Use of flow cytometry for determination of differential leukocyte counts in bovine blood

A flow cytometric method was developed to perform differential leukocyte counts on bovine blood. Blood specimens from 50 healthy Holstein cows were analyzed by use of a flow cytometer. The method entailed diluting blood with phosphate-buffered, hypotonic saline solution containing acridine orange, and performing a step-wise, 3-parameter analysis on the bases of cell size, cellular granularity, and granulocyte fluorescence. Initially, proportions of monocytes, granulocytes, and lymphocytes were determined by creating appropriate windows on dot plots of cell size (determined by forward light scatter) vs cellular granularity (determined by the logarithm of side light scatter). Eosinophils were resolved by analysis of granulocytes as dot plots of logarithms of green vs red fluorescence ascribed to acridine orange. Proportions of eosinophils and neutrophils were computed from data so generated. Microclumps of platelets spuriously affected counts of some granulocytes, particularly eosinophils. Differential leukocyte counts determined by flow cytometry generally compared favorably with those obtained by use of the conventional microscopic method, using Wright-stained blood films. Mean neutrophil and eosinophil counts determined by the 2 methods did not differ significantly, but lymphocyte counts determined by flow cytometry were significantly higher than those determined by microscopy (P less than 0.01). Correlation coefficients for counts of neutrophils, eosinophils, and lymphocytes determined by the 2 methods ranged from 0.519 to 0.833. Correlation between monocyte counts was low (r = 0.147), although mean monocyte counts determined by the 2 methods did not differ significantly.(ABSTRACT TRUNCATED AT 250 WORDS)     

Granulocyte isolation from whole blood of goat, sheep, cattle, horse, dog, pig, and man

A simple two step procedure for the isolation of caprine, ovine, bovine, equine, canine, porcine and human peripheral blood granulocytes is described. After enrichment of granulocytes by centrifugation, contaminating erythrocytes are lysed hypotonically. Recovery, purity, and viability of the granulocyte suspensions are determined. FACScan analysis of the cell suspensions measuring cellular size by forward and sideward light scatter is compared with the corresponding analysis of whole blood leukocytes. Constituencies of the isolated cell suspensions and loss of granulocyte subpopulations through isolation procedure is discussed with regard to granulocyte function assays.     

Flow cytofluorometric characterization of bovine blood and milk leukocytes

Flow cytometry and sorting proved to be a rapid method that facilitated the identification of different leukocyte populations in bovine blood and milk. After briefly incubating whole blood and milk samples in a hypotonic phosphate buffer, containing supravital acridine orange, 5 classes of leukocytes were found in the blood (lymphocytes, neutrophils, eosinophils, basophils, and monocytes) and 4 in the milk (lymphocytes, neutrophils, monocytes, and macrophages) by flow cytometry. Cells were morphologically identified by fluorescent microscopy after flow cytometric sorting and by light microscopy after Papanicolaous staining. Udder parenchymal and ductal tissue cells (secretory and epithelial cells) were not found in the milk samples evaluated. Large differences in the total and differential cell counts were found in the different milk secretions.     

Counting absolute number of lymphocytes in quail whole blood by flow cytometry

In a previous study, we reported a new method for counting quail blood cells. After quail blood cells were stained with fluorescent lipophilic dye (DiOC6(3)), absolute counts of erythrocytes, granulocytes, and monocytes were obtained by means of flow cytometry (FC). The FC method has the potential for application to avian blood cells count; however, the method was unable to distinguish between lymphocytes and thrombocytes. In the present study, we improved the FC method to obtain separate counts of lymphocytes using DiOC5(3). After quail blood cells were stained with DiOC5(3), the cells were measured with FC. Each blood cell type was distinguished by means of their typical FL-1 (green fluorescence) and SSC (side scatter). Absolute numbers of erythrocytes, granulocytes, monocytes and lymphocytes in whole blood were obtained. The improved FC analysis worked equally well with chicken (Gallus gallus) and goose (Anser cygnoides) blood.     

Validation of the Sysmex XT‐2000iV hematology system for dogs,cats, and horses. II. Differential leukocyte counts

Background: The Sysmex XT‐2000iV is a laser‐based, flow cytometric hematology system that stains nucleic acids in leukocytes with a fluorescent dye. A 4‐part differential is obtained using side fluorescence light and laser side scatter. Objective: The purpose of this study was to validate the Sysmex XT‐2000iV for determining differential leukocyte counts in blood from ill dogs, cats, and horses. Methods: Blood samples from diseased animals (133 dogs, 65 cats, and 73 horses) were analyzed with the Sysmex XT‐2000iV (Auto‐diff) and the CELL‐DYN 3500. Manual differentials were obtained by counting 100 leukocytes in Wright‐stained blood smears. Results: Leukocyte populations in the Sysmex DIFF scattergram were usually well separated in equine samples, but were not as well separated in canine and feline samples. Correlation among the Sysmex XT‐2000iV, CELL‐DYN 3500, and manual counts was excellent for neutrophil counts (r ≥.97) and good for lymphocyte counts (r ≥.87) for all three species. Systematic differences between the 3 methods were seen for lymphocyte and monocyte counts. The Sysmex reported incomplete differential counts on 18% of feline, 13% of canine, and 3% of equine samples, often when a marked left shift (>10% bands) and/or toxic neutrophils were present. Eosinophils were readily identified in cytograms from all 3 species. Neither the Sysmex nor the CELL‐DYN detected basophils in the 7 dogs and 5 cats with basophilia. Conclusions: The Sysmex XT‐2000iV automated differential leukocyte count performed well with most samples from diseased dogs, cats, and horses. Basophils were not detected. Immature neutrophils or prominent toxic changes often induced errors in samples from cats and dogs.     

Differential leukocyte counts determined in chicken blood using the Cell-Dyn 3500

Background: Automated hematology instruments commonly are used for mammalian blood analysis, but there is a lack of accurate automated methods available for avian leukocyte analysis. Objective: The aim of this study was to validate differential leukocyte counts in blood from chickens using the Cell-Dyn 3500 hematology system and avian-specific software.
Methods: Blood samples were collected in lithium-heparin tubes from 2 groups (n = 84 and n = 139) of laying hens. Manual 200-cell differential counts were done on routinely-stained blood smears, and manual total granulocyte counts (heterophils and eosinophils) were done using an eosinophil stain in a counting chamber. Automated differential counts were done using VET 2.3, a research and development version of avian-specific software for the Cell-Dyn 3500. Results were analyzed using Pearson's correlation and difference plots.
Results: Automated granulocyte counts from the Cell-Dyn were in good agreement with manual granulocyte counts ( r = 0.93 and 0.80 for the 2 study groups). No correlation was found between automated and manual lymphocyte counts. Correlation coefficients for monocyte counts were 0.70 and 0.43. Conclusion: Automated leukocyte results from the Cell-Dyn using VET 2.3 software were not fully accurate. Total granulocyte counts may be of clinical usefulness, but results obtained for other parameters were unreliable.     

Guinea pig erythrocyte rosette formation as a nonspecific cell surface receptor assay in the cat

The specificity of guinea pig erythrocyte (GPE) rosettes for feline peripheral blood lymphocytes was studied. Of the GPE rosette-positive cells from peripheral blood, 54% were monocytes, 29% were granulocytes, and only 17% were lymphocytes. Results were similar for rosettes incubated at 4 C and those incubated at 37 C. Mononuclear cells separated with polyvinylpyrrolidone-coated silica formed fewer monocyte rosettes (49%) and more granulocyte rosettes (34%) than did cells separated with sodium diatrizoate-Ficoll (60% monocyte rosettes and 18% granulocyte rosettes), whereas the percentage of lymphocyte rosettes was similar for both media. Mononuclear cells suspended in Eagle's minimum essential medium had a higher percentage of monocyte rosettes (75%) and a lower percentage of granulocyte rosettes (12%) than did cells suspended in RPMI 1640 medium (59% monocyte rosettes and 27% granulocyte rosettes). The percentage of lymphocyte rosettes was similar in the 2 media. Two sequential 45-minute plastic adherent cell depletions decreased monocyte rosettes to 51% and increased lymphocyte rosettes to 23% compared with 63% monocyte rosettes and 12% lymphocyte rosettes before adherent cell depletion. The granulocyte rosettes were unchanged by plastic adherent cell depletion. The percentage of rosette-positive cells (9%) was not significantly affected by incubation at 4 C or 37 C, cell separation with polyvinylpyrrolidone-coated silica or lymphocyte separation medium, or suspension in Eagle's minimum essential medium or RPMI 1640 medium. Plastic adherent cell depletion decreased the percentage of rosette-positive cells. Feline thymocytes were 38% to 80% GPE rosette-positive and a feline leukemia virus-infected lymphoblastic cell line (F422) was 88% GPE rosette-positive.(ABSTRACT TRUNCATED AT 250 WORDS)     

Influence of chronic caprine arthritis-encephalitis virus infection on the population of peripheral blood leukocytes

The influence of caprine arthritis-encephalitis (CAE) virus infection on the population of peripheral blood leukocytes in goats was evaluated. For this purpose two groups of adult dairy female goats were formed. The experimental group consisted of 17 goats, which had been naturally infected for many years. The control group comprised 29 non-infected goats, which originated from CAE-free herd. All goats were clinically healthy. Whole blood was collected and tested in hematological analyzer and light microscope to assess the total number of leukocytes and the percentage of four leukocyte populations--neutrophils, eosinophils, monocytes and lymphocytes. Then, flow cytometry with monoclonal antibodies against several surface antigens (namely CD14, CD2, B-B2, CD4, CD8h, TCR-N6, WC1-N2 and WC1-N3) was performed to assess the proportion of lymphocyte subpopulations. Statistically significant differences (alpha < or = 0.01) were observed only in the subpopulations of T lymphocytes--percentage of all subpopulations were significantly higher in the group of seropositive goats. No statistically significant differences were revealed with respect to the total number of blood leukocytes, the average percentage of blood leukocyte populations and proportions of both T and B lymphocytes.     

Effect of dexamethasone on bovine leukocyte functions and bovine herpesvirus type-1 replication.

In an attempt to elucidate the mechanism whereby dexamethasone could reactivate bovine herpesvirus type-1 the effect of dexamethasone on virus replication and leukocyte functions was assessed. No effect was detectable on either virus yield or in vitro replication kinetics. In contrast, dexamethasone influenced several leukocyte functions thought to be of importance in antiviral defense and maintenance of latency. In vitro exposure of peripheral blood polymorphonuclear neutrophilic granulocytes of normal animals to dexamethasone depressed their migratory and cytotoxic activities, but had no effect on Fc- and complement receptor expression. Dexamethasone also depressed lectin-induced lymphocyte proliferation and interleukin-2 generation in a dose-dependent manner. When cows were treated repeatedly with dexamethasone and their leukocytes assayed, suppression of phytohemagglutinin-induced lymphocyte proliferation, interleukin-2 generation, natural cytotoxicity of mononuclear cells and polymorphonuclear neutrophilic granulocyte functions were observed. In contrast, concanavalin A induced lymphocyte proliferation was increased following treatment.     

The variations of white blood cell count in Lipizzan horses

Total and differential leucocyte counts (lymphocytes, neutrophils, monocytes, eosinophils and basophils) were measured in 140 stallions, 101 mares and 25 foals of Lipizzan breed. The values fell in the normal ranges for warm-blooded horses. Differences between mares and stallions were not significant with the exception of foals, having higher white blood cell, neutrophil, lymphocyte, monocyte and basophil values in females than in males. Foals exhibited an age-related increase of total leucocyte count during the first 4 months of life, accompanied by a decrease in neutrophil and increase in lymphocyte and eosinophil counts. In mares and stallions, the total number of leucocytes, lymphocytes, monocytes and basophils significantly decreased but the number of neutrophils and eosinophils remained almost unchanged with age gain.     

Differences in leukocyte differentiation molecule abundances on domestic sheep (Ovis aries) and bighorn sheep (Ovis canadensis) neutrophils identified by flow cytometry

Although both domestic sheep (DS) and bighorn sheep (BHS) are affected by similar respiratory bacterial pathogens, experimental and field data indicate BHS are more susceptible to pneumonia. Cross-reactive monoclonal antibodies (mAbs) for use in flow cytometry (FC) are valuable reagents for interspecies comparative immune system analyses. This study describes cross-reactive mAbs that recognize leukocyte differentiation molecules (LDMs) and major histocompatibility complex antigens on DS and BHS leukocytes. Characterization of multichannel eosinophil autofluorescence in this study permitted cell-type specific gating of granulocytes for evaluating LDMs, specifically on neutrophils, by single-label FC. Evaluation of relative abundances of LDMs by flow cytometry revealed greater CD11a, CD11b, CD18 (β₂ integrins) and CD 172a (SIRPα) on DS neutrophils and greater CD14 (lipopolysaccharide receptor) on BHS neutrophils. Greater CD25 (IL-2) was identified on BHS lymphocytes following Concavalin A stimulation. While DS and BHS have similar total peripheral blood leukocyte counts, BHS have proportionately more neutrophils.     

Changes in peripheral blood leukocyte subpopulations in piglets co-infected experimentally with porcine reproductive and respiratory syndrome virus and porcine circovirus type 2

Porcine reproductive and respiratory syndrome virus (PRRSV) and porcine circovirus type 2 (PCV2) are pathogens, which can significantly affect the swine industry worldwide. Field surveys suggest that simultaneous PRRSV and PCV2 infection is common in pigs. The objective of this study was to measure the changes in peripheral blood leukocyte subpopulations in piglets co-infected experimentally with PRRSV and PCV2, in order to analyze the synergistic influence of co-infection on the immune system. Changes in peripheral blood leukocyte subpopulations were systematically measured by flow cytometry (FCM). The levels of antibodies to PRRSV and PCV2 were detected by indirect Enzyme-Linked ImmunoSorbent Assay (ELISA) and the indirect fluorescent antibody test (IFA), respectively. Serum viral loads were measured using real-time PCR. The results showed that piglets co-infected with PRRSV and PCV2 exhibited slower generation and lower levels of antibodies to PRRSV and PCV2, and increased amounts and a prolonged presence of both PRRSV and PCV2 in serum, in comparison to the piglets infected with either virus alone. The major finding in our study was that the total and differential leukocyte counts, including white blood cells (WBCs), monocytes, granulocytes and lymphocytes (T, B and NK cells, as well as T-cell subpopulations), dramatically decreased early during co-infection with PRRSV and PCV2 for about two weeks, in contrast with animals singly infected with either PRRSV or PCV2. These results suggest that PRRSV and PCV2 co-infection results in a synergistic decrease in immune cells in the peripheral blood of piglets. These data contribute to the understanding of the immunosuppressive effects resulting from PRRSV and PCV2 co-infection in pigs.     

Longitudinal evaluation of bovine mammary gland health status by somatic cell counting, flow cytometry, and cytology.

Bovine mastitis phases induced by Staphylococcus aureus were assessed in 6 lactating cows before challenge and at 1, 4-8, and 9-14 days postinoculation (dpi). Milk lymphocytes, macrophages, and polymorphonuclear cells (PMN) were counted by conventional (manual) cytology, identified by CD3+ and CD11b+ immunofluorescence and counted by flow cytometry (based on leukocyte forward and side light scatter values). Somatic cell counts (SCC) and recovery of bacteria were recorded at the same times. Preinoculation samples showed a lymphocyte-dominated composition. At 1 dpi, the percentage of PMN increased and that of lymphocytes decreased. At 4-8 dpi, PMN were predominant, but the percentage of mononuclear cells increased above that at 1 dpi and further increased by 9-14 dpi (when lymphocytes approached prechallenge values). Based on leukocyte percentages, 3 indices were created from the data: 1) the PMN/lymphocyte percentage ratio (PMN/L), 2) the PMN/macrophage percentage ratio (PMN/M), and 3) the phagocyte (PMN and macrophage)/lymphocyte percentage ratio (Phago/L). Significant correlations were found between cytologic and flow cytometric data in all of these indicators (all with P < or = 0.01). These indices identified nonmastitic, early inflammatory (1-8 dpi), and late inflammatory (9-14 dpi) animals. In contrast, SCC and bacteriology did not. Although sensitivity of the SCC was similar to that of Phago/L, the specificity of SCC was almost half that of the Phago/L index. Based on flow cytometry indicators, an algorithm for presumptive diagnosis of bovine mastitis was developed. Flow cytometry provides results as valid as those obtained by conventional (manual) cytology, shows greater ability to identify mastitic cases than does SCC, and may identify 3 mammary gland health-related conditions.     

Changes in peripheral blood leukocyte populations in pigs with naturally occurring exudative epidermitis

The objective of the study was to analyze changes in peripheral blood leukocyte subsets in cases of naturally occurring exudative epidermitis (EE) in pigs. Five of ten piglets developed the chronic clinical form of EE 2-5 days after weaning (PW). Blood samples were obtained at 7, 14 and 21 days from both normal and clinically affected piglets for routine haematology and for the determination of CD45, CD21, CD4, CD8 and gammadeltaTCR cell markers by flow cytometry. When compared with clinically normal piglets EE affected pigs showed significantly decreased values of monocytes at 14 and 21 days PW, and increased numbers of neutrophils and leukocytes at 21 days PW. The EE affected pigs also had an early significant CD4(+) and CD8(high+) T lymphocyte proliferative response at 7 days PW. However affected pigs had a significantly reduced number of B (CD21(+)) and gammadeltaTCR(+) T lymphocytes in blood at 21 days PW. Although all values remained within the normal range, the significant differences in some peripheral blood leukocyte subsets between the two groups of piglets suggest that the generalised cutaneous infection with Staphylococcus hyicus is severe enough to induce a systemic inflammatory and immune responses.     

Changes in peripheral blood leukocyte populations in pigs with natural postweaning multisystemic wasting syndrome (PMWS)

The objective of the present study was to analyze, by flow cytometry, changes in PBMC subsets in pigs having postweaning multisystemic wasting syndrome (PMWS), a new condition associated to porcine circovirus type 2 (PCV2) infection. Thirteen acutely PMWS affected pigs were selected from a farm seronegative to porcine reproductive and respiratory syndrome virus (PRRSV) and to Aujeszky's disease virus (ADV); 11 clinically healthy pigs were selected from a high health farm with no history of PMWS and free of the major swine pathogens, and used as a control group. All pigs were necropsied, and tissue samples were fixed in formalin; blood with EDTA anticoagulant was used to perform the flow cytometric analysis. PBMC were incubated with mAb against porcine CD3, CD4, CD8, CD25, CD45, IgM, SWC3, and SLA-Class II. Flow cytometric analysis showed substantial changes in leukocyte subsets in the peripheral blood of PMWS-affected pigs, which were characterized by an increase of monocytes, a reduction of T (mainly CD4(+)) and B-lymphocytes, and the presence of low-density immature granulocytes. Altogether, these changes would suggest an inability of acutely PMWS-affected pigs to mount an effective immune response.     

猪繁殖与呼吸综合征病毒实验感染SPF猪外周血单核细胞和肺泡巨噬细胞早期凋亡细胞的观察

采用Ａｎｎｅｘｉｎ Ｖ－ＦＩＴＣ／ＰＩ双染色法,用流式细胞仪检测了猪繁殖与呼吸综合征病毒（ＰＲＲＳＶ）实验感染ＳＰＦ猪不同时期外周血单核细胞和肺泡巨噬细胞感染Ａｎｎｅｘｉｎ Ｖ－ＦＩＴＣ＾＋／ＰＩ＾－细胞群（早期凋亡细胞群）。结果显示,ＰＲＲＳＶ感染猪外周血单核细胞和肺泡巨噬细胞Ａｎｎｅｘｉｎ Ｖ－ＦＩＴＣ＾＋／ＰＩ＾－细胞群的表达率均明显高于正常对照猪,感染后２４ｈ表达率达最高值。     

Pure white cell aplasia in a dog

A 3-year-old Irish Wolfhound was evaluated because of acute onset of lethargy and fever. Severe neutropenia (0/microL; reference interval 2500-11,200/microL) was associated with granulocyte aplasia in the bone marrow (myeloid:erythroid ratio, 0.009:1). Antineutrophil antibodies were assessed by an indirect immunofluorescence assay using flow cytometry. When normal canine leukocytes were incubated with the patient's serum and anti-IgG, a marked shift was observed in the forward-angle light scatter of the neutrophil population, and the monocyte cluster disappeared, possibly the result of fragmentation or lysis. Both neutrophil fluorescence intensity (309 +/- 11 median channel units [MCU], control values 107-152 MCU) and the percentage of neutrophils with increased fluorescence intensity (61 +/- 5%, control values 3.8-13.7%) were increased in the patient's serum, consistent with the presence of antineutrophil antibodies. Repeated episodes of neutropenia occurred while treatment with steroidal and nonsteroidal immunosuppressive therapy was initiated and modified. The neutrophil count eventually stabilized in the low-normal range, and the dog was maintained for the next 15 months on prednisone (0.4 mg/kg PO q 48 h) and azathioprine (2 mg/kg daily). During this period, the dog developed immune-mediated hemolytic anemia and thrombocytopenia, decubital ulcers, nasal aspergillosis, and eventually, multi-organ septicemia, which led to euthanasia on day 784. A diagnosis of pure white cell aplasia was made in this dog, based on the many similarities to human patients with pure white cell aplasia, including severe neutropenia with selective granulocyte aplasia, serum antineutrophil antibodies, remission dependent on treatment with immunosuppressive therapy, and recurrent bacterial infections.     

Depression of lymphocyte reactivity by granulocytes in equine whole blood culture

Tritiated thymidine uptake in response to Concanavalin A was recorded as stimulated counts per minute (SCPM) in equine whole blood cultured with an optimum concentration of 3 X 10(4) mononuclear cells/culture. A significant negative correlation was found between log10 SCPM and granulocyte level in culture, in a group of 27 adult horses (r = -0.745, P less than 0.001). Addition of isolated autologous granulocytes to such cultures resulted in a reduction of log10 SCPM of a magnitude similar to that predicted by the gradient of the log10 SCPM/granulocyte level regression line in the group of adult horses. It is concluded that the presence of granulocytes in culture directly depresses lymphocyte reactivity, and that a large part of the variability of SCPM between horses is due to variation of granulocyte levels.     

Interaction between Salmonella typhimurium and phagocytic cells in pigs. Phagocytosis, oxidative burst and killing in polymorphonuclear leukocytes and monocytes

Interactions between Salmonella typhimurium and peripheral blood leucocytes from healthy, Salmonella-free pigs were investigated in vitro. Both granulocytes and monocytes phagocytized FITC-labelled heat-killed Salmonella bacteria as shown by flow cytometry. Phagocytosis in whole blood and isolated leucocytes was measured as acquired fluorescence in the leukocytes and was both time and dose related. Living, serum-opsonized Salmonella bacteria induced a dose-dependent oxidative burst in PMNs and monocytes as measured by luminol-enhanced chemiluminescence (LC). When opsonized in normal serum the Salmonella bacteria, in the range of 2-5 x 10(7) cfu, induced a LC response in monocytes comparable to the level of responses induced by phorbol myristate acetate (PMA) and opsonized zymosan, and the Salmonella-induced response was only marginally reduced by superoxide dismutase (SOD). Intracellular killing of Salmonella by monocytes was assessed from plate colony counts of lysed monocytes and showed that Salmonella typhimurium was able to survive and proliferate in adherent monocytes in vitro despite a reduction in intracellular cfu during the first hour's incubation in cells from some pigs. Experiments with the exhaustion of oxidative burst in non-adherent monocytes were performed by prestimulation with PMA, heat-killed Salmonella or buffer. Prestimulation with PMA led to a strong reduction in oxidative burst induced by living opsonized Salmonella bacteria, whereas prestimulation with heat-killed bacteria gave rise to an enhanced response. In these experiments intracellular killing of the added living Salmonella gave variable results, in that monocytes from two out of three pigs showed no essential change in intracellular bactericidal activity, but with cells from one pig a less pronounced bactericidal activity was found after prestimulation with PMA.     

The use of cytochemistry,immunophenotyping, flow cytometry,and in vitro differentiation to determine the ontogeny of a canine monoblastic leukemia

We evaluated the utility of cytochemistry, immunophenotyping, flow cytometry, and in vitro culture with forced differentiation of leukemic cells as diagnostic aids to identify the malignant cell ontogeny in a dog with leukemia. A tentative diagnosis of monoblastic leukemia was established by microscopic examination of Romanowsky-stained blood smears and bone marrow aspirate smears. This diagnosis also was supported by the light scatter signature that identified the blast cells as large, non-granular monocytic cells using a CellDyn 3500 automated hematology analyzer; as well as by the detection of N-butyrate esterase and the lack of choloroacetate esterase or leukocyte peroxidase by cytochemical staining. Subsequently, leukemic cells were isolated from the dog's peripheral blood and placed into tissue culture or cryopreserved. The leukemic cells grew in suspension cultures and proliferated spontaneously for up to 4 days. By day 7, proliferation was negligible. Upon culture with conditioned supernatant using mitogen-stimulated human T cells as a source of cytokines, an increased proportion of cells entered S phase by day 2 of culture; however, proliferation declined markedly by day 4, at which time the cells had apparently differentiated to adherent, vacuolated macrophages. The cytokine-stimulated leukemic cells were positive for the monocyte/macrophage specific markers alpha-1-antitrypsin, alpha-1-antichymotrypsin, lysozyme, CD14, MHC class II, and calprotectin, an antigen found in differentiated macrophages and granulocytes. Despite the strong tendency of the leukemic cells towards monocytic differentiation, our results suggested that they retained some features of a myelomonocytic precursor. These data show that cytochemistry, immunophenotyping, flow cytometry, and in vitro differentiation of canine leukemia cells are useful tools for confirming the lineage of malignant hematopoietic cells.