Murine terminal deoxynucleotidyl transferase: cellular distribution and response to cortisone.

The mouse thymus contains two forms of terminal deoxynucleotidyl transferase (TdT) which are distinguishable by the salt concentration necessary to elute them from a phosphocellulose column, by their distrubtion among the thymocyte subpopulations, and by their sensitivity to cortisone treatment. In the whole thymus the later eluting peak (peak II) is the predominant one with about 3-10% of the total activity appearing in peak I. Both peak I and peak II activities are most sensitively assayed by the polymerization of dGMP onto an oligo(dA) primer. The minor population of thymocytes which is less dense and cortisone-resistant contains a higher specific activity of peak I TdT. The majority of TdT activity is, however, found in the major population of thymocytes which occurs in the center region of a bovine serum albumin gradient and is cortisone-sensitive. A very low level of an activity indistinguishable from peak II TdT activity is also detected in the mouse bone marrow. Other tissues, such as spleen, liver, heart, and brain lack detectable amounts of TdT activity.

Analysis of rat hemopoietic cells on the fluorescence-activated cell sorter. II. Isolation of terminal deoxynucleotidyl transferase-positive cells.

A method is described by which highly enriched populations of viable terminal deoxynucleotidyl transferase-positive (TdT+) cells can be isolated from rat bone marrow by use of the fluorescence-activated cell sorter. Such cells have been postulated to be progenitors of thymocytes and, possibly, of B lymphocytes, and may serve as the targets of neoplastic transformation in acute lymphoblastic leukemia. The separation procedure is based on differences in relative low-angle light scatter and relative fluorescence intensity for Thy-1 antigen between TdT+ cells and other lymphohemopoietic cell populations in bone marrow. Simultaneous sorting of bone marrow cells according to these two parameters resulted in a mean 87% purification of TdT+ cells. The morphological characteristics of the isolated TdT+ cells are described at the light and electron miscroscopic levels.

Defective lymphopoiesis in bone marrow of motheaten (me/me) and viable motheaten (mev/mev) mutant mice. I. Analysis of development of prothymocytes, early B lineage cells, and terminal deoxynucleotidyl transferase-positive cells.

This study identifies defects in the early stages of lymphopoiesis that may contribute to the abnormalities in the development and/or function of peripheral T and B lymphocytes in mice homozygous for the motheaten (me/me) and viable motheaten (mev/mev) mutations. The results indicate that in me/me and mev/mev mice prothymocytes in bone marrow are present in essentially normal numbers, as determined by intrathymic injection, but apparently lack the ability to home effectively to the thymus, as determined by intravenous transfer; early B lineage cells in bone marrow, identified by the B220 antigen, are markedly depleted, including immature B cells (sIg+), pre-B cells (cIg+, sIg-), and pro-B cells (B220+, cIg-, sIg-); TdT+ bone marrow cells, especially a subset that expresses the B220 B lineage antigen, are markedly depleted by two weeks of age; normal numbers of TdT+ thymocytes are present during the first 3 wk of postnatal life, but rapidly decrease thereafter. The results further indicate that neither the defective thymus homing capacity of prothymocytes nor the deficiency of TdT+ bone marrow cells is due to autoantibodies. The possible relationship of the defective development of lymphoid precursor cells to the premature onset of thymic involution and to the abnormalities of peripheral T and B lymphocytes in me/me and mev/mev mice is discussed; as are the results of in vitro studies (presented in a companion paper), which suggest that a primary defect in the stromal microenvironment of the bone marrow is responsible for the abnormal development of the lymphoid precursor cells.

High frequency of clonal immunoglobulin or T cell receptor gene rearrangements in acute myelogenous leukemia expressing terminal deoxyribonucleotidyltransferase.

Ig and T cell receptor rearrangements have been used as irreversible markers of lineage and clonality in the study of B- and T-lymphoid populations. We have addressed the issue of lymphoid lineage specificity of these rearrangements by analyzing a panel of 25 TdT- acute myelogenous leukemias, 13 TdT+ AMLs, and 4 TdT+ undifferentiated leukemias. We report that while gene rearrangements represent extremely rare events in classical TdT- AML (less than 8%), rearrangements of either the Ig or T beta locus or both were detectable in the majority of the TdT+ AMLs (greater than 60%), and rearrangements of both loci were detectable in all of the TdT+ undifferentiated leukemias. These data demonstrate a significant association between TdT expression and Ig or T beta gene rearrangements even outside the lymphoid lineage, further supporting a role for TdT in Ig and T cell receptor gene assembly. These data also indicate that a coordinated program of lymphoid gene expression involving TdT-CD7-expression and Ig/T beta rearrangements can be activated before myeloid commitment. Whether the activation of this program represents a normal, albeit rare, event in early myelopoiesis or a transformation-related event present only in leukemic cells remains to be determined.

A selective culture system for generating terminal deoxynucleotidyl transferase-positive (TdT+) lymphoid precursor cells in vitro. I. Description of the culture system.

A primary xenogeneic culture system has been devised that selectively generates undifferentiated TdT+ lymphoblasts from rat bone marrow under conditions that do not support the growth or maintenance of rat colony-forming unit-spleen (CFU-S) or granulocyte/macrophage colony-forming cells (GM-CFC). The culture system requires a mouse bone marrow feeder layer, and a serum supplement that has markedly reduced levels of cortisol. The growth of TdT+ cells can be significantly enhanced by the addition of mesodermalizing factors (e.g., fibroblast growth factor, guinea pig bone marrow extract) to the culture medium, and the serum supplement can be decreased by the addition of selenium, transferrin, and T3. The cultured TdT+ cells are antigenically "null" cells that further resemble their normal counterparts in bone marrow with respect to morphology, size, cortisone sensitivity, and pattern of TdT fluorescence. The TdT+ cells are generated with equal facility from bone marrow of normal and congenitally athymic rats, can be maintained in logarithmic growth for at least 10 mos by serial passage in vitro, and do not cause leukemia when infused into irradiated recipients. Although the lineage relationships of these immature lymphoid cells have not yet been established, our working hypothesis, based on preliminary evidence, is that the cultured TdT+ cells are primitive members of the T cell series.

I. Quantitative determination of the amount of DNA per nucleus by interference microscopy.

A method for the determination of the DNA content of isolated nuclei of different ploidy has been developed. It is based on measurement of the nuclear dry mass, with an integrating microinterferometer, before and after DNase treatment. The values found are slightly low, because, as indicated by biochemical determinations, consistently 5% to 8% of DNA is not extracted by DNase under these conditions. The average DNA values thus obtained for diploid and tetraploid nuclei of adult rat liver are 7.7 and 15.6 pg (10(-12) g), respectively. Definite advantages of this procedure are: i) comparisons with biochemical determinations to give DNA values for each class of ploidy, ii) comparisons with histophotometry of the Feulgen dye-DNA complex to give absolute values instead of arbitrary units.

Mammalian DNA polymerase: separation of binding from incorporation of deoxyribonucleoside triphosphates.

An assay for the binding of deoxyribonucleoside triphosphate with an enzyme-DNA complex has been developed. This binding requires active enzyme and magnesium ion, takes place equally with native or denatured DNA, and may proceed in the absence of demonstrable DNA synthesis. The binding reaction appears to be specific for deoxyribonucleoside triphosphates, and studies on competition indicate that one active site accommodates the four common triphosphates.

Reversion of murine sarcoma virus transformed mouse cells: variants without a rescuable sarcoma virus.

Murine sarcoma virus transformed mouse 3T3 cells, which are negative for murine leukemia virus and which yield sarcoma virus after superinfection with murine leukenmia virus, spotaneously give rise to flat variants front which murine sarcoma virus can no longer be rescued. The revertants support leukemia viruis growth and show an enhanced sensitivity to murine sarcoma superinfection and, like normal cells, do not release RNA-dependent DNA polymerase activity. Because revertants could be obtained with high frequency from progeny of single transformed cells, each cell that containts the sarconma virus genome seems to have the capacity to suppress or eliminate an RNA tumor virus native to its species of origin.

Induction of murine C-type viruses from clonal lines of virus-free BALB-3T3 cells.

Each clone of BALB/c mouse embryo cells that has been tested can be induced to form C-type virus. The individual cells therefore contain a complete copy of the genetic information for making the murine RNA tumor viruses.

Human DNA polymerase 3 (R-DNA polymerase): distinction from DNA polymerase I and reverse transcriptase.

DNA polymerase III is an enzyme activity in eukaryotic cells which under certain conditions shows strong preference for polyadenylic acid as template when primed by oligodeoxythymidylate. Its first complete separation from other DNA polymerases in human lymphoblasts is reported. This enzyme is biochemically and immunologically distinct from DNA polymerase I and from viral reverse transcriptase from a primtate type C virus.

Absence of polymerase protein in virions of alpha-type rous sarcoma virus.

Noninfectious particles of a mutant of Rous sarcoma virus failed to exhibit DNA polymerase activity even with the use of the most sensitive synthetic template-primer complexes. A neutralization blocking test against antibody to DNA polymerase revealed that these mutants did not contain protein immunologically related to the DNA polymerase.

Antigenic relationships in mammalian DNA polymerase.

Rabbit antibody was prepared against a high-molecular-weight DNA polymerase purified from the soluble fraction of calf thymus gland. This antibody does not inhibit terminal deoxynucleotidyl transferase isolated from that source, but does inhibit both low-molecular-weight and high-molecular-weight DNA polymerases isolated from cytoplasmic and nuclear fractions of a number of mammalian tissues (mouse L cells, calf thymus, phytohemagglutinin-stimulated human lymphocytes, rat liver, and rabbit bone marrow). The results suggest that (i) no antigenic relationship exists between terminal transferase and DNA polymerase, (ii) common antigenic determinants exist in the DNA polymerases from all mammalian sources, and (iii) multiple forms of DNA polymerase found in mammalian, cells are related by having polypeptide sequences or subunits in common.

Reverse transcriptases of primate viruses as immunological markers.

Antibodies were prepared against the DNA polymerases (reverse transcriptases) of three potentially oncogenic RNA viruses of primates. Two type C viruses, isolated from a woolly monkey fibrosarcoma and from a gibbon ape lymphosarcoma, have polymerases that are immunologically related to each other and are distinct from the type C viruses isolated from other mammals.

Terminal deoxynucleotidyltransferase distribution in neoplastic and hematopoietic cells.

In the present study, terminal deoxynucleotidyltransferase was examined in the peripheral blood and (or) bone marrow of 115 children with a variety of neoplastic, hematologic, and other unrelated disorders. Terminal deoxynucleotidyltransferase activity was present at 4.08+/-0.74 U/108 cells in 23 morphologicall normal bone marrow samples from childhood controls. Terminal transferase was present at greater than 23 U/108 nucleated cells and at greater than31 U/108 blasts in the bone marrow of all children with acute lymphoblastic leukemia studied at initial diagnosis and at disease relapse. Terminal deoxynucleotidyltransferase was detectable at low levels, less than 7.5 U/108 cells, in all remission marrow smaples. Bone marrow terminal transferase activity was markedly elevated in all untreated acute lymphoblastic leukemia patients, whereas low levels which were difficult to interpret were present in the peripheral blood samples of two patients at diagnosis and six patients at relapse who had low absolute lymphoblast counts. Because of greater variation in the lymphoblast content of peripheral blood, bone marrow assays are more reliable in detecting disease activity. Marrow terminal deoxynucleotidyltransferase values obtained during the active phase of acute lymphoblastic leukemia were significantly greater than those found in other types of leukemia, bone marrow malignancies, and hematologic disorders. Terminal transferase determinations in blast cells of two patients with leukemic conversion of non-Hodgkin's lymphoma and in tumor cells from one patient with Burkitt's lymphoma were within the control range. These dat further define the usefulness of terminal deoxynucleotidyltrnasferase assay in the differentiation and classication of hematologic malignancies.

A novel fluorescence-based system for assaying and separating live cells according to VDJ recombinase activity.

We describe two retroviral vector-based recombination substrate systems designed to assay for lymphoid VDJ recombinase activity in cultured cells. Both substrates incorporate a constitutive dominant marker gene (the simian virus promoter-driven neo gene) to allow selection of cells that stably integrate the substrate. Both substrates also include a second marker gene that becomes transcriptionally active only when inverted by a site-specific recombination event between flanking immunoglobulin variable-region gene segments. The first vector, similar in structure to previous retrovirus-based recombination substrates, utilizes the bacterial guanine-xanthine phosphoribosyltransferase gene (gpt) as its activatable marker; detection of inversion (VDJ recombinase activity) involves drug selection and Southern blotting analyses. We have used this vector to make a more extensive and quantitative survey of VDJ recombinase activity in B-lineage cell lines than has previously been performed with stable substrates, and we have compared our results with those of other studies that use transient recombination substrates. In the second vector, the activatable gene is the bacterial beta-galactosidase gene (lacZ). Detection for inversional activation of this gene is achieved by a fluorogenic assay, termed FACS-Gal, that detects beta-galactosidase activity in viable cells. The latter assay has the unique advantage of rapidly detecting cells that undergo recombination and also allows viable sorting of cells on the basis of the presence or absence of VDJ recombinase activity. We have used the lacZ vector to rapidly quantitate VDJ recombinase activity in B-lineage cell lines and compared the results with those obtained with the gpt vector. We have also used the lacZ vector to isolate variant pre-B-cell lines with low and high levels of VDJ recombinase activity.

Analysis of platelets from patients with thrombocythemia for reverse transcriptase and virus-like particles.

A preliminary analysis of an RNA-directed DNA polymerase was made and a C-type virus-like particle was identified in platelets from 2 patients with the myeloproliferative disorder thrombocythemia (primary, essential, hemorrhagic, or idiopathic thrombocythemia). Platelet homogenates were centrifuged through a sucrose equilibrium density gradient. Both endogenous and exogenous DNA polymerase activity was found at a density of 1.19 g/ml. No activity was seen at comparable densities in control gradients. Electron micrographs of thin sections of these platelets revealed a particle with the morphologic characteristics of a C-type virus; however, the diameter of this particle was about 80 nm, slightly lower than that commonly found for C-type particles. Critical-point dried specimens, from the fractions of the sucrose gradient at which DNA polymerase activity was found, contained particles of the same size and morphology as those in the thin sections.

Adenosine deaminase and terminal deoxynucleotidyl transferase: biochemical markers in the management of chronic myelogenous leukemia.

Serial determinations of adenosine deaminase (ADA) activity in 69 patients with chronic myelogenous leukemia provided a biochemical marker of disease activity. Eighty-nine % of patients in the accelerated phase had an elevation of ADA activity. This elevation was not a direct reflection of an increased absolute blast count. Furthermore, five of seven patients studied serially from the stable phase into the accelerated phase had an increase in ADA activity before the absolute blast count increased. This is the first investigation which clearly demonstrates the potential value of measuring serial ADA activities in a large number of patients with chronic myelogenous leukemia.

Estimation of tumor growth fraction in murine tumors by the primer-available DNA-dependent DNA polymerase assay.

The tumor growth fraction measured by the percentage labeled mitoses method has been determined in transplantable solid and ascites murine tumors, the latter being measured at different times after transplantation. These values were compared to an in vitro, autoradiographic assay that determines the fraction of cells in a given population (primer-available DNA-dependent DNA polymerase index) that have both nuclear DNA-dependent DNA polymerase and DNA capable of acting as primer-template. It appears that almost all cells with a short G1 phase duration (less than 19 hr) that are within the proliferative cycle are primer-available DNA-dependent DNA polymerase positive. The results of the comparison indicate that the primer-available DNA-dependent DNA polymerase index estimation of growth fraction is very nearly identical to the growth fraction measured by the percentage labeled mitoses method.

Mitoxantrone, vincristine, and prednisone in adults with relapsed or primarily refractory acute lymphocytic leukemia and terminal deoxynucleotidyl transferase positive blastic phase chronic myelocytic leukemia.

A combination of mitoxantrone, vincristine, and prednisone was used to treat 19 patients with acute lymphocytic leukemia. Of these, 12 were patients with acute lymphoblastic leukemia (ALL) (9 in first relapse and 5 primarily refractory to standard induction therapy with daunorubicin, vincristine, and prednisone), 2 had a phenotypic ALL relapse after an initial diagnosis of acute myelocytic leukemia and 5 had terminal deoxynucleotidyl transferase positive blastic phase chronic myelogenous leukemia (BCML). Eight patients with ALL (and of these, four with primarily anthracycline resistant disease), and two with BCML achieved complete remission. Five patients died in induction (three ALL from sepsis and two BCML from bleeding), and five had progressive disease. Median duration of response was 5 months, with two primarily refractory ALL patients remaining in continuing complete remission at 28 and 31 months. Treatment was well tolerated, with minimal nausea and vomiting, and oral mucositis. Posttreatment transient hepatic dysfunction was seen in 80% of patients. Mitoxantrone, vincristine, and prednisone are an active combination for the treatment of relapsed or refractory ALL and of terminal deoxynucleotidyl transferase positive BCML. The finding that four of five primarily refractory ALL patients were induced in complete remission supports the contention that mitoxantrone and anthracyclines are not cross-resistant.

The effect of different buffers on terminal deoxynucleotidyl transferase activity.

The polymerization of dATP, dCTP, and dGTP onto the defined length initiator, d(pA)10, has been carried out in four buffers. The relative effectiveness of the buffers for the polymerization of each deoxynucleoside triphosphate decreased in the order: cacodylate, 2(N-morpholino)ethane sulfonic acid, N-2-hydroxyethylpiperazine-N'-2-ethanesulfonic acid, and Tris. With the poorer buffers, activity could be increased by the addition of KCl: this effect is not primarily due to an increase in ionic strength. With dGTP as the substrate, but not with dATP or dCTP, activity increased when the concentration of the more active buffers was raised beyond 0.2 M.