophil granulocytes in rat. Acta path. microbiol, scand., Suppl. 140, 50, 1-118 (1960) Thorn, G.W., Forsham, P.H.: A test for adrenal cortical insufficiency J. Amer. med. Ass. 137, 1005-1009 (1948) Turnbull, L.W., Kay, A.B.: Eosinophils and mediators of anaphylaxis. Histamin and imidazole acetic acid as chemotactic agents for human eosinophiI leucocytes. Immunology 31,797-802 (1976) Van Duijn, P.: In: Pearse, A.G.E., Histochemistry. Theoretical and applied. IIIrd Edition p. 1334. Vol. 2. Edinburgh and London: Churchill Livingstone 1972
Received February 22, 1977
Histochemistry 9 by Springer-Verlag 1977
Localization of Eosinophils in the Thymus by the Peroxidase Reaction Edith Mfiller Institute of Anatomy, Universityof Ziirich (Director: Prof. Dr. G. T6ndury) GloriastraBe 19, CH-8006 Ziirich, Switzerland
Summary. Thymuses of human fetuses and infants and of young mice were investigated histochemically for peroxidase. Eosinophils were shown to be the only peroxidase-positive cells in the thymus. In human thymuses the eosinophilic cells were predominantly localized in medullar areas, with concentration of cell clusters at the cortico-medullar junction, around or inside Hassall's bodies and occasionally in high numbers in the intraseptal vessels of the cortex. In the normal mouse the eosinophils were evenly distributed throughout the medulla. Treatment with corticosteroids or X-rays produced a severe involution of the thymus with concommitant change in cellular pattern. The central areas of the thymus residue contained lymphocytes while the peripheral regions consisted of reticuloepithelia, macrophages and numerous eosinophils. Azathioprine did not change the morphology of the thymus. The numbers of eosinophils were slightly reduced, the distribution pattern remaining unchanged.
Introduction It is well known that the thymus contains numerous enzymes with activities that fluctuate as an expression of the functional state of the organ. Peroxidase(EC 1.11.1.7) activity, however, is not measurable (Arvy, 1973). Experiments with intraperitoneally injected horse-radish peroxidase showed to some extent the fate of the exogenously applied enzyme in the thymus (Gervin and Holtzman, 1972). The authors found that, within 30 min of injection, the peroxidase was localised in macrophages both in the blood vessels and adjacent connective tissue. In saline-injected controls, peroxidase activity was found in erythrocytes and occasionally in cells resembling eosinophil leucocytes.
274
E. Mtiller
However, it has been demonstrated that eosinophils contain large quantities of peroxidase (Ryt6maa, 1960; Archer, 1963; Bainton and Farquhar, 1970; Okun et al., 1970) which has different properties from myeloperoxidase (Archer, 1965). The aim of the present study was to analyse the normal distribution of peroxidase-active cells and the fluctuation of their numbers in the thymus after treatment with corticosteroids and X-rays, both of which influence the migration of eosinophils (Essellier etal., 1960; Lee and Domm, 1967; Henning etal., 1973; Blau, 1967; Ghossein et al., 1975). Materials and Methods The tissue specimens of h u m a n thymuses were obtained from stillborn fetuses (22, 25 and 27 weeks of gestation); from four children still-born at term; and from five children aged 1, 2~/2, 4, 5, and 12 m o n t h s who had died suddenly without previous manifestation of illness. For the experiments with immunesuppressive agents, thymuses from I C R mice, all male 9 weeks old, were used. They were divided into 3 groups of ten and each group was treated as follows: group a) sodium cortisoneacetate (1 rag) was injected i.p. daily for two days and the animals killed 24 h after the second treatment; group b) the entire bodies of the animals were irradiated for 16 min with 850 r (180 kv, 0.5 m m Cu-filter) and the mice were killed 24, 48 and 72 h after this single X-ray application; group c) azathioprine sodium (6 mg/kg, Imuran, Burroughs Wellcome & Co) was injected i.p. daily for 5 days and the animals killed three days later after the final injection. The control animals were killed at the same time intervals. Small pieces of thymuses were frozen at - 7 0 ~ with solid CO2 and cut at 8 g m in the cryostat. The sections were dried at 4 ~ C before incubation for enzy/ne demonstration or staining by the Giemsa-May-Grfinwald method. In light microscopy sections, peroxidase was demonstrated by the benzidine method of Van Duijn, 1955 (Pearse, 1972) with and without addition of L-DOPA. In electronmicroscopy studies the enzyme was visualized by the diaminobenzidine reaction of Grah a m and Karnovsky (Okun et al., 1970). Tyrosine was demonstrated by the method of Rappaport (Pearse, 1972). Control sections were incubated in media lacking HzO2, benzidine or L - D O P A respectively.
Results
The distribution of peroxidase-active cells in the human thymuses studied was variable. Usually they were diffusely spread in the medulla (Fig. 1) and were occasionally concentrated as cell clusters on the cortico-medullary border or around the Hassall bodies (Fig. 2). Intact and disrupted peroxidase-positive cells were also demonstrated within the Hassall bodies (Fig. 3). In the cortex, only single isolated peroxidase-active cells could be observed. Occasionally the intraseptal blood vessels of the cortex were surrounded by peroxidase-active cells, whereas none or few were found in the medulla. Fig. 1. T h y m u s from a 2-month-old child post mortem, incubated for peroxidase. Dark dots are cells with peroxidase-reaction product. Enzyme-positive cells spread mostly in the medullar parts, or as cell clusters on the cortico-medullar (cm) border, or along the septal vessels (sv). ca. x 28 Fig. 2. Medullar part from a thymus of a newborn child. Note the numerous peroxidase-active cells around one and inside of another Hassall's body (h). ca. x 93
Fig. 3. Detail of a medullar part from a t h y m u s of a newborn child. Two Hassall's bodies (h) loaded with peroxidase-active cells, x 375 Fig. 4. Eosinophil granulocyt in the mouse t h y m u s incubated for peroxidase. Heavy reaction-deposit in the granules (g). The cell nucleus (n) is visible in two parts, x 14,000
Fig. 5. Normal mouse t h y m u s incubated for peroxidase. Note the n u m e r o u s active cells spread almost regularly t h o u r o u g h the medulla. Only single cells with peroxidase-reaction product are in the cortex, x 56
Fig. 6. Corticosteroid-treated mouse thymus incubated for peroxidase. No cortico-medullary structure in the atrophied thymus. The peroxidase-active cells now mostly in the border and less in the middle of the organ, x 56
Eosinophils in the Thymus
277
After treatment of mice with corticosteroids and X-ray irradiation, the Tyrosinase (EC 1.10.3.1) had the same distribution as peroxidase (EC 1.11.1.7) in cells in the thymus. In m o u s e thymuses, the peroxidase-active cells were also confined to the medulla where they were regularly dispersed (Fig. 5). The EM-investigations of the thymuses incubated for peroxidase, showed enzyme-active granules characteristic of eosinophils (Fig. 4). After treatment of mice with corticosteroids and X-ray irradiation, the thymus involuted to a small residue, the central region of which contained densly packed lymphocytes and single peroxidase-active cells. The periphery showed a loose arrangement of reticular epithelia, numerous macrophages and a high number of peroxidase-positive cells (Figs. 6 and 7). In azathioprine-treated animals, the thymus retained normal structure but the peroxidase-active cells were less numerous than in controls. Most of the peroxidase-active cells were located as clusters on the cortico-medullary border or in the intraseptal vessels (Fig. 8).
Discussion
A stable histochemical method is a prerequisite for the study of peroxidaseactivity in the thymus. By adding L-DOPA (Okun et al., 1970), the quick-fading reaction product of the benzidine-peroxidase method (van Duijn, 1955) is stabilized. This allows an easy and reliable light microscopical study of the distribution of peroxidase-active cells in the tissue. At the EM-level, this method provides strong evidence that tile eosinophil contain peroxidase (Okun et al., 1970; Bainton and Farquhar, 1970). Recently, studies of c o r t i c o s t e r o i d influence on eosinophils (Thorn and Forsham, 1948; Essellier etal., 1954) have undergone a revival. Lee and Domm (1967) described, that within six hours after administration of 1.5 mg of hydrocortisone to rats, eosinophils become prominent either at the cortico-medullary border, or as perivascular clusters. Henning et al. (1973) found an extreme involution of rabbit thymus and a 2--3 fold increase of eosinophils at its borders 12 h after hydrocortisone treatment (10 mg/kg). In the present study the thymus of hydrocortisone-treated mice, displayed within 24 h a similar extreme involution and distribution pattern of eosinophils, to that observed by Henning et al. (1973). Eosinophils disappear from the blood stream after corticosteroid treatment and accumulate predominantly in the lymphoid organs (Essellier et al., 1954; Ryt6maa, 1960). This behaviour may be attributed to some eosinophilotactic substance mediated by lymphocytes, as reported by Litt (1964) and by Basten and Beeson (1970). Studies on the thymus-dysgenetic nude mouse Fig. 7. X-ray treated mouse thymus. Note here the same structure and peroxidase-active cell-pattern as in the corticosteroid-treated mouse Figure 6. x 56 Fig. 8, lmuran-treated mouse thymus incubated for peroxidase. Arrows indicate enzyme-positive ceil clusters (cO on the cortico-medullar border. In the medullar p a r t - t e s s peroxidase-active cells are present compared with the control mouse t h y m u s Figure 5. • 56
278
E. Mfiller
(Miller et al., 1976; Hsu and Hsu, 1976) and on thymic-depleted mice (Ponzio and Speirs, 1973) have indicated that the presence of T-lymphocytes is necessary for eosinophil migration into the affected tissue sites. As Blau (1967) has described, 48 h after X-ray-treatment the guinea-pig thymus undergoes an involution. This involution is paralleled by a concentration of X-ray-resistant lymphocytes in central parts and a disappearance of lymphocytes from the peripheral parts, which leads to the inverted appearance. Four days after the X-ray exposure the eosinophils become more prominent in Hassall's bodies. This observation of the involution of the thymus is confirmed by the results described here, with the exception that eosinophils become more prominent at the border of the thymus residue (Fig. 7). In conclusion, both corticosteroid- and X-ray-treatment lead to a decrease in the number of lymphocytes in the thymus, as a result of cell death (La Pushin and de Harven, 1971). The increased numbers of eosinophils in the thymus residue might be due to some eosinophilotactic substance, released in connection with the destruction of T-lymphocytes, which seems to influence migration of eosinophils (Turnbull and Kay, 1976; Basten and Beeson, 1970). Azathioprine, did not change the morphology of the thymus or the distribution pattern of the eosinophils, which were only slightly reduced in number. The action of this substance may be similar to that of methotrexate and cyclophosphamide, which both suppresse the eosinophil response in the animal on inoculation with parasites (Boyer et al., 1970). In the h u m a n thymuses studied, the distribution of eosinophils was variable (Figs. 1-3). This variability could be attributed to different illnesses and causes of death in children. In this case a stress situation leading to high cortisone blood levels before death might be assumed. In such circumstances the cortisone would lead to the eosinophil distribution pattern similar to that found shortly after cortisone administration in the animal (Blau, 1967). Thus, only in some cases could the distribution pattern of eosinophils in the thymus of children be correlated with pattern observed in the experimental animal studies.
Acknowledgments. I am grateful to PD Dr. G. Kistler, Institute of Anatomy, University Ztirich, for providing the electronmicrograph Figure 4, to Miss Walther, Dermatologic Clinic, Kantonspital Ziirich, for carrying out the X-ray radiation and to Miss Tunesi for technical assistance.
References Archer, G.T., Air, G., Jackas, M., Morell, D.B.: Studies on rat eosinophil peroxidase. Biochim. biophys. Acta (Amst.) 99, 96-101 (1965) Archer, G.T., Hirsch, J.G.: Isolation of granules from eosinophil leucocytes and study of their enzyme content. J. exp. Med. 118, 277~85 (1963) Arvy, L. : Enzymologyof the thymus. In: Handbuch der Histochemie,Vol. VIII, Part l, Supplements. Stuttgart: Gustav Fischer 1973 Bainton, D.F., Farquhar, M.G.: Segregation and packaging of granule enzymes in eosinophilic leukocytes. J. Ceil Biol. 45, 54-73 (1970) Basten, A., Beeson, P.B.: Mechanism of eosinophilia. II. Role of the lymphocyte. J. exp. Med. 131, 1288 1305 (1970) Blau, J.N. : The dynamic behaviour of Hassall's corpuscles and the transport of particulate matter in the thymus of the guinea-pig. Immunology 13, 281-292 (1967)
EosinophiIs in the Thymus
279
Boyer, M.H., Basten, A., Beeson, P.B. : Mechanism of eosinophilia. III. Suppression of eosinophilia by agents known to modify immune responses. Blood 36, 458-469 (1970) Essellier, A.F., Jeanneret, R.L., Morandi, L. : The mechanism of glucocorticoid eosinopenia. Contribution to the physiology of eosinophile granulocytes. Blood 9, 531-549 (1954) Gervin, S.W., Holtzman, E.: The fate of exogenous peroxidase in the thymus of newborn and young adult mice. J. Histochem. Cytochem. 20, 445-462 (1972) Ghossein, N.A., Bosworth, J.L., Stacey, R.T., Muggia, F.M, Krishnaswamy, V. : Radiation-related eosinophilia. Radiology 117, 413 417 (1975) Henning, W., M611mann, H., Kindler, J., Reisch, J., Alfes, H. : Einflug yon synthetischen Glucocorticoiden und von Metopiron auf ketosteroidhaltige Zellen des postnatalen Kaninchenthymus. Z. Zellforsch. 143, 37 44 (1973) Hsu, Ch.K., Hsu, S.H.: hnmunopathology of schistosomiasis in athymic mice. Nature (Lond.) 262, 397-399 (1976) La Pushin, R.W., Harven, E. de: A study of gluco-corticosteroid-induced pyknosis in the thymus and lymph node of the adrenalectomized rat. J. Cell Biol. 50, 583-597 (1971) Lee, R.E., Domm, L.V.: A histological and histochemical study on the effects of adrenal cortical steroids in the fetal and neonatal thymus. Anat. Rec. 157, 105-116 (1967) Litt, M.: Eosinophils and antigen-antibody reactions. Ann. N.Y. Acad. Sci. 116, 964-985 (1964) Miller, A.M., Colley, D.G., McGarry, M.P. : Spleen cells from schistosoma mansoni-infected mice produce diffusible stimulator of eosinophilopoiesis in vivo. Nature (Lond.) 262, 586 587 (1976) Okun, M.R., Edelstein, L.M., Or, N., Hamada, G., Donnellan, B., Lever, W.F.: Histochemical differentiation of peroxidase-mediated from tyrosinase-mediated melanin formation in mammalian tissues. Histochemie 23, 295-309 (1970) Pearse, A.G.E.: Histochemistry. Theoretical and applied. IIVd Edition, Vol. 2. Edinburgh and London: Churchill Livingstone 1972 Ponzion, N.M., Speirs, R.S.: Lymphoid cell dependence of eosinophil response to antigen. III. Comparison of the rate of appearance of two types of memory ceils in various lymphoid tissues at different times of priming. J. Imnmnol. 110, 1363-1370 (1973) Ryt6maa, T.: Organ distribution and histochemical properties of eosinophil granulocytes in rat. Acta path. microbiol, scand., Suppl. 140, 50, 1-118 (1960) Thorn, G.W., Forsham, P.H.: A test for adrenal cortical insufficiency J. Amer. med. Ass. 137, 1005-1009 (1948) Turnbull, L.W., Kay, A.B.: Eosinophils and mediators of anaphylaxis. Histamin and imidazole acetic acid as chemotactic agents for human eosinophiI leucocytes. Immunology 31,797-802 (1976) Van Duijn, P.: In: Pearse, A.G.E., Histochemistry. Theoretical and applied. IIIrd Edition p. 1334. Vol. 2. Edinburgh and London: Churchill Livingstone 1972
Received February 22, 1977
