Search

The c-Myb nuclear oncoprotein is phosphorylated in vitro and in vivo at an N-terminal site near its DNA-binding domain by casein kinase II (CK-II) or a CK-II-like activity. This in vitro phosphorylation reversibly inhibits the sequence-specific binding of c-Myb to DNA. The site of this phosphorylation is deleted in nearly all oncogenically activated Myb proteins, resulting in DNA-binding that is independent of CK-II. Because CK-II activity is modulated by growth factors, loss of the site could uncouple c-Myb from its normal physiological regulator.

The Philadelphia chromosome, widely implicated in human leukaemia, is the result of a reciprocal translocation t(9;22) (q34;q11) in which the abl oncogene located at 9q34 is translocated to chromosome 22q11, where it is fused head-to-tail with 5' exons of the bcr gene. In acute lymphoblastic leukaemia, some patients have a breakpoint within the major breakpoint cluster region of the bcr gene, whereas others have the break within its first intron. This second type of translocation results in the transcription of a 7.0-kilobase chimaeric bcr/abl messenger RNA translated into a bcr/abl fusion protein, p190, which has an abnormal tyrosine kinase activity and is strongly autophosphorylated in vitro. We have generated mice transgenic for a bcr/abl p190 DNA construct and find that progeny are either moribund with, or die of acute leukaemia (myeloid or lymphoid) 10-58 days after birth. This finding is evidence for a causal relationship between the Philadelphia chromosome and human leukaemia.

Vertebrate evolution has taken place against a background of constant retrovirus infection, and much of the mammalian genome consists of endogenous retrovirus-like elements. Several host genes have evolved to control retrovirus replication, including Friend-virus-susceptibility-1, Fv1, on mouse chromosome 4 (refs 3, 4). The Fv1 gene acts on murine leukaemia virus at a stage after entry into the target cell but before integration and formation of the provirus. Although restriction is not absolute, Fv1 prevents or delays spontaneous or experimentally induced viral tumours. In vitro, Fv1 restriction leads to an apparent 50-1,000 fold reduction in viral titre. Genetic evidence implicates a direct interaction between the Fv1 gene product and a component of the viral preintegration complex, the capsid protein CA (refs 7-9). We have now cloned Fv1: the gene appears to be derived from the gag region of an endogenous retrovirus unrelated to murine leukaemia virus, implying that the Fv1 protein and its target may share functional similarities despite the absence of nucleotide-sequence homology.

Triggering of mast cells and basophils by immunoglobulin E (IgE) and antigen induces various biochemical signals, including tyrosine kinase activation, which lead to cell degranulation and the release of mediators of the allergic reaction. The high-affinity receptor for IgE (Fc epsilon RI) responsible for initiating these events is a complex structure composed of an IgE-binding alpha-chain, a beta-chain and a homodimer of gamma-chains. It has been assumed that beta and gamma, which have extensive cytoplasmic domains, play an important but undefined role in coupling Fc epsilon RI to signal transduction mechanisms. Here we show that Fc epsilon RI engagement induces immediate in vivo phosphorylation on beta (tyrosine and serine) and gamma (tyrosine and threonine) by at least two different non-receptor kinases. We take advantage of unique features of this receptor system to demonstrate that the phosphorylation signal is restricted to activated receptors and is immediately reversible upon receptor disengagement by undefined phosphatases. Rapid phosphorylation and dephosphorylation may be a general mechanism to couple and uncouple activated receptors to other effector molecules. This could be particularly relevant to other multimeric receptors containing Fc epsilon RI gamma-chains or the related zeta and eta chains such as the T-cell antigen receptor (TCR) and the low-affinity receptor for immunoglobulin G (Fc gamma RIII, CD16).

Viruses isolated from leukaemic and preleukaemic tissues of the AKR mouse differ in their ability to induce thymic lymphomas. The Gross passage A virus, which originates from the thymus of a leukaemic AKR mouse1, and the MCF viruses, isolated from preleukaemic and leukaemic AKR tissues2,3, accelerate the onset of disease on injection into newborn AKR mice3,4 whereas the endogenous ecotropic Akv virus, which is expressed in AKR mice from mid-gestation onward, is nonleukaemogenic3,5,6. What are the structural features of the genome that account for the leukaemogenic activity of these viruses? The observation that the gp70 envelope glycoprotein of the leukaemogenic MCF virus contain ecotropic and xenotropic regions7–9 raised the possibility that the structure of the gp70 gene was the major determinant of leukaemogenic activity2. However, a cloned, in vitro passaged, leukaemogenic isolate of the Gross A virus very closely resembles that of the nonleukaemogenic Akv virus in the gag–pol and gp70 coding regions of the genome10. Most of the differences in structure of the Gross A and Akv viruses are located within the 3′-terminal 1,000 nucleotides—this raises the possibility that sequences outside the gp70 gene which are proximal to the 3′ end of the genome encode determinants of leukaemogenic activity. Due to the complex passage history of the Gross A virus we sought to isolate further leukaemogenic AKR viruses that had not been passaged in other animals to determine whether or not they resembled the Gross A virus. Here we report the isolation and characterization of new leukaemogenic viruses from a lymphoid cell line derived from a spontaneous AKR tumour. These isolates resemble the Gross A virus in the 3′ region of the genome. Our results suggest that this region encodes determinants of leukaemogenic potency and that major changes in the envelope glycoprotein are not necessary for the leukaemogenic activity of these viruses.

Chemically induced differentiation of cultured murine erythroleukaemia (MEL) cells results in a several hundred-fold increase in transcription of the adult mouse globin genes and thus serves as a model for gene activation during erythropoiesis. One approach to study gene regulation in this system has been to analyse the expression of foreign globin genes introduced into MEL cells. By introducing cosmid DNA containing the human adult(beta), fetal(gamma) and embryonic(epsilon)-globin genes, we have shown here that expression of the beta, but not the gamma or epsilon genes, is regulated during MEL differentiation. Regulated expression of the human beta-globin gene was observed when it was introduced either as part of the intact globin gene cluster or as an individual gene with 1.5 kilobases (kb) of 5' flanking DNA. Transcription from a herpes simplex virus (HSV) promoter adjacent to the thymidine kinase (tk) gene is also inducible in MEL cells.

A cell system has been developed to determine and dissect the controls that regulate normal myeloid cell growth, differentiation and malignancy, and to suggest a novel approach to the therapy of myeloid leukaemia.

For cytoplasmic glucocorticoid-receptor complexes to enter and accumulate in the nucleus a temperature-dependent event, 'activation' is required. Activation can be achieved in vitro by increased ionic strength, dilution or gel filtration and is manifested by an increased affinity of steroid-receptor complex for DNA and an altered elution profile from ion-exchange resins. Munck and Foley have shown that activated complexes isolated from thymocytes elute from DEAE-cellulose in a manner identical to complexes activated in vitro. We report here that DEAE-cellulose chromatography of steroid-receptor complexes from CEM-C7, a cloned human leukaemic T-cell line sensitive to the cytolytic action of glucocorticoids, and its steroid-resistant subclone 4R4 demonstrated that steroid receptors of clone 4R4 cannot form stable activated complexes. This defines a new defect in receptor action, activation lability (r+act1), which is unlike either the r-, r+nt-, or r+nti phenotypes previously described for mouse lymphoid variants.

Although the murine retrovirus SL3-3 is highly leukaemogenic, in both the structure of its genome and in its properties of replication in tissue culture it closely resembles the nonleukaemogenic retrovirus Akv (refs 3, 4). An earlier investigation of the properties of recombinant SL3-3-Akv viruses localized the major determinant of leukaemogenicity outside the env gene, in a region of the viral genome that includes the gag gene and the noncoding long terminal repeat (LTR). To localize the determinant of SL3-3's leukaemogenicity more precisely we have now construced a recombinant provirus containing the LTR of SL3-3 and the coding region of Akv. The leukaemogenicity of these recombinants demonstrates that the determinant of leukaemogenicity lies within the SL3-3 LTR. Nucleotide sequencing of the LTRs of SL3-3 and Akv shows that they differ by a set of changes in the region thought to contain a transcriptional enhancer element. We suggest that enhancer region sequences are the major determinants of leukaemogenicity in these viruses.

The long terminal repeats (LTRs) of retroviruses contain sequences necessary for the initiation and termination of retroviral transcription. These sequences include promoter elements, transcriptional termination signals and transcriptional enhancer elements. The enhancer elements of Moloney murine leukaemia virus (M-MuLV) are localized in a tandemly repeated region (approximately 75 base pairs (bp) long), which lies 5' to the CAT and TATA promoter elements in the U3 region of the LTR (see Fig. 1). We have shown that the tandem repeats are required both for LTR promoter activity, as measured by transient expression assays, and for biological activity, as measured by production of infectious virus. Furthermore, they can be replaced by transcriptional enhancers from the F101 host-range mutant of polyoma virus without loss of function. We report here that the addition of the polyoma (PyF101) enhancers to the M-MuLV LTRs (either with or without the M-MuLV tandem repeats) results in complete loss of viral leukaemogenicity, even though the virus can replicate to high titres in tissue culture fibroblasts and can establish infection in animals.

Two structural genes, β major and β minor, are present at the β-globin locus (Hbb) of a variety of mice phenotypically characterised by adult haemoglobin of the diffuse type1. These genes may be the result of duplication of an ancestral gene2, but there is increasing evidence that their expression is independently regulated. For example, in erythrocytes of the adult DBA/2 mouse, the ratio of the two nonallelic variant gene products is 4:1 in favour of the β-major chain whereas in Friend erythroleukaemia cells, clone 745, derived from the same mouse strain, the ratio varies from 3 to 0.6, depending on the inducer of erythroid differentiation used3–6. We reported previously that treatment of clone 745 cells for 6 d with 1×10−4 M haemin, an inducer of haemoglobin synthesis7, induces synthesis of the β minor, but not the β major globin chain5. This observation suggested the possibility that intracellular levels of haem (the reduced form of haemin) might act physiologically to selectively control the expression of one of the two β globin genes. To test this hypothesis, we determined the effect of allylisopropylacetamide (AIA) on the relative amounts of β globin chains present in induced erythroleukaemia cells. AIA is a strong inducer of δ-aminolevulinic acid synthetase in rat hepatocytes8 and in mouse erythroleukaemia cells9. Induction of the enzyme is due to breakdown by the drug of the haem moiety of the haemoproteins10. We report here that AIA treatment of Friend cells induced to differentiate by dimethylsulphoxide (DMSO) causes a reduction in the intracellular haem concentration and at the same time a marked decrease in the amount of the β minor globin chain. The ratio of the two β chains thus becomes very close to the ratio observed in normal erythrocytes of adult DBA/2 mouse.

DMSO and other cryoprotective agents produce a pronounced increase on the phase transition temperature of phospholipid membranes, indicating an increased stability. The effects of DMSO and the other cryoprotective agents, divalent cations, and local anaesthetics on the transition temperature of phospholipid membranes seem to correlate with their effects on the differentiation of Friend leukaemic cells in vitro. These studies suggest that the induction of differentiation by cryoprotective agents may be the result of the interaction of these agents with cell membranes.

Inoculation of susceptible strains of mice with the SL3-3 strain of murine leukaemia viruses induces T-cell lymphomas, whereas injection of the Akv strain does not1–3. Recombinant viruses that contain the long terminal repeat (LTR) of the SL3-3 virus and the gag, pol and env genes of the Akv virus are also leukaemogenic4. The cell-type specificity of leukaemias induced by viruses containing different LTR sequences is due in part to the ability of the virus to replicate in the appropriate cellular environment5–7. One explanation of the role of the LTR in determination of both cell tropism and leukaemogenic potential is that the LTR encodes tissue-permissive transcriptional elements. We report here that there are differences in the transcriptional activity of the SL3-3 and Akv LTR sequences in different murine cell types, and that the sequences present in the LTR of SL3-3 exhibit significantly enhanced transcriptional activity in T cells compared with the corresponding region of the Akv LTR. The results suggest that transcriptional elements are primary determinants of cell tropism and of leukaemogenicity of these viruses.

Retrovirus infection of embryonal carcinoma cells is blocked at the level of provirus transcription. De novo methylation of the input provirus occurs in embryonal carcinoma cells but not in permissive, differentiated teratocarcinoma. The kinetics of proviral methylation in embryonal carcinoma cells, however, suggest that while methylation may have an important maintenance role in controlling gene expression, additional mechanisms are used by the early embryo to initiate negative gene expression.

Embryo DNA of AKR mice contains several copies of intact mink cell focus-forming (MCF)-like env gene sequences with a proviral structure. In spontaneous thymic tumours, these MCF-like env sequences have regularly recombined with a specific region of the 3′ end of ecotropic env. This specific recombination is probably a critical event in the pathogenesis of spontaneous AKR tumours.

A study of the organization of light- and heavy-chain immunoglobulin genes in mouse splenic B cells, spleen-derived hybridomas and plasmacytomas has unequivocally demonstrated that aberrant rearrangements are common during normal B-cell development. The results support a probabilistic model for allelic exclusion of immunoglobulin gene expression.

THE membranes of lymphoid cells from tumours induced by oncornaviruses are antigenically altered. The new surface antigen(s) are readily recognised by isologous (histocompatible) hosts which mount both cellular and humoral immune responses1,2. It is essential to know whether or not antigens induced by the various strains of murine leukaemia viruses (MuLV) are cross-reacting immunologically.

Interleukin-3 (multi-CSF) is a multilineage haematopoietic growth regulator that initiates the proliferation and differentiation of multipotential stem cells. Complementary DNA clones encoding interleukin-3 (IL-3) have recently been isolated and the structure of the IL-3 gene determined. IL-3 is produced by T lymphocytes or T lymphomas only after stimulation with antigens, mitogens or chemical activators such as phorbol esters. The myelomonocytic leukaemia line WEHI-3B also produces IL-3 but its production is constitutive and the WEHI-3B cells do not appear to produce significant levels of any of the other lymphokines normally secreted by T lymphocytes after stimulation. It has been proposed that the genetic change leading to the constitutive expression of IL-3 may have been a key event in the development of this leukaemia. We report here that the constitutive synthesis of IL-3 by the WEHI-3B cell line is due to the insertion of an endogenous retrovirus-like element close to the 5' end of the gene. The insertion, an intracisternal A particle (IAP) genome, is positioned with its 5' long terminal repeat (LTR) close to the promoter region of the IL-3 gene, resulting in constitutive synthesis of IL-3.

A cell line that switches from mu to gamma 2b synthesis during growth in culture uses the same VH region for both heavy chains but retains two copies of the Cmu gene. This suggests that the mu to gamma 2b class switch can occur, at least in part, by an RNA processing mechanism. Regulatory variants of this cell line lose constitutive mu-chain synthesis but simultaneously acquire lipopolysaccharide (LPS)-inducible synthesis of that chain. This co-variation is allele-specific and is correlated to a large deletion of DNA in the JH--Cmu intron.

Analysis of chromosome translocations associated with B-cell derived tumours and studies of chromosome-15 trisomy in murine leukaemia support the theory, emerging from recent work on tumour viruses, that cancer can be associated with DNA rearrangements which result in the increased expression of normal cellular genes.

Friend murine leukaemia virus complex (FV) causes an immunosuppressive retrovirus-induced disease. In certain mouse strains, FV shows striking similarities to human immunodeficiency virus (HIV) infection in man in that infected mice have severe T-cell immunosuppression but also develop virus-neutralizing antibodies incapable of eliminating infected cells. Previously we noted the influence of mouse major histocompatibility complex (H–2) genes on both FV-induced immunosuppression1 and on ability to protect mice against FV by immunizing with a vaccinia–Friend murine leukaemia helper virus (F-MuLV) envelope (env) recombinant virus2. Here we show that different subregions of H–2 are involved in susceptibility to virus-induced immunosuppression (H–2D subregion) and protective immunization with a recombinant vaccinia virus (H–2K or I–A subregions). Thus, susceptibility to virus-induced immunosuppression does not preclude protection by vaccinia–Friend immunization. The mechanism of protection seems to involve priming of immune T cells, and not initial induction of neutralizing antibodies or cytotoxic T lymphocytes (CTL) (ref. 2). Subsequent virus challenge generates a secondary response, resulting in appearance of IgG antibodies and CTL. In human HIV infection there could also be host genetic influences on elements of disease pathogenesis, such as immunosuppression, and on the success of T-cell priming by potential protective vaccines.

The human alpha- and beta-globin genes have been activated in MEL X human fibroblast cell hybrids. However, even though the human gamma- and beta-globin genes are closely linked and were shown in these hybrid clones to be present in approximately equal numbers, no human gamma-globin mRNA was produced. Thus, the human beta- and gamma-globin genes in these cells are differentially regulated apparently by a positive regulatory factor(s) specific for individual globin genes.

NUMEROUS attempts have been made to solubilize tumour specific transplantation antigens (TSTA) of carcinogen-induced neoplasms and of virus-specified tumour antigens (VSTA) from the cell membranes of DNA and RNA virus-induced neoplasms1–6. Usually, yields of biologically active materials are low and the materials are labile. Soluble antigenic materials obtained from transplantable hepatomas1,2 and purified by gel filtration were found to have limited biological activity. Although immunological specificity was preserved, the assays used were of relatively low precision. Tumour rejection, a major function of both TSTA and VSTA, was not observed.

Natural killer (NK) cells, which occur at high levels in normal non-immune individuals of several species, have the capacity to lyse certain tumour cells in vitro, and have been proposed as a first level of defence against tumour growth in vivo1–3. Evidence for this comes mainly from the murine system, where certain FI hybrid or T-cell deficient mice with high NK activity resist growth of transplanted tumours better than do normal syngeneic controls with lower NK activity4–6. Recent studies of the beige mutation on the C57BL mouse strain background, previously proposed as an animal homologue of the human Chediak-Higashi syndrome7,8, have revealed a new model for analysing anti-tumour effector mechanisms in the intact syngeneic host9,10. The recessive beige gene (bgJ), which affects melanosome and granulocyte lysosome functions11,12, also causes a profound (but not total) depression of NK activity whereas other anti-tumour effector mechanism mediated by T cells and macrophages remain relatively intact13. It was important, therefore, to test the tumour susceptibility of beige mice as a first direct test of the hypothesis that NK cells are involved in surveillance against neoplasia. In partial confirmation of this hypothesis, we report here that failure of small threshold doses of one virally and one chemically induced transplantable leukaemia to grow out in syngeneic mice partly depends on a rapidly acting host defence mechanism. This mechanism is deficient in bg/bg mice, which develop palpable, progressively growing tumours faster and at a higher frequency than do phenotypically normal +/bg littermates.

THE high incidence of spontaneous leukaemias in AKR mice may be explained by the early appearance and high level of replication of a type C virus. It is often suggested, however, that in addition, AKR mice could be tolerant to their leukaemias. This tolerance would enhance the in vivo growth of malignant cells. In fact, AKR mice are not completely devoid of antileukaemic immune response1–5. It would be interesting, therefore, to determine if a quantitative or qualitative defect of one given type of antileukaemic response exists in AKR mice. Previous work has shown that an antibody response directed against the type C virus or against antigens of leukaemic cells can be demonstrated not only in hyperimmune1 but also in normal AKR mice4. A cell mediated reaction directed against G (Gross)6 leukaemia antigen-bearing cells was also detected in vitro5,7 but this cell-mediated reaction was characterised by inhibition of the growth of G+ cells in monolayers. It is now known, however, that this kind of inhibition can be due to non-T cells7–9 and could be antibody dependent10. A completely different cell-mediated antitumour reaction can be evidenced in the murine sarcoma virus (MSV) induced tumour system by the chromium release test (CRT) which detects a pure T-cell-mediated11,12 and antibody-independent13 cytolysis of tumour cells. Our work was performed to determine if AKR mice are capable of spontaneously developing such a kind of T-killer cell response.

Neoplastic cells are characterized by partial or total autonomy from the interactions that regulate the behaviour of normal cells in the intact animal. Despite the role of the host cellular environment in governing the proliferation and differentiation of both normal and malignant cells, little is known about these host factors. The characterization of host genes that influence both normal cellular processes, as well as susceptibility to tumour induction, is one approach to identifying such factors. Mice carrying two recessive mutations at the steel (Sl) locus have an environmental defect that affects both normal haematopoietic stem cell function as well as susceptibility to Friend leukaemia virus. In this study, we have used Sl/sld mice to examine whether malignant transformation by this RNA tumour virus results in a population of cells capable of proliferating even in the defective cellular microenvironment of Sl/sld mice. We report here that late after infection, the leukaemic spleens of Friend virus-infected mice contain cells which, unlike normal haematopoietic stem cells, are able to form macroscopic spleen colonies in irradiated mice of genotype Sl/Sld. This observation forms the basis for the first in vivo colony assay for leukaemic cells transformed by Friend virus.

IN many mouse strains, RNA tumour virus information is transmitted vertically (inherited) perhaps as part of the cell genome1,2. The activation of type C virus replication in cells from low leukaemia incidence mouse strains3 and from ‘virus-free’ cell lines4 by halogenated pyrimidines suggests that all mouse cells may contain murine leukaemia virus (MuLV) information, often in an unexpressed form. Indeed, proponents of the ‘oncogene’ hypothesis have predicted a complete copy of type C virus information in the genome of all somatic cells2.

NATURALLY-occurring antibodies against thymocytes, certain leukaemias and other tumour cells have been observed in various species including man1,2. In addition, natural cellular immunity—also known as cellular natural killing (CNK)—has been described recently in man3–5 and in mice6–9. The possible role of CNK in tumour surveillance and the nature of the effector cells are presently subject to extensive research. Relatively little is known, however, about the nature and sensitivity of the target cells in relation to CNK. We report here that human leukaemic T cells and human thymocytes are highly sensitive to CNK by normal human peripheral blood and that B cell lines are relatively insensitive. The response seems to be unrelated to HLA and expressed variably by different individuals.

Erythroleukaemia, until now defined by morphological criteria, has been observed in mice infected with several distinct virological entities, including the polycythaemia-inducing strains of Friend virus complex (FV-P), the anaemia-inducing strains of Friend and Rauscher virus complex (FV-A and RV-A, respectively) and various helper-independent virus isolates derived from the FV-A or RV-A complexes. Whereas erythroleukaemia develops rapidly (within 2-3 weeks) in mice infected with any strain of FV or RV, the helper-independent virus alone is pathogenic in mice only if they are infected neonatally. We now describe how two biological markers can be used to distinguish among the otherwise confusing array of virus-induced erythroleukaemias in mice. The evidence suggests that there are three different classes of this disease: (1) those (S+E+) from which both defective spleen focus-forming virus (SFFV) and erythropoietin-independent, erythroid colony-forming units (CFU-EI) can be recovered, (2) those (S+E-) from which only SFFV (but not CFU-EI) can be recovered, and (3) those (S-E-) from which neither SFFV nor CFU-EI can be recovered. There is a consistent association between the type of virus used to induce the erythroleukaemia and the class of the disease induced.

IN domestic cats, regression of tumours originally induced by inoculation of feline sarcoma virus (FeSV) as well as the resistance to feline leukaemia virus (FeLV)-induced lymphosarcoma in the natural environment are correlated with high titres of antibody to the feline oncornavirus-associated cell membrane antigen (FOCMA)1–4. FOCMA is present on FeLV-producing and nonproducing feline lymphosarcoma (LSA) cells and FeSV-infected fibrosarcoma cells but is absent from FeLV-infected normal cells. This finding is consistent with the interpretation that FOCMA represents an FeLV or FeSV-induced transformation-specific protein5–7. Other workers have initiated studies of the biochemical nature of FOCMA derived from cytoplasmic extracts of cells of a heterologous species (mink) nonproductively transformed by FeSV8–10. However, our primary interest has been to analyse FOCMA expressed on naturally occurring LSA cells from pet cats. We describe here the identification of a hitherto unrecognised protein on the surface of feline LSA cells and cultured transformed feline lymphoblastoid cells which exhibits antigenic determinants in common with FOCMA. The molecular weight of this protein is approximately 70,000. Our data agree with the accumulating evidence that FOCMA is distinct from all FeLV and RD114-coded virion antigens4,6–11.

SEVERAL rapid in vitro methods1–5 for the quantitative assay of murine lymphatic leukaemia viruses (MuLV) have replaced the once standard in vivo bioassay for leukaemogenicity6,7, which was slow, cumbersome, and relatively costly. The in vitro procedures, however, suffer from two limitations: first, some MuLV propagated serially in vitro may lose leukaemogenicity while retaining infectivity in vitro8, so the in vitro assays do not necessarily correlate with biological activity in vivo; second, some wild-type MuLV extracted from normal or neoplastic tissues in vivo seem to require a period of adaptation to in vitro growth conditions before they can be assayed reliably by in vitro methods9.

Activation of a cellular transforming gene in tumours induced by Abelson murine leukaemia virus

WE have found two distinct biological activities in cultures of peritoneal exudate cells rich in macrophages. One consists of an inhibitory effect on cell proliferation, described previously, caused by a low molecular weight product1; the other is a stimulatory effect brought about by a non-dialysable material. Both activities were found in the same culture fluid; and different kinds of cells exhibited different sensitivities to them. Our results offer a possible explanation for the discrepancies in the literature concerning the different effects of macrophages or their products on cells1–7.

DEFECTIVE type C RNA tumour viruses which are genetic recombinants between ecotropic and xenotropic viruses have been described and suggested to be the real transforming agents during the course of viral-induced lymphatic leukaemia1,2. The spleen focus-forming virus (SFFV-F) component of the Friend virus (FV) complex is a defective virus and is the erythroid-specific component of the complex which stimulates erythropoiesis in the mouse and is also involved in the relatively rare event of erythroid transformation3,4. In order to examine the role of the SFFV component during erythroblastosis and subsequent transformation of erythroid cells, we have synthesised a SFFV-specific probe using FV released from F4-6, a transformed erythroid cell line5,6. We report here that, using this probe, no xenotropic sequences are present in the SFFV-specific cDNA and that there is a low level of transcription of part of the SFFV sequences but not specifically during normal erythroid differentiation. Our results thus show that if SFFV is a recombinant virus, the xenotropic sequences are only present in the lymphatic leukaemia virus (LLV)-related common sequences of the FV genome. Furthermore, our experiments do not support a hypothesis whereby the specific action of SFFV can be explained by transcription of the SFFV-specific genome during normal erythroid differentiation.

Despite the fact that tumour cells with the potential for metastasis may circulate randomly, many demonstrate a preference for specific organs. Recently, several investigators have selected variant tumour cell lines with enhanced capacity to metastasize to specific organs, mainly using the spontaneously originating B16 melanoma cell line of the mouse and tumour variants with enhanced capacity to metastasize to the lungs1, brain2 and liver3. We previously reported4,5 the derivation of a liver-specific metastatic variant of a Marek's disease (MD) virus-transformed, non-producer4,6 lymphoma cell line. MD is a naturally occurring, herpes virus-induced, T-cell lymphoma of chickens which bears pathological and aetiological similarities to Burkitt's lymphoma in man. This makes MD a useful model for study7. One similarity is the pattern of metastasis in which both lymphomas induce a high incidence of ovarian and liver lesions8,9. We now report the existence of a cell-surface antigen, detectable by a monoclonal antibody, correlated with organ-specific metastasis.

The retroviruses human T-cell lymphotrophic virus-I (HTLV-I)1,2 and HTLV-III/LAV (lymphadenopathy-associated virus)3,4 are clearly linked to human diseases. Patients with HTLV-I-positive neoplasms may respond transiently to traditional chemotherapy, but are not cured5. For patients with acquired immune deficiency syndrome (AIDS) there is no curative therapy6. In retroviruses of different species, viral propagation crucially depends on reverse transcriptase, an enzyme not present in normal mammalian cells and different from mammalian DNA polymerases, making it a target for specific inhibition. Reverse transcriptase has been well conserved through evolution: an LAV isolate contained a 250-amino-acid-long domain, presumably the reverse transcriptase core sequence, which has 21% homology to Moloney murine leukaemia virus (MoMLV)7. Because HTLV-IH infects only humans and chimpanzees8, we substituted murine retroviruses for in vivo evaluation of candidate anti-AIDS drugs after ascertaining similar inhibition in vitro of HTLV-III and MLVs, which were chosen for their short incubation time. The triphosphate of 3'-azido-3'-deoxythymidine (AZT)9–11 is incorporated into complementary DNA by retroviral reverse transcriptase, causing premature chain termination. Here we show that chronic AZT treatment of mice infected with Rauscher murine leukaemia virus complex (RLV) prevents infection of splenocytes and development of splenomegaly, and suppresses viraemia if started soon after inoculation. Starting AZT late in the course of disease still leads to significant prolongation of life; anaemia, however is a significant side-effect. By analogy, AZT may have a role in preventing retroviral disease in humans if started early after infection, and it may lead to significant survival gains even if started later in the course of disease.

INOCULATION of BALB/c mice with high doses of Rauscher murine leukaemia virus (R-MuLV) results in rapid spleen enlargement due to the accumulation of immature erythroid cells, which subsequently infiltrate other haemopoietic organs1–3. If low doses of virus are injected, the animals overcome the initial erythroid phase of the disease, but usually die later of a lymphatic leukaemia. In newborn C57/Black mice, R-MuLV induces lymphomas4,5, whereas in (C57/B1 X DBA) F1, various types of reticulum cell sarcomas are formed6. Attempts to establish permanent cell lines from these reticulum cell sarcomas were successful, but similar attempts with leukaemic spleen or liver tissue failed repeatedly in BALB/c mice4. Instead, the blast cells matured into haemoglobin-containing cells in vitro7. We report here the induction of tumours in DBA mice using syngeneic transplants of R-MuL V-infected animals in the terminal phase of erythroid leukaemia. From this tumour, a new permanent cell line could be established, which could be induced to synthesise haemoglobin in vitro.

INTEREST in the regulation of T-cell-mediated cytotoxicity by the major histocompatibility complex (MHC) genes stems from the discovery of (1) an H–2 restriction of T-cell-mediated cytotoxicity, that is sensitising cells, effector cells and target cells need to be syngeneic at the H–2 locus, especially at the D and or K regions of the MHC1–5 for effective killing and (2) inhibition of T-cell-mediated cytotoxic reactions against syngeneic tumours with alloantibodies to H–2 antigens specific for the tumour cells6,7. Different origins have been proposed for the tumour-associated antigens (TAA), and the most favourable model is a modified H–2 antigen or an adaptor–antigen complex4–7. This issue has important implications for tumour immunology, and so we have evaluated the validity of the inhibition of cytotoxicity by serum at the effector phase level; investigated afferent interference by serum of the cell-mediated cytotoxic response, and studied the inhibition of cell-mediated cytotoxic reactions with soluble antigens. We report here that there is a lack of H–2 restriction of T-cell-mediated cytotoxicity to TAA, and that TAA seems to differ from H–2 antigens. This implies that T-cell recognition is not entirely regulated by the MHC genes.

TUMOUR-SPECIFIC antigens have usually been demonstrated with little knowledge of their molecular nature; for example by the residual reactivity with tumour shown by an anti-tumour serum which has been absorbed with normal tissue. Certain tumours of B lymphoid cells which secrete immunoglobulin (Ig) provide a promising situation for identification of the antigen in molecular terms1–3. Amino acid sequences in the variable (V) regions of Ig molecules determine not only antibody activity, but also the ‘idiotypic’ antigenic determinants of the molecule (for review see ref. 4). In a normal B lymphoid clone the V regions on the surface Ig of the early members are the same as those on the Ig exported by specialised protein-secreting descendants and are essentially unique to this clone5,6. The morphology and mode of Ig secretion in neoplastic cells are frozen at some stage in the range shown by their normally differentiating counterparts (Fig. 1). Tumours with cells having both surface and export Ig have provided Ig in the host serum in amounts suitable for the conventional raising of anti-V sera. These sera, by reacting specifically with the surface Ig of the corresponding tumour, demonstrated that the V regions can represent tumour-specific antigens2,3. Furthermore, Lynch et al.1 demonstrated that mice immunised so as to produce antibodies to the V regions of such a tumour (a syngeneic mouse myeloma) were thereby afforded a measure of specific protection against tumour challenge.

Endogenous avian1–3 and murine4–9 C-type viruses are genetic elements which are transmitted at several distinct chromosomal loci. Different patterns of virus activation have been observed in a variety of different mouse strains10–12. However, because there are multiple copies of closely related endogenous viruses in every mouse strain, the genetic basis of the differential expression of these genes is poorly understood. A new substrain of mice, BALB/Mo, was derived previously13 carrying the exogenous Moloney leukaemia virus (M-MuLV) genome on chromosome 6 (ref. 14). Virus activation occurs in lymphatic tissues of all BALB/Mo mice soon after birth12. We report here the derivation of three new substrains of mice each carrying a single M-MuLV genome on different chromosomal integration sites. Each locus was associated with a distinct phenotype of virus expression. Evidence is presented that apparently identical viral genomes show differences in spontaneous virus activation and that defective viral genomes can be carried in the germ line of mice.

Experiments predicting1,2 or demonstrating lymphomagenesis in mice undergoing graft-versus-host (G-v-H) or chronic immunological reactions have been reported3–5. In such cases, lymphomagenesis seemed to operate through the agency of activated murine leukaemia viruses (MuLV) and occurred only after very long latent periods, and most (but not all) lymphomas were of host origin4. No attempts were made in these studies to assess the relative role of T-cell subpopulations in lymphomagenesis. We have proposed elsewhere that certain MuLV-induced T lymphomas might be the progeny of that subset of normal thymocytes bearing antigen specific T-cell receptors for MuLV envelope antigens6–10. Further, the MuLV binding to these cells provides both a portal of infection and a mitogenic signal6,9,10. These cells and their thymic progeny would be subject to continuous restimulation by the endogenously produced MuLV envelope glycoproteins, and thus the apparent and “uncontrolled” proliferation of the cells would be due to their inability to resolve the viral infection and escape the resultant antigenic microenvironment. This receptor-mediated leukaemogenesis hypothesis predicts that any T-cell clone that cannot escape its stimulating antigens would appear to be lymphomagenic in that environment. Recent experiments in our laboratory have demonstrated that unique clones of strain A alloreactive T cells exist which can recognise hybrid mixed lymphocyte reaction (MLR)-stimulating determinants present on (C57B16 × A/J)F1 (B6A) stimulator spleen cells11–15. A variety of such clones have been isolated and characterised13–15. We reasoned that, using these clones, it might be possible to generate alloreactive T-cell lymphomas by in vivo injection of these cells into hybrid B6A mice. We report our successful results here.

Two types of filaments have been described in the cytoplasm of non-muscular cells—microfilaments (40–70 A in diameter) and 100-A filaments (90–110 A). Microfilaments have been found in almost every type of animal cell1, and they are known to be actin-like2 and believed to have a role in many contractile processes1,3. In contrast, the 100- A filaments have been observed only in a limited number of cell types, such as tissue culture cells2,4, smooth5 and skeletal6 muscle cells, nerve cells7,8 and tumour cells9,10, and are thought not to be actin-like2,3,10,11. The function of 100- A filaments is not clear although they have been suggested to act in the intracellular transport of organelles in fibroblasts4, the axoplasmatic transport in nerve cells12,13 and the spreading of cells3, or to serve as a cytoskeleton14. Induction of 100-A filaments by colcemide6 and cytochalasin B and colcemide15,16 has also been demonstrated. A clue to their function may lie in the demonstration, reported here, that the distribution of the 100-A filaments differs in resting and locomotive cells.

Growth and development of haematopoietic cells in vitro require the presence of specific regulatory molecules. Some of these molecular species appear to have a broad specificity, being able to promote the proliferation and differentiation of multipotential cells, as well as megakaryocytic, erythroid and granulocytic-progenitor cells1–3. Such factors are present in medium conditioned by the growth of lectin-stimulated mouse spleen cells or WEHI-3 myelomonocytic leukaemia cells (WEHI-CM)1–5. Using WEHI-CM, we6 and others 7–9 have been able to obtain permanently growing, non-leukaemic cell lines of a granulocytic or mast cell nature. Significantly, we have found that the factor in WEHI-CM necessary for the growth of these cells has co-purified with the multi-lineage stimulating activity present in WEHI-CM, suggesting that one molecule may be concerned in the development of multiple cell types3. We have now used these cells to investigate the mode of action of this haematopoietic cell growth factor and have found that the requirement of this factor for survival and growth may lie in its ability to modulate ATP levels within the cells.

The transforming gene of avian myelocytomatosis virus MC29, v-myc, causes a variety of malignancies in chickens. A cellular homologue, c-myc, has been implicated in B-cell malignancies in mice and humans but is also expressed in many normal cell types and may be important in the control of normal cell proliferation. c-myc is highly conserved in vertebrates. We have been investigating the relationship between c-myc expression and the terminal differentiation of cultured mouse erythroleukaemia (MEL) cells. We find that the level of c-myc messenger RNA shows a rapid biphasic change in MEL cells induced to differentiate by dimethyl sulphoxide or hypoxanthine. The changes occur during the first few hours of the differentiation programme and require active protein synthesis. These data suggest that changes in c-myc expression may be important in the irreversible commitment of MEL cells to terminal erythroid differentiation.

A molecular recombinant of Rous sarcoma virus and murine amphotropic leukaemia virus, src(MoMuLV), where the avian src oncogene has been placed under the influence of a murine virus promoter sequence, has been reported. Infection of long-term marrow cultures with this virus led to a dramatic change in the relative numbers of stem cells, granulocyte-macrophage progenitor cells and mature cells found in normal haematopoietic cell development. However, although the balance between self-renewal, differentiation and development was disturbed, injection of the cultured cells into irradiated syngeneic recipients did not lead to the development of leukaemia. Thus, although the control had been 'loosened', the host regulatory mechanisms were sufficient to impose a restraint on unlimited growth of the cells. We now show that the stem cells from the src-infected cultures show a remarkably increased capacity for self-renewal in vitro in situations which are inimical to the maintenance of self-renewal in normal uninfected stem cells and that self-renewal/differentiation can be modified by the culture conditions.

IT may be of great help in the elucidation of regulatory systems in mammalian cells if we could isolate temperature-conditional mutants and identify their mutated functions. I report here that a temperature-sensitive mutant of L5178Y murine leukaemic cells has a thermolabile L-alanyl-tRNA synthetase.

Selective killing of malignant cells in a leukaemic rat bone marrow using an antibody–ricin conjugate

THE possibility that mammary tumour virus (MTV) and lymphatic leukaemia virus (LLV) may interfere with each other in mice was suggested by the findings that: (1) RILL mice, a strain carrying MTV in which there is a high incidence of mammary tumours, also carry “inhibitors” of MTV in their milk1, show a spontaneous tendency to lose MTV2, and develop a number of lymphoid tumours when free from MTV3. (2) Factors are present in the RIII strain which delay the progression of mammary tumours. Tumours of this strain show long periods of “pregnancy dependence”, that is, a very slow progression towards autonomous growth4–6. In this respect RIII mice differ significantly from MTV-carrying C3H strain, the tumours of which rapidly progress7,8. (3) Factors which delay progression of mammary tumours in the RIII strain are extrachromosomal and are transmitted through the milk together with MTV. When low-tumour BALB/c mice, supposedly free from virus, are foster-nursed by RIII and then maintained as an inbred line for generations, this line (BALB/cf(RIII)) shows in breeding females a high incidence of mammary tumours the behaviour of which closely resembles that of RIII tumours9,10. Conversely, the behaviour of mammary tumours in BALB/c mice foster-nursed by C3H (BALB/cf (C3H)) closely approaches that of C3H tumours11. (4) A high incidence of lymphatic leukaemia has unexpectedly been observed in BALB/cf(RIII) mice, especially in males and virgin females12–14. In contrast, however, the incidence of leukaemia in BALB/c, C3H and BALB/cf(C3H) mice is low14. (5) In BALB/cf(RIII) females there is repulsion between mammary tumours and leukaemia14–15. (6) A cell-free leukaemogenic agent has been demonstrated in BALB/cf(RIII) mice15.

Cheek Pouch of the Syrian Hamster and Immunity to Heterotransplantation of a Murine Leukaemia