Complement-fixation tests with cell lines derived from Burkitt's lymphoma and acute leukemias

Armstrong, Donald (Children's Hospital of Philadelphia, Philadelphia, Pa.), Gertrude Henle, and Werner Henle. Complement-fixation tests with cell lines derived from Burkitt's lymphoma and acute leukemias. J. Bacteriol. 91:1257-1262. 1966.-Cells of various lines isolated from Burkitt's lymphomas and acute leukemias and disintegrated by freezing and thawing or sonic treatment were found to react in complement-fixation tests with a considerable proportion of human sera. At least 10(7) cells per milliliter were required for antigenic activity. All but one of 13 sera from Burkitt lymphoma patients were positive, with titers ranging from 1:8 to 1:320. About 20% of sera from American children and 60% of sera from adults, regardless of diagnosis, showed titers in a similar range. Sera giving positive tests with one of the neoplastic white cell antigens usually reacted also with many if not all of the others, but rarely with antigens derived from normal peripheral leukocyte cultures and not at all with HeLa or other human nonleukocytic cells. Various observations indicate that the complement-fixation test measures mainly antigens which are different from those detected by immunofluorescence. The nature of the reactions described remains obscure.     

Human leukemia-associated antigens: detection on cells of established lymphoblastoid lines.

Primate and rabbit antisera to different morphologic classes of human leukemia cells, after appropriate absorptions, detected leukemia-associated antigens present on cultured lymphoblastoid cell lines derived from leukemia patients. The primate antisera distinguished antigens on cells derived from myeloid leukemia patients from those on cells derived from lymphocytic leukemia patients. Of particular interest was the fact that antigens of myeloid leukemia, but not of lymphatic leukemia, were detected on lymphoid cell lines established from blood of patients with myeloid leukemia. One of four lymphoblastoid cell lines derived from normal donors expressed antigens of lymphatic leukemia. Leukemia-associated antigens were not found on the HRIK lymphoblastoid line derived from a Burkitt's lymphoma patient on skin fibroblasts or HeLa cells. Expression of these antigens on cultured cells derived from leukemia patients could not be related to the presence of the EB virus or the EB virus genome. Rabbit antisera detected antigens common to cells from patients with myeloid and lymphocytic leukemia. Absorption experiments demonstrated that the antigens detected on cell lines derived from leukemia patients are similar to those detected by the primate and rabbit antisera on fresh peripheral blood leukemic cells. The serologic detection of leukemia-associated antigens on lymphoblastoid cell lines indicates that some of these cultures contain cells with antigenic properties similar to those of human peripheral blood leukemic cells.     

Burkitt's lymphoma. Distinction of subgroups by morphometric analysis of the characteristics of 55 cell lines

In an attempt to clarify the controversy about the distinction between Burkitt's and non-Burkitt's small noncleaved lymphomas, 55 cell lines derived from 48 Burkitt's lymphoma patients were characterized by morphometry on plastic-embedded sections. The results of the measurements permitted the identification of five main cytologic types, with regard to nuclear size, nuclear area dispersion and irregularity of nuclear profiles. The presence of the Epstein-Barr virus (EBV) and the geographic origin of the tumors seemed to play essential roles in the determination of nuclear size, with a significantly larger size seen in EBV-positive cell lines, and especially in the African lines among these. Immunoglobulin profile and monoclonal antibody expression also correlated with the nuclear size. Two conclusions may be drawn from this analysis. An in vivo transformation of the cells of Burkitt's lymphoma can be postulated to explain the wide morphologic spectrum of lymphomas presenting a rearrangement of chromosome 8. The fact that typical and atypical Burkitt's lymphomas cannot be differentiated by study of their derived cell lines raises the question as to the validity of the distinction between the two subtypes of small noncleaved lymphomas.     

Immunotoxin-mediated elimination of clonogenic tumor cells in the presence of human bone marrow

An immunotoxin was synthesized with pokeweed antiviral protein and an IgG1 monoclonal antibody directed against human B and pre-B cells. The B43 murine monoclonal antibody does not react with normal human bone marrow precursor cells. The immunotoxin bound to all Burkitt's lymphoma cell lines that were tested but not to human peripheral blood T cells. The ability of antibody-toxin conjugate to inhibit human lymphoblast cell lines was checked in a clonogenic assay system. The immunotoxin in the presence of chloroquine elicited 5.8 logs of killing of Burkitt's lymphoma cells (B-ALL). The efficient inhibition of clonogenic growth of B-ALL cells was not affected by the presence of normal bone marrow cells. The immunotoxin was not very toxic to pluripotent stem cells; less than 50% of the stem cells were lost under conditions where 5.6 logs of clonogenic lymphoma cells were eliminated from a 100-fold excess of normal marrow cells. Further, when assayed by long-term human bone marrow cultures, immunotoxin treatment did not result in a significant loss of pluripotent precursor cells.     

Biological properties and viral surface antigens of Burkitt lymphoma- and mononucleosis- derived strains of Epstein-Barr virus released from transformed marmoset cells.

Three strains of Epstein-Barr virus (EBV), two from Burkitt lymphoma (BL) and one from infectious mononucleosis (IM) were used to transform separate cultures of the same batch of primary marmoset leukocytes, and the viruses released from the transformants were compared. The three viruses shared properties of the transforming biotype of EBV, namely, stimulation of DNA synthesis and immortalization of cord blood leukocytes, and failure to induce "early antigen" in lymphoblast lines. All viruses produced more virus in transformed marmoset cells than in transformed human cells, as measured by the number of EBV genomes detected by complementary RNA/DNA hybridization, by virus capsid antigen expression, or by released virions and biologically active virus. Reference human sera and sera from primary EBV infections were used to compare the three virus strains in a virus neutralization test based on inhibition of stimulation of DNA synthesis. Specimens taken late in convalescence from patients with mononucleosis and sera from marmosets experimentally infected with virus from a patient with mononucleosis neutralized the homologous virus, as well as the two virus strains isolated from patients with BL. This finding indicates that viral antigens that elicit neutralizing antibodies are shared among the strains. However, in certain sera the neutralizing-antibody titer against one strain was consistently higher than against another strain. Furthermore, sera taken early after onset of IM contained low levels of neutralizing antibody against IM-derived virus, but failed to neutralize BL-derived virus. These latter findings suggest the existence of heterogeneity among surface antigens of EBVs. The results emphasize the biological and antigenic similarity of EBV isolates from BL and IM and do not suggest major subtype variations. It remains to be determined whether antigenic diversity such as described or virus genome variation detectable by other means is epidemiologically significant.     

Identification of Epstein-Barr virus terminal protein 1 (TP1) in extracts of four lymphoid cell lines, expression in insect cells, and detection of antibodies in human sera.

The terminal proteins TP1 and TP2 are putative products of Epstein-Barr virus (EBV) genes expressed during the latent cycle of the virus. They are predicted to code for 53- and 40-kilodalton integral membrane proteins. We used the baculovirus Autographa californica nuclear polyhedrosis virus as an expression vector to produce TP1 in large amounts in insect cells. The DNA sequences used to express TP1 originated from a TP1 cDNA derived from an M-ABA/CBL1 cDNA library. Rabbit antisera raised against procaryotic TP1 fusion proteins recognized a monomer and a dimer of the recombinant TP1 protein in the infected insect cells. Immunofluorescence studies of living insect cells showed that the recombinant protein is located in the plasma membrane. The insect cells infected with the recombinant baculovirus producing TP1 provided a test system to screen human antisera for TP1 antibodies. A total of 168 human EBV-positive and EBV-negative antisera were studied. TP1 antibodies were detected only in sera from nasopharyngeal carcinoma patients (16 out of 42). Rabbit antiserum raised against the recombinant TP1 protein expressed in the baculovirus system specifically recognized a protein of about 54 kilodaltons in the lymphoblastoid cell lines M-ABA and M-ABA/CBL1 and in the Burkitt's lymphoma cell lines BL18 and BL72. This protein could be located in the total membrane fraction of M-ABA cells and is upregulated by treating the cells with 12-O-tetradecanoylphorbol-13-acetate.     

Production and characterization of a monoclonal antibody that binds Reed-Sternberg cells

A murine monoclonal antibody, termed HeFi-1, was produced after immunization with the L428 Hodgkin's disease tissue culture cell line. HeFi-1 selectively stained only the Reed-Sternberg or Hodgkin's cells in 18 of 18 cases of Hodgkin's disease, including the nodular sclerosis, mixed cellularity, and lymphocyte-depleted histologic subtypes. HeFi-1 did not stain any cells in normal lung, brain, salivary gland, thyroid, gall bladder, pancreas, liver, testis, breast, endometrium, or kidney. Rare large cells at the edge of the lymphoid follicles were stained in normal tonsil, colon, and hyperplastic thymus. There was no staining of any cells in 14 cases of B cell non-Hodgkin's lymphoma; however, the malignant cells in three of 11 cases of non-Hodgkin's lymphoma which appeared to express T cell markers were also stained with HeFi-1. Tissue culture cell lines including the T cell acute lymphocytic leukemia lines MOLT4 and CEM, the histiocytic cell line U-937, and the amniotic cell line WISH were not stained. Seven Epstein Barr virus (EBV)-positive lymphoblastoid cell lines were stained with HeFi-1, but there was no staining of three EBV+ African Burkitt's lymphoma cell lines or three EBV- American Burkitt's cell lines. HeFi-1 did not block the ability of the L428 cells to stimulate a mixed lymphocyte reaction or function as accessory cells for mitogen-induced human T cell proliferative responses. Modulation of the HeFi-1 cell surface antigen on the L428 cells was not observed. HeFi-1 specifically immunoprecipitated a cell surface protein of approximately 120,000 daltons from both the L428 and EBV+ lymphoblastoid cell lines. HeFi-1 monoclonal antibody should prove useful not only in the diagnosis, staging, and potential therapy of Hodgkin's disease, but also for determining the cell of origin of the Reed-Sternberg cell.     

A notable phenomenon recapitulated. A fusion product of a murine lymphoma cell and a leukemia virus-neutralizing antibody-producer host plasma cell formed spontaneously and secreting the specific antibody continuously

In the mid-1960s the #620 cell passage line of a murine lymphoma-leukemia was developed at the Section of Clinical Tumor Virology and Immunology, Department of Medicine, The University of Texas M.D. Anderson Hospital in Houston, TX. The diploid lymphoma cells released a retrovirus and were antigenic in young adult Swiss (YAS) mice. Small lymphoma cell inocula were rejected with immunity acquired against large inocula of lymphoma cells. Tissue sections revealed the "starry sky" configurations. In one of the tissue cultures set up from lymphoma #620, a cell line consisting of large round poly- or tetraploid cells arose and was referred to as lymphoma cell line #818. The #818 cells grew in suspension cultures and in the form of large, lethal ascitic tumors in YAS mice. Diploid #620 lymphoma cells stained for retroviral antigens; #818 cells stained both for retroviral antigens and immunoglobulins. Fluids withdrawn from #818 cultures neutralized the leukemia virus in spleen focus assays. Immunoglobulin precipitated from #818 suspension culture fluids strongly and specifically neutralized the leukemia virus. The growth of #620 or #818 cells in YAS mice treated with rabbit anti-lymphoma cell immune sera was strongly inhibited but culture fluids of #818 cells showed weak and insignificant inhibition against leukemia-lymphoma #620 (in one experiment, unpublished). In two experiments #620 lymphoma cells were co-inoculated with immune spleen cells into the peritoneal cavities of YAS mice. The immune spleen cells derived from mice that rejected #620 cell inocula or were actively immunized with a photodynamically inactivated mouse leukemia virus vaccine. In the peritoneal cavities of mice co-inoculated with #620 cells and immune spleen cells, clones of large round cells emerged with tetra- or polyploid chromosomal modes. These cells stained for leukemia viral antigens and immunoglobulins. When passaged in YAS mice these cells induced lethal ascites tumors. It was concluded as early as in 1968-69 that an immune plasma cell can fuse with a lymphoma cell, if the lymphoma cell expresses retroviral antigens against which the plasma cell is producing a specific antibody. Some human lymphoma-leukemia cells express retroviral antigens and/or budding retroviral particles, whether due to the acquisition of new env sequences by incomplete resident endogenous retroviral genomes or due to the entry of exogenous retroviruses into lymphopoietic stem cells. In the Discussion illustrations are provided for the human cell line #778 established from a patient with "lymphosarcoma cell leukemia" in 1966. The malignant cells released unidentified retrovirus-like particles and fused with one another and with reactive lymphoid cells of the host. It should be investigated further if human lymphoma-leukemia cells could fuse with an immune plasma cell of the host and thus alter the clinical course of the disease.     

Defective RNA-mediated c-myc gene silencing pathway in Burkitt's lymphoma

Keeping in view the fact that molecular basis of Burkitt's lymphoma (BL) is poorly understood, we attempted to explore the small interfering RNA (siRNA) mediated c-myc gene regulation using BL-derived EB-3 cell line as archetype cellular model. Such a study revealed that EB-3 cells possess 4-fold higher expression of Dicer gene coupled with 2-fold higher activity of RNA polymerase III than that observed in normal human lymphocytes. siRNAs derived from EB-3 cells had the inherent capacity to suppress c-myc gene expression in normal cells but not in native cells. Based on these findings we have proposed a novel RNA-mediated c-myc gene regulation pathway that may be responsible for BL.     

Replacement of the Epstein-Barr virus plasmid with the EBER plasmid in Burkitt's lymphoma cells

Transfection of an Epstein-Barr virus (EBV)-encoded plasmid containing EBER caused a substantial decrease in the level of plasmid containing EBV in Akata and Mutu Burkitt's lymphoma (BL) lines, but failed to do so in other BL lines. The results suggest that EBER could replace the role of EBV, but other EBV products also play a role in the growth of BL.     

Persistent infection of K562 cells by encephalomyocarditis virus.

Infection of human erythroleukemic K562 cells by encephalomyocarditis virus readily resulted in establishment of persistently infected cultures. In contrast to the usual typical lytic infection by encephalomyocarditis virus, in which trypan blue staining of cells reaches close to 100% by about 15 h postinfection, K562 cell cultures required 3 to 4 days postinfection to reach a maximum of about 80 to 90% cell staining. The proportion of K562 cells taking up stain gradually decreased to about 10% of those present by about 13 days postinfection; during this time, virus yield per day measured by either plaque or hemagglutination titration fell about 10-fold. The decrease in percent staining was followed by waves of increased staining accompanied by increased virus production. Virus-producing cultures were maintained for over 3 months. Evolution of both virus and cells accompanied establishment of persistence in that plaque size changed from about 7 mm in diameter for the original virus to less than 1.5 mm by day 20 postinfection and most of the cells cloned from persistently infected cultures were resistant to superinfection with the original virus. Resistance was due, at least in part, to reduced virus attachment in that binding of 3H-labeled virus to cloned resistant cells was about 2% of that to uninfected cells.     

Epstein-Barr virus-transformed B lymphocytes produce low molecular mass molecules with autocrine growth factor and competence factor activity

A human Epstein-Barr virus (EBV)-positive lymphoblastoid B cell line, named BA-D10-4, produces a factor of a molecular mass less than 10 kDa that promotes cell proliferation of both BA-D10-4 cells and other human T or B lymphoid cell lines, either EBV-positive or -negative. The factor synergizes with higher molecular mass autocrine growth factors and makes both BA-D10-4 cells and B cell lines from Burkitt's lymphoma, but not cells from T cell leukemia, more responsive to interleukin-1 and interleukin-6. Therefore, this low molecular mass factor seems to be an autocrine growth factor per se and to have the characteristics of a competence factor.     

Burkitt's lymphoma - a human tumor model system for immunological studies.

Burkitt's lymphoma occurs mainly in parts of tropical Africa and has attracted the attention of experimental workers due to its epidemiological and clinical features, which indicate a viral etiology and a host immune response to the tumor. As a result of virological studies, Epstein-Barr virus (EBV) DNA has been demonstrated in almost all tested biopsies of African BL. This contrasts to the absence of EBV in all, or almost all, of the non-African Burkitt's lymphoma-like tumors, even though the number of tested tumors in this group is small, and to the lack of EBV in all other types of lymphoma or leukemia. Immunological studies have revealed the presence of antibodies to different EBV-associated antigens in all African patients with Burkitt's lymphoma. However the antibodies are not specific for Burkitt's lymphoma but are found in most adults all over the world, although at lower levels. They cannot therefore serve diagnostic purposes, but they can give prognostic information and occasionally give clues to the mechanisms behind late tumor recurrences, and possibly guide so-called immunotherapy. Burkitt's lymphoma patients contrast to appropriate control groups where some of the persons are anti-EBV seronegative, and this, together with the presence of EBV in Burkitt's lymphoma biopsies and the absence of EBV in other lymphomas, even though the cell type involved may be infectable by EBV in vitro and the tumor may arise in an EBV-carrying person, favors an etiological role in EBV in Burkitt's lymphoma and speaks against the "passenger" hypothesis, according to which EBV is picked up by the Burkitt's lymphoma cell which happens to be particularly suitable for EBV persistence. To explain the geographical distribution, a cofactor, such as certain forms of malaria, has been implied.     

Epstein-Barr virus latent membrane protein: induction of B-cell activation antigens and membrane patch formation does not require vimentin.

The latent membrane protein (LMP) of Epstein-Barr virus (EBV) forms patches associated with the vimentin intermediate filament system in EBV-transformed lymphoblastoid cell lines, EBV-infected Burkitt's lymphoma cells, and LMP-transfected, EBV-negative Burkitt's lymphoma cells. By gene transfer, LMP induces the expression of vimentin and B-cell activation antigens in EBV-negative Burkitt's lymphoma cells. We have now expressed LMP in an EBV-positive Burkitt's lymphoma cell line, Daudi, which does not express any LMP or vimentin. In these Daudi transfectants, LMP still formed plasma membrane patches in the absence of vimentin. LMP did not resist nonionic detergent extraction in Daudi cells as it does in vimentin-expressing cells. LMP still retained functional activity as judged by induction of B-cell activation antigens. These data indicate that LMP can form plasma membrane patches and induce B-lymphocyte activation independent of vimentin association.     

Enhancement of hepatitis C virus replication by Epstein-Barr virus-encoded nuclear antigen 1.

Based on our recent observation that Epstein-Barr virus (EBV) is detected in 37% of the tissues of hepatocellular carcinoma, and especially frequently in cases with hepatitis C virus (HCV), the effect of EBV infection on the replication of HCV was investigated. EBV-infected cell clones and their EBV-uninfected counterparts in cell lines MT-2 (a human T-lymphotropic virus type I-infected T-cell line), HepG2 (a hepatoblastoma cell line) and Akata (a Burkitt's lymphoma cell line) were compared in terms of their permissiveness for HCV replication following inoculation of HCV derived from patients who were HCV carriers. The results indicated that EBV-infected cell clones, but not their EBV-uninfected counterparts, promoted HCV replication. EBV-encoded nuclear antigen 1 (EBNA1), which is invariably expressed in EBV-infected cells, supported HCV replication. Deletion analysis of the EBNA1 gene showed good correlation between transactivation activity and the activity supporting HCV replication. The present findings suggest that EBV acts as a helper virus for HCV replication.     

Induction of bcl-2 expression by Epstein-Barr virus latent membrane protein 1 protects infected B cells from programmed cell death.

Epstein-Barr virus (EBV) not only induces growth transformation in human B lymphocytes, but has more recently been shown to enhance B cell survival under suboptimal conditions where growth is inhibited; both effects are mediated through the coordinate action of eight virus-coded latent proteins. The effect upon cell survival is best recognized in EBV-positive Burkitt's lymphoma cell lines where activation of full virus latent gene expression protects the cells from programmed cell death (apoptosis). Here we show by DNA transfection into human B cells that protection from apoptosis is conferred through expression of a single EBV latent protein, the latent membrane protein LMP 1. Furthermore, we demonstrate that LMP 1 mediates this effect by up-regulating expression of the cellular oncogene bcl-2. The interplay between EBV infection and expression of this cellular oncogene has important implications for virus persistence and for the pathogenesis of virus-associated malignant disease.     

Epstein-Barr virus-positive Burkitt's lymphoma in a German woman during pregnancy

A fatal case of a Burkitt's lymphoma which occurred in a 34-year-old German woman during pregnancy is described. Nearly all organs showed either diffuse or nodular infiltration by tumor cells. Placenta and fetus were free of detectable tumor tissue. The patient had extremely high antibody titers (1 : 2056), both against Epstein-Barr virus capsid antigen (VCA) and the early antigen complex (EA). Within the tumor cells the Epstein-Barr virus-specific nuclear antigen EBNA and viral DNA was detected. A cell line established from a tumor biopsy displayed a translocation involving chromosomes 2 and 8. The role of Epstein-Barr virus in the development of Burkitt's lymphoma is discussed.     

Proliferation inhibitory effect of human alpha interferon on primary explants of Burkitt lymphoma: inverse relationship to patient survival

Eleven biopsies from 9 patients with Burkitt's lymphoma were tested for their sensitivity to the cell multiplication inhibitory activity of interferon. Three were resistant to interferon while 8 were sensitive to various degrees. Different biopsies from the same patient did not differ in interferon sensitivity. These results indicate that Burkitt's lymphoma cells might be resistant to interferon already in vivo as previously shown for some derived cell lines tested in vitro. The results imply an inverse relationship between patient survival and interferon sensitivity of the tumor cells.