Cytochemical Characterization of Leukemic Cells from 20 Dogs

Transcription

1 Vet. Pathol. 22: (1985) Cytochemical Characterization of Leukemic Cells from 20 Dogs N. R. FACKLAM and G. J. KOCIBA Department of Veterinary Pathobiology, The Ohio State University, Columbus, Ohio Abstract. The hematopoietic cells in blood and/or bone marrow from 20 leukemic dogs and 22 control dogs were characterized using a battery of cytochemical stains. The results of cytochemical staining led to modification of the diagnoses based on clinical, hematologic and histologic findings in seven (35%) of the s. Sudan black B and chloroacetate esterase served as granulocytic markers in both the control and leukemic groups. Peroxidase activity was present in the granulocytes and monocytes of control animals but not the blasts of leukemic dogs. Alkaline phosphatasepositive staining of granulocytic precursors was a consistent finding in granulocytic and myelomonocytic, and alkaline phosphatasepositive lymphoblasts were seen in 38% of lymphocytic s. Diffuse alpha naphthyl butyrate esterasepositive staining marked monocytes in both control and leukemic dogs. Cytochemical staining was found to be a valuable diagnostic aid in the classification of s in the dog. The neoplastic cell type in the majority of s in dogs can be classified with WrightGiemsa stained blood and bone marrow films. When primitive cells predominate, however, the morphologic differences are indistinct. Various leukocytes contain specific enzymes which may aid in the identification of the cells. Cytochemical stains of human blood and bone marrow cells have been shown to improve the accuracy of diagnosis of hematologic diseases in man. Reports of cytochemical characterization of hematopoietic cells of dogs are limited and most are confined to the normal anima1. o 5 The objectives of this study were to characterize the cytochemical staining patterns of normal and neoplastic hematopoietic cells of dogs using a battery of enzyme stains and to compare the cytochemical results with the diagnoses based on clinical, hematologic and histologic findings. Materials and Methods Bone marrow specimens were aspirated from the proximal humerus or femur from 20 leukemic dogs presented to the OSU Veterinary Teaching Hospital. Samples also were aspirated from 22 control dogs, including dogs with nonhematologic diseases. In some cases, blood samples collected on the same day as the marrow aspirate also were stained. A few samples were obtained from a private veterinary laboratory. Blood films and bone marrow coverslip smears were prepared and airdriedgenerally within one hour of sample collection. Films were stained with WrightGiemsa, alpha naphthyl butyrate estera~e, ~ and periodic acidschiff. Cytochemical staining kits (Sigma Chemical Co., St. Louis, Mo.) were used 363 for Sudan black B, acid phosphatase with and without tartrate, alkaline phosphatase, and chloroacetate esterase procedures. Positive staining was defined as distinct staining of a minimum of 10% of the abnormal cells. A positive control was included in each batch of stains. If staining could not be done within eight hours, the films were stored unfixed at room temperature in the dark with the exception of the slides for acid phosphatase and alkaline phosphatase which were fixed in accordance with the manufacturer s instructions (Sigma Chemical Co., St. Louis, Mo.) and stored at 4 C. The corresponding histologic sections of representative tissues from dogs that were necropsied along with the WrightGiemsa blood and bone marrow films were reviewed by one of the authors (GK). Diagnosis was based on cell morphology, anatomic distribution of lesions, and histologic findings. To simplify data analyses, the lymphocytic s were subdivided into lymphoblastic, lymphoblastic associated with lymphoma,22 and welldifferentiated lymphocytic on morphologic criteria only. These results were compared with the corresponding cytochemical findings. Results Cytochemical staining of normal canine blood and bone marrow The cytochemical staining characteristics of normal canine blood and bone marrow cells are summarized in table I. While the cells in peripheral blood samples gave the same reaction as those in bone marrow for all stains tested, the cells in the marrow stained with greater intensity. In the normal dog, peroxidase activity was detected

2 364 Facklam and Kociba Table I. Cytochemistry of normal canine blood and bone marrow cells Alpha Bone Sudan phosphatasea Alkaline Chloroacetate naphthyl Periodic marrow Peroxidase black cell B phosphatase esterase butyrate acidschiff phosphataseb esterase M yeloblast 1 Progranulocyte I occ Neutrophilic myelocyte f 51 metam yelocyte f f/ band f f/ segmented f / Eosinophilic myelocyte I metam yelocyte I band I segmented I f sl Basophilic myelocyte I metam yelocyte / band I segmented / Lymphocyte occ/ occ foc Monocyte sl / dif sl Rubriblast / Prorubricyte / Rubricyte / Metarubricyte / Megakaryocyte / sl sl phosphatasea = without tartrate; phosphataseb = with tartrate; = positive; = negative; & = some cells positive, others negative; occ = occasionally positive; st = strongly positive; sl = slightly positive; foc = focally positive; dif = diffusely positive. St sl sl sl in late myeloblasts and was most intense in the progranulocytes (table I). Moderately granular staining generally was present in neutrophilic myelocytes and in the more mature neutrophilic leukocytes. Eosinophils of all stages of maturity stained more strongly than neutrophils and in the segmented eosinophil, the stain outlined the specific granules. Monocytes were lightly positive, and had a fine dustlike granulation. Lymphocytes, erythrocytes, and megakaryocytes were negative. Sudan black B was a better stain than peroxidase for granulocytic cells because neutrophils of all stages were consistently positive. Eosinophils were strongly positive and monocytes slightly positive. All other cells were negative for Sudan black staining. Positive chloroacetate esterase staining was present in late myeloblasts, was most intense in progranulocytes, and decreased as the neutrophils matured. Megakaryocytes and basophils were slightly positive. Eosinophils, lymphocytes, and monocytes were negative for chloroacetate esterase. Myeloblasts and progranulocytes did not stain with periodic acidschiff. Positive staining was seen in neutrophilic myelocytes and increased in intensity with cell maturation. Eosinophils, monocytes and megakaryocytes were slightly positive. Basophils, erythrocytes, and lymphocytes were negative. Strongest acid phosphatase staining was seen in progranulocytes. Staining intensity was inversely related to maturity in the neutrophilic cell line and less so in the eosinophils and basophils. Monocytes generally had some positive granulation as did occasional lymphocytes. The erythroid cell line had positive granular acid phosphatase staining from the rubriblast to the diffusely basophilic erythrocyte. Megakaryocytes were also diffusely positive. Only the eosinophilic and basophilic cell lines remained positive when films were stained in the presence of tartrate. Positive alkaline phosphatase staining was seen in some but not all progranulocytes. Strongest staining was seen in eosinophilic myelocytes and the reaction diminished slightly with cell maturation. All other cells were negative. Monocytes and megakaryocytes incubated with al

3 ~ ~ ~ Cytochemistry in Leukemia 365 pha naphthyl butyrate esterase had a diffuse staining pattern and were the only consistently positive cells. Occasional lymphocytes were positive with a focal staining pattern. Cytochemical staining of leukemic cells in the dog The results of cytochemical staining of leukemic cells from dogs with various s are presented in table 11. The preliminary diagnoses, based on clinical, routine hematologic, and histologic findings, were compared with the final diagnoses based on the preliminary diagnoses plus the cytochemical findings. Thirteen (65%) of the s were diagnosed as lymphoid in origin, five of which (dogs 610) were associated with lymphoma. Ten of the cases were classified as lymphocytic using morphologic criteria and three additional cases were included after cytochemical characterization. Of these 1 3 lymphocytic s, five were negative for all of the cytochemical stains. phosphatasepositive cells were present in four dogs with lymphocytic, but the activity was tartrateresistant in only one instance. Distinctive focal alpha naphthyl butyrate esterase staining was seen in two dogs (fig. 1) and light granular chloroacetate esterase staining in another. Five of the dogs had alkaline phosphatasepositive lymphoblasts (fig. 2). The alkaline phosphatase staining intensity ranged from slight to strong. The cells in one of the alkaline phosphatase Table 11. Cytochemical staining results in cells of canine s phos Alka Alpha Sudan Dog Preliminary Per phatasea line Chloro naph Periodic black phos acetate tyl bu acidnumber diagnosis oxidase B pha esterase tyrate Schiff phos tase esterase phataseb Lymphoblastic Lymphoblastic Lymphoblastic Lymphocytic (welldifferentiated) Lymphocytic (welldifferentiated) Lymphoblastic Lymphoblastic Lymphoblastic Lymphoblastic Lymphoblastic Leukemiauncertain cell type (lymphocytic/monocytic) Leukemiauncertain cell type (monocytic/lymphocytic) Leukemiauncertain cell type (monocytic/lymphocytic) Leukemiauncertain cell type (monocytic/lymphocytic) Leukemiauncertain cell type (lymphocytic/monocytic) Granulocytic Monocytic Monocytic N.D. sl N.D. sl occ sl / / I / I / I / I / I / / / / occ occ st occ foc occ dif foc foe foe dif occ dif dif dif Final diagnosis Lymphoblastic Myelomonocytic Ly mphoblastic Ly mphoblastic M yelomonocytic M yelomonocytic Myelomonocytic 19 Malignant histiocytosis I 20 Erythremic myelosis occ / occ = same as preliminary diagnosis. ACPA = without tartrate; ACPB = with tartrate; = negative; = positive; occ = occasional positive cells; foc = focally positive; N.D. = not done; st = strongly positive; sl = slightly positive; dif = diffusely positive.

4 366 Facklam and Kociba.,... ".... r..., 1 2 I: 3 4 Fig. 1: Distinctive focal alpha naphthyl butyrate esterase staining pattern in immature lymphocytes from blood of dog with lymphocytic. Fig. 2: Alkaline phosphatasepositive lymphoblasts in blood of dog with lymphocytic. Fig. 3: Diffuse alpha naphthyl butyrate esterase staining pattern in immature monocytes from bone marrow of dog with myelomonocytic. Fig. 4: Sudan black Bpositive cells from blood of dog with myelomonocytic. positive cases (a dog with associated lymphoma) were also positive for fine, dustlike Sudan black B. There were four cases of myelomonocytic. Based on morphologic characteristics, the preliminary diagnosis was monocytic (two cases) and either monocytic or lymphocytic (two cases). The cells in all four cases were positive for chloroacetate esterase and diffuse alpha naphthyl butyrate esterase (fig. 3). Three dogs had cells positive for Sudan black B (fig. 4). All but one of these dogs had cells positive for alkaline phosphatase. The alkaline phosphatase staining patterns of the blasts differed with apparent cell maturation. Those cells appearing less differentiated had a distinctly focal alkaline phosphatase staining pattern. The alkaline phosphatase staining pattern in the more differentiated cells was granular and dispersed throughout the cytoplasm. The diagnosis in the single case of granulocytic leu kemia was based on morphologic criteria and corroborated by cytochemical stains. The cells, predominantly myeloblasts and progranulocytes, were positive for Sudan black B, tartratesensitive acid phosphatase, alkaline phosphatase and chloroacetate esterase. No peroxidase activity was detected. In a single case of malignant histiocytosis, the neoplastic cells were diffusely positive for Sudan black B, acid phosphatase, periodic acidschiff, and tartrateresistant acid phosphatase. The cells were negative for alpha naphthyl butyrate esterase activity. In a single case of erythremic myelosis, the blasts were occasionally positive for Sudan black B, tartratesensitive acid phosphatase and alkaline phosphatase. The acid phosphatase staining pattern was in the form of coarse blocks and heavy granulation. Seven (35 %) of the preliminary diagnoses were modified after consideration of the cytochemical staining

5 Cytochemistry in Leukemia 367 results. Reclassification was based on the presence or absence of enzyme activity that is considered relatively specific for a particular cell type. In five of the seven cases, the morphology of the cells was consistent with either lymphoblasts or monoblasts. Diffuse alpha naphthy1 butyrate esterase activity was considered to be specific evidence for cells of monocytic lineage. Chloroacetate esterase staining was considered to be quite specific for granulocytic cells. When positive staining for alpha naphthyl butyrate esterase occurred in association with chloroacetate esterase, the disease was classified as myelomonocytic. Cytochemical characterization was of greatest benefit in cases where cells were poorly differentiated. The frequency of positive staining of the neoplastic cells in the various s is presented in table 111. Five (38%) of the lymphocytic s had alkaline phosphatasepositive cells. All of the myelomonocytic s were positive for chloroacetate esterase and diffuse alpha naphthyl butyrate esterase, and 75 % had cells positive for Sudan black B. Discussion The cytochemical characteristics of hematopoietic cells in man are welldocumented and cytochemical stains are used extensively in human medicine to identify blasts that appear poorly differentiated with Wright Giemsa stain. Comparatively little work has been reported for dogs, but as the interest in chemotherapy of canine s has grown, so has the need for more accurate identification. In development of a battery of cytochemical stains, Sudan black B, and stains for alkaline phosphatase, chloroacetate esterase, and alpha naphthyl butyrate esterase would be the minimal panel for classification of various s. After consider ing the cytochemical staining results, seven (35%) of the preliminary diagnoses in this study were modified. The reclassification usually was based on results of the alpha naphthyl butyrate esterase and chloroacetate esterase stains. These two stains were most helpful in the identification of cells of monocytic and granulocytic lineage, respectively. Sudan black B and chloroacetate esterase served as markers for the neutrophilic cell line in both normal and leukemic dogs. Sudan black B also marked monocytes, but the staining was less intense and involved fewer granules compared to granulocytes. Peroxidasepositive blasts were not seen in any of the s, including those of granulocytic and monocytic origin. These findings imply that the peroxidase stain is not sensitive enough to identify leukemic myeloid blasts in the dog and are consistent with previous report^.'^ A few chloroacetate esterasepositive cells were noted in one dog with lymphoblastic, and associated lymphoma and a few Sudan black Bpositive cells in another. Chloroacetate esterasepositive staining has been detected in human monocytes and lymphocytes with localization in the endoplasmic reticulum, mitochondria, and nuclear en~elope.~ The significance of the Sudan black Bpositive staining of lymphoid cells in the one dog remains unclear. In neutrophils, the major compound to react with the periodic acidschiff reagent is glycogen.' Periodic acidschiff staining of neutrophils in the normal dog was restricted to myelocytes and the more mature stages, and did not serve as a marker for immature leukemic granulocytic cells. Alkaline phosphatase has been shown ultrastructurally to be contained within the specific granules of rabbit heterophils4 and extragranular but membranebound in human ne~trophils.~ Specific granules are Table 111. Cytochemical staining patterns in leukemic cells from 18 dogs Number of dogs with positive staining of leukemic cells Alpha Number Perox Sudan ph0sphatasea Alkaline Chloro acetate naphthyl Periodic black idase phosphatase butyrate acidschiff B ester as e phosphataseb esterase Lymphocytic foc 0 Granulocytic M yelomonoc ytic dif 1 phosphatasea = without tartrate; phosphataseb = with tartrate; foc = focal staining pattern; dif = diffuse staining pattern.

6 368 Facklam and Kociba formed at the distal face of the Golgi apparatus beginning at the myelocyte ~tage.~~ The alkaline phosphatase stain has proven useful in human medicine in distinguishing chronic granulocytic from leukemoid reaction.20 Increased levels are seen in infection and decreased levels in granulocytic. The mature neutrophils of the control dogs in both blood and bone marrow were completely lacking in alkaline phosphatase in both clinically normal patients and in those with a variety of inflammatory conditions resulting in a neutrophilia (unpublished data). Positive alkaline phosphatase staining of both mature and immature neutrophils has been detected in dogs with granulocytic2 and myelomonocytic. While alkaline phosphatasepositive mature neutrophils were not seen in any leukemic cases in this study, alkaline phosphatasepositive blasts were noted in granulocytic and myelomonocytic. The alkaline phosphatase staining pattern in neoplastic myeloblasts differed with apparent cell maturation. The distinctly focal alkaline phosphatase staining pattern seen in cells appearing less differentiated with WrightGiemsa possibly reflects early enzyme production at the convex face of the Golgi apparatus. The staining pattern in the more differentiated cells was granular and dispersed throughout the cytoplasm. The focal alkaline phosphatase activity was not seen in any other type of blast and may serve to mark primitive myeloid cells not recognizable with WrightGiemsa. Of particular interest was the alkaline phosphatase staining of blasts noted in five cases of lymphocytic. The cells were negative for Sudan black B and chloroacetate esterase. Cytochemical stains of lymph node frozen sections in man have revealed alkaline phosphatasepositive cells in the Bcell areas. Alkaline phosphatasepositive Bcell lymphomas have been documented in man. The use of this stain in combination with immunologic markers may eventually identify a subclass of canine lymphocytic. Ultrastructural studies have shown that alpha naphthy1 butyrate esterase is an ectoenzyme occurring over the external surface of monocytes. Staining begins early in monocytic differentiation, appearing before the cells acquire phagocytic capability. Diffuse alpha naphthyl butyrate esterase staining served as a marker for monocytes in both the normal and leukemic dog. Focal positive alpha naphthyl butyrate esterase staining was seen in two dogs with lymphocytic. Frozen sections of canine lymph node and thymus show the same type of focal alpha naphthyl butyrate esterase positive staining in cells of the Tcell areas.23 The reaction product corresponds with ultrastructural membranebound organelles called Gall bodiesthought to specifically mark Tcells, those able to bind the Fc portion of immunoglobulin M.6 phosphatase is contained in the primary granules of myeloid cells. phosphatasepositive granules arise from the proximal face of the Golgi apparatus during the progranulocyte stage and their numbers are reduced through This stain was not suitable for differentiating blasts because virtually all hematopoietic elements were positive to some degree for tartratesensitive acid phosphatase. In the normal dog, only the eosinophilic and basophilic granulocytes contained tartrateresistant acid phosphatase. This staining was seen in the neoplastic cells of malignant histiocytosis and in a single case of lymphoid. Similar results are seen in the hairy cells of leukemic reticuloendotheliosis in man While there is, to date, no canine equivalent of the human hairy cell, the continued use of the tartrateresistant and phosphatase stain may eventually help identify such a. The results of this study indicated that cytochemical staining allowed more precise cellular characterization. Accurate identification of leukemic cells is a prerequisite for development of more specific therapeutic regimens and for clinical characterization of various s. Additional investigations are needed to characterize the cytochemistry of leukemic cells in granulocytic, monocytic, and myelomonocytic which were underrepresented in this study. References ACKERMAN, G.A.: Histochemical differentiation during neutrophil development and maturation. Ann NY Acad Sci 113:537565, 1964 BAINTON, D.F.: FARQUHAR, M.G.: Origin of granules in polymorphonuclear leukocytes: Two types derived from opposite faces of the Golgi complex in developing granulocytes. J Cell Biol 28:277301, 1966 BAINTON, D.F.; FARQUHAR, M.G.: Differences in enzyme content of azurophil and specific granules of polymorphonuclear leukocytes. 1. Histochemical staining of bone marrow smears. J Cell Biol39:286298, I968 BAINTON, D.F.; FARQUHAR, M.G.: Differences in enzyme content of azurophil and specific granules of polymorphonuclear leukocytes. 11. Cytochemistry and electron microscopy of bone marrow cells. J Cell Biol 39: , 1968 BOYCE, R.; TEICHBERG, S.; WEISELBERG, L.; VINCI GUERRA, V.; HOPE, E.; DE~~NAN, T.: Ultrastructural localization of naphthol ASD chloroacetate esterase (CLE) in human leukocytes. (Abstract) Fed Proc 43:607, 1984

7 Cytochemistry in Leukemia BOZDECH, M.J.; BAINTON, D.F.: Identification of anb of acid phosphatase isoenzyme in leukemic reticuloenesterase as a plasma membrane ectoenzyme of monocytes dotheliosis. Cancer 29: , 1972 and as a discrete intracellular membranebounded organ 17 MATUS, R.E.; LEIFER, C.E.; MACEWEN, E.G.: Acute lymelle in lymphocytes. J Exp Med 153: , 1981 phoblastic in the dog: A review of 30 cases. J 7 BRETZ, U.; BAGGIOLINI, M.: Biochemical and morpholog Am Vet Med Assoc 183:859862, 1983 ical characterization of azurophil and specific granules of 18 NABA, K.; JAFFE, E.S.; BRAYLAN, R.C.; SOBAN, E.J.; BERhuman neutrophilic polymorphonuclear leukocytes. J ARD, C. W.: Alkaline phosphatasepositive malignant lym Cell Biol63:251269, 1974 phoma. A subtype of Bcell lymphomas. Am J Clin Pathol 8 HARVEY, J.W.; TERREL, T.G.; HYDE, D.M.; JACKSON, R.I.: 68:535542, 1977 Welldifferentiated lymphocytic in a dog: Long 19 PRUCE, A.: A Manual for Histologic Technicians, pp. 18 I term survival without therapy. Vet Pathol 18: 374 1, I Little, Brown & Co., Boston, HOROWITZ, D.A.; ALLISON, A.C.; WARD, P.; KIGHT, N.: 20 ROSENBLUM, D.; PETZOLD, S.J.: Neutrophil alkaline phos Identification of human mononuclear leucocyte popula phatase: Comparison of enzymes from normal subjects tions by esterase staining. Clin Exp Immunol 30:289 and patients with polycythemia Vera and chronic myelog 298, 1977 enous. Blood 45:335343, JAIN, N.C.: Peroxidase activity in leukocytes of some 21 SCHALM, O.W.; JAIN, N.C.; CARROL, E.J.: Veterinary Heanimal species. Folia Haematol (Leipz) 88:297304, I967 matology, 3rd ed., pp Lea and Febiger, Phila 11 JAIN, N.C.: Alkaline phosphatase in the canine and feline delphia, 1975 granulocytes. Vet Rec 81:266267, SQUIRE, R.A.; BUSH, M.; MELBY, E.C.; NEELY, L.M.; 12 JAIN, N.C.: Alkaline phosphatase activity in leukocytes of YARBROUGH, B.: Clinical and pathologic study of canine some animal species. Acta Haematol (Basel) 39:5 159, lymphoma: Clinical staging, cell classification, and ther 1968 apy. JNCI 51:565572, JAIN, N.C.: Alkaline phosphatase in leukocytes of dogs 23 WULFT, J.C.; SALE, G.E.; DEEG, H.J.: STORB, R.: Nonand cats. Blut 22:133143, 1970 specific acid esterase activity as a marker for canine 14 JAIN, N.C.: A comparative cytochemical study of leuko Tlymphocytes. Exp Hematol 9: , 1981 cytes of some animal species. Folia Haematol (Leipz) 24 YAM, L.T.; LI, C.Y.; CROSBY, W.H.: Cytochemical iden 94:4963, 1970 tification of monocytes and granulocytes. Am J Clin 15 JAIN, N.C.; MADEWELL, B.R.; WELLER, R.E.; GEISSLER, Pathol , 1970 M.C.: Clinicalpathological findings and cytochemical 25 YAM, L.T.; LI, C.Y.; LAM, K.W.: Tartrateresistant acid characterization of monomyelocytic in 5 dogs. phosphatase isoenzyme in the reticulum cells of leukemic J Comp Pathol91:1731, 1981 reticuloendotheliosis. N Engl J Med 284: , KATAYAMA, I.; LI, C.Y.; YAM, L.T.: Histochemical study Request reprints from Dr. Gary J. Kociba, Department of Veterinary Pathobiology, The Ohio State University, 1925 Coffey Road, Columbus, Ohio (USA).