A method for assessing efficiency of bacterial cell disruption and DNA release.

BACKGROUND: DNA-based testing is becoming the preferred method both for identifying microorganisms and for characterizing microbial communities. However, no single DNA extraction method exists that is suitable for all types of microorganisms because bacteria are variable in their susceptibility to lysis by available extraction procedures. To develop and test new DNA extraction procedures, it would be helpful to determine their efficiencies. While the amount of extracted DNA can readily be measured by different methods, calculation of true efficiency requires knowledge of the initial amount of DNA in the starting bacterial sample, which cannot be done with precision by any existing method. In the process of developing a new extraction procedure, we developed a method that can be used to determine the total amount of both DNA and RNA in bacteria. The amount of DNA can be calculated from the amount of purines released after mild acid and alkali treatment. The amount of RNA in the same extract can also be calculated from the amount of ribonucleoside monophosphates. The released purines and ribonucleoside monophosphates can be quantified by absorbance using HPLC, with reference to appropriate standards. RESULTS: The acid/HPLC method was used to measure the efficiency of commonly used bead-beating and chemical protocols for releasing DNA from a particularly hardy organism, Mycobacterium smegmatis as well as several other species (Bacillus subtilis vegetative cells and spores; Francisella philomiragia; Pseudomonas aeruginosa; Moraxella catarrhalis; Bacillus thuringiensis; Staphylococcus aureus). Surprisingly large differences in efficiency between methods were found. CONCLUSIONS: The acid/HPLC method is a new tool to determine DNA extraction efficiencies and should aid in the development of improved protocols for releasing DNA from a broad range of microorganisms.

Effect of dietary vitamin E on spontaneous or nitric oxide donor-induced mutations in a mouse tumor model.

BACKGROUND: Vitamin E, an antioxidant, has been investigated for its effect on cancer incidence in humans, but no firm conclusions about a protective effect can be drawn from these studies. Recently, we reported a statistically significant correlation in the Mutatect mouse tumor model between the number of neutrophils and the frequency of mutation at the hypoxanthine phosphoribosyltransferase (hprt) locus. We have now used this model to investigate vitamin E's effect on the hprt mutation rate. METHODS: Mutatect cells were grown in mice as subcutaneous tumors for 2-3 weeks, the tumor cells were recovered, and 6-thioguanine-resistant (i.e., hprt mutant) colonies were scored. Myeloperoxidase activity was used as a measure of neutrophil infiltration. Vitamin E (2 IU/kg body weight) was provided in the diet for 3-4 weeks. In some experiments, glyceryl trinitrate (100 mg/kg body weight) was also administered as a source of nitric oxide. All statistical tests were two-sided. RESULTS: Mouse tumors from the Mutatect MN-11 cell line exhibited a 3.2-fold higher median mutation frequency than the same cells in culture (P:<. 0001); vitamin E reduced this frequency by 24.9% (P: =.01). Mutatect TM-28-derived tumors (which secrete interleukin 8) were heavily infiltrated with neutrophils and had a correspondingly high mutation frequency; in two separate experiments, vitamin E reduced the median mutation frequency by 68.9% (P: =.0019) and 84.1% (P: =.011) and myeloperoxidase levels by 75.3% (P: =.0002) and 75.5% (P: =.026), respectively. Glyceryl trinitrate increased the mutation frequency in MN-11 tumors, and vitamin E reduced the median frequency by 61.4% (P: =.058). CONCLUSIONS: Dietary vitamin E afforded strong protection against both spontaneously arising and nitric oxide-induced mutations. Two separate protective mechanisms by vitamin E may be operating: scavenging of a nitric oxide-related genotoxic species and altering the infiltration of neutrophils into tumors.

Fluorometric method for rapid detection of DNA strand breaks in human white blood cells produced by low doses of radiation.

DNA strand breaks can be detected with great sensitivity by exposing crude cell lysates to alkaline solutions and monitoring the rate of strand unwinding. As little as one strand break per chromosome can be detected. Previous methods for measuring strand unwinding have required physical separation of single- from double-stranded molecules. We now describe conditions under which unwinding can be monitored directly using a fluorescent dye, thus greatly simplifying the analysis. Breaks due to irradiation of blood samples by 60Co gamma-rays at doses as low as 0.05 to 0.1 gray (5 to 10 rads) were detectable. Rapid rejoining of strand breaks during in vitro incubation at 37 degrees could readily be observed following a dose of one gray. Since the procedure is very rapid and cells can be analyzed directly without the requirement for culturing or radiolabeling, the procedure could be useful in cancer chemotherapy if in vivo damage is to be monitored or for testing the in vitro sensitivity of cells to drugs.

Triggering of DNA strand breaks by 45 degrees C hyperthermia and its influence on the repair of gamma-radiation damage in human white blood cells.

Human peripheral white blood cells, freshly isolated from normal individual donors, were exposed to hyperthermia. Heat-generated DNA strand break damage and white blood cell capacity to repair radiation-induced breaks were determined by a fluorometric alkaline unwinding assay. Strand breaks could be readily detected when white blood cells were incubated in a physiological salt solution at temperatures between 41 degrees and 46 degrees C, for times up to 90 min. The time course of strand break induction at 45 degrees C was characterized by a short initial lag, followed by a period of rapid break induction and subsequently a lower rate. Evidence is presented which suggests that the induction of DNA damage involved a "triggering" mechanism; a short treatment at 45 degrees C (10 to 20 min) initiated a cellular event which led to a rapid increase in the number of strand breaks during subsequent incubation of 37 degrees C. Continuous incubation at 45 degrees C produced less DNA damage than an initial period at 45 degrees C followed by incubation at 37 degrees C. This apparent "triggering phenomenon" was not due to a triggering of the respiratory burst in phagocytic cells, since no O2- could be detected; in fact, a 30-min treatment at 45 degrees C largely blocked the capacity of the cells to respond normally to a soluble stimulator of the respiratory burst. Unlike gamma-ray-induced breaks, 45 degrees C hyperthermia-induced breaks did not rejoin during subsequent incubation for up to 1 h at 37 degrees C. Additionally, 45 degrees C hyperthermia treatment progressively inhibited the ability of the cells to repair subsequent gamma-ray-induced breaks (4 Gy). This inhibition occurred during the period in which 45 degrees C heat rapidly induced strand breaks. Hyperthermia (41 degrees C), which did not trigger strand breaks, did not cause detectable inhibition of this repair capacity. There was no indication that hyperthermia sensitized cells to radiation-induced strand breaks.

Prevention of G:C pairing in mouse DNA by complete blocking of guanine residues with glyoxal. Availability of cytosine, adenine and thymine for hydrogen bonding with added unmodified polynucleotides.

We have developed conditions for the reaction of single-stranded DNA with glyoxal which permit blocking of essentially all guanine residues. This procedure effectively prevents base pairing involving these guanine residues, yet permits cytosine, thymine and adenine residues in the DNA to pair with added polynucleotides. Modification of DNA with glyoxal greatly reduces intrastrand helical regions, resulting in a very low binding to hydroxyapatite, as compared to unmodified DNA. Annealing of modified DNA with some synthetic polynucleotides of restricted composition (but not others) leads to a significant increase in binding, presumably because mouse DNA has sequences not containing guanine which are capable of hydrogen bonding to the added polynucleotides. This relatively simple procedure may allow isolation and further study of these guanine-free sequences in DNA.

Polypyrimidine sequences found in eukaryotic DNA have been conserved during evolution.

L-cell DNA contains an unexpectedly large amount of long pyrimidine tracts. Hydroxyapatite chromatography has been employed to show that these polypyrimidines hydridize extensively to the reiterated DNA of a large number of eukaryotes but fail to hybridize to prokaryotic DNA. This reaction is sequence specific and not the result of a special property of polypyrimidines since random 3H-labelled poly(dC-dT) shows poor hybridization to eukaryotic DNA. The hybrids formed by L-cell polypyrimidines and heterologous repeated sequences have a higher thermal stability than the corresponding hybrids of total repeated DNA indicating that sequences related to these polypyrimidines have been conserved during evolution. Furthermore, at least some of these tracts are transcribed because they are capable of reacting extensively with total cellular RNA. Although the function of these sequences is not yet known, the fact that they are widely conserved in evolution and also transcribed leads us to speculate that they play an an important role in eukaryotic cells.

Expression of serum- and glucocorticoid-regulated kinase (sgk) mRNA is up-regulated by GM-CSF and other proinflammatory mediators in human granulocytes.

Stimulation of human peripheral blood granulocytes with the proinflammatory cytokine, granulocyte-macrophage colony-stimulating factor (GM-CSF), increases incorporation of [3H]uridine into RNA. We investigated the nature of the RNA synthesized under these conditions. Using transcription inhibitors, gel electrophoresis, and high-salt precipitation, it was concluded that as much as 90% of this radiolabeled RNA represents polymerase II transcripts. Differential display reverse transcription-polymerase chain reaction was used to identify and clone GM-CSF-responsive mRNAs. Serum- and glucocorticoid-regulated kinase (sgk) mRNA was identified that could be up-regulated 10- to 20-fold by > or =0. 1 ng/mL recombinant human GM-CSF. The 2.6-kb sgk mRNA was induced rapidly (within 30 min) by GM-CSF and remained at high levels for at least 12 h. Up-regulation was blocked completely by the transcription inhibitor, actinomycin D, but not by the translation inhibitor, cycloheximide, nor by the tyrosine kinase inhibitor, genistein. Up-regulation did not appear to be caused by enhanced mRNA stability. Other inflammatory mediators could also increase sgk mRNA levels (GM-CSF > > lipopolysaccharide > fMLP = tumor necrosis factor alpha). The function of sgk in granulocytes remains unknown.

Expression of interleukin-8 promotes neutrophil infiltration and genetic instability in mutatect tumors.

Neutrophils represent a potential source of genotoxic reactive oxygen and nitrogen species in the tumor microenvironment. Using Mutatect cell lines, which can form subcutaneous tumors in syngeneic C57BL/6 mice, we have previously established that the number of spontaneously infiltrating neutrophils correlates with the number of mutations at the hypoxanthine phosphoribosyltransferase (Hprt) locus. We now describe the properties of four lines that express different levels of the neutrophil chemokine, interleukin-8 (IL-8), from a tetracycline (TET)-responsive promoter. In a series involving 45 animals, IL-8-expressing lines produced tumors with a higher neutrophil content than the control line. Analysis of the 45 tumors revealed that the neutrophil level again strongly correlated with hprt mutant frequency (MF) (P<.0001, r=0.88). Administration of TET was effective in lowering the neutrophil content of low IL-8-expressing tumors, but not high IL-8-expressing tumors. Although the IL-8 transgene was stable in all lines in vitro, high IL-8-expressing lines completely lost the transgene in vivo whereas low IL-8-expressing lines showed no evidence of transgene instability. These results provide further evidence, based on the study of an endogenous gene (hprt) and an IL-8 transgene, that neutrophils may contribute to genetic instability in tumors.

Simultaneous protective and damaging effects of cysteamine on intracellular DNA of leukocytes.

Incubation of human leukocytes with cysteamine can lead to the induction of DNA strand breaks. The induction of breaks is biphasic with increasing concentration of scavenger. The number of breaks increases in a dose-dependent manner to a maximum and then decreases at higher concentrations. Catalase has been shown to prevent the production of breaks, indicating an involvement of hydrogen peroxide. Cysteamine reacts with oxygen to generate hydrogen peroxide but at higher concentrations it also reacts with hydrogen peroxide. Thus, the biphasic effect of cysteamine on leukocyte DNA may be due to the sum of two separate reaction pathways. (i) Cysteamine reacts with oxygen to generate hydrogen peroxide which leads to DNA strand breakage. (ii) At higher concentrations, it eliminates hydrogen peroxide by reacting with it, thereby protecting the cellular DNA. Other antioxidant scavengers such as WR2721, acetylcysteine and ascorbate can also autooxidize to produce strand breaks. Thiourea and tetramethylurea do not. When tested for their ability to protect cells against DNA damage from added H2O2, the agent which most damaging by itself, cysteamine, was also the most protective.

Elimination of non-viable 6-thioguanine-sensitive T cells from viable T cells prior to PCR analysis.

The study of T cell clones at the genomic level is expanding our understanding of their role in diseases such as rheumatoid arthritis (RA) and multiple sclerosis (MS). We have been carrying out genotypic analysis by PCR of hypoxanthine phosphoribosyltransferase (hprt) mutations in these cells. Mutant T cells in the population can be cloned on the basis of their resistance to the cytotoxic drug, 6-thioguanine-(6-TG). A difficulty is that the majority of primary human T cells are capable of only limited growth ex vivo, even in the presence of 'feeder' cells. PCR analysis of DNA from such clones is made difficult by the limited number of viable mutant (drug-resistant) T cells and the large number of dead (drug-sensitive) mononuclear cells and feeder cells. DNA from the 'dead' cells remains sufficiently intact for many weeks in culture and can represent a significant source of background in PCR analysis. Here we describe a method employing hypotonic shock and micrococcal nuclease that reliably eliminates non-viable 6-TG-sensitive cells, allowing the study of the hprt gene in < 200 T cells by PCR.

Acid treatment of Drosophila deoxyribonucleic acid.

In cytologic preparations of chromosomes, acid-treated deoxyribonucleic acid (DNA) is found largely in the native state. However, acid treatments widely used for chromosome preparations produce significant amounts of depurination in DNA. DNA is similarly sensitive to depurination in intact cells or as purified DNA. If treated with alkali, these apurinic gaps can be converted to single strand breaks. Acid treatment has widely different effects on specific fractions of DNA. In Drosophila melanogaster 3% of the DNA is composed of very long tracts of pyrimidines (polypyrimidines) which are resistant to acid hydrolysis. The implications of these results for molecular cytogenetics are discussed.

Characterization of a polyclonal antibody to human thymidylate synthase suitable for the study of colorectal cancer specimens.

Measurement of thymidylate synthase (hTS) using immunohistochemical techniques has been reported in several clinical studies. However, its value as a prognostic indicator is still not clear. To pursue this, we have developed a new rabbit polyclonal antibody, hTS7.4. The antigen was recombinant hTS containing an N-terminal His(6)-tag. Antiserum hTS7.4 detected recombinant hTS by ELISA at a titer of 1:100,000. Western blot analysis of several human cell lines showed a single band of the expected 36-kD molecular size. HeLa cells treated with the TS inhibitor 5-FUdR showed the expected additional band corresponding to the ternary complex of hTS-dFUMP-reduced folate. hTS7.4 detected TS in bacterial, rat, mouse, and monkey cell extracts, and hTS8.3 (a closely related antiserum) immunoprecipitated a 36-kD [(35)S]-methionine-labeled protein from HeLa extracts. TS was detectable by indirect immunofluorescence in HeLa cells. Proliferating normal human fibroblasts in culture showed staining, but nonproliferating cells did not. Lymphocytes in the germinal center of human tonsil tissue, which are known to be proliferating, stained with hTS7.4 and also with monoclonal antibody TS106. TS may therefore be useful as an immunohistochemical marker of cell proliferation. Normal colon mucosa showed weak staining, whereas some colorectal cancer specimens stained very strongly with hTS7.4. A clinical study of colorectal cancer using this antibody is in progress. (J Histochem Cytochem 47:1563-1573, 1999)

In vivo induction of tumor variants by phorbol 12-myristate 13-acetate.

Malignant cells within solid tumors commonly exhibit phenotypic changes such as alterations in karyotype and acquisition of the ability to invade and to metastasize. We have used a fibrosarcoma, grown subcutaneously in C57BL/6 mice, to study the mechanisms underlying this phenotypic instability. Tumor-bearing animals were injected with phorbol myristate acetate (PMA) and then the tumors were transplanted to animals without further PMA treatment. These tumor cells were tum+, that is, unlike the parental tumors, they were able to grow in animals immunized against the parental tumors. This property was maintained for at least 6 tumor passages after the initial PMA injections. Thus, PMA appears to be able to induce an unstable tum+ phenotype in these cells at relatively high frequency.

DNA clastogenic activity of diethylstilbestrol.

Incubation of human leukocytes with the synthetic estrogen and known human carcinogen, diethylstilbestrol (DES), for 40 min caused extensive DNA strand breakage (clastogenesis), as measured by a fluorometric assay. The level of DNA clastogenesis was dose dependent above an apparent threshold of 10 microM. Clastogenesis was increased by addition of cysteamine, a reducing agent and hydroxyl radical scavenger, and was blocked by low concentrations of plasma. DES epoxide, a weakly estrogenic derivative, was about one-tenth as potent as a DNA clastogen. Unexpected and paradoxical findings were observed when cells were treated with DES in the presence of a hydrogen peroxide-generating system plus a peroxidase. At the subthreshold concentration of 10 microM DES, the oxidizing system increased DNA clastogenicity, yet at 30 microM DES the oxidizing system decreased clastogenicity. The addition of superoxide dismutase to the oxidizing system increased clastogenicity at both concentrations of DES. DNA damage was largely blocked by arsenite, N-ethylmaleimide, iodoacetamide and bromophenacyl bromide. These experiments provide further indication of the complex nature of reactions involving DES which can lead to DNA damage and which may be relevant to DES-induced carcinogenesis.

Inhibition by superoxide dismutase-mimetic copper complexes of phorbol ester-induced respiratory burst in human granulocytes.

Two superoxide dismutase-mimetic lipophilic copper complexes, Cu(II)2(indomethacin)4 [Cu(II)2(indo)4] and Cu(II)2(3,5-diisopropylsalicylate)4 [Cu(II)2(3,5-DIPS)4], were tested for their effects on the respiratory burst of intact human granulocytes and on xanthine oxidase, under conditions where superoxide and hydrogen peroxide were generated. The effect of the copper complexes on these enzyme systems (as opposed to their dismutase effect on superoxide) was determined by measuring oxygen uptake with an oxygen meter. It was found that, after a short delay, both systems were inhibited markedly by micromolar amounts of these complexes. This inhibition was prevented by treatment with EDTA or catalase if added prior to starting the reaction. Similar inhibitory effects were seen using copper sulfate. It appears that these lipophilic SOD-mimetic compounds can, in the presence of H2O2 and O2-, give rise to a species that can inhibit some component of the respiratory burst oxidase or protein kinase C in intact granulocytes and xanthine oxidase in solution. The observed decrease in O2- levels observed upon addition of these compounds is likely due to inhibition of the source and not to their SOD-mimetic properties.

Potentiation of retinoic acid-induced U-937 differentiation into respiratory burst-competent cells by nitric oxide donors.

All-trans retinoic acid (tretinoin) is a known inducer of differentiation of the human monoblastic cell line, U-937. We now report that the ability of retinoic acid (RA) to induce differentiation of U-937 cells into cells possessing respiratory burst activity is enhanced by the known nitric oxide-donating drugs glyceryl trinitrate, molsidomine and CAS 936, and by tetranitromethane in combination with cysteine. RA alone was a strong inducer of U-937 differentiation as indicated by the following responses to 12-O-tetradecanoylphorbol-13-acetate (TPA) stimulation: (1) increase in the percentage of cells staining with nitroblue tetrazolium (NBT); (2) increase in the total amount of formazan (the product of NBT reduction by O2-.) as determined spectrophotometrically; (3) increase in hexose monophosphate shunt (HMPS) activity as assessed by [14C]CO2 released from D-[1-14C]glucose. RA was also able to increase mRNA levels for two respiratory burst-related genes and for glucose-6-phosphate dehydrogenase (G6PD), an HMPS enzyme. Other indications of differentiation were reduced cell proliferation, increased adherence and altered nuclear morphology. The observed increase in formazan production and HMPS activity and the reduction of cell proliferation due to RA were augmented by co-treatment with either glyceryl trinitrate, molsidomine, CAS 936 or tetranitromethane plus cysteine. Glyceryl trinitrate alone increased HMPS activity and G6PD mRNA levels and also reduced cell proliferation. Glyceryl trinitrate, molsidomine and CAS 936 are presumed to release nitric oxide and increase intracellular cGMP levels by stimulation of soluble guanylate cyclase. The mechanism of action of tetranitromethane is less certain, although it may also generate reactive nitrogen intermediates. These data suggest that a NO./cGMP pathway may augment a retinoic acid-mediated pathway to enhance maturation of U-937 cells with respect to the respiratory burst. Glyceryl trinitrate may act additionally by another pathway.

Fluorometric determination of DNA in agarose gels: usefulness for measurement of double-strand breaks in nonlabeled cells by pulsed-field gel electrophoresis.

Quantitative measurement of DNA in agarose gels, particularly as needed for measurement of double-strand breaks induced by agents such as radiation, usually involves the use of radioactively labeled DNA. Thus its usefulness is usually limited to growing cells which incorporate radiolabeled thymidine into DNA. To circumvent this problem, we have developed a fluorometric technique for quantitative estimation of DNA in the presence of large amounts of agarose. Gel slices are solubilized with concentrated sodium perchlorate and DNA is selectively precipitated with cadmium chloride. The amount of DNA can then be estimated with 3,5-diaminobenzoic acid. Determination of DNA is linear in the range 10 ng to 1 microgram or more. We describe the application of this technique to the measurement of 60Co gamma-ray-induced double-strand breaks by pulsed-field gel electrophoresis. Our results are essentially identical to those obtained using radiolabeled DNA.

Rapid rejoining of DNA strand breaks in resting human lymphocytes after irradiation by low doses of 60Co gamma rays or 14.6-MeV neutrons.

The production and repair of DNA strand breaks was studied in human lymphocytes by means of a sensitive fluorometric technique. Lymphocytes were isolated by conventional methods and air-equilibrated suspensions were irradiated by low doses (less than or equal to 2 Gy) of either 60Co gamma rays or 14.6-MeV neutrons at 0 degree C. The apparent yield of initial strand breaks induced by neutrons was only 36% of that induced by gamma rays, in agreement with the observations of other workers. Resting lymphocytes were found to be proficient in their ability to rejoin gamma-induced strand breaks at 37 degrees C; rejoining followed biphasic kinetics, with 70% of the breaks disappearing with a half-life of about 3 min. Although the initial number of breaks induced by neutron irradiation was low, after 20 min of incubation the residual number of breaks was very similar for the two forms of radiation.

Oligodeoxynucleotides capable of binding to GC-rich regions in DNA are inappropriately synthesized during random hexanucleotide-primed labeling reactions.

Random hexanucleotide labeling is a commonly used and powerful technique for preparing radiolabeled nucleic acid probes. Because of the dependence of DNA polymerase upon single-stranded DNA as a template, the technique should theoretically also be useful for detecting small amounts of single-stranded DNA in the presence of a larger amount of double-stranded DNA. In the course of such an experiment, we identified an unexpected reaction that could contribute to background problems often encountered during Southern and Northern blotting. In the complete absence of template, random hexanucleotides appear capable of functioning both as template as well as primer, giving rise to labeled oligonucleotides as large as 20-mers. Some of these bind very tightly to GC-rich regions in target DNA immobilized on membranes, resisting high-stringency washing conditions. By choosing shorter incubation times and including sufficient single-stranded template, the problem may be minimized.