Search

We have identified two promoters of the Escherichia coli phr gene by DNA deletion mapping, S1 mapping of transcripts and sequence homology. The weaker promoter, P2, located approximately 530 bp upstream from the start codon, extends beyond the previously known nucleotide sequence. The stronger, P1, lies 90 bp from the gene and is distinct from three previously described promoter-like sequences nearby. beta-Galactosidase production from a plasmid-borne gene, promoted by a synthetic copy of P1, increases after DNA damage, but the increase does not depend on the SOS-box-like sequences normally present in the vicinity of P1. Induction still requires intact recA and lexA genes, and also intact sulA.

DNA photolyase, a DNA repair enzyme encoded by the phr gene of Escherichia coli, is normally regulated at 10 to 20 active molecules per cell. In purA mutants deprived of adenine, this amount increased sixfold within 2 h. Operon fusions placing lacZ under transcriptional control of phr promoters indicated no change in transcription rate during adenine deprivation, and gene fusions of phr with lacZ showed a nearly constant level of translation as well. Immunoblot analysis indicated that the total amount of photolyase protein remained constant during enzyme amplification. On the other hand, treatment of cells with chloramphenicol during the adenine deprivation prevented any increase. DNA regions lying 1.3 to 4.2 kb upstream of the phr coding sequences were necessary for this amplification to occur and for this purpose would function in trans. These results suggest that adenine deprivation leads to a posttranslational change, involving synthesis of protein encoded by sequences lying upstream of phr, which increases photolyase activity. The amplification in activity was found to be reversible, for when adenine was restored, the photolyase activity declined before cell growth resumed.

The phr gene of Escherichia coli K-12 encodes the light-dependent, DNA repair enzyme photolyase, which removes UV light-induced pyrimidine dimers from cellular DNA. From Southern hybridization analysis of several strains containing successively extended phr deletions, we have determined the direction of transcription of the phr gene on the E. coli K-12 chromosome. Northern (RNA) hybridization analysis suggests that the phr gene is cotranscribed with a previously identified gene of unknown function (orf169) into two messages of different lengths. S1 nuclease mapping analysis indicates that the two transcripts share a single termination site but initiate at two different sites. Finally, we have determined that the presence of orf169 is not necessary for phr gene activity in vivo.

Bacillus subtilis strains UVSSP-42-1 (hcr42 ssp1) and UVSSP-1-1 (hcr1 ssp1) are ultraviolet (UV) radiation sensitive both as dormant spores and as vegetative cells. These strains are unable to excise cyclobutane-type dimers from the deoxyribonucleic acid (DNA) of irradiated vegetative cells and fail to remove spore photoproduct from the DNA of irradiated spores either by excision (controlled by gene hcr) or by spore repair (controlled by gene ssp1). When irradiated soon after spore germination, these strains excise dimers, but not spore photoproduct, from their DNA. This process, termed germinative excision repair, functions only transiently in the germination phase and is responsible for the high UV resistance of germinated spores and for their temporary capacity to host cell reactivate irradiated phages infecting them. The recA1 mutation confers higher UV sensitivity to the germinated spores, but does not interfere with dimer removal by germinative excision repair.

A survival assay with chick-embryo fibroblasts was used to study photoreactivation of ultraviolet (UV) irradiation-induced damage. The kinetics of the photoreactivation was studied as a function of the length of a post UV dark incubation period of from 0 to 18 h at 38.5 degrees C. The logarithmic survival curve with no photoreactivation had a Do of about 4.3 J/m2 giving approximately 0.8% relative plating efficiency after a UV dose of 21 J/m2. At this dose the efficiency of photoreactivation (survival increase per unit blacklight dose) increased with post UV incubation time reaching a maximum at 4-6 h, then declining until there was little photoreactivation observed for times longer than about 11 h. The possibility that this effect was produced by pre-UV perturbations of the cell cycle was eliminated by the fact that the same results were achieved after several rather different trypsinization protocols. The shape of the photoreactivation vs. blacklight curve at the time of peak efficiency showed a threshold up to about 3 kJ/m2, a rising portion and a plateau after 12-16 kJ/m2 when the survival increased by a factor of roughly 8.

A genetic technique has been developed for isolating nuclear hybrids which are products of fusion between karyoplasts and whole cells. Nuclear hybrids have greatly reduced mitochondrial component obtained from the nuclear donor and exhibit differences when compared to the corresponding whole-cell hybrids, such as better long-term survival and accelerated chromosome segregation. These results suggest that divergent mitochondria in a cell-fusion product may have adverse effects on survival.

Two Escherichia coli K-12 Hfr strains have been constructed which transfer a recA deletion, which is highly linked to a Tn10 insertion conferring tetracycline resistance, early during conjugation. These strains transfer the recA deletion in opposite directions with different origins of transfer, allowing for preservation of desirable recipient strain markers either clockwise or counterclockwise of recA.

Colony-forming ability of Potorous tridactylus-kidney (PtK-2) cells was measured after exposure to 254-nm ultraviolet (UV) light and cycloheximide. Addition of 5 microM cycloheximide after exposure of the cells to UV light greatly decreased cell survival. Maximum effect of the drug was obtained by 24-h exposure after irradiation. The cycloheximide sensitivity of irradiated cells was eliminated if addition of the drug was delayed for 8-10 h after irradiation, or if the cells received photoreactivating light treatment prior to cycloheximide exposure. Thus, a major component of pyrimidine dimer removal may involve a cycloheximide-sensitive mechanism.

The DNA of UV-irradiated Bacillus subtilis spores, which contains 5-thyminyl-5,6-dihydrothymine (TDHT) as the major thymine photoproduct, is known to be repaired during germination by two complementary mechanisms: (I) the well-known excision repair, and (2) a special process, "spore repair", which destroys TDHT in situ without rendering it acid-soluble. In the absence of both mechanisms TDHT is not removed, and spores are highly UV-sensitive. When either of two mutations (pol-59 and pol-151) giving defective DNA polymerase, or one (rec-A1) giving a recombination deficiency are introduced into strains defective in one of these known TDHT removal processes, the chemically measured elimination of TDHT from spore DNA is unaltered, but spore UV-sensitivity is increased. The pol mutations produce their greatest sensitivity increase in spores of strains already deficient for the in situ destruction of TDHT, while the rec mutation gives its maximum sensitivity increase to spores of strains lacking excision. These facts argue that the pol mutations interfere mostly with excision repair (presumably its later resynthesis step), shile the rec mutation impairs "spore repair" in some step occurring subsequent to the TDHT destruction in situ. With either of these impairments of the later repair steps, DNA of UV-irradiated and germinated spores is considerably degraded, unless germination is carried out in the presence of chloramphenicol.

DNA-membrane complexes have been obtained from Escherichia coli by using a freeze-thaw lysis procedure that avoids lysozyme and detergents. Complexes made in this manner and containing DNA near the origin of replication are uniquely sensitive to ionic strength, Pronase, and trypsin. There is approximately one such complex per chromosomal origin. The sensitivities suggest that origin-specific binding is mediated by a protein. By using these unique characteristics to distinguish origin-specific complexes from the majority of DNA-membrane binding sites, it was found that the origin-specific binding persists after termination of chromosomal replication.

We have determined the nucleotide sequence of a 2039-base pair segment of Escherichia coli chromosomal DNA containing the phr gene, which encodes deoxyribopyrimidine photolyase. The coding region of phr is 1416 base pairs and is preceded by regions homologous to consensus sequences for E. coli promoters and ribosome binding sites. The phr gene is preceeded by an open reading frame of 169 codons (orf169) which is transcribed in the same direction. The proximity of orf169 to phr suggests that both are members of a single operon containing one or more internal promoters allowing differential expression of phr. An unusually large number of rare or infrequently used codons are utilized in phr, which may contribute to the low copy number of photolyase. The sequence at the NH2 and COOH termini and the overall amino acid composition of mature photolyase, determined using purified protein, agrees with predictions based upon the nucleotide sequence. Photolyase consists of 471 amino acids and has a calculated molecular weight of 53,994.

Exponentially growing cells of the PtK-2 line (ATCC No. CCL56, from the marsupial Potorous tridactylus) require protein and RNA synthesis in a limited period following UV-radiation damage for optimal recovery as colony formers [Overberg et al. (1988) Mutat. Res. 194, 83-92]. Overall behavior suggests the operation of damage-induced recovery processes. The capacity of confluent cell monolayers for infection with unirradiated herpes simplex virus 1 (HSV-1) is sharply reduced by UV-irradiation. We have followed capacity changes in exponentially growing cells after irradiation and varying amounts of photoreactivation by means of an infectious center assay. These changes closely parallel changes of colony formation. Spontaneous recovery of capacity in the dark occurs over approximately the same time period that the UV sensitivity of colony formation depends on macromolecular synthesis. The effect of photoreactivation is complementary rather than additive to this recovery, suggesting that the dark recovery in this period concerns pyrimidine dimers in cell DNA.

The absolute action spectrum of Escherichia coli DNA photolyase was determined in vitro. In vivo the photoreactivation cross-section (epsilon phi) is 2.4 X 10(4) M-1 cm-1 suggesting that the quantum yield (phi) is about 1.0 if one assumes that the enzyme has the same spectral properties (e.g. epsilon 384 = 1.8 X 10(4) M-1 cm-1) in vivo as those of the enzyme purified to homogeneity. The relative action spectrum of the pure enzyme (blue enzyme that contains FAD neutral semiquinone radical) agrees with the relative action spectrum for photoreactivation of E. coli, having lambda max = 384 nm. However, the absolute action spectrum of the blue enzyme yields a photoreactivation cross-section (epsilon phi = 1.2 X 10(3) at 384 nm) that is 20-fold lower than the in vivo values indicative of an apparent lower quantum yield (phi approximately equal to 0.07) in vitro. Reducing the enzyme with dithionite results in reduction of the flavin semiquinone and a concomitant 12-15-fold increase in the quantum yield. These results suggest that the flavin cofactor of the enzyme is fully reduced in vivo and that, upon absorption of a single photon in the 300-500 nm range, the photolyase chromophore (which consists of reduced FAD plus the second chromophore) donates an electron to the pyrimidine dimer causing its reversal to two pyrimidines. The reduced chromophore is regenerated at the end of the photochemical step thus enabling the enzyme to act catalytically.+

Photoreactivating enzyme (PRE) from yeast (as semi-crude extract, or in highly purified form) shows increased activity if it is illuminated with near UV or short wavelength visible light prior to its use for photoenzymatic repair of UV-induced pyrimidine dimers in transforming DNA in vitro. This effect results from an alternation in PRE molecules changing those with low activity in the light-dependent step of the reaction to a higher activity. Light-induced activation of PRE preparations is slowly lost by dark storage for several hours to 1 day (faster at 23 degrees C than at 5 degrees C), but can be recovered repeatedly by renewed preillumination. The action spectrum for these preillumination effects generally resembles that for the photoenzymatic repair reaction itself, having its maximum in the same 355-385 nm region as the latter, but light of somewhat longer wavelengths (546 nm) is still effective. Preilluminated PRE is also more stable to thermal inactivation (65 degrees C) than untreated enzyme.

A rapid method has been developed for isolation of ultraviolet-sensitive mutants of Escherichia coli, by inducing delay in the growth and/or division of repair-deficienct cells with low fluences of far-ultraviolet radiation, and killing with penicillin the repair-proficient cells, which continue to grow and divide. With this technique, we have achieved about a 3,000-fold enrichment for photoreactivation less (phr) cells and have isolated and characterized three phr mutants.

A previously reported salt-sensitive binding of deoxyribonucleic acid (DNA) to the cell envelope in Escherichia coli, involving approximately one site per chromosome near the origin of DNA replication, is rapidly disrupted in vivo by rifampin or chloramphenicol treatment and by amino acid starvation. DNA replication still initiates with this origin-specific binding disrupted, even when the disruption extends over the period of obligatory protein and ribonucleic acid synthesis that must precede initiation after release of cells from amino acid starvation. Thus the origin-associated membrane-DNA interaction is not necessary either for the initiation event itself or for the maturation of a putative initiation apparatus in E. coli.

The biologic effects of visible and UV light result from photochemical changes in cell components. The amount of photochemical change induced in a small non-self-shadowing structure is proportional to the number of photons traversing it per unit area normal to the direction of propagation, summed over all component beam directions. Within an optically complex, absorbing, and scattering structure, this quantity is difficult to determine, but for skin it is approximately proportional to the total number of photons per unit area entering its outer surface. The magnitude of some photobiologic effects depends on the total amount of photochemical change induced, whereas others depend on the rate of photoproduct formation or on a more complex relation. The nature of the dependence must be determined before light measurements can be related to the magnitude. The effect of a polychromatic illumination depends on its wavelength distribution, weighted by the effectiveness of each wavelength (the action spectrum) under the conditions employed. Until the latter is known, no dosimetric characterization of the light is possible. The wavelength distribution can be determined by spectroradiometric measurement, with the weighting performed numerically, or (more conveniently, though less accurately) by the use of an analog reaction with an action spectrum like that of the photobiologic effect.

The ultraviolet sensitivity of Potorous tridactylus male kidney (PtK-2) cells is markedly increased by post irradiation treatment for 24 h with 5 microM emetine (which inhibits protein synthesis by acting on the 40S ribosomal subunit), or with 5 microM cycloheximide (which inhibits by interaction with the 60S subunit), or with the RNA polymerase II inhibitor 5,6-dichloro-1-beta-ribofuranosylbenzimidazole at 50 microM. All 3 treatments give the same sensitivity, while unirradiated cells are little affected. Shortening the time of treatment, or delaying application of the drugs decreases their effect on the same time schedule. Preiirradiation of cells, with no drug treatment in the following 8 h, diminishes the sensitivity to a subsequent irradiation with protein synthesis blocked afterwards. Photoreactivation immediately following such preirradiation eliminates its desensitizing effect. Inhibiting protein synthesis after irradiation also markedly reduces the frequency of UV-induced mutants in the surviving population. These facts suggest that gene expression in the period following irradiation facilitates recovery from radiation damage, with an increased probability of mutation, reminiscent of the "SOS response" in Escherichia coli.

A new technique is developed for physically enriching recombinant DNA molecules in an in vitro recombination mixture. UV-irradiation of the donor DNA before recombination enables photoreactivating enzyme (PRE) (deoxyribodipyrimidine photolyase, EC 4.1.99.3) to attach to the donor segments in recombinant molecules. This attached protein causes retention of the recombinant molecules on a nitrocellulose filter, while molecules not containing donor DNA pass through. The bound DNA is repaired of its UV damage and released for insertion into cells by exposure to photoreactivating light in situ, yielding approx. 350-fold enrichment. Although applicable to any gene, this procedure has been used in cloning the Escherichia coli phr gene itself, permitting 100-fold amplification of the gene product in vivo.

Ultraviolet-inactivated Hemophilus influenzae transforming DNA recovers its activity when mixed with cell-free extracts of bakers' yeast and exposed to visible light. The active agent in the extract is not used up in the reaction, and purification has not separated it into more than one non-dialyzable component. It differs from the agent in Escherichia coli extract, which produces very similar photoreactivation, but which can be resolved into non-dialyzable and dialyzable components, the latter being used up during illumination. The yeast agent can be salted out of solution and recovered quantitatively; it is inactivated by crystalline trypsin and chymotrypsin and by brief heating at 60 degrees C.-all facts suggesting that it is an enzyme for which ultraviolet lesions in the DNA serve as substrate. The kinetics of recovery are also consistent with such an assumption. This enzyme is unusual both because it is involved in a light-dependent reaction and because it has a non-destructive action on DNA outside an intact cell.

The photoenzyme from bakers' yeast which repairs ultraviolet-inactivated transforming DNA is mechanically bound to ultraviolet-irradiated DNA in the dark, but not to unirradiated DNA. In the bound condition it is stabilized against inactivation by heat and heavy metals. Both the mechanical binding and stabilization are eliminated by illumination. These observations are consistent with the reaction scheme suggested by kinetic studies, in which the enzyme combines with the ultraviolet lesions in DNA and the complex absorbs light, producing repair and subsequent liberation of the enzyme. The approximately exponential decrease of heat stabilization during illumination gives the first order rate constant for the light-dependent step at the corresponding light intensity. This quantity in turn sets limits on the possible magnitude of the molar absorption coefficient of the enzyme-substrate complex and on the quantum yield of the process.

Previous genetic analysis indicated that at least two genes determine the ultraviolet (UV) sensitivity of Bacillus subtilis spores. The present study shows that these genes independently control two distinguishable processes for removing UV-induced spore photoproduct (5-thyminyl-5,6-dihydrothymine, or TDHT) from spore deoxyribonucleic acid. The first, is a spore repair mechanism by which TDHT is removed rapidly without appearing in acid-soluble form. This mechanism, which is demonstrated in both UV-resistant and excision-deficient strains, operates to a certain extent during germination without requiring vegetative growth. The second, demonstrated in a mutant which lacks the first mechanism, removes TDHT relatively slowly and only if germinated spores are allowed to develop toward vegetative cells. The latter mechanism appears identical to excision-resynthesis repair, since the mutation abolishing it renders the irradiated vegetative cells incapable of removing cyclobutane-type pyrimidine dimers. Blocking either one of these mechanisms only slightly affects the UV sensitivity of spores, but blocking both prevents TDHT removal and gives high UV sensitivity.

As previously reported, ultraviolet-inactivated bacterial transforming DNA can be restored to activity by an enzyme-like agent from bakers' yeast which requires light for its activity. Kinetics of this reaction, in the presence and absence of inhibitors, are found consistent with the Michaelis-Menten reaction scheme, with the sites of ultraviolet damage on the DNA serving as substrate and the repaired structure as product. Kinetic studies with different light intensities suggest that the necessary illumination causes photolysis of the enzyme-substrate complex with concurrent repair of the DNA. Competitive inhibition of irradiated transforming DNA repair, which occurs when irradiated non-transforming DNA is present in the same reaction mixture, permits ultraviolet damage (of the kind capable of being photoreactivated) to be detected in any type of DNA.