Search

RNA- and reverse transcriptase (R.T.)-containing viruses have been etiologically related to leukemia in several species of animals. There is no convincing report of isolation of a similar virus from man. In the past few years, however, biochemical evidence for three components thought to be characteristic of type C viruses has been detected in human acute leukemic cells. This includes the following: 1) An enzyme with physical and biochemical properties so far indistinguishable from viral R.T. has been purified from these cells. It was also recently shown that this enzyme is immunologically closely related to R.T. purified from type C primate RNA tumor viruses, much less to murine virus R.T., and not at all to avian or feline. 2) RNA has been detected in these cells with some nucleotide sequence relatedness to the genomic RNA of some murine and primate type-C viruses. 3) The DNA product of the endogenous R.T. of human leukemic cells has nucleotide sequences related to murine and especially primate type C sarcoma viruses. Thus, certain virus-like components exist in human leukemic cells in a functional form within ribonucleoprotein particles which resemble animal type C viruses. However, if they represent virus, they are defective—not released under ordinary conditions—and a causal relationship to human neoplasia, if any, has not been shown and may be impossible to prove.

[ 3 H]DNA copies of avian, feline, murine, and primate RNA tumor virus genomes were synthesized in vitro by an RNA-dependent DNA polymerase reaction. These DNAs were hybridized to 60-70S RNA that had been purified from the viruses. The amount of the [ 3 H]DNA hybridized yielded a measure of the genetic relatedness among the DNA preparations synthesized by the viruses. When many combinations of DNA and RNA were analyzed, the pattern of hybridization showed in some cases that the DNA copies of the viral RNA were related to each other in the same way that the natural hosts of the viruses are phylogenetically related. This pattern was observed only among the RNA leukemia viruses. The sarcoma component in sarcoma-leukemia viruses from rats and primates appeared to be unusually closely related. The mouse mammary carcinoma virus and two unclassified viruses (MPMV and Visna) appeared to be genetically distinct. A similar analysis of DNA synthesized by an RNA-dependent DNA polymerase associated with a viral-like particle obtained from the cytoplasm of human leukemic white blood cells demonstrated that this DNA occupied a space in the affinity pattern of leukemia viruses which is expected of a nucleic acid from a primate-type-C RNA tumor virus. This observation strengthens earlier evidence that components of RNA tumor viruses are associated with human leukemia.

Particles with the morphology of type C virus have been identified from primate placentas by electron microscopy. A reverse transcriptase (RNA-dependent DNA polymerase) was isolated and purified from microsomal pellets of two fresh placentas of rhesus monkeys in the early stages of gestation. This enzyme was biochemically similar yet immunologically distinct from the reverse transcriptases of known tumorigenic type C RNA viruses isolated from primates, but was immunologically related to a reverse transcriptase isolated from a type C virus obtained from normal baboon placenta. These particles may represent endogenous viruses and may function in the transfer of genetic information during embryogenesis.

This report describes a technique for isolating and separating DNA polymerases in mammalian cells infected with an RNA tumor virus. In whole cell extracts of cultured human lymphoblastoid cells infected with simian sarcoma virus, four DNA polymerases were resolved. Three of the enzymes were cellular, and one was viral in origin. The viral reverse transcriptase and cellular DNA polymerase II were eluted from a DEAE-cellulose column at 50 mM KCl, while cellular DNA polymerases I and III (R-DNA polymerase) were removed by 0.3 M KCl. These DNA polymerase mixtures were then resolved on phosphocellulose and DNA-cellulose columns. The identity of the separate polymerases was further confirmed by primer template utilization, molecular size determination, and response to antibody to the viral enzyme. This procedure was designed for further application to the study of DNA polymerases in human leukemic cells and other human tumors.

We have studied the effects of N 6 -(Δ 2 - isopentenyl)adenosine and related cytokinin analogs on the stimulation and inhibition of phytohemagglutinin-induced growth of human lymphocytes. We have found that N 6 -(Δ 2 - isopentenyl)adenosine exerts inhibitory effects on RNA synthesis, uridine transport and uridylate kinase synthesis. On the basis of the interaction of N 6 -(Δ 2 - isopentenyl)adenosine with cyclic AMP in this system, we propose that the observed stimulatory and inhibitory effects may be due to the involvement of N 6 -(Δ 2 - isopentenyl)adenosine in cyclic AMP metabolism.

(1) Synthesis of deoxythymidine by either direct transfer of deoxyribosyl to thymine (pyrimidine deoxyribosyltransferase) or by a coupled deoxynucleoside phosphorylase mechanism is approximately twofold greater with normal leukocyte extracts (55 to 88% granulocytes) than with extracts prepared from leukocytes obtained from patients with chronic myelogenous leukemia. Activities in lymphocytes (normal or leukemic) are one-fifth the activity of normal granulocytes.(2) The lower activity in chronic myelogenous leukemia remains at 50% of normal even when patients are in hematologic remission with a normal per cent mature granulocytes in the peripheral blood.(3) The leukemic enzyme could not be distinguished from the normal by pH optima, thermal stability, or kinetic properties. The Km's for the deoxyribosyl acceptor and deoxyribosyl donors were identical for both enzymes. Both are subject to substrate inhibition by thymine and to inhibition by purine bases with similar Ki's. In addition, the transferase component of both the leukemic and the normal cell enzyme is activated by phosphate and arsenate. It appears, therefore, that there is no qualitative difference between the enzyme obtained from leukocytes of patients with chronic myelogenous leukemia and the enzyme obtained from normal leukocytes, suggesting that the difference in total cell activity is due to an actual decrease in amount of enzyme in chronic myelogenous leukemia or to a mixed cell population, one with a normal quantity of enzyme and the other with little or no active enzyme.(4) In both the normal cell and the leukemic cell extracts, transferase and phosphorylase activities could not be separated. The ratio of the two activities remained constant over a 140- and a 230-fold purification in normal and leukemic cell extracts, respectively. These and other observations indicate that transferase and phosphorylase activities are associated with the same protein.(5) The metabolism of pyrimidine and purine deoxynucleosides is similar for normal and leukemic cells. Catabolism of all deoxynucleosides tested was by direct phosphorolysis, except for deoxyadenosine which required initial deamination to deoxyinosine before phosphorolysis. In contrast to the greater rates of pyrimidine deoxynucleoside synthesis and cleavage with normal leukocyte extracts, the rates of purine deoxynucleoside synthesis and cleavage were approximately twofold greater with extracts prepared from cells of patients with chronic myelogenous leukemia. There was no significant difference in the rate of phosphorolytic cleavage of pyrimidine nucleosides (uridine) between the CML and normal leukocyte extracts.

Certain rifamycin derivatives have been found to be more toxic for fresh human leukaemic blood cells than for normal blood cells. These particular derivatives also inhibit the reverse transcriptases of both human and viral origin.

The excitement aroused by the idea that transformation of cells by RNA tumour viruses might depend on the activity of a reverse transcriptase enzyme(s) specific to the virus has stimulated much research since the enzyme was discovered in 1970. But the questions of the physical identity and catalytic activities of the enzyme(s), its function in transformation and whether it is unique to RNA tumour viruses remain unanswered.

Rifamycin antibiotics, which inhibit RNA-directed DNA polymerase of Rauscher leuckemia virus, prevent the leukemogenic activity of the virus, and the effect of leukemogenesis correlates with the magnitude of inhibition of the purified enzyme. This inhibition of enzyme activity by rifamycin SV derivatives is due to a relatively tight binding between the enzymes and the inhibitors, yet the binding can be reversed by nonionic detergent. The results of this study suggest that RNA-directed DNA polymerase is essential for induction of leukemia by exogenous virus and correlate with the previous observation that the same derivatives block viral transformation in vitro .

A DNA polymerase activity present in the 60,000 × g pellet of human acute myelocytic leukemic blood cells catalyzes an endogenous RNA-dependent (RNase sensitive) DNA synthesis. After purification, this enzyme transcribes heteropolymeric regions of 70S RNA from various RNA tumor viruses but prefers mammalian over avian viral RNA. DNA transcribed from the viral RNA is, like that from RNA tumor virus reverse transcriptase, hydrogen bonded and covalently attached to the RNA template-primer. The DNA transcript, after alkaline hydrolysis, hybridizes specifically to the template-primer RNA. This activity is not decreased in the presence of DNA polymerases from normal cells, suggesting that the inability of the latter polymerases to catalyze transcription of viral RNA is not due to the presence of inhibitor(s). This DNA polymerase is easily distinguished from a terminal addition deoxynucleotidyl transferase and has previously been shown to be immunologically closely related to primate type-C virus reverse transcriptase.

Ribonuclease-sensitive DNA synthesis is demonstrated in a cytoplasmic particulate fraction of normal human blood lymphocytes stimulated with phytohemagglutinin, but not in unstimulated lymphocytes. DNA polymerase purified from this fraction does not transcribe the heteropolymeric regions of 70S RNA from RNA tumor viruses, thus distinguishing this enzyme from the RNA-directed DNA polymerase (reverse transcriptase) found in oncogenic RNA viruses and human leukemic cells.

Avian myeloblastosis virus (AMV) 4 S RNA was tested for amino acid acceptor activity for 18 of the 20 amino acids. A nonrandom distribution of viral tRNAs was found compared with tRNA from normal liver or from AMV-infected leukemic myeloblasts, confirming previous reports. Methionine and proline tRNAs were considerably enriched, whereas glutamic acid, glutamine, serine, tyrosine, and valine tRNAs were markedly depleted in AMV relative to homologous cellular tRNAs. The seven AMV tRNAs with the greatest amino acid acceptance capacities, which were in order methionine, proline, lysine, arginine, histidine, isoleucine, and threonine tRNAs, were compared with homologous tRNAs from leukemic myeloblasts and liver by reversed-phase 5 chromatography. Of the 25 isoaccepting chromatographic fractions identified, no tRNA species unique to AMV was detected. Only methionyl-tRNA showed a substantial quantitative variation in isoaccepting species compared with the host cell. Thus, viral selectivity for amino acid-specific tRNAs is not, generally, paralleled by selectivity for individual isoaccepting tRNA species. Qualitative differences in arginyl- and histidyl-tRNA isoaccepting species were discovered in virus and leukemic myeloblasts compared with liver. This indicates the existence of structural differences in these tRNA species which could be related to virus replication or expression.

Two DNA polymerases have been purified from blood lymphocytes of normal human individuals. In contrast to a DNA polymerase isolated from lymphoblasts from individuals with leukemia, neither of the DNA polymerases from normal lymphocytes transcribes the heteropolymeric regions of tumor-virus RNA into DNA.

The results of a study to identify the transfer RNA's for every amino acid in human tissues are reported. The isoaccepting species of the tRNA of human lymphoblasts (normal and leukemic) were fractionated by reverse-phase partition column chromatography. In every case, comparisons were made between the aminoacyl-tRNA profile of normal and leukemic lymphoblasts by co-chromatography using 3 H for one label and 14 C for another and the homologous amino acid-activating enzyme. Differences between the profiles of normal and leukemic cells were re-evaluated by reversal of the labeled amino acid and by acylation with the heterologous enzyme. The results indicate that (1) there are at least 56 species of tRNA fractionated from both the normal and the leukemic cells; (2) for most cases the profiles of the leukemic cells were very similar if not identical to the profiles of normal cells; (3) small but reproducible differences were found for leucyl-, seryl-, threonyl- and prolyl-tRNA; (4) the most pronounced differences were with tyrosyl-tRNA and glutaminyl-tRNA. The difference in the glutaminyl-tRNA was an extra isoaccepting species found only when normal lymphoblast enzyme preparation was used. Therefore, in this case the difference was due to the enzyme preparation rather than the tRNA itself.

In a cell-free system prepared from human leukocytes, synthesis of deoxythymidine is catalyzed by a transfer of deoxyribosyl from a deoxynucleoside to thymine (pyrimidine deoxyribosyltransferase) and by a coupled deoxynucleoside phosphorylase mechanism in which deoxythymidine is synthesized from the deoxyribose-1-P formed after phosphorolytic cleavage of either purine or pyrimidine deoxynucleosides. Both enzymatic mechanisms participate in deoxythymidine synthesis with pyrimidine deoxynucleosides as the deoxyribosyl donor. However, the proportion synthesized by either mechanism is dependent on the concentration of deoxynucleoside. At concentrations above 5 mm the transferase is the predominant enzymatic mechanism. Purine deoxynucleosides, on the other hand, participate in deoxythymidine synthesis only by a coupled deoxynucleoside phosphorylase mechanism. Rates of deoxythymidine synthesis with purine deoxynucleosides as deoxyribosyl donors were approximately one-tenth to one-thirtieth of the rates with pyrimidine deoxynucleosides. The constant ratio of transferase to phosphorylase through a 140-fold purification and during heat inactivation studies suggests that both activities are the function of one protein. In the presence of thymine and both deoxyuridine and deoxyribose-1-P, synthesis of deoxythymidine occurs independently by both enzymatic mechanisms, indicating that there are separate sites for each deoxyribosyl donor.

Treatment of lymphocytes from three patients with chronic lymphocytic leukemia (CLL) with phytohemagglutinin (PHA) resulted in a fivefold increase in transfer RNA (tRNA) methylase activity. Evidence was obtained which suggests that the increase included enzyme species not present prior to the induction. As with normal lymphocytes, the time of induction coincided with the formation of fully transformed PHA cells, which is markedly delayed compared with that occurring in normal lymphocytes stimulated with PHA. Although all three patients had an identical pattern of tRNA methylase induction, initial and induced absolute values were higher in the cells of one patient who had 10 per cent immature peripheral blood lymphocytes. The levels of transfer RNA methylase activity were similar in normal and CLL lymphocytes. These data suggest that PHA induces quantitative and qualitative changes in tRNA methylase enzymes of CLL lymphocytes similar to those previously reported in normal lymphocytes but that the sequence of events of PHA interaction with CLL lymphocytes leading to the subsequent enzyme induction is delayed.

Assays were performed that measured both the rate and extent (or total capacity) of transfer RNA methylases in extracts of lymphocytes cultured in the presence and absence of phytohemagglutinin (PHA). The tRNA methylases of human peripheral blood lymphocytes undergoing blastogenesis in culture with PHA had a five- to sixfold increase in rate and a three- to sevenfold increase in extent of methylation of heterologous tRNA. These data suggest that PHA transformed lymphocytes not only contain elevated levels of tRNA methylases, but that the increase includes qualitatively different enzymes from those found in normal peripheral blood lymphocytes. Experiments in which lymphocytes were incubated for various times with PHA revealed that tRNA methylase induction occurred late in or after DNA synthesis and after morphologic transformation, but prior to mitosis. Rate and extent of tRNA methylation increased simultaneously. PHA induction of tRNA methylase activity was dependent on the synthesis of new RNA in lymphocytes cultured from 40 to 45 hours. The increase was not due to different levels of inhibitors or activators or preferential degradation of reaction components. The data suggest that quantitative and qualitative changes occur in the tRNA methylases of the normal human peripheral blood lymphocyte stimulated by PHA to undergo transition to an undifferentiated cell "in cycle." The possible significance of these findings to control of protein synthesis in PHA transformed lymphocytes is discussed.

Precipitation with cetyltrimethylammonium bromide followed by precipitation with ethanol:salt provides a quick, simple, and effective method for preparing labelled nucleic acid products of DNA polymerase reactions for analysis by Cs 2 SO 4 equilibrium density gradient centrifugation. Results obtained using this treatment show that this approach is superior to other methods of preparing products of endogenous DNA polymerase reactions from relatively crude particulate fractions such as disrupted virions.

The chromatographic characteristics of uridine- and methyl-labeled low molecular weight RNA (4S and 5S) of human lymphoblasts from Burkitt lymphoma, infectious mononucleosis, and EB-virus transformed lymphocytes were compared with lymphoblasts from normal donors (cell line) and phytohemagglutinin (PHA)-transformed peripheral blood lymphocytes. No significant differences in these RNA profiles were found between the lymphoblast cell line from a normal donor and the lymphoblast cell line derived from a patient with infectious mononucleosis. The profiles of uridine-labeled RNA but not methyl-labeled RNA of Burkitt lymphoma and EB-virus transformed cells differed from the normal control cells, i.e., the lymphoblast cell line derived from a normal donor and PHA-transformed lymphocytes, respectively. The findings suggest that the elution pattern of low molecular weight nonmethylated RNA of Burkitt lymphoma and EB-virus transformed cells are similar but differ from normal and infectious mononucleosis lymphoblasts. These differences apparently reflect variations in the relative amount of certain species of nonmethylated low molecular weight RNA that probably represents 5S RNA and/or messenger RNA fragments.

Comparative studies of aspartyl-, leucyl-, seryl- and tyrosyl-tRNA by a reverse phase column co-chromatography have been performed between a human normal lymphoblast cell line (RPMI 1788) and a human myeloma cell line (RPMI 8226). The cell lines were derived from normal peripheral blood and peripheral blood of a patient with multiple myeloma and produce IgM and lambda-type light chain (Bence Jones protein), respectively. Similar chromatographic patterns were obtained for leucyl- and seryl-tRNA. Minor differences in aspartyl-tRNA and larger differences in tyrosyl-tRNA were observed.

Molecular hybridization with radioactive polyuridylic acid has been used to detect regions of polyadenylic acid in virus RNA. On the average, RNA from tumour viruses is twenty-five to fifty-fold richer in polyadenylic acid than RNA from non-oncogenic viruses. Polyacrylamide gel electrophoresis permitted a qualitative distinction between RNA preparations from the two virus classes.

Polyuridylic acid inhibited DNA polymerases purified from three species of oncornaviruses as well as three out of seven DNA polymerases purified from cells. Viral and cellular DNA polymerases could not be distinguished by polyuridylic acid inhibition, but were easily distinguished by their template preferences in the presence of magnesium.

Allopurinol and 6-mercaptopurine, two clinically useful purine antagonists, are inhibitors of deoxythymidine phosphorylase and pyrimidine deoxyribosyltransferase from human leukocytes. The inhibition of these enzyme activities is only in the direction of nucleoside synthesis and is not competitive with either substrate (thymine or deoxyribose-1-phosphate). 6-Mercaptopurine, but not allopurinol, also inhibits uridine phosphorylase.

Effective uptake of exogenous 3H-labelled tRNA into the active metabolic tRNA pool of mouse L1210 cells was demonstrated by detecting aminoacyl-(3H) tRNA synthesized in the intact, pulsed cells.

A reverse transcriptase isolated and purified from a case of human acute myeloblastic leukaemia cells has both biochemical and immunological properties similar to that of known type C leukaemia virus reverse transcriptases and, in particular, to the enzyme from primate type C virus.

Terminal deoxynucleotidyltransferase was isolated and partially purified from nuclei from white blood cells of a patient with chronic myelogenous leukemia. The terminal transferase has a sedimentation value of 3.4 S, a pH optimum of 7.5, a Mn2+ optimum of 0.1 mm, and a Mg2+ optimum of 5 to 7 mm. The purified enzyme efficiently utilizes a number of DNA primers but not RNA primers. The presence of this enzyme in human myelogenous leukemic cells suggests that its presence may be associated with a particular stage of the disease rather than being specific for a particular form of leukemia or for thymic cells.

No significant hybridization was detected of DNA from simian virus 40 or polyoma virus, and of 70 S RNA from avian myeloblastosis virus, murine leukemia virus (Rauscher), murine sarcoma virus (Kirsten), RD-114B, simian sarcoma virus-1, or Mason-Pfizer virus.

Visna virus 70S RNA contains long stretches of polyadenylic acid [poly(A)]. The homogeneity in length of poly(4) regions is observed in 70S RNA from visna virus and all RNA tumor viruses tested, and not with other types of RNA. By this criterion visna virus resembles RNA tumor viruses.

Reverse transcriptase activity has been found in a type C virus isolated from a cell line from a patient with lymphoma which showed antigenic characteristics distinct from other mammalian type C viruses. These findings open several possibilities for determining the relationship of this virus to the pathogenesis of human lymphoma.

Two DNA polymerases purified from normal human lymphocytes are distinguishable from the viral reverse transcriptases of avian myeloblastosis virus and Mason-Pfizer monkey virus by their relative affinity for select templates. In this respect, the activity of the two normal human lymphocyte polymerases closely resembles the activity of Escherichia coli DNA polymerase 1. The viral and cellular DNA polymerases are equally active with the nonspecific template, poly(rA) · poly(dT). Criteria for distinguishing the activity of viral reverse transcriptase are discussed.

INTEREST in the enzyme L-asparaginase has been stimulated by the observation that it inhibits the growth of experimental and human tumours1–3. Tumours sensitive to L-asparaginase require exogenous L-asparagine whereas tumours not sensitive synthesize L-asparagine from L-aspartic acid and L-glutamine. Moreover, if L-asparaginase-sensitive tumours are made resistant using suboptimal doses of the enzyme, the resistant variant possesses L-asparagine synthetase and no longer requires exogenous L-asparagine for growth3,4. The critical questions now concern the regulation of L-asparagine synthetase activity.

Antibodies to a DNA:RNA Hybrid in Systemic Lupus Erythematosus measured by a Cellulose Ester Filter Radioimmunoassay

Some rifamycin SV derivatives are inhibitors of purified murine type-C virus RNA-directed DNA polymerase, although other polymerases are also affected. We have previously grouped these compounds into three classes (A, B, and C) based on their relative inhibitory effect on the viral polymerase. The concentration for 50% inhibition by these three classes are 5, 20, and 100 μg/ml, respectively. In this report we examine the mechanism of inhibition of the viral enzyme. The inhibitory effect is independent of the choice of template-primer but is sensitive to the presence of non-ionic detergent. Potent inhibitors bind to the viral enzyme as shown by co-sedimentation of the inhibitor with purified enzyme during glycerol gradient centrifugation. At least two molecules of potent (class C) inhibitors bind to each molecule of the viral enzyme. This binding is reversible with a relatively low concentration of Triton X-100 (0.05%) as judged by a complete restoration of the enzyme activity. Similar results were found with bacterial DNA polymerases but much firmer binding was found between Escherichia coli RNA polymerase and rifamycin derivatives since only 20% recovery of enzyme activity was obtained at a concentration of Triton X-100 as high as 5%. With respect to the purified viral DNA polymerase in decreasing order the rifamycin derivative interact with: the free form of the enzyme, the enzyme—template-primer complex, the enzyme—substrate complex, the initiation complex. The inhibitors were non-competitive with the deoxyribonucleoside triphosphate substrates with K i equivalent to 10−6 M. The inhibition appears to be at the step before initiation. Therefore, conditions which facilitate initation interfere with the effectiveness of these compounds, and the order of addition of each component becomes critical in assessing their relative inhibitory activity.

An RNA-directed DNA polymerase was isolated from the peripheral blood leukocytes of a patient with acute myelomonocytic leukemia by successive purification of a particulate cytoplasmic fraction with endogenous, ribonuclease-sensitive DNA polymerase activity. Like RNA-directed DNA polymerase from mammalian type-C virus, the human leukemic cell enzyme efficiently utilized (A) n ·(dT) 12-18 and (C) n ·(dG) 12-18 and had an approximate molecular weight of 70,000. Further, the leukemic cell enzyme was strongly inhibited by antisera to RNA-directed DNA polymerase of primate type-C virus in a fashion similar to that noted with an extensively purified RNA-directed DNA polymerase from a person with acute myelogenous leukemia [Todaro, G.J. & Gallo, R.C. (1973), Nature 244, 206]. By these biochemical and immunological results the leukemic cell enzyme could be differentiated from all other known cellular DNA polymerases but could not be distinguished from RNA-directed DNA polymerase of primate type-C virus. We interpret these data, combined with observations published elsewhere, to indicate that human acute myelogenous leukemia cells contain components related to primate type-C virus. The parameters used in this study may provide the specificity and sensitivity required for determining the presence or absence and (if present) the relatedness of RNA-directed DNA polymerase in other cases and types of human leukemia.

Nascent DNA synthesized from RNase-sensitive endogenous DNA polymerase activities of microsomal and mitochondrial fractions from phytohemagglutinin-stimulated normal human blood lymphocytes is covalently attached to a small RNA molecule (

In this brief review, we will begin by outlining a plausible rationale for the development of inhibitors of the DNA polymerases of RNA tumor viruses (reverse transciptases). Then, rather than exhaustively describing all agents which have been shown to inhibit these enzymes, we will present a classification of these inhibitors based on the biochemical properties of the enzymes and the mechanism of action of the inhibitors, offer a critique of existing inhibitors, and suggest directions for future research.

The presence of reverse transcriptase in RNA tumor viruses (oncornaviruses) is well documented (1,2). We recently described the purification and characterization of reverse transcriptase activity from the blood leukocytes of some patients with acute lymphocytic and acute myelocytic leukemia (3). The enzyme isolated from human acute leukemic cells has both biochemical and immunological characteristics of the reverse transcriptase from oncornaviruses. For example, (a) the enzyme is located in a particulate fraction, requiring high salt and/or non-ionic detergent for solubilization; (b) it catalyzes a ribonuclease-sensitive endogenous DNA synthesis; (c) the DNA product of the endogenous reaction is a DNA·RNA hybrid and is covalently attached to RNA; and (d) the purified enzyme transcribes heteropolymeric portions of 70S RNA isolated from avian myeloblastosis virus (AMV)1 and Rauscher leukemia virus (RLV) and prefers (rA) n ·(dT)12–18 over (dT)12–18 ·(dA) n as a template-primer. The purified enzyme from the human acute leukemic cells does not possess any terminal transferase activity. Recently, in collaboration with G. Todaro, we showed that the enzyme we isolated from human acute leukemic cells with the biochemical properties of viral reverse transcriptase is also immunologically related to the enzyme from mammalian type-C leukemia virus, especially to reverse transcriptase from a primate type-C virus, gibbon ape virus (4).

1. 1. Seven aminoacyl-tRNAs from control (early passage) and polyoma-transformed BN rat embryo culture cells were compared by reversed-phase-2 chromatography. No differences were noted in the elution profiles of seryl-, tyrosyl-, leucyl-, asparaginyl-, or glutaminyl-tRNA. A consistent quantitative difference in phenyl-alanyl-tRNA was demonstrated. Aspartyl-tRNA profiles from the polyoma-transformed rat cell line showed a substantial additional isoaccepting fraction. 2. 2. A major difference in aspartyl-tRNA (asp-tRNA), analogous to that observed in rat cells, was also demonstrated by reversed-phase-2 and reversed-phase-5 chromatography in polyoma- and SV40-transformed Al/N mouse embryo culture cells in comparison to various other Al/N mouse cells and tissues ( vide infra ). The greater resolution achieved by reversed-phase-5 chromatography suggests that the differences in Asp-tRNA are due to marked quantitative variations in isoaccepting species rather than to “all-or-none” differences. 3. 3. The following sources had a similar Asp-tRNA elution profile pattern which was substantially different from the pattern observed in DNA oncogenic virus transformed mouse embryo culture cells: (a) early passage and established mouse embryo culture cells; (b) spontaneously tranformed and Moloney sarcoma virus-transformed mouse culture cells; (c) tumors formed by the injection of polyoma-transformed mouse cells into syngeneic hosts; (d) mature mouse liver; and (e) whole mid-term mouse embryos. 4. 4. It is suggested that the pattern of Asp-tRNA is due to selective cellular gene expression. The latter may be related to the function of the DNA oncogenic virus genome in transformed cells, and/or the production of polyoma- and SV40-specific proteins or cellular alterations in transformed cells may require the variant Asp-tRNA composition. The Asp-tRNA pattern is not strictly related to the property of neoplasia or to the level of cellular differentiation.

Before beginning the topic of this paper, we want to present some introductory comments on the background that led us into studies on transfer RNA and unique components of tRNA, such as isopentenyladenosine. Our laboratory has been interested in the biochemical mechanisms involved in the conversion of normal human leukocytes to leukemic cells. From recent autoradiographic and cell kinetic studies (Astaldi and Mauri, 1953; Gavasto et al., 1964; Craddock and Nakai, 1962), it has become increasingly evident that human leukemia, in particular, acute leukemia, may be regarded as a disorder of the normal process of cellular differentiation rather than primarily a proliferative disorder. The evidence for this is that when leukemic blast cells of the bone marrow are compared to the blast cells of normal bone marrow, rates of DNA synthesis and mitosis are quite comparable (Astaldi and Mauri, 1953; Gavasto et al., 1964; Craddock and Nakai, 1962). In fact, if anything, the normal “blast” cells may proliferate at a rate faster than the leukemic cell. How then do leukemic cells accumulate? In the normal bone marrow, 5 % or less of the cells are primitive cells or so-called “blast” cells. There is a continuous process of maturation, during which the primitive cells are converted to mature leukocytes. During the progression to the granulocyte (the fully mature leukocyte), two major morphological changes occur.

A particle of discrete biochemical composition was purified from fresh, unfrozen peripheral blood leukocytes of human patients with acute myeloblastic leukemia. This particle endogenously synthesized DNA by use of an RNA primer and template. About half of the DNA sequences synthesized in the presence of actinomycin D hybridized to RNA isolated from type-C sarcoma viruses of primates or mice; lower annealing values were obtained with RNA isolated from other sarcoma or leukemia viruses. The results confirm and extend previous results from molecular hybridization experiments related to the existence in human leukemia of components of RNA tumor viruses.

The DNA polymerase from the Mason-Pfizer monkey virus (M-PMV), an RNA tumor virus not typical type-C or type-B, has been purified a thousand-fold over the original crude viral suspension. This purified enzyme is compared to a similarly purified DNA polymerase from the primate woolly monkey virus, a type-C virus. The two enzymes have similar template specificities but differ in their requirements for optimum activity. Both DNA polymerases have a pH optimum of 7.3 in Tris buffer. M-PMV enzyme has maximum activity with 5 mM Mg 2+ and 40 mM potassium chloride, whereas the woolly monkey virus optima are 100 mM potassium chloride with 0.8 mM Mn 2+ . The apparent molecular weight of the M-PMV enzyme is approximately 110,000, whereas the woolly monkey virus polymerase is approximately 70,000. The biochemical properties of these two enzymes were also compared to a similarly purified enzyme from a type-C virus from a lower mammal (Rauscher murine leukemia virus). The results show that more similarity exists between the DNA polymerases from viruses of the same type (type-C), than between the polymerases from viruses of different types but from closely related species.

The plant cytokinin isopentenyladenosine, a component of yeast and mammalian transfer ribonucleic acid, is both a potent inhibitor and stimulator of DNA synthesis, transformation, and mitosis of the phytohemagglutinin-stimulated lymphocyte. The stage of the cell cycle and the concentrations used are critical for these effects. The addition of isopentenyladenosine within the first 12 hours after phytohemagglutinin at a concentration above 10(-6) molar results in inhibition, while lower concentrations (between 10(-7) and 10(-6) molar), added at 24 hours or later, have a stimulatory effect.

Molecular hybridization analyses of DNA products from endogenous and from reconstituted avian myeloblastosis virus (AMV) reverse transcriptase reactions show that the poly (rA) region of AMV 70S RNA is not transcribed unless oligo (dT) is added as primer and that at least some of the poly (dT) synthesized in the oligo (dT)-stimulated reaction is covalently bound to heteropolymeric transcripts. These results indicate that 3′-OH groups of the resident primers do not reside within the poly (rA) regions.

A particulate fraction from human leukaemic blood cells contains an RNAase-sensitive endogenous DNA polymerase system. This DNA polymerase was purified and it transcribes heteropolymeric regions of AMV 70S RNA. These and other template characteristics are similar to those of RNA tumour virus reverse transcriptase and appear to distinguish this leukaemic DNA polymerase from other normal cellular DNA polymerases.