This report describes six cases of feline large granular lymphocyte lymphoma identified by light microscopy on the basis of their characteristic azurophilic granulation in Giemsa-stained plastic sections and by electron microscopy on the basis of their typical granules. Although the granules of all the tumor cells were negative for peroxidase activity, they all demonstrated chloroacetate-esterase and acid phosphatase activity. All the tumors reacted with cross-reacting antibodies against the CD3 antigen (epsilon chain) and did not react with a cross-reacting monoclonal antibody directed against epitopes on cytoplasmic domains of the CD20 antigen. Three tumors had a positive reaction with a monoclonal human CD57-like antibody. This is highly suggestive of either a cytotoxic T cell or a natural killer cell origin of the neoplasias. In three cats, although other abdominal organs were affected to a variable extent, the main neoplastic lesions were localized in the gastrointestinal tract and the jejunal lymph nodes. In contrast, in the other three cats, organ involvement was more widespread, affecting the lung (two), myocardium (two), precardiac mediastinum (one), salivary gland (one), and spinal cord (one); in addition, leukemia was present in two of these cats. The data presented indicate that tumors made up of large granular lymphocytes occur more frequently in cats than previously assumed and that they share many characteristic features with specific subtypes of clonal disorders of large granular lymphocytes in humans.

Histopathological findings of renal biopsies in cats and dogs with diffuse nephropathies generally lead to an exact diagnosis and facilitate prognostic judgements. Complications following renal biopsy are usually slight, provided the biopsy is performed properly. Routine renal laboratory data have been compared with histopathological findings. High urine protein values are often the result of glomerular lesions, whereas high creatinine values are frequently related to tubulointerstitial lesions. In general, there is no relationship between different types of nephropathy and age; nevertheless animals with chronic tubulointerstitial nephritis were, on average, older than animals with glomerulopathies.

Species-specific antibodies against numerous canine and feline homologues of human leucocyte markers are commercially available. In some cases, they correspond to CD (clusters of differentiation) antibodies. 6,7 In addition, crossreactions between antibodies against human leucocyte antigens and canine or feline leucocytes have been repeatedly described. 1,9,11,12,15 The usefulness of some of these antibodies for diagnosing lymphoid tumors has already been demonstrated. 4,8,13,16,17,19 We tested numerous antibodies on feline and canine lymphatic tissue (Table 1). All investigations were carried out on paraffin-embedded sections of formalin-fixed tissue from 2 dogs and 2 cats. Paraffin-embedded sections and cryostat sections from human tonsils served as positive controls. As per data supplied by the manufacturer, antibodies against formalin-sensitive epitopes (see Table 1) and those that did not furnish unequivocal results in initial trials were further tested on cryostat sections and on paraffin-embedded sections that had been pretreated via microwave heating (antigen retrieval). Material for paraffin sections was immersion-fixed for 24 hours in 7% buffered formalin and routinely embedded in paraffin wax. Paraffin-embedded sections 3 µm thick were mounted on poly-L-lysine-coated glass slides, deparaffinized with 2 changes of xylene, and rehydrated through graded alcohol baths to distilled water. Endogenous peroxidase was removed by a 15-minute wash in 1% hydrogen peroxide followed by 3 5-minute rinses with 0.01 M phosphate-buffered saline (PBS), pH 7.2. Enzymatic pretreatment was carried out to the extent recommended by the manufacturer for antibodies A452 (CD3), A426 (anti-p-chain), A262 (anti

Canine and feline platelet cytocentrifuge preparations (CCPs), cryostat and paraffin-embedded bone marrow sections were used in this study. We evaluated whether platelets, megakaryocytes and megakaryocyte precursor cells could be labelled by monoclonal antibodies (Y2/51, CLB-thromb/1, HPL1) against human platelet membrane glycoprotein GP IIIa and the GP IIb/IIIa complex or by the following 10 biotinylated lectins: concanavalin A (Con A), Lens culinaris agglutinin (LCA), Pisum sativum agglutinin (PsA), wheat germ agglutinin (WGA), peanut agglutinin (PNA), Phaseolus vulgaris lectin (PHA-L), Ricinus communis agglutinin 120 (RCA120), Ulex europaeus agglutinin-I(UEA-1), soybean agglutinin (SBA) and Dolichos biflorus agglutinin (DBA). Monoclonal antibodies Y2/51 and HPL1 cross reacted with platelets and megakaryocytic cells from both species, whereas CLB-thromb/1 was unreactive with canine preparations. Only Y2/51 labelled megakaryocytic cells in paraffin-embedded samples. LCA, PSA, WGA and PHA-L labelled feline and canine platelets and different numbers of morphologically identifiable megakaryocytes and numerous other, mostly myeloid, cells. Immunoblots of dog and cat platelet lysates using Y2/51 visualized a single protein of 95 kDa (unreduced), a mol.wt value within the range of those reported for GP IIIa. Some of the platelet (but not necessarily megakaryocyte) glycoproteins reacting with LCA, PSA and WGA could be identified in lectin blots following one- or two (nonreduced/reduced)-dimensional sodium dodecyl sulphate-polyacrylamide gel electrophoresis (SDS-PAGE). Thus in dogs and cats, the immunohistochemical detection of GP IIIa (and eventually GP IIb/IIIa) rather than lectin binding patterns could be important for the diagnosis of megakaryoblastic leukaemias.

No abstract