THE ANATOMICAL RECORD 196:119-127 (1980) Genesis, Ultrastructure and Cytochemical Study of the Cat Eosinophil B. PHESENTEY, 2. JERUSHALMY, M. BEN-BASSAT AND K. PERK Rogof- Wellcome Medical Research fnstitu.te, Tei-Auiu Uniuersity Medical School, J . Casper Department of Patholog,y, Reilinson Hospital, Petah Tikun, Hematological Department, District Lahoratory, Kupath Holim, Rehouot, and Department of Animal Anatomy and Ph,ysiulog,y, The Hehrew Uniuersity, Rehouot, Campus, Israel ABSTRACT Eosinophils from cat bone marrow and peripheral blood were studied by electron microscopy and cytochemical procedures. The maturation of eosinophils and formation of typical granules were described. Contrary to the accepted opinion that the core of animal's eosinophilic specific granules have a crystal-like structure, our observations revealed that the core has a myelin-like cylindrical appearance, whose layered formation proceeds from the inside outwards. Electron microscopic observations revealed that localization of reaction product to potassium pyroantimonate and phosphotungstic acid and to acid phosphatase activity was similar to that of eosinophils of man and other animals. Antimonate deposits and acid phosphatase activity were detected between the layers of the myelin-like structure of the core. Eosinophil granules failed to yield a positive reaction for peroxidase activity. The secretory activity of the eosinophil is discussed. MATERIALS AND METHODS Use of histochemical techniques together with electron microscopy permits determination of the location of enzymes and other constituents within eosinophilic granules (Ackerman and Clark, '71; Ackerman, '72; Bainton and Farquhar, '70; Geger et al., '70; Hardin and Spicer, '71; Hardin and Spicer, '70; Kelenyi and Zombai, '64). Tnese granules have been designated a s lysosomes (Baintonand Farquhar, '70; Faller, '66; Fawcett, '69; Zucker-Franklin, '68; Geyer et al., '70; Kelenyi, '68; Leder and Paper, '71; Presentey, '69; Scott and Horn, '70),on the basis of their high enzyme content, particularly of acid hydrolases (Archer and Hirsch, '63; Bainton and Farquhar, '70; Geyer et al., '70; Presentey, '69). In eosinophils of different animals, the enzyme content of the granules varies (Jain, '68; Nakamura, '55; Presentey, '781, as does also the relative dimensions of the inner and outer regions of the granule. Developing and mature cat neutrophils were studied by Ackerman ('68). The present study traces the course of development in the cat eosinophil, starting from the undifferentiated blast stage, the ultrastructural features, and cytochemical reactions. 000-3276x180/1962-0119$02.00 0 1980 ALAN R. LISS, INC Five mature cats with eosinophilia in excess of 15% were selected for this study. Blood samples were withdrawn into heparinized test tubes and kept a t room temperature for 30 min; the buffy coat was collected and processed according to Watanabe ('59). Aspirated bone marrow was processed according to Ackerman and Clark ('71). To test the reaction to phosphotungstic acid IPTA) during dehydration, the procedure of Hudson ('66) was used. Peroxidase activity was detected using the technique of Ackerman and Clark ('711, and acid phosphatase using that of Seeman and Palade ('67). Reaction to pyroantimonate was detected in nonglutaraldehyde-treated buffy coat or bone-marrow according to the method of Hardin and Spicer ('71). Samples were fixed for 1hr in 2.57~buffered glutaraldehyde, postfixed for 1 hr in osmium, dehydrated in graded ethanol, and embedded in epon. Sections examined for enzymic activity Received September 26, 1978; accepted July 3, 1979. This work 1s part of Ph. D. thesis submitted by B. Presentey to the Tel-Aviv University. Reprint requests to B. Presentey, Head, District Laboratories, Kupat Hohm, P 0. Box 1300, Rehovot, Israel. 119 120 B. PRESENTEY, Z. JERUSHALMY, M. BEN-BASSAT AND K. PERK and thin sections treated with PTA or pyroantimonate were unstained. Ultrathin sections were double stained with uranyl acetate and lead citrate and examined with a SEM-7 electron microscope. RESULTS Electron microscopic observations The blast cell stage was characterized by a large nucleus occupying the major part of the cell. The nucleolus was well defined; small amounts of heterocromatin could be observed near the nuclear membrane. The cytoplasm was rich in endoplasmic reticulum and occasionally displayed dilatations. Mitochondria and Golgi apparatus were also evident. The transitional cells, the early promyelocytes (Fig. l), were characterized by their large size and by a large nucleus with more heterochromatin than in the blast cell, possessing a large nucleolus and several granules composed of a homogeneous, electron-dense substance. In some of the granules, a differentiation in two parts was evident. Endoplasmic reticulum were numerous and their cisternae were more distended than in the blast cell; mitochondria were also r a t h e r numerous (Fig. 2) . The number of granules increased during maturation. Granules were composed of two parts: a darker one-core-and a lighter one-matrix (Fig. 2). With cell maturation and transition to myelocyte, the nucleus diminished; the nucleolus became for the most part undetectable and the amount of heterochromatin increased. There was less endoplasmic reticulum and fewer mitochondria in mature cells as compared to more immature stages. However, the number of granules increased as did the dimensions of their core (Fig. 3 ) . The mature cat eosinophil was characterized by a segmented nucleus, cytoplasm which abounded in granules and contained very few mitochondria, and by endoplasmic reticulum. In the specific granules, matrix was very well detected; the core had a layered myelin-like structure. Occasionally, however, the granules contained only a few myelin-like fragments (Fig. 4). Some granules were lacking normal core structure. The latter became lighter from the matrix, giving the impression that its content had been released (Fig. 5 ) . In addition, the mature eosinophil contained microgranules, round or oval i n shape, and sometimes double-walled (Fig. 5 ) , similar to those described in man (Parmley and Spicer, '74; Nichols and Bainton, '73). Reaction of eosinophils to phosphotungstic acid IPTA) While a positive reaction to FTA was usually observed in the matrix of the granules, in a few cases this reaction did not take place, though the cells did not appear to have suffered any damage (Fig. 6). Reaction of eosinophils to potassium pyro-antimonate IPPA) A strong positive reaction to PPA occurred mainly in the matrix of the granules, but positive reactions were also observed between the layers of the inner, myelin-like part (Fig. 7). Granules without any evidence of destructive changes occasionally failed to react positively. As in human eosinophils, there was a positive response in the region of heterochromatin material in some of the nuclei. Reaction of eosinophils to peroxidase Cat eosinophils did not react positively for peroxidase; this is in agreement with findings using cytochemical procedures and light microscopy (Fig. 8). Localization of acid phosphatase activity in eosinophils A strong positive reaction for acid phosphatase was observed in the matrix, in the layers of the inner, myelin-like region, and in the center of the eosinophil granules (Fig. 9). In some cells, a positive reaction was observed in the nucleus and the intergranular cytoplasm, while the granules themselves displayed positive or negative reaction (Fig. 10).In other cells, positive reactions occurred only in the intergranular cytoplasm and in the nucleus, in which heterochromatin was not detected. No destructive changes could be observed in the nonreacting granules (Fig. 11). DISCUSSION Even in cases of pronounced eosinophilia, the number of eosinophils in bone marrow remains small. Consequently, investigators of cat eosinophils have so far limited themselves to a thorough study only of the mature eosinophil (Bargmann and Knopp, '58; Fawcett '69; Ward et a1 '72). We studied the ultrastructure of the bone-marrow eosinophils, including all stages of blast maturation. The granules first appear in the promyelocyte, and core and matrix are formed. Simultaneously, the process of granule GENESIS, ULTRASTRUCTURE OF CAT EOSINOPHIL Fig. 1. An early promyelocyte-probably a n eosinophilic one. Large nucleus, delicately wrought, with heterochromatin a t the periphery and a prominent nucleolus. The cytoplasm contains many profiles of rough endoplasmic reticulum (some dilated), mitochondria, and granules of varying electron density. Uranyl acetate and lead nitrate x 3,720. Fig. 2. An immature eosinophil-a late promelocyte with large nucleus and nucleolus. Abundant endoplasmic reticulum with expansions. Increase of specific granules. Core and matrix are identified in all of them. Uranyl acetate and lead nitrate x 4,185. 121 122 B. PRESENTEY, 2. JERUSHALMY, M. BEN-BASSAT AND K. PERK Fig. 3. Three eosinophils at different stages of maturation. One l:ite promyelocyte, one myelocyte, and one methamyelocyte. In the more mature cells the nucleolus has disappeared, the nucleus became smaller, and the hekrochromatin more abundant. Cytoplasm is filled up with typical granules. Uranyl acetate andlead nitrate x 2,325. Fig. 4. Portion of mature eosinophil. Layered structure of myelin-like core of specific granules is discernihle Uranyl acetate and lead nitrate x 37,200. GENESIS, ULTKASTRUCTURE OF CAT EOSINOPHIL 123 Fig. 5. Portion of mature eosinophil. In three typical granules the core is lighter than the matrix (thick arrows). The intergranular cytoplasm contains round or elongated microgranules, some of which are doublewalled (thin arrows). Uranyl acetate and lead nitrate x 18,600. Fig. 6. Mature eosinophil following treatment with phosphotungstic acid during dehydration. Positive reaction isdiscernihle in the matrix of specific granules without additional staining. x 6,510. 124 B. PRESENTEY, 2. JERUSHALMY, M. BEN-BASSAT AND K. PERK Fig. 7. Detail of mature eosinophil following treatment with potassium pyroantimonate. Positive reaction discernible in outer part of typical granules and between the layers of the myelin-like inner part, without additional staining. x 15,810. Fig. 8. Bone marrow following incubation to detect peroxidase. Positive reaction in neutrophil, but not in eosinophi. Uranyl acetate and lead nitrate X 3,720. Fig. 9. Positive acid phosphatase reaction in matrix of specific granules. Positive reaction discernible between the layers of the myelin-like structure of the inner part. Without additional staining. x 15,000. Inset x 50,000. Fig. 10. Mature eosinophil following treatment with acid phosphatase. Positive reaction in some of the granules, but many of the granules remain negative, although they show no visible degernerative changes. Positive reaction detectable in the intergranular cytoplasm and in the heterochromatin of the nucleus. It is suggested that the enzyme evident a t these sites originates from the granules, and that such granules, which have released their enzyme contents, are consequently rendered negative to acid phosphatase. x 7,000. Fig. 11. Mature eosinophil with a trilobate nucleus (N).Strong positive reaction to acid phosphatase in the intergranular cytoplasm and in the nucleus, suggesting acid phosphatase release. Unlike Figure 10, no heterochromatin discernible in nucleus. All the granules appear intact (arrows). x 6,000. formation encompasses changes in both regions. Mature specific granules reveal a myelin-like structure of the core. The data presented indicate that it is not feasible to attempt to distinguish different categories of typical granules; contrary to the cat, in human eosinophilic granules, the matrix forms later than the core (Bessis and Thiery, '57; Capone et al., '64; Fedorko, '68; Miller et al., '66; Scott and Horn, '70). Treatment with phosphotungstic acid a t times concealed the layer structure of the inner part of the granules, resulting in a myelin-like appearance. Quintarelly et al. ('71) confirm that staining with phosphotungstic acid is not specific for phospholipids (Presentey and Jerushalmy, '79; Presentey, '78). The response of the granules to potassium pyroantimonate is similar to that already reported by other investigators of human eosinophils and of various laboratory animals (Ackerman, '72; Hardin and Spicer, '71; Hardin, '70). The positive reaction observed between the layers of the core a t its center suggests that formation of this myelin-like structure proceeds gradually outwards. In addition to N a + cations, potassium pyroantimonate precipitates also Ca+ and Mg cations, but this can be partly suppressed under certain conditions + ++ (Bulger, '69; Clark and Ackerman, '71; Klein et al., '72). Responses to acid phosphatase in the cat eosinophil follow a similar pattern to those found in other animals and man (Geyer et al., '70; Miller andHerzog, '69; Seeman and Palade, '67; Wetzel et al., '67), but are stronger and in agreement with findings by histochemical techniques (Presentey, '78). In several instances a positive response was recorded in the cytoplasm and nucleus, with some of t h e granules responding positively and others negatively (Fig. 10). We also observed cells where all the granules failed to react, although the intergranular cytoplasm and the nucleus reacted positively (Fig. 11).These findings reveal t h a t t h e enzyme released from t h e granules accumulates within the cytoplasm and the nucleus, in the initial stage in the heterochromatin region and later in the entire nucleus causing damage to its structure. This observation confirms that intergranular response to acid phosphatase is not an artifact, as it has been previously interpreted (Seeman and Palade, '67). Secretory activity by eosinophils had been postulated and was confirmed by Scott and Horn ('701, who showed that release of entire granules can be associated with release of the GENESIS, ULTRASTRUCTURE OF CAT EOSINOPHIL granule content-either of the core or of the matrix. We found that negative responses by granules was due to secretory processes (Figs. 1 0 , l l )and not t o degenerative changes, such as described by Kelenyi ('68) and Kelenyi and Nerneth ('72). Basic proteins in the cat eosinophil have been found to be the primary constituent of the core of the granules and are not present in the outer part (Lewis et al., '78), as claimed by Faller ('66).Contrary to Faller ('66),we found no evidence of the presence of various unsaturated acids in the granules. LITERATURE CITED Ackerman, G.A. (1968) Ultrastructure and cytochemistry of the developing neutrophil. Lab. Invest., 19:290-302. Ackerman, G.A., and M.A. Clark (1971) Ultrastructural localization of peroxidase activity in normal human hone marrow cells. Z. Zellforsch., I 1 7,465-475. Ackerman, G.A. (1972) A cytmhemical evaluation ofpyroantimonate binding to the plasmalemma of blood and hone marrow cells and its relationship to cellular maturation. J. Histochem. Cytochem., 20:88&895. Archer, G.I., and J.G. Hirsch (1963) Isolation of granules from eosinophil leukocytes and study of their enzyme content. J. Exp. Med., 118:277-286. Bainton, D.F., and M.G. Farquhar (1970) Segregation of packing of granule enzymes in eosinophilic leukocytes. J. Cell Biol., 45154-73. Bargmann, E., and A. Knopp (1958)Uber das granulum des Eosinophilen. Z. Zellforsch., 48: 13@136. Hessis, M., and J.P. Thiery (1957) Electron microscopy of human white blood cells and their stem cells. Int. Kev. Cytol., 12:19%241. Bulger, R.E. (1969) Use of potassium pyroantimonate in the localization of sodium ions in rat kidney tissue. J. Cell Biol., 40:7%94. Capone, R.J., E.L. Weinreh, and G.B. Chapman (1964) Electron microscope studies on normal human myeloid elements. Blood, 23:30&320. Clark, M.A., and G.A. Ackerman (1971) A histochemical evaluation of the pyroantimonate-osmium reaction. J. Histochern. Cytochem., 29:727-737. Faller, A. (1966) Zur Frage von Struktur and Aufban der Eosinophilen Granula. Z. Zellforsch., 69,551-565. Fawcett, D. (1969) The Cell: Its Organelles and Inclusions. An Atlas of Fine Structure. W.B. Saunders, Philadelphia and London. Fedorko, M.E. (1968)Formation of cytoplasmic granules in human eosinophilic myelocytes: An electron microscope autoradiographic study. Blood, 31 :18%194. Geyer, G., P. Schaaf, W. Linss, H.J. Stibenz, K.J. Kalbhuber, R. Quade, and A. Christner (1970) Enzyme distribution patterns in the granules of eosinophil leukocytes of the mouse. Acta Histochem., .?8:18%191. Hardin, J.H., and S.S.Spicer (1971)Ultrastructural localization of dialyzed iron-reactive mucosubstance mucosubstance in rabbit heterophils, hasophils, and eosinophils. J. Cell Biol., 48:36%386. Hardin, J.H., and S.S. Spicer (1970)An ultrastructural study of human eosinophil granules: Maturational stages and pyroantimonate reactive cation. Amer. J . Anat ., 128:283- 297. Hudson, G. (1966) Eosinophil granules and phosphotungstic acid: an electron microscope study of guinea-pig hone marrow. Exp. Cell Res.,41:265-273. Jain, N.C. (1968) Peroxidase activity in leukocytes of some animal species. Folia Heamatologica (Leipzig), 88:297304. 127 Kelenyi, G., and E. Zomhai (1964) Electron microscopy of the henzidine peroxidase reaction in the p-anulesofeosinophil leukocytes. Europ. Regional Confer. on Electr. Micr., Publ. House of the Czechosl. Acad. Sci., pp. 207-208. Kelenyi, G. (1968) Degranulation of tissue eosinophil leukocytes. Acta Morphol. Acad. Sci. (Hung.),16:265-269. Kelenyi, G., and A. Nerneth (1972) Tissue eosinophil leucocytes 11. Electron microscopy and histochemistry of the uterine eosinophil leucocytes in eostrogen-treated spayed rats. Acta Biol. Acad Sci. (Hung.), 23:253-267. Klein, R.Z., Shyne-Shong Yen, and A.T. Klein (1972) Critique on the K-pyroantimonate method for semiquantitative estimation of cations in conjunction with electron microscopy. J. Histochem. Cytochem., 20:65-78. Leder, L.D., and B. Paper (1971) Cytological and cytochemical investigations on the origin of human eosinophilic granulocytes. Beitr. Path., 143:241-248. Lewis, D.M., J.C. Lewis, D.A. Loegering, and G.J. Gleich (1978) Localization of the Guinea pig eosinophil major basic protein to the core of the granule. J. Cell Biol., 77:702-713. Miller, F., E. de Harven, and G.E. Palade 11966) The structure of eosinophil leukocyte granules in rodents and in man. Z. Cell Biol., 31:34%362. Miller, F.E., and V. Herzog (1969) Die Lokalisation von Peroxidase und saurer Phosphatase in Eosinophilen Leukocyten wehrend der Reifung. Z. Zellforsch., 97:84110. Nakamura, M. (1955) Variations in the peroxidase reaction intensities of blood cells in various vertebrates. Okajimas Folio Anat. Jpn., 27:l-16. Nichols, B.A., and Bainton, D.F. (1973) Differentiation of human monocytes in bone marrow and blood: Sequential formation of two granule populations. Lab. Invest., 29: 27-40, Parmley, R.T., and S.S. Spicer (1974)Cytochemical and ultrastructural identification of small type granule in human late eosinophils. Lab. Invest., 30:557-567. Presentey, B. (1968)A new anomaly of eosinophilic granulocytes. Amer. J. Clin. Pathol., 49:887-890. Presentey, B. ( 1969) Cytochernical characterization of eosinophils with respect to a newly discovered anomaly. Amer. J. Chn. Pathol., 51:451-457. Presentey, B., and Z. Jerushalmy (1979) On the specificity of various methods used to identify phospholipids in tissues. Submitted Presentey, B. (1978) Study of eosinophils of humans and of animals. Ph. D. Thesis, Sackler School of Medicine, TelAviv University, Tel-Aviv. Presentey, B., 2. Jerushalmy, and U. Minz (1978) Eosinophilic leukemia: Morphology. Cytochemical and electron microscopic studies. J. Clin. Pathol., (In press.) Quintarelly, G., R. Zito, and J.A. Cifonelli 11971) On phosphotungstic acid staining. IV. Selectivity of staining reaction. J . Histochem. and Cytochem., 21:155160. Scott, R.E., and R.G. Horn (1970)Fine structural features of eosinophile granulocyte development in human bone marrow. J . Ultrast. Res., 33:1&28. Seeman, Ph. M., and G.E. Palade (1967) Acid phosphatase localization in rabbit eosinophils. J. Cell Biol., 34;74L 756. Ward, J.M., J.F. Wright, and G.H. Wharran (1972) Ultrastructure of granulocytes in the peripheral blood of the cat. J. Ultrast. Res., 39.38LL-396. Watanabe, I. (1959) Watanabe, I., S. Donahue, and N. Hoggatt (1967)Method for electron microscopic studies of circulating human leukocytes and observations on their fine structure. J. Ultrast. Res., 20:366382. Wetzel, B.K., S.S. Spicer, and R.G. Horn (1967) Fine structural localization of acid and alkaline phosphatases in cells of rabbit blood and hone marrow. J. Histochem. Cytochem., 15:311-334. Zucker-Franklin, D. 11968) Electron microscopic studies of human granulocytes: Structural variation related to function. Semin. Hematol., 5:10%133.