Search

Nonphagocytic cells bearing receptors for complement and IgGFc in blood samples from four boys with panhypogammaglobulinaemia who lacked B lymphocytes were positively stained for human macrophage antigen (HuMA) and intracellular peroxidase. Similar non-T, non-B cells were found in adult blood preparations but in lower numbers than in the four boys above and in two of three other boys with various immunodeficiencies. Because of their monocyte characteristics, it is improbable that these cells are B-lymphocyte precursors; this is consistent with our view that there is a central failure of B-cell induction in boys with panhypogammaglobulinaemia who lack lymphocytes with surface immunoglobulin.

While it is certain that reactive microglia (phagocytes in damaged brain tissue) originate from monocytes, the derivation of resting microglia (argentophilic cells in undamaged brain tissue) is not clear. In order to examine a possible lymphocytic and/or monocytic origin for resting microglia, cryostat sections of presumably normal human brain tissue were treated with anti-lymphocyte, anti-monocyte and anti-beta 2-microglobulin sera and then stained with the immunoperoxidase technique. No positive reaction was obtained. It must be concluded that resting microglia are not derived from mononuclear blood cells.

Most human lymphoid malignancies preserve a pattern of gene expression reflecting their proliferative activity and the development level of clonal expansion and maturation arrest. Characteristics of leukemia and other cancer cells frequently considered to reflect aberrant differentiation may more often reflect clonal selection of cell types that are normally infrequent and transitory. The differentiation status of progenitor or mature lymphoid cells influences which genetic elements are at risk of being exploited, via mutation, recombination, or deletion, for clonal advantage. These alterations may frequently arise spontaneously as a consequence of the unique developmental and functional programs of lymphoid cells and have as a major phenotypic consequence the stabilization of transitory cellular phenotypes.

Human bone marrow contains plastic-adherent hemopoietic progenitor cells whose plating efficiency is increased by brief (2 h) exposure to methylprednisolone (MP). When subsequently covered with methylcellulose medium, they form colonies of monoblastoid cells. Colony size, but not number, and mature cell production are increased by erythropoietin (epo) and granulocyte-macrophage colony-stimulating factor (GM-CSF). However, colonies do not grow under serum-free conditions. The resistance of plastic-adherent progenitors to treatment with 5-fluorouracil (5FU), their growth pattern, and their capacity to produce granulocytic and erythroid colonies on replating, suggest that they may be similar to the primitive, 5FU-resistant, plastic-adherent progenitor cells (HPP-CFC) in murine marrow.

An antiserum raised against myelomonocytic and monocytic leukaemia cells has been used to define a cell surface antigen shared by normal and leukaemic cells of the granulocyte and monocyte cell lines. The quantitative expression of this antigen was investigated using the Fluorescence Activated Cell Sorter and was shown to correlate with the degree of morphological maturation.

Binding of purified cholera toxin to cell surface receptors has been visualized by an indirect immunofluorescence procedure. Normal nucleated cells from blood, bone marrow and lymphoid tissues, express these receptors with the possible exception of erythroid precursors. Cells from patients with chronic lymphoid or myeloid leukaemias have a normal receptor expression but acute leukaemic cells showed a marked deficiency in cholera toxin binding. Insertion of purified Gm ganglioside into membranes of acute leukaemic cells provided cellular binding sites for the toxin.

Sera from 120 children and young adults with acute leukaemia (59), various other tumours (53) and histiocytosis X (eight) were studied for the presence and characteristics of circulating immune complexes (CIC). Serial and parallel testing was performed using: C1q binding (solid phase), Raji cell radioimmunoassay and anti-C3 (solid phase). CIC were detected in 36 of 56 (64%) patients with acute lymphoblastic leukaemia (ALL) and in 62% of other tumour subjects. In the ALL sera, the mean positive C1q binding was 5.4 s.d., Raji cell 4.2 s.d. and anti-C3 4.4 s.d. In 12 ALL sera CIC were characterized for molecular size by sucrose gradient centrifugation. Most samples showed high molecular weight (19S) complexes but intermediate (11-14S) and smaller (8-9S) complexes were also detected. There was no apparent relationship between the presence, amount or physical size of the detectable CICs and clinical course of the patients studied; 12 patients with ALL in long term remission showed presence of CIC at some time during their course. Immune complexes precipitated from leukaemic sera were also examined for the presence of common ALL antigen (cALL) and Ia(DR) antigens utilizing rabbit antisera and mouse monoclonal antibodies. Experiments with isolated immune complexes from ALL sera provided no positive evidence for the presence of cALL antigen or Ia antigen within immune complex materials from ALL patients.

The Thy-1 antigen is a cell surface glycoprotein found in neural tissue of all mammalian species so far studied. The distribution and amount of this antigen has been measured on 4 human neuronal and 2 neuroglial cell lines and on fresh tumour cells of neuronal origin. In 3 out of 4 neuronal lines (LAN-1, TR14, CHP 212) more than 90% of cells were Thy-1+, however, LAN-1 cells showed only weak immunofluorescence and bore on average 2.4 times fewer molecules of Thy-1 per cell than those of either TR14 or CHP 212. The mean number of Thy-1 molecules per TR14 cell was shown to be approximately 2.25 x 10(5). In contrast, only 66% of cells in the fourth neuronal line (CHP 100) were Thy-1+, although these showed strong immunofluorescence. Both glial cell lines, UCH-203 and H314/123, showed strong Thy-1 immunofluorescence on more than 90% of cells. Similarly, with fresh neuronal tumour cells, although approximately 80% of tumours were Thy-1+ (essentially 100% of cells in these being positive) there were considerable differences in the intensity of labelling by immunofluorescence between different tumours. Such heterogeneity in cell lines and malignancy may reflect normal in vivo variation. Different phenotypes might therefore represent separate neural cell lineages, or simply differences in maturational status within a lineage. The very low frequency of Thy-1+ cells in normal bone marrow (less than 0.1% of nucleated cells) indicates that anti-Thy-1 antibodies may be valuable in both the diagnosis and subsequent treatment of neuroblastoma.

We had been able to classify 70 cases of acute lymphoblastic leukemia (ALL) into different immunological subtypes based on surface marker analysis with the aid of a panel of monoclonal antibodies. The relative proportions of different ALL subclasses appears to be similar to those reported from Europe and the U.S.A. Our results also indicated that monoclonal antibodies OKT11, UCHT-2 and WT1 were very useful in defining pre-T or early T subtypes of ALL. The latter leukaemias lack most maturation-linked T-cell markers but share with the other T-ALL the characteristic of low density or undetectable insulin receptors.

Monoclonal antibodies have been raised against a dimeric cell surface antigen (p75/150) which is specifically associated with the tumorigenic phenotype in human fibroblast X HeLa hybrids. During biosynthesis, a precursor molecule (p70/140), was associated with microsomal membranes in vivo but possessed no detectable cytoplasmic domains. At this stage, each p70 monomer contained 3 "high-mannose" type N-linked glycans which were subsequently processed into endoglycosidase H-insensitive complex oligosaccharides on the mature cell surface forms. Cleavage of this cell surface form with endoglycosidase F yielded non-N-glycosylated polypeptides of Mr = 60,000/120,000. All the monoclonal antibodies identified similar non-N-glycosylated polypeptides in cells grown in the presence of tunicamycin. p75/150 could be weakly labeled with [3H]palmitic or myristic acid. In vivo, p75/150 was found to be phosphorylated on serine residues. Immunoprecipitates of p75/150 from HeLa or tumorigenic hybrid cell lysates exhibited protein kinase activity in vitro, which phosphorylated p75/150 itself, also on serine residues. We were unable to detect this kinase activity in normal fibroblasts and in the nontumorigenic hybrid cells. Furthermore, we were unable to detect p75/150 or its precursors by either cell surface labeling, metabolic labeling, or Western blotting in nontumorigenic cell hybrids; p75/150 thus represents a tumor-specific marker in this system. Tryptic peptides of highly purified p75/150 have been generated, but their amino acid sequences did not reveal any significant homology with previously described proteins.

The leukaemia-associated cell surface antigen p24/BA-2 is a single polypeptide chain with a molecular weight of 24,000. Treatment with glycosidases or exposure of cells to tunicamycin failed to show any change in the molecular weight of the antigen when examined by SDS-polyacrylamide gel electrophoresis. In addition, it failed to bind to lectin affinity columns of concanavalin A, lentil lectin or ricinus communis lectin. This is consistent with the absence of N-asparagine linked oligosaccharide chains on the antigen. Pulse-chase labelling of protein p24 shows a post-translational modification resulting in a molecular weight increase of approx. 500-1000. Alkaline treatment resulted in a decrease in molecular weight of approximately the same amount, suggesting that p24 contain some O-glycosidically linked oligosaccharide. Protein p24 has a basic pI of 7.3 which is unchanged after neuraminidase treatment. Protein P24/Ba-2 cannot be labelled by either the lipophilic photoactivatable nitrene reagent, hexanoyldiiodo-N-(4-azido-2-nitrophenyl)tyramine, or with [32P]phosphate. This suggests that the molecule is non-integral in nature and that it does not form an intimate association with the lipid matrix. Identical molecular weights, when reduced and non-reduced antigens were compared, suggest that it contains no internal disulphide linkages and failure to detect any other band on gradient gel SDS-polyacrylamide gel electrophoresis from 5-15% suggests that is is not strongly associated with any other structure.

The neutral glycolipids and gangliosides of sheep thymus and of sheep peripheral lymphocytes were compared. The patterns of both of these major classes of glycolipids were more complex in thymus than in the lymphocytes. The incorporation of radioactivity from D-[1-14C]galactose into the individual glycolipids of control and concanavalin A-stimulated sheep peripheral lymphocytes was also studied. A marked enhancement of incorporation into trihexosylceramide and an alteration of the pattern of incorporation into gangliosides were noted in the mitogen-treated cells. The results suggest that significant alterations of glycolipid composition and metabolism may occur during at least certain stages of lymphocyte differentiation.

Four distinct rat monoclonal antibodies against the common ALL antigen (CALLA, gp 100) were obtained in a single fusion. Rat AL2, AL3, AL4, AL5 and the previously reported mouse J5 monoclonal antibodies identified the same subsets of leukaemic cells. AL2 and AL3 reacted weakly with terminal transferase-positive cells in normal bone marrow and foetal liver, as well as with a minority of mature granulocytes in blood. Immunoprecipitation experiments and competitive binding assays demonstrated that the four rat antibodies and J5 bound to the same glycoprotein of approximately 100,000 mol. wt. This set of rat monoclonal antibodies directed against CALLA has not only a diagnostic usefulness but may also be of therapeutic value.

An investigation was performed to elucidate the problems of converting a demonstrator expert system for leukaemia diagnosis into a large high-performance system. Analysis of the logical completeness of the demonstrator system revealed only minor errors and a limited representation of the domain. Most of its test cases were quite simple and correction of the minor errors raised the performance to 94%. However, the remaining cases were very complex and it is doubtful whether the system could handle them without fundamental changes to its representation of the domain. Demonstrator systems can function with an imperfect view of the domain since the small sacrifice of accuracy is of little consequence; but in a high-performance system the imperfections must be corrected before extra detail is added. The use of default mechanisms for making specific conclusions was harmful because they concealed errors, and a numerical method for handling uncertainty contributed nothing of value to the system.

A library of monoclonal antibodies plus "conventional' markers (e.g. anti-TdT) have been used to explore the detailed phenotypes of leukaemic cells in relation to normal haemopoietic differentiation. This analysis reveals that consistent, composite phenotypes of different subclasses of lymphoid malignancies closely mimic those of corresponding normal cells at equivalent levels of maturation. It is suggested that three major target cell populations are available for lymphoid malignancy: the pluripotential stem cell (e.g. "lymphoid' blast crisis of CGL), lymphoid progenitors or stem cells in the bone marrow (non-T ALL) or thymus (T-ALL, T-NHL) and long lived, mature and immunocompetent T and B cells (T-CLL, PLL, Sézary and B-CLL, lymphoma respectively). The major phenotypes documented in different leukaemias represent the level of "maturation arrest' imposed on the dominant subclone; this is determined by, but not necessarily synonymous with, the "target cell' and associated clonogenic cell population in the leukaemia. No consistent major abnormalities of gene expression are revealed by this investigation and although they may exist (e.g. loss or acquisition of antigens with malignant progression) we suggest that they are irrelevant to the central issue of what alterations are essential and sufficient for the evolution of clonogenic leukaemic cells. We propose instead that subtle changes, which uncouple proliferation and differentiation, are all that is required.

Because differences in hexosaminidase isoenzyme profiles of granulocytes and lymphoyctes suggested that such profiles might help to distinguish between various types of leukemia, was examined leukocyte extracts from 55 untreated children and 12 controls by automated anion-exchange chromatography. In 23 of 27 cases of the common form of childhood acute lymphoblastic leukemia, the activity of hexosaminidase component I was greatly increased; the activity ratio of hexosaminidase I to hexosaminidase A was greater than 0.5 in 18 of these cases, whereas for other types of leukemia and for all normal cells tested, the ratio was less than 0.2. A raised hexosaminidase I was shown to be associated with leukemic cells by the finding of normal isoenzyme profiles of bone-marrow cells from 17 patients in remission who had such an increase either at diagnosis or in relapse. Hexosaminidase isoenzyme analysis had diagnostic value and may provide a new marker for study of leukocyte differentiation.

Acquired immune deficiency syndrome (AIDS) is characterized by opportunistic infections and by 'opportunistic neoplasms' (for example, Kaposi's sarcoma). Persistent generalized lymphadenopathy (PGL) is epidemiologically associated with AIDS, especially in male homosexuals. A subset of T lymphocytes positive for the CD4 antigen (also termed T4 antigen), is depleted in AIDS and PGL patients. A retrovirus found in T-cell cultures from these patients is strongly implicated in the aetiology of AIDS because of the high frequency of isolation and the prevalence of specific antibodies in the patients. Here we have detected cell-surface receptors for the AIDS retrovirus (human T-cell leukaemia virus-III (HTLV-III) and lymphadenopathy-associated virus-1 (LAV-1) isolates) by testing the susceptibility of cells to infection with pseudotypes of vesicular stomatitis virus bearing retroviral envelope antigens, and by the formation of multinucleated syncytia on mixing virus-producing cells with receptor-bearing cells. Receptors were present only on cells expressing CD4 antigen; among 155 monoclonal antibodies tested, each of the 14 anti-CD4 antibodies inhibited formation of syncytia and blocked pseudotypes. Productive infection of CD4+ cells with HTLV-III or LAV-1 markedly reduced cell-surface expression of CD4. In contrast, receptors for HTLV-I and HTLV-II were not restricted to CD4+ cells, were not blocked by anti-CD4 antibodies; cells productively infected with HTLV-I and HTLV-II expressed surface CD4. Hence, we conclude that the CD4 antigen is an essential and specific component of the receptor for the causative agent of AIDS.

A detailed analysis of lymphoid leukaemia cells with a panel of monoclonal antibodies plus other markers indicates a striking conservation of phenotype that closely mimics that of normal lymphoid precursor or progenitor cells and reflects the imposition of maturation arrest. The composite phenotypes detected show some minimal deviations or asynchronies (with respect to maturation "stages") but are faithful with respect to cell lineage. These observations are suggested to shed light on the available "target" populations for lymphoid malignancy in humans and challenge the view that dedifferentiation or qualitatively aberrant gene expression is common and important in malignancy. They are further taken to support the view that normal gene products that regulate or couple proliferation and maturation are critically involved in the successive clonal selection events underlying the evolution of leukaemia. No qualitatively unique and consistent leukaemia "markers" may exist or be required. The only exception to this rule are the nonrandom chromosome changes, which, it is suggested, involve genetic loci critical to the regulation of growth and maturation in particular normal cell types.

Combined immunologic assays for TdT enzyme and membrane markers show that TdT+ cells in nonleukemic human bone marrow carry ALL-associated and Ia-like antigens but no thymocyte markers or surface Ig. These cells could be precursors involved in acute lymphoblastic leukemia of the "common" or non-T, non-B type and in lymphoid blast crisis of Ph' positive chronic myeloid leukemia. A few TdT+, Ia+ cells express cytoplasmic IgM, indicating that some pre-B cells may be TdT positive.

A non-specific acid alpha-naphthyl-acetate esterase activity was investigated in human lymphoid cell populations from tonsils, blood, thymus and in different leukaemias. Four patterns have emerged. (i) The vast majority of T lymphocytes (peripheral blood, tonsils) showed a localized, intense reaction product ("T-like"). (ii) A thymocyte subpopulation expressed faint, localized enzyme activity ("Thy-like) but the majority of thymocytes showed no esterase activity. Some cells from acute lymphoblastic leukaemias with thymocyte surface characteristics had a "Thy-like" appearance. (iii) Most B lymphocytes and cells from chronic lymphocytic leukaemias were esterase negative. No activity was seen in mitogen activated peripheral blood T and B lymphoblasts from peripheral blood and in leukaemic blasts from the common form of childhood acute "lymphoblastic" leukaemia. (iv) Myeloid cells, including peripheral blood monocytes, and blast cells from acute myeloblastic and chronic granulocytic leukaemias showed an intense, diffuse reaction product (M-like). It is concluded that the modified esterase technique is a convenient marker for small mature human T lymphocytes in tissue sections and smears. It may also be helpful in the differential diagnosis of acute lymphoblastic leukaemia with T characteristics, the common form of acute "lymphoblastic" leukaemia and acute myeloblastic leukaemia.

Rabbit antisera were produced against pooled living lymphocytes from 25 patients with active systemic lupus erythematosus (SLE). Lymphocytes collected at plasmapheresis or venipuncture were frozen in liquid nitrogen and later coated with rabbit antibody to normal human tonsils and normal thymocytes immediately before intravenous immunization of rabbits. Antisera were subsequently extensively absorbed with normal human tonsillar cells, thymocytes, peripheral blood lymphocytes, erythrocytes, and leukocytes from patients with myelogeneous and lymphatic leukemia until residual base-line immunofluorescent staining of normal human lymphocytes using F(ab)2' of whole antisera averaged less than 5%. Absorbed pepsin-digested antisera detected membrane antigens which were markedly increased (mean 32%) on lymphocytes from patients with active SLE (P less than 0.05). Membrane antigens reacting with absorbed, pepsin-digested antisera were present on both T and B cells but, in most instances, predominated on T cells. Control observations using absorbed pepsin-digested antisera to normal human lymphocytes or peripheral blood lymphocytes from patients with rheumatoid arthritis showed no similar specificity. SLE patients treated with moderate or high dose corticosteroids or immunosuppressive agents (cytoxan or azathioprine) appeared to lose lymphocyte antigens detected by these reagents. Control studies with other connective tissue disease patients, miscellaneous hospitalized subjects, or normal controls showed low levels of reactivity (2-5%). SLE lymphocyte membrane antigens uniquely increased during active disease; this may represent neoantigens or alterations associated with the disease itself.

Monosomy 7 was the sole karyotypic abnormality seen in a case of biphenotypic leukaemia involving both the myeloid and lymphoid lineages and in a case of de novo ALL undergoing a phenotypic shift to acute myeloid leukaemia at relapse. These observations suggest that monosomy 7 can be associated with transformation of a common lymphoid-myeloid progenitor cell.

The proportion of lymphocytes from patients with non-Hodgkin lymphoma (NHL) expressing transferrin receptors in biopsy material from lymph-nodes assayed by binding of the monoclonal antibody OKT9 correlated significantly with the histological class of the tumour. High-grade lymphomas contained 22.5% (range 3-57%) OKT9-positive lymphocytes, and low-grade lymphomas contained 2.5% (range less than 1-22). Since histological class of lymphoma is an important factor in survival, it was expected and confirmed that OKT9 levels were related to survival in these patients. Transferrin receptor is expressed by growing cells. The correlates observed in this series of NHL suggests a relation between transferrin receptor and either the growth fraction, or factors affecting the growth fraction, of the tumour.

Transient pancytopenia preceded the onset of acute leukaemia in eight of 360 (2%) of children with acute lymphoblastic leukaemia (ALL) but did not occur in 70 cases of acute non-lymphoblastic leukaemia. The patients developed overt leukaemia within 5-38 weeks of first presentation with features of marrow failure. Immunological classification of blast cells was performed in six of the eight patients and all had phenotype of common-ALL (c-ALL). We conclude that the sydnrome of pre-leukaemia aplasia in childhood is a feature of c-ALL.

Of 597 cases of acute leukaemia in adults (greater than 16 years) seen at St. Bartholomew's Hospital, London, between May 1973 and January 1982, 412 were diagnosed as AML, 103 as ALL and 58 as Philadelphia chromosome positive blast crisis of CML (13 presenting as acute leukaemia and 45 having a prior chronic phase). The remaining 24 cases were considered to be acute undifferentiated leukaemia. Twenty-one of the latter were investigated using a panel of immunological markers at diagnosis and/or retrospectively using frozen cell suspensions. Eighteen out of 21 were shown to have a predominantly 'lymphoid' phenotype which comprised 12 cases of common ALL (two of whom were Ph1 positive), three cases of null-ALL, one case with a probable early thymic phenotype, and two cases with a monoclonal B lymphoblast phenotype. One 'common ALL' and one 'null-ALL' had a significant proportion of pre-B (cytoplasmic mu chain+) cells. One other case reacted with anti-myeloid sera. Leukaemic blasts from two patients were unreactive with all markers tested. No cases of glycophorin positive erythroleukaemia or anti-platelet (glycoprotein I) positive leukaemia were detected. These observations suggest that the overwhelming majority of acute leukaemias have an identifiable affiliation to the lymphoid or myeloid lineages and that patients diagnosed haematologically as 'AUL' might benefit by therapy appropriate for their leukaemic cell type.

A series of monoclonal antibodies and additional selective markers (e.g. TdT) have been used to describe the detailed composite phenotypes expressed by cells in 172 cases of T cell leukemia. The results provide further evidence for a striking correlation between phenotype, maturation arrest, and clinical subgroup, and have important implications for our understanding of the origins of T cell malignancy.

Analysis of leukemic cell phenotypes using cell surface antigens and various enzymes indicates that acute lymphoblastic leukemia (ALL) is a biologically heterogeneous disease consisting of four major subclasses with additional subsets existing within these subclasses. These different types of ALL appear to reflect sequential stages of early lymphocyte ontogeny. There is a strong association between cell phenotype and first remission duration in ALL (p trend less than 0.0001) and an equally strong correlation between remission duration and white blood cell count at presentation. If common ALL and thymic ALL (T-ALL) are compared after adjustment of white blood cell counts, then the prognostic differences between these two major subclasses almost disappear (p = 0.38). It is suggested, therefore, that an immunological (and enzymatic) phenotype of ALL subclasses may not be an independent correlate of prognosis but nevertheless is linked to other differentiation-linked features, especially growth rate and sites of clonal expansion (e.g., marrow versus thymus), which critically influence the size of the clonogenic leukemic population and its associated evolutionary status with respect to drug resistant mutants at the time of diagnosis and introduction of therapy. An extensive library of monoclonal antibodies has been used to further define the phenotypic heterogeneity of T-and non-T All. Several of the antigenic structures identified by these monoclonal antibodies have been isolated and characterized. T-ALL can be dissected into several subsets corresponding to stages of intrathymic differentiation. Non-T ALL (null-ALL, common ALL, and B-ALL) all have a phenotype indicative of B-lineage affiliation indicating that "non-T, non-B" ALL may originate from B-cell precursors in bone marrow. A cell type is identified in normal bone marrow which has the same identical monoclonal antibody-defined phenotype as common ALL and may provide the target cell for this disease.

Between January 1980 and May 1981, 1966 marrow or blood samples from leukaemia patients were tested for terminal deoxynucleotidyl transferase (TdT) using nuclear immunofluorescence. The cells were also tested with a panel of immunological markers including monoclonal antibodies. Of 869 TdT positive cases detected, 555 were diagnosed as ALL and 32 as blast crisis of CGL; 226 were provisionally diagnosed as 'acute leukaemia' and finally diagnosed as ALL partly on the basis of immunological data; 56 TdT+ cases were provisionally diagnosed as acute non-lymphocytic or myeloid leukaemia; 266 cases of AML and 177 cases of CGL in blast crisis were TdT negative. Eleven of the above 'AML' cases were anti-cALL+ as well as TdT+ and were re-diagnosed and treated successfully as cALL. The remaining 45 were anti-cALL negative and finally diagnosed and treated, at least initially, as AML. Eleven of these cases had only 5-10% TdT+ cells which could have been normal, non-myeloid cells. Twenty cases had 11-50% TdT+ cells and 14 cases had 50-100% TdT+ cells. Of these latter two groups, details on 28 patients were available for evaluation. Three cases on review had no definitive myeloid cytochemistry and were haematologically AUL with a null-ALL phenotype (TdT+ DR+ cALL-). In 14 cases there was a large overlap (greater than 75%) of the proportion of cells with myeloid cytochemistry (Sudan black, peroxidase or esterases) and TdT; individual blast cells were therefore expressing these markers concurrently. In the remaining cases, mixtures of TdT negative myeloid and TdT+ (lymphoid?) cells may have coexisted although this was not proven unequivocally. Twenty-two cases of newly diagnosed TdT+ 'AML' received induction chemotherapy for AML (DAT regime) and only six (37%) obtained a complete remission. It is concluded that TdT positive 'myeloid' leukaemias do occur, albeit infrequently (approx. 5%) and may have a relatively poor prognosis.

A permanent hematopoietic cell line, designated NALM-1, was established from the peripheral blood of a patient who was in blastic crisis of Ph1-positive chronic myleocytic leukemia. By means of a panel of specific xenoantisera, the NALM-1 cells were found to express a specific antigen of acute lymphoblastic leukemia and blast leukemia-associated antigen. The cells exhibited no cell-surface receptors for sheep erythrocytes, IgG, or complement; neither cell-surface immunoglobulins nor cytoplasmic immunoglobulin were observed. Furthermore, normal T-cell or B-cell antigens, detectable by the antisera used in this study, were not found in the NALM-1 line.

As tested by DNase I hypersensitivity, the chromatin structure of the IgH enhancer region in human B cell precursor cell lines is in an open or accessible conformation. All T cell lines, with either germline or rearranged IgH genes, were also hypersensitive to DNase I but in contrast to B cell precursors showed no detectable Cmu expression. Normal thymocytes similarly had a hypersensitive IgH enhancer site. In contrast to lymphoid cells, all myeloid cell lines tested, as well as normal granulocytes, were not DNase I hypersensitive and did not express Cmu. A putative lymphomyeloid progenitor cell line KG1, although having a germline configuration of Ig genes, produced Cmu transcripts and was hypersensitive to DNase I in the IgH enhancer region. After induction of myeloid differentiation the Ig enhancer region of KG1 cells is no longer hypersensitive or transcriptionally active. Two HhaI restriction sites on either side of the IgH enhancer were not methylated in all Cmu-expressing lines but methylated in non-expressing cell lines. These results show that an open chromatin structure around the heavy chain enhancer is necessary but insufficient for initiating transcription from unrearranged IgH genes and further suggests this region may be in an open or accessible configuration prior to lineage commitment and closed following adoption of the myeloid lineage.

Monoclonal antibody 38.13 was raised against the Burkitt lymphoma cell line Daudi and has been previously shown to recognize a globotriaosylceramide (Pk) determinant expressed selectively on Burkitt and a few other B-cell lymphomas. We report here that 38.13 binds to a subset of normal B lymphocytes in germinal centres of secondary lymphoid follicles as well as to blood vessels. In malignant populations equivalent to various developmental compartments of the B-cell lineage, only those with a follicular phenotype express the Pk determinant at detectable levels. The expression of this tumour-associated antigen, in common with many others, is therefore regulated in concert with cell lineage and maturation status.