Y. Aida

Molecular structures of cattle T-cell receptor gamma and delta chains predominantly expressed on peripheral blood lymphocytes.

T lymphocytes express T-cell receptors (Tcr) composed of either o~/~ (Tcr-a/b) or 3/5 (Tcr-g/d) heterodimers in association with CD3 complexes. Both Tcrs share the same immunoglobulin-like genetic structure, and have been shown to be generated by somatic rearrangement of variable (V), diversity (D), joining (J), and constant (C) gene segments, except that Tcra and Tcrg genes apparently lack the D segments (Kronenberg et al. 1986; Toyonaga and Mak 1987; Raulet 1989). The Tcr-a/b chains are found on the majority of peripheral blood T cells and appear t ° be responsible for the recognition of antigens in the context of cell surface proteins encoded by class I and class II genes of the major histocompatibility complex (MHC). Chains of the other heterodimer, Tcr-g/d, are expressed on a small fraction of peripheral blood and thymic T cells, although they are expressed on the majority of T cells in certain epithelial locations such as gut and skin (Groh et al. 1989; Goodman et al. 1988; Casorati et al. 1989; Itohara et al. 1990). In a striking contrast to humans and mice, in which T cells bearing Tcr-g/d constitute a minor population (1%-5%) in the peripheral T cells, ruminants usually possess a large population (15%-30%) of Tcr-g/d T cells in blood-borne T cells (Clevers et al. 1990; MacKay and Hein 1989; Hein et al. 1990b; Hein and MacKay 1991). In a previous study (Takeuchi et al. 1992), we isolated and characterized the molecular structure of cattle Tcrg and Tcrd genes; however, functional and fulllength cDNA clones were not isolated from )~gtl0 li-

B-1a, B-1b and conventional B cell lymphoma from enzootic bovine leukosis

In order to characterize the phenotypes of tumor cells and to clarify from which B cell lineage the lymphomas were derived, ten cows with enzootic bovine leukosis were examined by means of immunohistologic staining and flow cytometry. The tumor cells expressed mainly major histocompatibility complex (MHC) class II+ (10/10), BoCD11b+ (9/10), IgG1+ (8/10), B-B2+ (8/10) BoCD5+ (7/10), and lambda light chain+ (7/10). Tumor cells from only one animal expressed sIgM+ (1/10). Tumor cells from all ten animals were negative for IgG2, BoCD3, BoCD4, BoCD8, WC1-N2, and IL-2R alpha. The phenotypes of these tumor cells were all slightly different, suggesting that bovine leukemia virus (BLV)-induced lymphoma expresses phenotypic diversity. Moreover, tumor cells from seven cattle coexpressed BoCD5 and BoCD11b (B-1a cells). On the other hand, tumor cells from two of them only expressed BoCD11b (B-1b cells), and those from one were negative for both BoCD5 and BoCD11b (conventional B cells). Therefore, we concluded that BLV-induced lymphoma cells can be derived from B-1a, B-1b and conventional B cells.

Phenotypic Analysis of Neoplastic Cells from Calf, Thymic, and Intermediate Forms of Bovine Leukosis

Immunohistochemistry and flow cytometric analysis were used with monoclonal antibodies to examine the phenotype of neoplastic cells from cattle with sporadic bovine leukosis (three cases of calf form, two cases of thymic form, and three cases of intermediate form). Three cases of calf form and two cases of intermediate form were positive for B cell lineage in immunohistologic examination and in flow cytometric analysis for B-B2+, sIgM+, and major histocompatibility class II+. Two cases of thymic form and one case of intermediate form were CD2+, CD5+, CD6+, and CD8+ in immunohistologic examination and in flow cytometric analysis. The results show that neoplastic cells develop from B and T cell lineages in sporadic bovine leukosis.

CATTLE CDNA CLONE ENCODING A NEW ALLELE OF THE MHC CLASS II DQA1 GENE

A cDNA clone encoding the cattle MHC (BoLA) class II DQ c~ chain was isolated and sequenced. A cDNA library from the cattle lymphoid cell line BLSC-KU-1 was constructed in the expression vector pCDM8 (Aida et al. 1994) and the plasmid DNA prepared from the cDNA library was amplified by polymerase chain reaction (PCR) with two oligonucleotide primers, forward (5-CCGAATTCTGAGACCACCTTGAGAACAGG-3) and reverse (5CCTCTAGAACCTTCCTTCTGGAGTGTGAC-Y). These primers correspond to the 5-flanking sequence of the sheep DQA1.2 cDNA clone (GenBank AC M93430) and the 3flanking sequence of the cattle partial genomic DQA1 clone W1 (van der Poel et al. 1990) and the cattle partial DQA1.2 cDNA clone ~5 (Xu et al. 1993). The underlined sequences show artificial enzyme sites. PCR products were isolated and subcloned into pBluescript II SK (+) (Stratagene, Heidelberg, Germany), and the nucleotide sequences of both strands were determined with the BcaBEST TM sequencing kit (Takara Shuzo, Shiga, Japan). Sequences of a BoLA-DQA cDNA clone NQ1 are 76~3 nucleotides long, with one open reading frame encoding a polypeptide of 255 amino acids, 23 of which were deduced to form a signal peptide and 232 to form the mature polypeptide (Fig. 1). In addition, we obtained another clone which is identical to the NQ1 sequence. The NQ1encoded proteins exhibited the highest degree of overall identity with the previously published DQA1 sequence W1 (98.2% protein and 99.3% of nucleotide identities) and c~5 (90.0% protein and 92.0% of nucleotide identifies) and only

Predominant p53 mutations in enzootic bovine leukemic cell lines

The role of the p53 tumor suppressor gene in bovine lymphosarcomas, a fragment of about 100 bp corresponding to approximately 97% of the open reading frame of the p53 gene was first amplified from single-strand cDNA originated from calf thymus by polymerase chain reaction PCR) and sequenced to obtain the bovine wild-type p53 gene. At the amino acid level, the omologies of the bovine p53 gene with the human, mouse, chicken and cat p53 genes were 0.9%, 72.8%, 52.7% and 82.3%, respectively. Moreover, eight bovine leukemic cells lines were studied for alterations in the p53 gene. These lines showed no significant somatic alterations in southern blot analysis, and expressed 2.5 kb p53-specific transcripts in Northern blot analysis. In mutation analysis using the reverse transcriptase-PCR technique, we detected three missense point mutations in four of these bovine leukemic cell lines. These mutations occurred in the hotspots of the p53 gene. Thus p53 mutations predominantly occur in BLV-transformed cell lines and seem to be necessary for development of enzootic bovine leukosis (EBL).

FURTHER ANALYSIS OF THE PHENOTYPE AND DISTRIBUTION OF TUMOR CELLS IN SPORADIC B-CELL AND T-CELL LYMPHOMAS IN THE LYMPH NODE AND SPLEEN OF CATTLE

Immunohistologic studies were performed to identify the phenotype and distribution of neoplastic lymphocytes in the spleens of BLV-negative animals examined by PCR and diagnosed as having sporadic bovine leukosis. Tumor cells from three cases of sporadic bovine leukosis were identified as of B-cell lineage. Tumor cells from three additional cattle were identified as CD3+ CD4- CD8+, CD3+ CD4- CD8-, and CD3+ CD4- WC1+, respectively. The last case was diagnosed as a gamma/delta T-cell lymphoma. Differences in morphology proliferative characteristics were recognized between B- and T-cell type lymphomas. The tumor cells in B-cell type lymphoma were characterized as follows: medium or large in size, round or polymorphic nucleus with rough chromatin with some tumor cells containing a convoluted nucleus. These tumor cells of B-cell type lymphoma were present in the red pulp and periarteriolar lymphoid sheath. Tumor cells of the T-cell type lymphoma were uniformly smaller than B-cell type and present around arteries or replaced red pulp of the spleen.

Immunohistologic Studies on Subpopulations of Lymphocytes in Cattle with Enzootic Bovine Leukosis

The distribution of subpopulations of lymphocytes in lymph nodes and tumors from cattle with enzootic bovine leukosis (EBL) was examined by immunohistochemistry using a panel of monclonal antibodies against leukocyte differentiation molecules of EBL. The lesions in lymph nodes could be divided into three types based on the extent of infiltration and proliferation of neoplastic cells with provirus and differential expression of leukocyte differentiation molecules. The number of B-B2+, sIgM+ cells was reduced in frequency in follicles during the neoplastic cell proliferation. CD4- and CD8-positive α/β T cells and γ/δ T cells positive for WC1 (workshop cluster designation) were also reduced in frequency in areas infiltrated with neoplastic cells. Almost all neoplastic cells were B-B2- and IgM-positive. However, there were a few B-B2- and/or IgM-negative cells or cells stained faintly in all cases. WC1+ cells were not observed in tumor tissues. However, CD4+ and CD8+ cells were observed throughout tumor tissues, suggesting a role for these cells in tumor immunity.

Antitumor Effect of Diphtheria Toxin A-Chain Gene-containing Cationic Liposomes Conjugated with Monoclonal Antibody Directed to Tumor-associated Antigen of Bovine Leukemia Cells

Monoclonal antibody c143 against tumor‐associated antigen (TAA) expressed on bovine leukemia cells was conjugated to cationic liposomes carrying a plasmid pLTR‐DT which contained a gene for diphtheria toxin A‐chain (DT‐A) under the control of the long terminal repeat (LTR) of bovine leukemia virus (BLV) in the multicloning site of pUC‐18. The specificity and antitumor effects of the conjugates were examined in vitro and in vivo using TAA‐positive bovine B‐cell lymphoma line as the target tumor. In vitro studies with the TAA‐positive cell line indicated that luciferase genecontaining cationic liposomes associated with the c143 anti‐TAA monoclonal antibody caused about 2‐fold increase in luciferase activity compared with cationic liposomes having no antibody, and also that the c143‐conjugated cationic liposomes containing pLTR‐DT exerted selective growth‐inhibitory effects on the TAA‐positive B‐cell line. Three injections of pLTR‐DT‐containing cationic liposomes coupled with c143 into tumor‐bearing nude mice resulted in significant inhibition of the tumor growth. The antitumor potency of the c143‐conjugated cationic liposomes containing pLTR‐DT was far greater than that of normal mouse IgG‐coupled cationic liposomes containing pLTR‐DT as assessed in terms of tumor size. These results suggest that cationic liposomes bearing c143 are an efficient transfection reagent for BLV‐infected B‐cell lymphoma cells, and that the delivery of the pLTR‐DT gene into BLV‐infected B‐cells by the use of such liposomes may become a useful technique for gene therapy of bovine leukosis.

Antigenic regions defined by monoclonal antibodies on tumor-associated antigens of bovine leukemia virus-induced lymphosarcoma cells

Tumor-associated antigens (TAAs) expressed on enzootic bovine leukosis tumors were divided previously into three types by use of 13 monoclonal antibodies (MAbs): common TAA, partially common TAA and individually distinct TAA. Since MAb-defined epitopes on the common TAA were conserved on both soluble TAA prepared from bovine B-lymphoma cells and untreated viable same cells, all the MAbs that bound to the soluble TAA also bound to untreated viable cells. By contrast, MAb-defined epitopes on the partially common and individually distinct TAAs varied according to the test systems used. Two of seven MAbs were found to bind to both the soluble TAA and viable cells and one MAb bound to the soluble TAA but not to the viable cells.

The γδ T Cell Population in Sheep Experimentally Infected with Bovine Leukemia Virus

Bovine leukemia virus (BLV) is an exogenous retrovirus and associated with enzootic bovine leukosis (EBL)4 which often includes persistent lymphocytosis and development of B-cell lymphosarcoma/leukemia after a long latent period. Natural cases of malignant lymphomas are rare in sheep. Under experimental conditions, however, sheep are easily infected with BLV7 and develop B-cell lymphosarcoma/leukemia at a higher frequency and within a shorter period (1 to 3 years) than cattle. Sheep also show serologic and pathologic responses to BLV similar to those in The lymphoid systems of sheep and cattle contain a large number of the CD4CD8“double-negative” T lymphocyte s~bpopu la t ion~-~ in striking contrast to the lymphoid systems of human and mouse. These double-negative lymphocytes are known to express the y6 chains of T cell receptor (TCR)2.3 as well as a surface molecule termed T19.6 In spleen and lymph nodes, y6 T cells are localized in regions of cellular traffic5 and they exhibit cytotoxic activity against allogenic cells. In order to elucidate the role of y6 T cells in BLVinfected sheep, we examined, by flow cytometry, the population of y6 T cells in the peripheral blood mononuclear cells (PBMC). The distribution of these cells in the superficial cervical lymph nodes of BLV-infected sheep was examined by immunohistology using the monoclonal antibody (MoAb) against the y6 TCR. Fourteen 6-month-old sheep (Suffolk, Corridale and crossbreed), negative for BLV-specific antibodies, were hypodermally inoculated with 1.0 ml of blood from BLV-infected cattle. BLV-infected sheep were classified according to clinical and hematological observations, and the host genome integration pattern of the BLV provirus in PBMC or lymph nodes. For histological observation, lymph node sections were obtained from two BLV-infected but healthy sheep and five sheep with lymphosarcoma. For flow cytometry, blood samples were obtained from seven BLV-infected but healthy sheep and two sheep with lymphosarcoma. Seven sheep (Corridale and Suffolk) were kept for 3 to 4 years without inoculation, two were sacrificed as controls for immunohistological observation, and five were used for flow cytometry. Immunodiffusion test was performed for detection of antibodies against BLV, using the BLV env glycoprotein gp60 and the gag protein p24. Monoclonal antibodies 86D (anti+ TCR) and 197 (antiT I9 antigen) were kindly provided by Dr. c . R. Mackay (Basel Institute for Immunology, Basel, Switzerland). PIg45A MoAb (anti-IgM) was purchased from Veterinary Medical Research and Development Inc. (Pullman, WA). Pieces approximately 1 cm thick of each superficial cervical lymph node were frozen rapidly by immersion in liquid nitrogen and preserved at -80 C. Cryosections were stained by the avidin biotin peroxidase complex technique in accordance with the manufacturer’s instructions (Vector Laboratories, Burlingame, CA) with either 86D MoAb or normal mouse serum as the negative control. Each section was observed by microscope, using an apchromatic 40 x objective and widefield 10 x ocular lens (Olympus BHS, Tokyo). The number of the 86D MoAb-positive cells was counted inside photographic field (1 80 x 120 pm square). Five different fields were randomly examined in each section. The percentage of 86D MoAb-positive cells out oftotal cells analyzed was determined. A mean value and the standard deviation were calculated. Differences in the number of positive cells of lymphocytes among the three groups of sheep were evaluated by the Mann-Whitney U-test for differences between frequency distributions. PBMC from BLV-infected sheep were examined by flow cytometry for reactivities with 86D, 197 and pIg-45A MoAbs (Table 1). Increases in both the proportion and the actual number of surface immunoglobulin M (sIgM)-bearing cells were observed in PBMC of all BLV-infected sheep. The proportion of sIgM-positive cells in PBMC from BLV-free normal sheep was approximately 23.3%, while that in the BLVinfected but healthy sheep was 55.6%. The proportion was much higher in the case of PBMC from sheep with lymphosarcoma (88.0% and 99.7%). In contrast, the proportion and the actual number of y6 T cells and those of T19-bearing cells in most BLV-infected but healthy sheep were almost the same as those in BLV-free normal sheep. Although two sheep with lymphosarcoma had considerably lower proportions of y6 T cells and T 19-bearing cells than BLV-free normal sheep, the actual numbers of these cells were not significantly different from those in BLV-free normal sheep (P < 0.05). In order to determine the distribution of76 T cells in lymph nodes, we stained cryosections of superficial cervical lymph node with the 86D MoAb (Figs. 1-4). In two BLV-free normal sheep, y6 T cells were prominent in the marginal sinus of the cortex and were scattered in the perifollicular area in the cortex, but were not in the follicles (Fig. 1). The proportions of y6 T cells determined in the defined area of the