Unraveling avian and reptilian hematology: An optical and electron microscopic study of the buffy coat

DOI: 10.1111/vcp.12640 ORIGINAL RESEARCH Unraveling avian and reptilian hematology: An optical and electron microscopic study of the buffy coat Filipe Fontes Pinto 1 Célia Lopes 2 Fernanda Malhão 2 Ricardo Marcos 2 1 Exoticvets, Porto, Portugal 2 Cytology Diagnostic Services, Laboratory of Histology and Embryology, Institute of Biomedical Sciences Abel Salazar, University of Porto, Porto, Portugal Correspondence R. Marcos, Lab Histology and Embryology, Institute of Biomedical Sciences Abel Salazar, University of Porto, Porto, Portugal. Email: rmarcos@icbas.up.pt Background: Blood centrifugation and buffy coats are at the cornerstone of hematology. In mammals, the buffy coat has a layered disposition (from bottom to top) with neutrophils on top of erythrocytes, followed by monocytes/lymphocytes, and platelets. In nonmammals, this distribution is unknown. Recently, the cell tube block (CTB) technique was developed to study the buffy coat, but it was never applied to nonmammal buffy coats. Objectives: This study aimed to evaluate using the CTB technique to study reptilian and avian buffy coats and to propose its use for clinical applications. Methods: Blood from five birds and eight reptiles of different species was obtained to make CTBs that were processed for optical/electron microscopy. H&E, Sirius red, and immunohistochemistry staining against CD3 (to label T lymphocytes) were applied to the CTBs. Results: In birds, the buffy coat had a layered appearance with the granulocyte layer containing granulocytes (heterophils and eosinophils) and nucleated erythrocytes followed by a mononuclear cell layer containing lymphocytes, monocytes, and thrombocytes. In some animals, a nucleated erythrocyte layer was observed admixed with the granulocyte/mononuclear cell layer. A small clot within the buffy coat was seen in seven reptiles, and less definition of layers occurred in reptiles, with only one or two layers. Lymphocytes appeared toward the top of the buffy coat. Conclusions: From a comparative hematology perspective, the buffy coat of mammals differs from that of birds and more from that of reptiles. The CTB technique can be used to study these differences in avian and reptilian hematology, especially to study atypical circulating cells, hemoparasites, or blood cell proportions in health and disease. KEYWORDS Birds, buffy coat, cell block, hematology, reptiles 1 INTRODUCTION It is widely recognized that the blood of birds and reptiles is particularly noted for their nucleated erythrocytes and thrombocytes (the latter equivalent to platelets of mammals). Moreover, their leukocytes also differ from most mammals, since these species have heterophils, which have the same function as neutrophils but are cytochemically and morphologically different. 1 This variability and the nucleated erythrocytes and thrombocytes have precluded the use of automated hematology analyzers and flow cytometry methods in birds and reptiles. 1 One of the cornerstones of hematologic techniques in mammals corresponds to the centrifugation of blood in capillary tubes, in which erythrocytes are tightly packed in a pellet referred to as the packed cell volume, whereas leukocytes and platelets are layered within the buffy coat. These cells have an orderly distribution in this layer, with Vet Clin Pathol. 2018;47:407 414. wileyonlinelibrary.com/journal/vcp 2018 American Society for Veterinary Clinical Pathology 407

408 PINTO ET AL. (bottom to top): granulocytes adjacent to erythrocytes, then monocytes/lymphocytes, and lastly, platelets next to the plasma interface. 2 The buffy coat height has been used to roughly estimate leukocyte numbers in mammals, with each 1% increase corresponding to 20 000 leukocytes per μl. 2 In reptiles and birds, a buffy coat is also formed after centrifugation. However, no correlation appears to exist between buffy coat height and leukocyte number. 3,4 The distribution of cells in reptile and avian buffy coat layers is unknown to date. It is assumed to be similar to that of mammals, even in the presence of nucleated erythrocytes and thrombocytes. Recently, techniques, such as the cell tube block (CTB), have been developed for embedding the buffy coat, both for optical and electron microscopy. 5,6 Despite the proven utility of these techniques in dogs and cats, to the authors knowledge, they have never been applied to nonmammals. This technique would be useful to study the distribution of cells within the buffy coat of birds or reptiles. Moreover, it could be a valuable tool to diagnose hemoparasites or leukemic conditions reported in those species. 7 In this study, we wanted to evaluate the CTB technique in the blood of reptiles and birds and to test its utility in health and disease. Moreover, within the view of comparative hematology, we aimed to study the distribution of blood cells within the buffy coat layer. 2 MATERIALS AND METHODS 2.1 Animals Blood was collected from nine birds of five species (Accipiter gentilis, Aquila pennata, Buteo buteo, Falco peregrinus, and Falco rusticulos x cherrug) and eight reptiles of different species (Aspidites ramsayi, Boa constrictor, Chamaeleo calyptratus, Iguana iguana, Pogona vitticeps, Testudo horsfieldii, Tupinambis merianae, Varanus acanthurus) as a part of a yearly checkup. In the case of birds, blood was collected from the left jugular vein into 1 ml tubes containing EDTA 3K (Vacuette, Vila do Conde, Portugal), whereas in reptiles, blood was collected into 0.5mL heparin tubes (Vacuette) from the heart in serpents (A ramsayi, B constrictor), caudal vein in lizards (C calyptratus, I iguana, P vitticeps, T merianae, V acanthurus), and jugular vein in the turtle (T horsfieldii). One animal per species was evaluated, except for B buteo (five individuals). This checkup included routine hematologic analyses and biochemical panels. Animals with biochemistry values within their respective reference intervals were included in this study. Regarding the hematologic analyses, conventional blood films were evaluated, and the packed cell volume was assessed. Screening for parasites, inclusions, and other abnormalities, which could affect the distribution of cells in the CTB or buffy coat, was also performed. 2.2 Converting buffy coats to cell tube blocks Plain capillary tubes (Marienfeld, Lauda Königshofen, Germany) were filled with blood and spun in a microhematocrit centrifuge (International Equipment Co., IEC MB Centrifuge, Needham, MA, USA) at 12 700g for 5 minutes. The centrifugation time and speed used in bird and reptile samples were similar to that of mammalian blood, as recommended in the literature. 1 Tubes were broken at the plasmabuffy coat interface using a diamond pen, and the remaining portion of the capillary tubes (containing the buffy coat) were immersed in either 10% formalin for 24 hours for optical microscopy, or 2.5% glutaraldehyde in 0.1 M sodium cacodylate buffer (ph 7.2) for 2 hours for electron microscopy, as detailed elsewhere. 6 2.3 Processing and staining for optical microscopy The tubes were routinely processed in an automated tissue processor (Leica TP1020, Nussloch, Germany) to obtain the CTBs. 6 Paraffin embedded CTBs were sectioned (3 μm) and routinely stained with H&E. Histochemical (Sirius red) and immunohistochemical (antibodies against CD3) procedures were evaluated. Sirius red F3B (Sigma Aldrich, St. Louis, MO, USA) was prepared as described by Wehrend et al. 8 Sections were deparaffinized, hydrated, and stained in hematoxylin (1 minute), washed in tap water and differentiated in 70% ethanol, and then stained with Sirius red for 24 hours. This stain is considered selective for eosinophils in mammals and heterophils and eosinophils in avian and reptile species. 9 Sections from mare cervical tissue were used as positive controls. 8 In the CTBs of reptiles, the Fraser Lendrum staining method for fibrin was applied. 10,11 In this case, sections were deparaffinized, hydrated, and placed in Zenker's fixative in the oven for 1 hour. After removal of mercury pigments, slides were rinsed in water and sequentially stained with Celestine blue, Mayer's hematoxylin, orange G picric solution, and acid fuchsin (each for 5 minutes, intermingled with tap water rinses). Sections were then differentiated in orange G picric solution (diluted in 80% alcohol), rinsed in water, and placed in MacFarlane's working solution for 5 minutes. Finally, they were washed in tap water and stained with Light green (1 minute) (Merck, Darmstadt, Germany). Sections of a fibrinous pleuritis were used as positive controls. As for immunohistochemistry, antibodies against CD3 [polyclonal A 0452 (Dako, Glostrup, Denmark) from rabbit against the human CD3ε chain at a 1:50 dilution] were applied. This antibody has been shown to highlight T lymphocytes in turtles and gallinaceous species. 12 Antigen retrieval was performed in a 100 C water bath, and a polymer detection system (Novocastra, Newcastle, United Kingdom) was used. Diaminobenzidine chromogen and hematoxylin counterstain were used, and positive (a canine lymph node section) and negative controls (CTB sections in which antibody was omitted) were included. 2.4 Processing for electron microscopy Samples were removed from the glass tube using a clip, and the CTBs were postfixed in phosphate buffered 1% OsO 4 in 0.1 M sodium cacodylate buffer (ph 7.2) for another 2 hours. After dehydration in ethanol and propylene oxide, CTBs were embedded in an epoxy resin. Semi thin sections were obtained with glass knives and stained with methylene blue azure II. Ultrathin sections were

PINTO ET AL. 409 obtained with a diamond knife (Diatome, Hatfield, PA, USA) and stained with uranyl acetate and lead citrate, following routine procedures. 5 Afterward, they were observed in a transmission electron microscope (JEOL 100CXII at 60 kv). The photographic record was made using a digital camera (SC200 CCD, Gatan Orius, Pleasanton, CA, USA). 3 RESULTS All of the animals had normal hematologic results, except for two animals. One B buteo had hemoparasites (Leukocytozoon spp.) in the blood smear; the B constrictor had inclusion bodies in heterophils and lymphocytes similar to what has been previously described, 12 and inclusion body disease (IBD) was diagnosed. The B constrictor was euthanized, and IBD was confirmed at necropsy (multiple inclusion bodies in the liver and lymphoid tissues). Regarding CTBs in birds, a layered appearance was noticeable. From bottom to top, nucleated erythrocytes were followed by a granulocyte layer (with heterophils, eosinophils, and basophils) and a mononuclear cell layer, comprising agranulocytes and thrombocytes (Figure 1). In some birds (eg, A gentilis and F peregrinus), a layer of nucleated erythrocytes was observed between the granulocyte and mononuclear cell layers (Figure 2). Considering their smaller size and larger nuclei, these were interpreted as immature erythrocytes. The Sirius red stain clearly marked granulocytes: heterophils were much more abundant and intensely stained with rod shaped granules, whereas eosinophils were less frequent and not as intensely stained as heterophils (Figure 3). Notably, a small number of heterophils appeared between erythrocytes, whereas rare cells were present in the mononuclear cell layer. With CD3 immunostaining, the perimembranous staining of lymphocytes was noticed; this is typical of T lymphocytes. 13 These cells either appeared on top of the buffy coat (eg, A gentilis) over thrombocytes or appeared dispersed in the mononuclear cell layer (eg, F peregrinus) (Figure 4). It should also be noted that in B buteo (five analyzed birds), the general features of the buffy coat were similar among the animals. The buffy coat of reptiles also presented a layered appearance, similar to that of birds, but with three major differences: (1) No distinct layer of immature erythrocytes was noticeable between granulocytes and the mononuclear cell layer, even if smaller erythrocytes appeared throughout these layers. (2) A clot was seen in the mononuclear layer (Figure 5) in all species except for I iguana. This was small (occupying less than 10% of the buffy coat area) (Figure 5) in most cases, but in A ramsayi and T horsfieldii it accounted for 15% and 25% of the area of the buffy coat, respectively. As assessed using the Fraser Lendrum stain for fibrin, this clot comprised erythrocyte debris and fibrin strands. (3) The separation of layers was less evident, with some heterophils present among erythrocytes and mononuclear cells, and erythrocytes present in the buffy coat. In I iguana and T merianae (Figure 5), two layers (granulocytes and mononuclear cells) appeared, even though heterophils were observed throughout the layers. In others, only a single layer was observed. In A B FIGURE 1 Capillary tube after formalin fixation (A), with the typical whitish buffy coat (BC) above the column of erythrocytes. (B) The general appearance of the cell BC tube block of a B buteo bird after routine paraffin embedding, sectioning, and staining. (C) Two zones were recognized in this species, namely with granulocytes (bottom) and with mononuclear cells (top). H&E, bar = 200 μm (B), 25 μm (C) FIGURE 2 Avian buffy coat cell tube block section of the buffy coat of A gentilis in which a layer of immature erythrocytes appeared between the granulocyte (highlighted by the Sirius red stain, left in the inset) and the mononuclear cell layers (right in the inset). Sirius red, bar = 150 μm (25 μm inset) the T horsfieldii turtle, the formation of layers was less evident, since erythrocytes appeared admixed with heterophils, lymphocytes, and thrombocytes. Regarding the Sirius red stain, it highlighted heterophils which appeared either lobed (typical of lizards) or nonlobed (seen in most reptiles) (Figure 6). The CD3 immunostaining highlighted T lymphocytes, which appeared mostly in the buffy coat subjacent to the plasma (Figure 7). Semi thin sections allowed a clear distinction between granulocytes and mononuclear cells and the details of these cells could be well appraised with electron microscopy (Figure 8). The CTBs were also useful in the two diseased animals since Leukocytozoon sp. were observed throughout the buffy coat in B buteo with optical C

410 PINTO ET AL. microscopy, and inclusion bodies were seen in the B constrictor with electron microscopy (Figure 9). 4 DISCUSSION FIGURE 3 Avian heterophils typically have bilobed nuclei and rod shaped granules that are intensely stained with Sirius red, whereas eosinophils have round and less intensely stained granules (arrows). (The rare basophils did not stain with the Sirius red stain.) Sirius red, F rusticulos x cherrug; bar = 15 μm A In this study, we applied the CTB technique for the first time to study the distribution of cells in the buffy coat of birds and reptiles, and to document pathological conditions, such as hemoparasite infection and IBD (showing that inclusions could be detected). Techniques used to study blood cells in optical and electron microscopy have been described in mammals 5,6,14 and prove useful in diagnosing and characterizing the pathologic conditions of chronic lymphocytic leukemia, 6 leukemic dissemination of multiple myeloma, 14 and hemoparasites in the dog. 15 However, it was unknown if such techniques could be applied to animals with nucleated erythrocytes. It should be stressed that flow cytometry methods, which are the standard for hematology in humans, are not readily applied in animals with nucleated erythrocytes and thrombocytes. In this study, we tested the CTB technique in nonmammals to generate material for optical and electron microscopy. Regarding the technique, we successfully recovered cells for optical and electron microscopy, and cellular details were comparable with existing literature. 9 Histochemical and immunohistochemical procedures could be easily applied to CTBs. Therefore, we propose that the CTB technique should be used in pathologic cases, such as in the diagnosis of leukemic dissemination 6 or when atypical cells are present in the blood smear. 14 As to electron microscopy processing, we used a protocol similar to that published for dogs and cats, 5 but simpler than that previously reported for reptiles, such as the Bobtail lizard 16 or Yellow Bellied Slider Turtle. 17 Still, it rendered acceptable ultrastructural leukocyte detail and enabled us to report, for the first time, the ultrastructure of inclusion bodies in the blood of a B constrictor using the CTB technique. These inclusion bodies, which were observed in heterophils, lymphocytes, and monocytes, had electron dense material compatible with a viral protein. 18 Considering that the CTB procedures B FIGURE 4 Avian buffy coat cell tube block sections immunostained with CD3 to highlight T lymphocytes. These cells either predominated over the top of the mononuclear cell layer, as in A gentilis (A) or were dispersed throughout this layer, as in F peregrinus (B). In the inset, the typical perimembranous pattern of CD3 is depicted; arrowhead: the transition between granulocyte/ mononuclear cell layer. Diaminobenzidine, hematoxylin counterstain; bar = 200 μm (25 μm inset) FIGURE 5 Reptilian buffy coat cell tube block section of T merinae with a small clot within the mononuclear cell layer, magnified in the inset; it is composed of erythrocyte debris and fibrin strands (light orange and red in the Fraser Lendrum stain, respectively). H&E and Fraser Lendrum (inset); bar = 95 μm (30 μm inset)

PINTO ET AL. 411 A B C FIGURE 6 A: Heterophils (and eosinophils) appear more distributed in reptiles than in birds, as seen in the buffy coat of P viticeps (A). Heterophils in reptiles are either round, as in V varanus and admixed in erythrocytes (B) or lobed, as in I iguana, and admixed in the mononuclear cell layer (C). Sirius red, bar = 130 μm (A) and 15 μm (B, C) A B FIGURE 7 Reptilian buffy coat cell tube block section of T merinae immunostained with CD3. The T lymphocytes appear dispersed in the mononuclear cell layer but predominate over the top of this layer. Typical perimembranous staining is present (B is a magnification of the square in A). Diaminobenzidine, hematoxylin counterstain; bar = 100 μm (A) and 15 μm (B) are easy to implement in a veterinary clinic, the technique could be used to confirm IBD, either by the use of immunohistochemistry (with recently developed antibodies) in paraffin sections 18,19 or by electron microscopy. In this manner, the technique could complement that of PCR (which is more sensitive) for the diagnosis of IBD, especially in research settings. Likewise, when hemoparasites appear in blood smears, the CTB technique can be used to generate material for immunohistochemical studies. From a comparative hematology point of view, we observed differences in birds and reptiles compared with mammals (Figure 10). It is well known that the position of cells in the buffy coat layer is related to cell densities, with heavier cells in the bottom (granulocytes have 1.070 and 1.080 g/cm 3 density) and lighter cells at the top (numerous platelets of mammals have 1.050 g/cm 3 density). 20,21 In mammals, the mixing of elements between layers is negligible, and younger erythrocytes have been described below granulocytes. 20 In some birds, we observed this layer, but above granulocytes, and there was extensive mixing of lymphocytes and thrombocytes (Figure 10). As a side note, avian thrombocytes are much less abundant than platelets in mammals, 20 30 000/ mm 3 in birds compared with more than 200 000/mm in mammals. 1,3 To the best of our knowledge, the distribution of cells in the buffy coat of selected reptile and nonpoultry avian species has never been studied in

412 PINTO ET AL. A B C D E F FIGURE 8 Semi thin section (A) and electron microscopic features (B F) of the cells in avian buffy coat tube blocks. (A) Heterophils (arrows) are next to erythrocytes in the lower part of the buffy coat from A. gentilis, semi thin section, Methylene blue azure II. (B) Erythrocyte (e), thrombocyte (t), and (C) heterophil in A pennata. (D) Eosinophil in a B buteo and (E) monocyte in A gentilis. (F) Azurophil in B constrictor this is a cell type unique in reptiles, Uranyl acetate, and Lead citrate; bar = 7.5 μm (A), 2.5 μm (B, C), and 2 μm (D, E, F) A B C D FIGURE 9 Leucocytozoon spp. parasites (A, B) in the buffy coat of a B buteo. These were observed in the mononuclear cell layer (arrows), and their appearance resembled that seen in blood smears (inset of B). Inclusion bodies (C, D) in the buffy coat of B constrictor, which appear in lymphocytes (black arrowheads and D) and in heterophils (white arrowhead in C); Paraffin section, H&E (A, B), semi thin section, Methylene blue azure II (C) and electron microscopy (D); bar = 15 μm (A and inset B), 10 μm (B), 7.5 μm (C), and 1 μm (D)

PINTO ET AL. 413 detail. Still, our data are in accordance with a classical report based sepa- to the use of heparin as an anticoagulant. Although heparin is rec- ration of cells by Percoll gradients.22 This study by Mills and Wilcox22 ommended as an anticoagulant for most reptilian species, it has been with chicken blood reported that thrombocytes appear admixed with reported to produce leukocyte and thrombocyte clumping in blood lymphocytes in lighter fractions (which corresponds to the upper buffy smears.1 Presumably, the clot may be due to platelet clumping, but coat parts), whereas denser fractions contained mostly heterophils. further studies (eg, using EDTA, at least in Squamata) are needed to These cells were also present among nucleated erythrocytes, and to a elucidate this issue. lesser extent in lighter fractions of lymphocytes. Notably, these authors Despite being a first look at using buffy coat CTBs on selected reported that immature chicken erythrocytes had the same density as species of nonpoultry birds and reptiles, our study has limitations. granulocytes22 as we highlighted here. First, except for B buteo, we only used one animal per species of the Only a few studies have been devoted to the separation of cells nonpoultry birds. Even if animals had normal hematological findings in reptilian blood. In testudines (turtles), it has been reported that (except for one B buteo and B constrictor), the use of a single individ- cell separation based on cell density is dependent on the anticoagu- ual to be a representative of a species could be considered far lant used. Nonetheless, this separation seems less effective com- reaching. Still, buffy coat CTBs seems to be a reasonable method for pared with mammals.23 For instance, erythrocyte (primarily immature highlighting the general features of the buffy coat in selected species erythrocytes) and thrombocyte contamination occur throughout all of nonpoultry birds and reptiles. In addition, using a variety of birds layers and specifically in sea turtles, for which a predominance of and reptiles provides a solid basis and a first assessment for the use granulocytes in lighter layers has been reported,23 which agrees with of CTBs on buffy coats in these species. For an in depth characteri- our data, even if we were unable to observe granulocytes over zation of a particular species, a higher number of individuals should mononuclear cells in the T horsfieldii turtle. be evaluated. Second, we only used CD3 for immunohistochemistry The presence of a clot in reptilian buffy coats was a new and of T lymphocytes because this antibody has been shown to be highly surprising finding. This appeared in most species, was composed of conserved across many species of birds and reptiles.13 In chicken the erythrocyte debris and fibrin and tended to be small. We have peripheral blood, B lymphocytes account for 5% 15% of lympho- no explanation for this finding, except that it may be due to the cytes.24 In our study, we presumed that the fewer B lymphocytes high speed centrifugation causing erythrocyte lysis, or it may be due would be positioned intermingled with T lymphocytes in the buffy F I G U R E 1 0 Schematic overview of the buffy coat (BC) in mammals (A), birds (B), and reptiles (C). In mammals, the distribution comprises (from bottom to top): erythrocytes, immature erythrocytes, and BC with three layers (granulocytes, lymphocytes/monocytes, platelets). In birds, two layers are recognized in the BC (granulocytes, mononuclear cell layer). The layers are less defined and vary with species; in some, granulocytes appear followed by immature erythrocytes in the BC (left side of B, as seen in A gentilis), whereas in others the opposite occurs (right side of B, as seen in B buteo). Afterward, a mononuclear cell layer follows, and variation among species also occurs: lymphocytes either appear admixed with thrombocytes (left side of B, as seen in F peregrinus) or predominate over the top (right side of B, as seen in A gentilis). In reptiles, the BC appears less defined, with only a single layer in many species. Erythrocytes are larger than in birds, and heterophils appear round (in most species) or lobed (in lizards). Lymphocytes appear dispersed in the mononuclear cell layer but predominate over the top of the BC

414 PINTO ET AL. coat column. Although this assumption seems plausible, and we fulfilled our main objectives to prove that buffy coat CTBs could be used for immunohistochemistry in birds and reptiles with appropriate and validated antibodies, a more detailed evaluation of the buffy coat of a particular species should include immunohistochemistry against both B lymphocytes (using antibodies against CD79α or PAX 5) and thrombocytes (eg, antibodies against CD41/CD61). Such studies would require a detailed and cumbersome evaluation of immunohistochemistry protocols since antibody cross reactivity has been reported to vary substantially among bird species 25 and is presumed to be even more variable in reptiles. This study's findings could be useful for clinical and research applications. Besides the use of the CTB for diagnostic purposes, we have shown that the appearance of cells in the buffy coat is slightly and highly variable in birds and reptiles, respectively. Considering that heterophils can be found admixed with erythrocytes and that these cells can be positioned inside the buffy coat, the use of the height of the buffy coat layer as a rough indicator of leukocyte number is not warranted, especially in reptiles. In conclusion, we have shown that CTB can be applied to the blood of various avian and reptilian species, enabling the use of histochemical and immunohistochemical procedures. And, although we have highlighted its value in studying buffy coats in health and disease, the fact that CTBs can be used to identify infectious agents suggests that further studies are needed to characterize CTB technology in these nonmammalian species. ACKNOWLEDGMENTS We deeply acknowledge the time and effort devoted by the anonymous reviewers of the manuscript. We are also in debt to Rosária Seabra for the technical assistance in the histochemical staining and Prof. Marta Santos for the comments on the manuscript. DISCLOSURE The authors have indicated that they have no affiliations or financial involvement with any organization or entity with a financial interest in, or in financial competition with, the subject matter or materials discussed in this article. REFERENCES 1. Campbell TW. Exotic Animal Hematology and Cytology. Oxford, UK: John Wiley & Sons; 2015:37 87. 2. Villiers E. Introduction to hematology. BSAVA Manual of Canine and Feline Clinical Pathology. Gloucester, UK: British Small Animal Veterinary Association; 2007:23 32. 3. Mader DR. The reptilian hematocrit. Proceedings of the Association of Reptilian and Amphibians Veterinarians.2000:159-160. 4. Walberg J. White blood cell counting techniques in birds. Sem Avian Exotic Pet Med. 2001;10:72 76. 5. Yabuki A, Sawa M, Chang H-S, Yamato O. A practical technique for electron microscopy of buffy coats in dogs and cats. Anat Histol Embryol. 2014;44:317 320. 6. Marcos R, Santos M, Marrinhas C, Caniatti M. Cell tube block: a new technique to produce cell blocks from fluid cytology samples. Vet Clin Pathol. 2017;46:195 201. 7. Schilliger L, Selleri P, Frye FL. Lymphoblastic lymphoma and leukemic blood profile in a red tail boa (Boa constrictor constrictor) with concurrent inclusion body disease. J Vet Diagn Invest. 2011;23:159 162. 8. Wehrend A, Huchzermeyer S, Bostedt H. Distribution of eosinophils and mast cells in the cervical tissue of non gravid mares in dioestrus. Reprod Dom Anim. 2004;40:562 663. 9. Kelényi G, Németh Á. Comparative histochemistry and electron microscopy of eosinophil leucocytes of vertebrates. Acta Biol Acad Hung. 1969;20:405 422. 10. Lendrum AC, Fraser DS, Slidders W, Henderson R. Studies on the character and staining of fibrin. J Clin Pathol. 1962;15:401 413. 11. Fraser Lendrum Method. WebPath: Internet Pathology Laboratory. https://library.med.utah.edu/webpath/webpath.html. Accessed January 25, 2018. 12. Banajee KH, Chang LW, Jacobson ER, Rich GA, Royal AB. What is your diagnosis? Blood film from a Boa constrictor. Vet Clin Pathol. 2012;41:158 159. 13. Muñoz FA, Estrada-Parra S, Romero-Rojas A, et al. Identification of CD3 + T lymphocytes in the green turtle Chelonia mydas. Vet Immunol Immunopathol. 2009;131:211 217. 14. Marcos R, Canadas A, Leite-Martins L, et al. What is your diagnosis? Cutaneous nodules and atypical blood cells in a dog Vet Clin Pathol. 2018;47:317 319. 15. Marcos R. Cell tube: a new technique for making cell blocks from needle rinses. Vet Clin Pathol. 2013;42:401 402. 16. Moller CA, Gaál T, Mills JN. The hematology of captive Bobtail lizards (Tliqua rugose): blood counts, light microscopy, cytochemistry and ultrastructure. Vet Clin Pathol. 2016;45:634 647. 17. Hérnandez JD, Orós J, Artiles M, Castro P, Blanco A. Ultrastructural characteristics of blood cells in the Yellow Bellied Slider Turtle (Trachemys scripta scripta). Vet Clin Pathol. 2016;45:106 109. 18. Hetzel U, Sironen T, Laurinmäkl P, et al. Isolation, identification, and characterization of novel arenaviruses, the etiological agents of Boid Inclusion Body Disease. J Virol 2013;87:10918 10935. 19. Chang L-W, Wozniak E, Chow M, et al. Immunohistochemical detection of a unique protein within cells of snakes having inclusion body disease, a world wide disease seen in members of the families Boidae and Pythonidae. PLoS ONE. 2013;8:e82916. 20. Zipursky A, Bow E, Seshadiri RS, Brown EJ. Leukocyte density and volume in normal subjects and in patients with acute lymphoblastic leukemia. Blood. 1976;48:361 371. 21. Sutton DW, Chen PC, Schmid-Schönbein GW. Cell separation in the buffy coat. Biorheology. 1988;25:663 673. 22. Mills JN, Wilcox GE. Separation of phagocytic leukocytes from the peripheral blood of chickens. Avian Pathol. 1993;22:343 352. 23. Harms CA, Keller JM, Kennedy-Stoskopf S. Use of a two step Percoll gradient for separation of Loggerhead Sea Turtle Peripheral Blood Mononuclear cells. J Wildlife Dis. 2000;36:535 540. 24. Burkhardt E. Scanning electron microscopy of peripheral blood leukocytes of the chicken. Cell Tissue Res. 1979;204:147 153. 25. Sinclair KM, Hawkins MG, Wright L, et al. Chronic T cell lymphocytic leukemia in a black swan (Cygnus atratus): diagnosis, treatment, and pathology. J Avian Med Surg. 2015;29:326 335. How to cite this article: Pinto FF, Lopes C, Malhão F, Marcos R. Unraveling avian and reptilian hematology: An optical and electron microscopic study of the buffy coat. Vet Clin Pathol. 2018;47:407 414. https://doi.org/10.1111/vcp.12640