Transformation and Deoxyribonucleic Acid Size: Extent of Degradation on Entry Varies with Size of Donor

The fate of label introduced as donor deoxyribonucleic acid (DNA) into competent cells of Diplococcus pneumoniae was determined immediately after entry at 25 C, as a function of the size of the donor DNA. Part of the label is found to be acid soluble, part has been incorporated into chromosomal DNA, apparently through reincorporation of degraded donor DNA, and part is found in single strands of length smaller than that of the input donor DNA strands. The last fraction apparently constitutes the precursor for integration of intact donor genetic markers and is referred to as the intact fraction. For large donor DNA the intact fraction contains over 80% of the total intracellular label, but the median strand length has been reduced to 2.2 x 10(6) daltons. For small donor molecules (1 x 10(5) to 6 x 10(5) daltons per strand) the fraction intact increases with donor size from 10 to 50% of the total intracellular label, and the median strand length of this fraction is half that of the donor strands. By combining these results with earlier data on the size dependence of the yield of transformants per unit of total intracellular donor label, we have calculated the probability that a marker in the intact fraction will be integrated, as a function of the length of the donor strand after entry. This probability has a linear dependence on strand length for activities below 40% of maximum, and extrapolates to zero activity at 77,000 daltons per strand.     

Mismatch excision and possible polarity effects result in preferred deoxyribonucleic acid strand of integration in pneumococcal transformation.

Heteroduplex deoxyribonucleic acid molecules having a drug resistance marker on one strand and its wild-type allele on the other have been used as donors in pneumococcal transformation. Opposite strands are not equally effective in producing transformants, and this strand bias is not the same, either in direction or magnitude, for various different genetic markers. Selective excision of mismatched base pairs is probably responsible for the large differences in strand efficiency seen with discriminating (hex+) strains, for when the recipient is nondiscriminating (hex-), and therefore presumably lacking an excision enzyme system, strand bias is drastically reduced or altered. The evidence also indicates that excision occurs after integration, as it is provoked by specific donor-recipient mismatch and not by the same mismatch when introduced within donor heteroduplex molecules. Excision can extend to include a neighboring linked marker which would otherwise not be excised, thus altering its intrinsic strand bias as well as its efficiency in transformation. There is a small bias in relative strand efficiency for some markers, not caused by mismatch excision, which perhaps is due to polarity in the integration process itself.     

Donor deoxyribonucleic acid length and marker effect in pneumococcal transformation.

The efficiency of transformation of point mutations depends upon base pair mismatches during the recombination process. For low-efficiency markers, the genetic information carried on the donor deoxyribonucleic acid is preferentially lost. To understand this elimination process, we investigated the effect of the size of donor deoxyribonucleic acid on the relative efficiency of low-efficiency point mutations. The deoxyribonucleic acid was shortened either by mechanical shearing or by restriction enzyme treatments. The results indicate that transformation by low-efficiency markers was not affected by shortening the distance between them and the end of the molecule any more than was transformation by the other markers. Moreover, no lethal event could be detected for either cell or chromosomal marker survival. These data do not exclude the double-strand-break hypothesis that was proposed to explain the loss of genetic information for low-efficiency markers, but they offer no support for it.     

Marker rescue studies of the transfer of recombinant DNA to Streptococcus gordonii in vitro, in foods and gnotobiotic rats

A plasmid marker rescue system based on restoration of the nptII gene was established in Streptococcus gordonii to study the transfer of bacterial and transgenic plant DNA by transformation. In vitro studies revealed that the marker rescue efficiency depends on the type of donor DNA. Plasmid and chromosomal DNA of bacteria as well as DNA of transgenic potatoes were transferred with efficiencies ranging from 8.1 x 10(-6) to 5.8 x 10(-7) transformants per nptII gene. Using a 792-bp amplification product of nptII the efficiency was strongly decreased (9.8 x 10(-9)). In blood sausage, marker rescue using plasmid DNA was detectable (7.9 x 10(-10)), whereas in milk heat-inactivated horse serum (HHS) had to be added to obtain an efficiency of 2.7 x 10(-11). No marker rescue was detected in extracts of transgenic potatoes despite addition of HHS. In vivo transformation of S. gordonii LTH 5597 was studied in monoassociated rats by using plasmid DNA. No marker rescue could be detected in vivo, although transformation was detected in the presence of saliva and fecal samples supplemented with HHS. It was also shown that plasmid DNA persists in rat saliva permitting transformation for up to 6 h of incubation. It is suggested that the lack of marker rescue is due to the absence of competence-stimulating factors such as serum proteins in rat saliva.     

Genetic Studies of Recombining DNA in Pneumococcal Transformation

The results of genetic fine structure experiments, performed on the amiA locus of Pneumococcus are summarized. The peculiar feature of transformation genetics is that a given donor marker mutation transforms with an efficiency characteristic of the mutated site. In spite of this difficulty, mapping procedures have been devised and quantitative recombination studies performed. It is concluded from these studies that transformation, in this locus, is the consequence of frequent, and essentially random exchanges occurring between donor DNA and the chromosomal DNA of the recipient cell. The average length of uninterrupted donor DNA polynucleotide strand which could be inserted into the chromosome of a transformed cell is estimated, from genetic data, to be probably not greater than 3·10⁵ daltons (for a double-stranded insertion). It is proposed, on the basis of genetic evidence, that following essentially random exchanges between donor DNA and recipient chromosome, a revision process, specific for certain types of mutated sites, occurs. The revision process appears to remove preferentially donor DNA sequences from the primary recombinant structure, and allow repair along the chromosomal template, leading to low efficiency in the genetic integration of these sites. A mechanism for this "destruction-choice" process is presented, and evidence in support of this mechanism discussed.     

Intermediate in adenovirus type 2 replication.

Replicating chromosomes, called intermediate DNA, have been extracted from the adenovirus replication complex. Compared to mature molecules, intermediate DNA had a greater buoyant density in CsCl gradients and ethidium bromide-cesium chloride gradients. Digestion of intermediate DNA with S1 endonuclease, but not with RNase, abolished the difference in densities. These properties suggest that replicating molecules contain extensive regions of parental single strands. Although intermediate DNA sedimented faster than marker viral DNA in neutral sucrose gradients, single strands longer than unit length could not be detected after alkaline denaturation. Integral size classes of nascent chains in intermediate DNA suggest a relationship between units of replication and the nucleoprotein structure of the virus chromosome. Adenovirus DNA was replicated at a rate of 0.7 x 10-6 daltons/min. Although newly synthesized molecules had the same sedimentation coefficient and buoyant density as mature chromosomes, they still contained single-strand interruptions. Complete joining of daughter strands required an additional 15 to 20 min.     

Fate of Recipient Deoxyribonucleic Acid During Transformation in Haemophilus influenzae

During genetic transformation of Haemophilus influenzae, segments of the host deoxyribonucleic acid (DNA) corresponding to the integrating donor DNA were degraded and liberated into the medium. This degradation was detected by the release of the radioactive label from host DNA during a time period matching the time of development of maximal linkage between donor and host markers. The host label released above that released from nontransformed, control cultures was equivalent to about 2% of the host genome or 16 x 10(6) daltons of DNA. The released, labeled material was acid-soluble and dialyzable. The label release from control cultures was unaffected at 30 C; at this temperature, the recombination-specific release from transformed cells was suppressed. High molecular weight fragments of host DNA corresponding in size to the donor fragments could not be found free within the cell, weakly bound to other host DNA, or bound to non-integrated donor DNA by a reciprocal cross mechanism.     

Base specificity of mismatch repair in Streptococcus pneumoniae

DNA sequence analysis was undertaken to investigate the structural basis of mutations showing different integration efficiencies in Streptococcus pneumoniae. Wild type, mutant and revertant sequences at two sites in the amiA locus were determined. It appears that markers which transform efficiently or inefficiently can result from single base pair changes. A low efficiency (LE) marker corresponds to a C:G to T:A change and a high efficiency (HE) marker to a G:C to T:A change. In the latter case, two mismatches, G/A and T/C, can exist at the heteroduplex stage in transformation; only T/C appears to be recognized by the hex system which controls transforming efficiencies in pneumococcus. Each of the recognized mismatches, T/G and C/A, which result from transitional change, and T/C appears to involve at least one pyrimidine. It is proposed that the mismatch repair system of S. pneumoniae is directed against mismatched pyrimidines. DNA sequence analysis also reveals that short deletions (33 or 34 bases long) behave as very high efficiency markers, confirming that deletions are not recognized by the hex system.     

Genetic Integration in the Heterospecific Transformation of Haemophilus influenzae Cells by Haemophilus parainfluenzae Deoxyribonucleic Acid

The in vivo chemical linkage of Haemophilus parainfluenzae deoxyribonucleic acid (DNA) with the H. influenzae genome has been found to occur at a much higher level than is suggested by the low efficiency of the heterospecific transformation of an antibiotic resistance marker. This linkage, about 60% of the level with homospecific DNA, was found to involve alkali-stable bonding. The amount of host DNA label released (about 60%) was about the same as that released during homospecific transformation. Also, over 60% of the H. influenzae cells adsorbing H. parainfluenzae DNA could not form colonies upon plating. This lethality of the heterospecific transformation was not immediate but followed considerable metabolic activity of the host cells. These data are presented to show that the "limited-pairing" hypothesis may be only a partial explanation for the low efficiency of heterospecific transformation. Another hypothesis is presented which takes into account the lethal effect of this kind of transformation.     

A Transforming Marker That Produces Merodiploids with High Efficiency and Stable Transformants with Low Efficiency in Streptococcus

A mutation (ery-r8) conferring a high level of resistance to erythromycin in the Challis strain of Streptoccus sanguis can be transferred to wild-type erythromycin-sensitive recipients via single molecules of donor DNA. The transformants thus produced are of two types: (1) cells slightly more resistant to erythromycin than wild-type and capable of segregating (at a frequency of 2 X 10(-4)/bacterium/generation) either wild-type or highly-resistant cells like the original donor type; (2) cells phenotypically and genotypically identical to the original donor type. The unstable diploids (ery-r8/+) occur with a frequency equivalent to that obtained with high-efficiency (HE) markers, whereas the stable donor-type (ery-r8) transformants occur with about five hundred times lower frequency. Penetration of the wild-type recipient by more than one molecule of DNA bearing the ery-r8 marker increases by as much as seven times the incidence of stable transformants. UV-irradiation of molecules bearing the ery-r8 marker diminishes their ability to cooperate in producing a stable transformant, although the UV sensitivity of stable transformant production by a single DNA molecule is not different from that of diploid production. Hence, stable transformants do not appear to be produced by a process typical of low efficiency (LE) markers, which are generally highly sensitive to ultraviolet irradiation. Moreover, stable ery-r8 transformants are produced with equally low frequencies in strains of S. pneumoniae that discriminate (hex+) and fail to discriminate (hex--) between HE and LE markers. We postulate that all transformations by the ery-r8 marker result in ery-r8/+ diploids, and that segregation results in the infrequent stable transformants of the original donor type. This hypothesis is supported by the observations that rifampin treatment of ery-r8/+ populations increases the frequency of segregation and similar treatment of wild-type recipients under-going transformation by the ery-r8 marker increases the frequency of stable transformants.--In producing the ery-r8/+ transformant the r8 allele is integrated close to the site of its wild-type homolog, since single molecules of DNA from this transformant can be shown to carry both alleles. Segregation of either the ery-r8 or + allele is not detectably enhanced by acridine orange or thymidine deprivation.--The ery-r8 marker occurs close to a site of mutation (ery-r2) which confers erythromycin resistance upon ribosomes. When the r2 and r8 markers are jointly transferred, ery-r2-r8/+ genomes are produced in which the r2 marker is stably integrated but the r8 marker is unstably adjoined to its wild-type homolog. Thus, the duplicated region can be quite short. When the ery-r8 marker is stably integrated, the region of the marker is refractory to subsequent transformation. Markers with properties like ery-r8 are not particularly rare, being found with a frequency of about 4% among spontaneous mutations to erythromycin resistance.     

Mechanism of Inactivation of Haemophilus influenzae Transforming Deoxyribonucleic Acid by Sonic Radiation

Transforming deoxyribonucleic acid (DNA) from Haemophilus influenzae was exposed to sonic radiation of various durations. Reductions in transforming ability of the DNA, cellular DNA uptake, and integration into the genome, and single- and double-stranded molecular weights of the transforming DNA were measured and compared. We conclude that (i) sonic radiation causes DNA strand breaks (almost always double-strand breaks with relatively few alkaline-labile bonds), the number increasing with exposure until the double-stranded molecular weight is reduced to less than 10(6) daltons; and (ii) since transformation is reduced about as much as integration and much more than uptake, inactivation of transforming DNA by sonic radiation appears to be caused mostly by failure of Haemophilus cells to integrate the transforming DNA that is taken into the cells. These results are similar to those for inactivation by X radiation but differ from those for ultraviolet radiation. A strand break caused by sonic radiation, however, does not necessarily inactivate the transforming DNA, whereas in the case of ionizing radiation it may. The results may be fit by the model proposed by Cato and Guild. From our data and the equation of Lacks, the minimum active site of DNA necessary for transformation and the frequency of exchanges between donor and recipient strands upon integration of transforming DNA were estimated as 0.35 x 10(6) to 0.7 x 10(6) daltons and 0.15 to 0.4 switches per 10(6) daltons, respectively.     

Construction of a marker rescue system in Bacillus subtilis for detection of horizontal gene transfer in food

A marker rescue system based on the repair of the kanamycin resistance gene nptII was constructed for use in Gram-positive bacteria and established in Bacillus subtilis 168. Marker rescue was detected in vitro using different types of donor DNA containing intact nptII. The efficiency of marker rescue using chromosomal DNA of E. coli Sure as well as plasmids pMR2 or pSR8-30 ranged from 3.8 x 10(-8) to 1.5 x 10(-9) transformants per nptII gene. Low efficiencies of ca. 10(-12) were obtained with PCR fragments of 792 bp obtained from chromosomal DNA of E. coli Sure or DNA from a transgenic potato. B. subtilis developed competence during growth in milk and chocolate milk, and marker rescue transformation was detected with frequencies of ca. 10(-6) and 10(-8), respectively, using chromosomal DNA of E. coli Sure as donor DNA. Although the copy number of nptII genes of the plant DNA exceeded that of chromosomal E. coli DNA in the marker rescue experiments, a transfer of DNA from the transgenic plant to B. subtilis was detectable neither in vitro nor in situ.     

A unique subgenomic species of adenovirus 2 DNA generated under high multiplicities of infection.

We have identified a novel subgenomic viral DNA in KB cells infected with adenovirus 2 (Ad2) under high multiplicities of infection. KB cells were infected with Ad2 at multiplicities of infection greater than 100 PFU/cell. 32P-labeled viral DNA was selectively extracted by a modification of the method of Hirt (8) from the infected cells and analyzed by electrophoresis on agarose gels. In addition to full-length DNA (33 to 23 x 10(6) daltons), a unique subgenomic DNA species of about 12 to 13% (2.6 x 10(6) daltons) of full-length DNA in size was found in the infected cells. This subgenomic DNA was found to be double stranded and was not packaged inside the virus particles. This DNA could be isolated in large amounts (30 to 50% of total viral DNA) from infected cells. When cleaved with restriction endonuclease KpnI, the subgenomic DNA yielded two fragments, each corresponding to about 6% and 7% of the full-length genome in size.     

Effect of food processing on the fate of DNA with regard to degradation and transformation capability in Bacillus subtilis

Soymilk, tofu, corn masa, and cooked potato were produced from transgenic raw materials and the effect of processing on the degradation of DNA was studied. Major degrading factors were for soymilk and tofu the mechanical treatment of soaked soybeans and for corn masa and cooked potatoes the thermal treatment. In the processed foods no DNA fragments > 1.1 kb were detected. We included in our studies the effect of the size of donor DNA and length of the homologous sequence on the marker rescue transformation of B. subtilis LTH 5466, which was monitored by restoration of deleted nptII. When DNA fragments (168, 414, 658, and 792 bp) of nptII and linearized plasmid DNA (pGEM-T-1, 3168 bp and pGEM-T-2, 3792 bp) containing the 168 bp or 792 bp fragments, respectively, were used as donor DNA, it was observed that the efficiency of marker rescue decreased with decreasing length of homologous sequence. The use of a larger plasmid (pMR2, 5786 bp) containing the 792 bp fragment revealed higher efficiency of marker rescue compared to pGEM-T-2. The nptII fragments resulted in lower efficiencies compared to plasmid DNA containing the same fragment. For the 792 bp fragment and the linearized plasmid pMR2 a first-order dependency of the frequency of marker rescue transformation on the DNA concentration was observed. Based on the acquired data, the hypothetical frequency of transformation of transgenic DNA to B. subtilis in cooked potatoes was calculated to be equal to 8.5 x 10(-19) and 1.2 x 10(-27) for homologous and illegitimate recombination, respectively. These data permit to roughly estimate the time after which a person (10(8) years) or the world population (15 days) is exposed to one transformant generated by homologous recombination event, when the daily consumption per person is 130 g of cooked potatoes.     

The size and structure of the DNA genome of symbiont xenosome particles in the ciliate Parauronema acutum

The size and structure of the DNA genome of xenosomes, bacterial endosymbionts of the marine hymenostome ciliate, Parauronema acutum 110-3, were investigated. Renaturation kinetic measurements, determined optically and by hydroxyapatite chromatography, suggested a genome size of 0.34 x 10(9) daltons. Sedimentation rate measurements of DNA gently released from the symbionts yielded molecules of comparable size. The analytical complexity, determined chemically, was 3.03 x 10(9) daltons. Consistent with these and other data is a model for the structure of the symbiont genome in which the DNA exists in the form of nine circularly permuted, double-stranded DNA molecules of unique sequence, each of molecular weight 0.34 x 10(9). It is suggested that xenosomes and certain symbionts found in ciliated protozoa may be extant forms of once free-living bacteria that have adapted to the intracellular environment.     

Integration and Repair of Ultraviolet-Irradiated Transforming Deoxyribonucleic Acid in Haemophilus influenzae

The extent of association between donor transforming deoxyribonucleic acid (DNA) and recipient DNA in Haemophilus influenzae as a function of ultraviolet (UV) dose to the transforming DNA has been measured by isopycnic analysis of lysates of (3)H-labeled recipient cells exposed to DNA labeled with (32)P and heavy isotopes. Except for doses above 15,000 ergs/mm(2), the results of these measurements are in good agreement with previous estimates made by another technique. Experiments with a mutant temperature sensitive for DNA synthesis and another mutant defective in excision of pyrimidine dimers suggest that the discrepancy between the methods of high doses results from DNA synthesis, in which portions of the associated donor DNA containing pyrimidine dimers are excised and broken down, and the components are reutilized for synthesis.Repair of UV-irradiated, transforming DNA during incubation of recipient cells is observed as an increase in transforming ability when fractions from CsCl gradients of cell lysates are assayed on excision-deficient cells. When transforming DNA containing markers of different UV sensitivities is used, repair of the UV-resistant nov marker by excision proficient cells takes place exclusively in the donor DNA that is associated with recipient DNA, and this repair is observed even in the absence of DNA synthesis. However, no repair is observed in the case of the more UV-sensitive str marker, possibly because excision events may remove a large fraction of the integrated str markers in addition to repairing a small fraction of the integrated DNA containing this marker.     

An international collaborative study of ''genetic drift'' in Salmonella typhimurium strains used in the Ames test

The efficiency of Weigle reactivation of ultraviolet light-irradiated single and double-stranded phi X174 DNA by wild-type and excision repair-defective E. coli hosts was determined. After limited exposure to ultraviolet light, the efficiency of Weigle reactivation by an ultraviolet light-irradiated wild-type host was greater for double-stranded phi X174 DNA than for its single-stranded counterpart. However, the efficiency of inducible recovery of the double-stranded DNA molecule decreased as its exposure to ultraviolet light increased until it became constant at a value 1.5 times less than that for single-stranded form of phi X174 DNA. The efficiency of Weigle reactivation of the single-stranded DNA molecule by the same host, however, was independent of the dose to the DNA, as were the efficiencies of reactivation for both forms of phi X174 DNA by ultraviolet light-irradiated excision repair-deficient hosts. In excision repair-defective hosts the efficiency of Weigle reactivation of double-stranded phi X174 DNA was also 1.5 times less than that for the single-stranded molecule. These results suggest that the Weigle reactivation of double-stranded phi X174 DNA is mediated in part by an excision repair process, and that this component of Weigle reactivation eventually can be saturated by ultraviolet light-induced DNA damage leaving other repair processes, such as trans-damage synthesis, responsible for the remaining inducible reactivation.     

On the Nature of Recombinants Formed during Transformation in Hemophilus influenzae

During the process of transformation in Hemophilus influenzae integration of donor DNA, i.e. the formation of recombinant DNA, involves the incorporation of single-stranded DNA. Evidence was obtained from cesium chloride density gradient centrifugation of DNA from donor-recipient complexes that integration was accompanied by the formation of hybrid DNA with a density intermediate with respect to heavy, ²H, ¹⁵N, donor and light, ¹H, ⁴N recipient DNA. On denaturation the position of the heavy donor DNA moved closer to, but not all the way toward, the density position of the original donor DNA. In addition to supporting the idea of single-stranded incorporation, this evidence suggested that the integrated donor DNA was covalently linked to light recipient DNA. The DNA was taken up in the double-stranded form and no detectable amounts of denatured DNA could be found during the transformation process. However, during the process of integration an amount of donor atoms, equivalent to the amount of hybrid DNA formed, appeared in recipient DNA, and indicated that while one strand of DNA was integrated the other was broken down and resynthesized. The density of the hybrid DNA, as well as rebanding of denatured hybrid, indicated that the size of the integrated piece of DNA was large, approximately 6 x 10⁶ daltons.     

Fate of Donor Deoxyribonucleic Acid in a Highly Transformation-Deficient Strain of Haemophilus influenzae

A transformation-deficient strain of Haemophilus influenzae (efficiency of transformation 10⁴-fold less than that of the wild type), designated TD24, was isolated by selection for sensitivity to mitomycin C. In its properties the mutant was equivalent to recA type mutants of Escherichia coli. The TD24 mutation was linked with the str-r marker (about 30%) and only weakly linked with the nov-r2.5 marker. The uptake of donor deoxyribonucleic acid (DNA) was normal in the TD24 strain, but no molecules with recombinant-type activity (molecules carrying both the donor and the resident marker) were formed. In the mutant the intracellular presynaptic fate of the donor DNA was the same as that in the transformation-proficient (wild-type) strain, and the radioactive label of the donor DNA associated covalently with the recipient chromosome in about the same quantity as in the wild type. However, many fewer donor atoms were associated with segments of the mutant's recipient chromosome as compared with segments of the wild-type chromosome. In the mutant the association was accompanied by complete loss of donor marker activity. The lack of donor marker activity of the donor-recipient complex of DNA isolated from the mutant was not due to lack of uptake of the complex by the second recipient and its inability to associate with the second recipient's chromosome. Because the number of donor-atom-carrying resident molecules was higher than could be accounted for by the lengths of presynaptic donor molecules, we favor the idea that the association of donor DNA atoms with the mutant chromosome results from local DNA synthesis rather than from dispersive integration of donor DNA by recombination.