Genesis, ultrastructure and cytochemical study of the cat eosinophil

Morphologic, cytochemical, and ultrastructural features of gray eosinophils in nine cats

BACKGROUND

Gray eosinophils, resembling those in sighthound dog breeds, have not previously been reported in cats.

OBJECTIVES

We aimed to provide a morphologic, cytochemical, and ultrastructural description of gray eosinophils in cats.

METHODS

Blood films examined as part of routine hematology profiles in cats from May 2015 to July 2018 were evaluated for the presence of gray eosinophils. When identified with modified Wright stain, cells were morphologically assessed and additionally stained with Diff-Quik, ALP, Luna, and Luxol fast blue stains and compared with feline controls. Two cases were prepared for transmission electron microscopy (TEM) and compared with a feline control.

RESULTS

Gray eosinophils were identified in 9 of 2641 cats during the study period. Compared with typical feline eosinophils, these cells contained abundant round granules instead of the characteristic rod-shaped specific granules. These granules lacked the characteristic intense pink/red staining with Romanowsky stains and did not stain with ALP, Luna, or Luxol fast blue stains. On TEM, the classical electron-dense core of these granules was replaced by a core with fragmented or amorphous internal material. Typical eosinophils were not identified in any cat in which gray eosinophils were identified.

CONCLUSIONS

The distinct morphologic, cytochemical, and ultrastructural changes in gray feline eosinophils might be associated with a reduction or lack of major basic protein (MBP) in specific granule cores. Similar to canine gray eosinophils, accurate recognition of these cells is essential to prevent their misclassification as toxic neutrophils.

Characterization of biological activities of feline eosinophil granule proteins

OBJECTIVE

To characterize eosinophil granule-derived proteins in cats.

SAMPLE POPULATION

Eosinophils collected via peritoneal lavage from 2 cats.

PROCEDURE

The cats were infested orally with Toxocara canis eggs and subsequently challenge-exposed with T. canis antigen injected IP to induce peritoneal eosinophilia; eosinophils were collected via peritoneal lavage. Eosinophil granule proteins were acid-extracted, separated by gel-filtration chromatography, and examined for their peroxidase, ribonuclease, and bactericidal activities; the N-terminal sequence of some of these proteins was determined and compared with homologue proteins from other species.

RESULTS

3 protein peaks were separated in the chromatogram. The first peak had both peroxidase and bactericidal activities. The second peak had ribonuclease and bactericidal activities, and the N-terminal sequence of the major protein was homologous with that of proteins of the ribonuclease A superfamily, including eosinophil ribonucleases from humans and other animal species. The third protein peak had bactericidal activity, and the N-terminal sequence of the major protein was homologous with that of human and murine major basic proteins.

CONCLUSIONS AND CLINICAL RELEVANCE

Results indicated that feline eosinophil granules contain major basic protein and eosinophil-associated ribonuclease and the granule proteins have peroxidase, ribonuclease, and bactericidal activities. In cats, characterization of eosinophil granule proteins may be useful in elucidation of the mechanism of tissue damage in eosinophil-associated diseases and development of new treatment options for those diseases. In addition, the identification of conserved structure and function of eosinophil granule proteins in cats is relevant from an evolutionary viewpoint.

Piecemeal degranulation (PMD) morphology in feline circulating eosinophils

Buffy coat preparation from six cats with 600-4560 circulating eosinophils/microL was collected by either blood centrifugation or sedimentation, fixed in 2.5% glutaraldehyde, post-fixed in either 1% osmium or in 1.5% potassium ferrocyanide-reduced osmium, ultra-sectioned and examined by transmission electron microscopy. Ultrastructural changes of piecemeal degranulation (PMD), which is a mechanism of eosinophil granule contents release indicative of eosinophil activation, were observed in specific granules from all the samples examined. The spectrum of PMD included coarsening of the granule matrix, budding vesicles, fragmented granule cores and lucent granules. The number of presumably activated eosinophils with ultrastructural evidence of PMD did not correlate with the level of eosinophilia. The lack of correlation suggested that, analogously with humans, blood eosinophil count might not represent the best criterion to evaluate the contribution of eosinophils to tissue damage in certain feline eosinophil-associated diseases.

Ultrastructure of developing and mature caprine leukocytes

Transmission electron microscopy demonstrated that, based on the structural uniqueness of the granules, caprine granulocytes are easily distinguishable from each other from the promyelocyte stage onwards. The neutrophils had the smallest granules which varied in size and were, in mature cells, either spherical to dumb-bell in shape; in mature cells the granule contents were compact and finely granular. The primary granules were smaller than the secondary granules. The eosinophil granules were large and typically had internal crystalloid structures; a second group of spherical granules with moderately coarse non-crystalloid sub-structure was present in smaller numbers in promyelocytes and myelocytes only. The basophil granules were also large, lacks crystalloids but showed variation in coarseness of granule substance, ranging from finely granular to markedly coarse. Mature granulocytes lacks Golgi apparatus and rough endoplasmic reticulum and ribosomes which were present in promyelocytes, myelocytes, metamyelocytes and bands. The monocytes had moderate numbers of spherical granules, rough endoplasmic reticulum, mitochondria and ribosomes, as well as prominent Golgi apparatus, and the cytoplasm had many small vacuoles.

Granulated round cell tumor of cats

Morphologic and biologic features of five feline granulated round cell tumours were compared with those previously reported to be of globule leukocyte and large granular lymphocyte origin. The five cats ranged from 6 to 9 years of age and presented with nonspecific gastrointestinal signs. Four of the five cats were tested for feline leukemia virus and were negative by enzyme-linked immunosorbent assay. The neoplastic process involved the abdominal cavity in all cases, with a predilection for the distal small intestine and mesentery. The liver and peripheral and thoracic lymphoid tissues were also sporadically affected. Neoplastic round cells contained 0.5-1.5-microns eosinophilic cytoplasmic granules that were difficult to discern on causal observation with hematoxylin and eosin stain but were deep blue and easily visualized when stained with phosphotungstic acid-hematoxylin. In two cases, epithelium in the affected ileum and liver contained unusually large numbers of apparently normal globule leukocytes. Ultrastructurally, the tumor granules tended to cluster at one nuclear pole and were spindle to round in shape with variably dense contents. Some granules contained a dense "cap" at one end or internal crystalloid bars that distorted the granule membrane. The tumors reported herein are similar to all three of the previously reported feline granulated round cell tumors and probably have a common cellular origin.

Granulocyte turnover in the feline intestine

The objective of this study was to determine the turnover rate of the extravascular pool of granulocytes in different regions of the feline gastrointestinal tract. Leukocyte emigration from the vasculature was prevented over a 48-h period by repeated intravenous injections of a monoclonal antibody (MAb IB4) directed against the leukocyte adhesion glycoprotein complex CD11/CD18. Tissue-associated myeloperoxidase (MPO) activity was used to monitor the total tissue granulocyte pool at 0.5, 12, 24, and 48 h after MAb IB4 administration. The mucosal layer of the duodenum, jejunum, ileum, and colon exhibited different kinetics of granulocyte clearance, with average life-spans (t1/2) ranging between 6.9 (colon) and 10.4 h (duodenum). Granulocyte clearance rates of 0.5 x 10(6) and 2.4 x 10(6) cells/h/g tissue were estimated (from measured values of t1/2 and tissue granulocyte pool) for the small bowel and colonic mucosae, respectively. The submucosal layer of the intestine exhibited a biphasic reduction in tissue MPO activity following immunoneutralization of CD11/CD18, with an initial t1/2 < or = 0.5 h followed by a t1/2 of 36-60 h. The initial rapid decline in tissue MPO suggests that a significant fraction of granulocytes in the submucosa is localized in a readily exchangeable pool (e.g., marginated cells within the vasculature). The results of this study indicate that the average life-span of resident granulocytes varies significantly between different regions of the gastrointestinal tract, with the intestinal mucosa exhibiting a t1/2 comparable to that previously reported for circulating feline neutrophils (approximately 8 h).

Eosinophils: a review

The eosinophil was discovered by Jones in 1846 (Dessein and David, 1982) but its proclivity to stain with aniline dyes was first described by Paul Ehrlich in 1879 (Hirsch and Hirsch, 1980). Recognized and named for this quality, eosinophils possess an abundance of highly basic proteins within their granules which confer their affinity for acidic dyes (Gleich and Loegering, 1984). Eosinophils are traditionally viewed as killer-effector cells in parasitic infestations and as modulators of Type I hypersensitivity reactions (Butterworth and David, 1981; Kay, 1985). The eosinophils' reserve of cationic proteins and enzymes which imparts their profound parasiticidal effects (Butterworth and David, 1981) contrasts with this leukocyte's purported regulatory function in inflammation (Kay, 1985; Fechter et al., 1986). The opposing functions possessed by this leukocyte exemplify the enigma of the eosinophil. Recent evidence suggests that although the eosinophil does posses some regulatory capabilities, its presence is, in fact, a harbinger of tissue destruction (Gleich and Adolphoson, 1986, Wardlaw and Kay, 1987; Spry, 1988). Nor does the presence of the eosinophil automatically infer IgE mediated hypersensitivity, as evidenced by studies examining the interaction of the eosinophil with the cellular arm of the immune system (Basten and Beeson, 1970; Ruscetti et al., 1976; Beeson and Bass, 1977; Raghavachar et al., 1987; Ohnishi et al., 1988). The purpose of this review is to provide a brief overview of the structure and biology of the mammalian eosinophil and to emphasize the fact that eosinophils fulfil a paradoxical role as effectors of tissue damage and as benign modulators of inflammation.

The eosinophilic leukocyte: structure and function

The evidence reviewed here indicates that the eosinophil has the ability to kill many species of helminths and likely does so during worm infection. This toxic ability appears to be regulated by several other cells including mast cells, monocytes, and T lymphocytes. Eosinophils kill helminths through their ability to generate potent oxidants and through their content of cationic proteins, which likely achieve high concentrations at points of granule deposition. Eosinophils also participate in inflammation in human disease especially asthma, skin diseases, and heart disease. Though present concepts hold that the mast cell is the cornerstone of the allergic inflammatory response (450), the findings that eosinophils bind IgE and are activated by antigen-IgE complexes and that the eosinophil can elaborate many inflammatory mediators raise the possibility that the eosinophil might also be involved in the initiation of inflammatory responses. Finally, an eosinophil-related protein appears to play an undefined role in human reproduction.