Terminal deoxynucleotidyltransferase of 60,000 daltons from mouse, rat, and calf thymus. Purification by immunoadsorbent chromatography and comparison of peptide structures

Terminal deoxynucleotidyl transferase: the story of a misguided DNA polymerase

Nearly every DNA polymerase characterized to date exclusively catalyzes the incorporation of mononucleotides into a growing primer using a DNA or RNA template as a guide to direct each incorporation event. There is, however, one unique DNA polymerase designated terminal deoxynucleotidyl transferase that performs DNA synthesis using only single-stranded DNA as the nucleic acid substrate. In this chapter, we review the biological role of this enigmatic DNA polymerase and the biochemical mechanism for its ability to perform DNA synthesis in the absence of a templating strand. We compare and contrast the molecular events for template-independent DNA synthesis catalyzed by terminal deoxynucleotidyl transferase with other well-characterized DNA polymerases that perform template-dependent synthesis. This includes a quantitative inspection of how terminal deoxynucleotidyl transferase binds DNA and dNTP substrates, the possible involvement of a conformational change that precedes phosphoryl transfer, and kinetic steps that are associated with the release of products. These enzymatic steps are discussed within the context of the available structures of terminal deoxynucleotidyl transferase in the presence of DNA or nucleotide substrate. In addition, we discuss the ability of proteins involved in replication and recombination to regulate the activity of the terminal deoxynucleotidyl transferase. Finally, the biomedical role of this specialized DNA polymerase is discussed focusing on its involvement in cancer development and its use in biomedical applications such as labeling DNA for detecting apoptosis.

Interaction of DNA polymerases with phospholipids

Effects of various phospholipids on the in vitro reactions of eukaryotic DNA polymerases alpha, beta and gamma were tested systematically. When phospholipids were added directly to the reaction mixture, neither stimulation nor inhibition was produced. However, when phospholipids were preincubated with enzymes in the absence of template-primer, some of them showed strong inhibition. Cardiolipin strongly inhibited the reactions of all three DNA polymerases and also of terminal deoxynucleotidyl transferase. Phosphatidylinositol selectively inhibited the reaction of DNA polymerase gamma. Phosphatidic acid moderately inhibited DNA polymerase alpha and strongly inhibited DNA polymerase gamma. The inhibition of DNA polymerase gamma by cardiolipin was nearly competitive with template-primer. Since the inhibition was reversed by the addition of 0.05% Triton-X 100 during preincubation, the phospholipid might interact with enzyme protein at the hydrophobic region in competition with template-primer. These results suggest a possible involvement of phospholipids in DNA replication in mitochondria and in nucleus through interaction with DNA polymerase.

Analysis of human terminal deoxynucleotidyl transferase cDNA expressible in mammalian cells

Human Molt3 cDNA library was constructed using pcD vector system which permits the expression of cDNA inserts in mammalian cells. Nearly full-length human terminal deoxynucleotidyltransferase (TdT) cDNA was cloned using a fragment of bovine TdT cDNA as a probe. The human TdT cDNA contains an open reading frame of 1,557 bp coding for 519 amino acids, including 31 bp and 341 bp from 5' and 3' untranslated regions, respectively. The TdT cDNA was transfected into COS7 monkey fibroblasts directed the synthesis of enzymatically active protein of Mr 59,495. The cloned TdT cDNA hybridized with poly A+ RNAs of 2,100 b and 3,300 b from stable T-cell leukemia Molt3 and Molt4 cells.

Genetic relatedness of human DNA polymerase beta and terminal deoxynucleotidyltransferase

The Protein Identification Resource (PIR) protein sequence data bank was searched for sequence similarity between known proteins and human DNA polymerase beta (Pol beta) or human terminal deoxynucleotidyltransferase (TdT). Pol beta and TdT were found to exhibit amino acid sequence similarity only with each other and not with any other of the 4750 entries in release 12.0 of the PIR data bank. Optimal amino acid sequence alignment of the entire 39-kDa Pol beta polypeptide with the C-terminal two thirds of TdT revealed 24% identical aa residues and 21% conservative aa substitutions. The Monte Carlo score of 12.6 for the entire aligned sequences indicates highly significant aa sequence homology. The hydropathicity profiles of the aligned aa sequences were remarkably similar throughout, suggesting structural similarity of the polypeptides. The most significant regions of homology are aa residues 39-224 and 311-333 of Pol beta vs. aa residues 191-374 and 484-506 of TdT. In addition, weaker homology was seen between a large portion of the 'nonessential' N-terminal end of TdT (aa residues 33-130) and the first region of strong homology between the two proteins (aa residues 31-128 of Pol beta and aa residues 183-280 of TdT), suggestive of genetic duplication within the ancestral gene. On the basis of nucleotide differences between conserved regions of Pol beta and TdT genes (aligned according to optimally aligned aa sequences) it was estimated that Pol beta and TdT diverged on the order of 250 million years ago, corresponding roughly to a time before radiation of mammals and birds.

Poly(ADP-ribosyl)ation of terminal deoxynucleotidyl transferase in vitro

The activity of purified bovine thymus terminal deoxynucleotidyl transferase was markedly inhibited when the enzyme was incubated in a poly(ADP-ribose)-synthesizing system containing purified bovine thymus poly(ADP-ribose) polymerase, NAD+, Mg2+ and DNA. All of these four components were indispensable for the inhibition. The inhibitors of poly(ADP-ribose) polymerase counteracted the observed inhibition of the transferase. Under a Mg2+-depleted and acceptor-dependent ADP-ribosylating reaction condition [Tanaka, Y., Hashida, T., Yoshihara, H. and Yoshihara, K. (1979) J. Biol. Chem. 254, 12433-12438], the addition of terminal transferase to the reaction mixture stimulated the enzyme reaction in a dose-dependent manner, suggesting that the transferase is functioning as an acceptor for ADP-ribose. Electrophoretic analyses of the reaction products clearly indicated that the transferase molecule itself was oligo (ADP-ribosyl)ated. When the product was further incubated in the Mg2+-fortified reaction mixture, the activity of terminal transferase markedly decreased with increase in the apparent molecular size of the enzyme, indicating that an extensive elongation of poly(ADP-ribose) bound to the transferase is essential for the observed inhibition. Free poly(ADP-ribose) and the polymer bound to poly(ADP-ribose) polymerase were ineffective on the activity of the transferase. All of these results indicate that the observed inhibition of terminal transferase is caused by the poly(ADP-ribosyl)ation of the transferase itself.

Molecular biology of terminal transferase

Terminal transferase is an unusual deoxynucleotide polymerizing enzyme found only in prelymphocytes. The protein was purified to homogeneity from calf thymus glands in 1971 as a 32 kDa protein with a two peptide structure. Subsequent biochemical and immunological analyses of terminal transferase protein in crude extracts from a number of animal species showed a single peptide with a molecular weight of about 58,000. The two peptide structure found earlier was caused by proteolysis. Homogeneous 58 kDa terminal transferase has now been produced from human lymphoblastoid cells and calf thymus glands by immunoaffinity chromatography. In vitro phosphorylation studies showed that the terminal transferase protein contains one phosphorylation site near one end of the polypeptide chain, and the phosphorylation of the enzyme has been confirmed by in vivo labeling experiments. Unambiguous demonstration of the molecular weight of the human terminal transferase was obtained by translation of the cloned human terminal transferase DNA sequence to a 58,308 Da protein. The translated amino acid sequence also provided a possible phosphorylation site near the amino-terminus of the protein. Preliminary analysis of the genomic structure shows a simple intron/exon pattern with the total human terminal transferase gene spanning at least 65 Kb.

Uracil-DNA glycosylase in HeLa S3 cells: interconvertibility of 50 and 20 kDa forms and similarity of the nuclear and mitochondrial form of the enzyme

Uracil-DNA glycosylase activity in HeLa S3 cells was found in nuclei (70%), mitochondria (15%) and cytosol (15%) after fractionation in hypotonic buffers. After fractionation in isotonic buffers the activity in cytosol was increased, apparently as a consequence of leakage from the nuclei. Both in the nuclear and the mitochondrial fraction, a major 50 and a minor 18 kDa form were found after gel filtration in the presence of 0.5 M NaCl. However, after glycerol, gradient sedimentation or gel filtration in the presence of 2 M NaCl or 20% glycerol most of the 50 kDa form dissociated into a 22 kDa form, which was also the smallest catalytically active form found after partial trypsin digestion. The dissociation of the 50 kDa form was reversible. Biochemical properties of the nuclear and mitochondrial forms were very similar. Thus, they had similar apparent Km values, pH optima, heat sensitivities and activation energies, and were stimulated 2-5-fold by 40-60 mM monovalent salt.

Disappearance of terminal deoxynucleotidyltransferase in human malignant T-lymphoblasts treated with a human alpha interferon preparation

Human T-lymphoblastoid cell lines RPMI 8402, MOLT-3, and CCRF-CEM were treated with interferon (IFN) to determine if the treatment would result in the disappearance of cellular terminaldeoxynucleotidyltransferase (TdT), a possible differentiation marker for T-lymphocytes. Incubation of RPMI 8402 cells in the presence of IFN preparation caused a decrease in the number of TdT-positive cells and in TdT activity of the cell extract. The inhibition of cell multiplication was dose dependent. The anticellular effect of IFN preparation was cytostatic, not cytocidal. The IFN preparation modified neither the TdT content nor proliferation of MOLT-3 and CCRF-CEM cell lines. The effects of IFN preparation thus varied with the cell line.

Intracellular localization and metabolism of DNA polymerase alpha in human cells visualized with monoclonal antibody

Indirect immunofluorescence microscopy with monoclonal antibody against DNA polymerase alpha revealed the intranuclear localization of DNA polymerase alpha in G1, S, and G2 phases of transformed human cells, and dispersed cytoplasmic distribution during mitosis. In the quiescent, G0 phase of normal human skin fibroblasts or lymphocytes, the alpha-enzyme was barely detectable by either immunofluorescence or enzyme activity. By exposing cells to proliferation stimuli, however, DNA polymerase alpha appeared in the nuclei just prior to onset of DNA synthesis, increased rapidly during S phase, reached the maximum level at late S and G2 phases, and was then redistributed to the daughter cells through mitosis. It was also found that the increase in the amount of DNA polymerase alpha by proliferation stimuli was not affected by inhibition of DNA synthesis with aphidicolin or hydroxyurea.

Structural polymorphism of mouse complement C2 detected by microscale peptide mapping: linkage to H-2

Complement C2 was isolated from 17 mouse strains by immunoprecipitation and sodium dodecyl sulfate-polyacrylamide gel electrophoresis and examined for structural polymorphism by using micro-peptide mapping. By comparing the peptide maps of tryptic digest of C2 from various strains, two allotypic variations were detected. B10 and 14 other mouse strains demonstrated C2.1 type, while a wild mouse line (M.Mol-Ohm) and one B10 congenic strain, B10.MOL.OHM, which carries the H-2 derived from M.Mol-Ohm, demonstrated C2.2 type. (B10 X B10.MOL.OHM)F1 demonstrated codominantly expressed C2 type (C2.1.2). Desialation of mouse C2 did not abolish the observed variation of mouse C2. It is concluded that an H-2-linked codominant locus controls the structure of mouse complement C2, further confirming the extensive homology of the major histocompatibility complex among higher vertebrate species.

DNA primase associated with 10 S DNA polymerase alpha from calf thymus

Among multiple subspecies of DNA polymerase alpha of calf thymus, only 10 S DNA polymerase alpha had a capacity to initiate DNA synthesis on an unprimed single-stranded, circular M13 phage DNA in the presence of ribonucleoside triphosphates (DNA primase activity). The primase was copurified with 10 S DNA polymerase alpha through the purification and both activities cosedimented at 10 S through gradients of either sucrose or glycerol. Furthermore, these two activities were immunoprecipitated at a similar efficiency by a monoclonal antibody directed against calf thymus DNA polymerase alpha. These results indicate that the primase is tightly bound to 10 S DNA polymerase alpha. The RNA polymerizing activity was resistant to alpha-amanitin, required high concentration of all four ribonucleoside triphosphates (800 microM) for its maximal activity, and produced the limited length of oligonucleotides (around 10 nucleotides long) which were necessary to serve as a primer for DNA synthesis. Covalent bonding to RNA to DNA was strongly suggested by the nearest neighbour frequency analysis and the DNAase treatment. The DNA synthesis primed by the RNA oligomers may be carried out by the associating DNA polymerase alpha because it was strongly inhibited by araCTP, resistant to d2TTP, and was also inhibited by aphidicolin but at relatively high concentration. The primase preferred single-stranded DNA as a template, but it also showed an activity on the double-stranded DNA from calf thymus at an efficiency of approx. 10% of that with single-stranded DNA.

Two proteolytic degradation products of calf-thymus poly(ADP-ribose) polymerase are efficient ADP-ribose acceptors. Implications for polymerase architecture and the automodification of the polymerase

Two polypeptides with molecular masses of 76 and 59 kDa were found to copurify with poly(ADP-ribose) polymerase from calf thymus, and to be as efficient acceptors of ADP-ribose as the polymerase itself. Analysis of their CNBr fragments by sodium dodecylsulfate/polyacrylamide gel electrophoresis revealed that the polypeptides were derived from the 112-kDa polymerase. Isolation of poly(ADP-ribose) polymerase in the absence of protease inhibitors resulted in a loss of more than 90% of the polymerase activity and an increased proportion of the 76-kDa and 59-kDa polypeptides in the final polymerase preparation. When the polymerase and the two polypeptides were separated by gel filtration or polyacrylamide gel electrophoresis in 5% acetic acid, no polymerase activity was found associated with the two fragments. Analysis of the CNBr fragments of the three polypeptides after incubation of the enzyme preparation with [32P]NAD showed that most of the fragments were radioactive, indicating multiple ADP-ribosylation sites. Several ADP-ribosylated fragments were found to be common to all three polypeptides, or to two of them.

Characterization of human cytomegalovirus-induced DNA polymerase and the associated 3'-to-5', exonuclease

A DNA polymerase activity induced by human cytomegalovirus (HCMV) was separated from host cell DNA polymerase and purified by phosphocellulose and DNA-cellulose column chromatography. The DNA polymerase activity was strongly inhibited by phosphonoacetic acid, aphidicolin, araATP, and N-ethylmaleimide, but it was resistant to 2',3'-dideoxyTTP. The sensitivity of HCMV-induced DNA polymerase to these reagents resembles that of host cell DNA polymerase alpha. However, HCMV-induced DNA polymerase activity was stimulated several fold by 100 mM ammonium sulfate, by which DNA polymerase alpha activity was strongly inhibited. Furthermore, it was found that a 3'-to-5' exonuclease activity was tightly associated with the HCMV-induced DNA polymerase. The exonuclease was also stimulated by ammonium sulfate, was inhibited by phosphoacetic acid, and it preferred single-stranded DNA as a substrate. The results suggest that the 3'-to-5' exonuclease may play a role in proofreading in the polymerization process as an integral part of the HCMV-induced DNA polymerase.

Growth of a cultured leukemia subline was promoted by conditioned medium of thymic reticuloepithelial-like cells (B6TE)

A murine leukemia subline (L17R) was selectively developed in the presence of conditioned medium of a thymic reticuloepithelial-like cell line (B6TE). Cytotoxicity tests and immunofluorescence microscopy showed that L17R cells were negative in the expression of Thy 1.1, Lyt 1.2 and terminal deoxynucleotidyl transferase (TdT), however, 35% positive in Lyt 2.1 phenotype, and 95% positive in the expression of peanut agglutinin (PNA) receptor. B6TE conditioned medium had no activity of interleukin 1 (IL 1), interleukin 2 (IL 2), interleukin 3 (IL 3) and granulocyte/macrophage colony-stimulating factor (GM-CSF). When L17R leukemic cells were plated at a low cell density, their growth was accelerated 40 times by the addition of concentrated B6TE culture supernatant. This growth activity, tentatively designated leukemia-growth-promoting factor (LGPF), was heat sensitive, and its mol. wt was estimated to be approx. 25,000 from the elution pattern of Sephadex G-100 chromatography.