No full-text available
To read the full-text of this research,
you can request a copy directly from the author.
Analysis of protein—nucleic acid filter binding using the poisson distribution: A method to estimate co-operativity
... In 0.68 M NaCI, NaClO,, or NaF, the binding affinity is reduced, and the binding curves appear to have a sigmoidal nature (Fig. 3). This may be the result of cooperativity observed earlier in this system by other means (3,32). In 0.35 M NaF and 0.68 M NaF, the initial protein fluorescence is reduced by the fluoride ion quenching, prior to addition of fd-DNA or poly(dT). ...
... Accordingly, the protein should have the capacity to form dimers. This we have found by analytical ultracentrifugation, and evidence suggestive of it has been presented earlier by others (4,32,34). We propose here, as a working hypothesis, that the monomer-dimer equilibrium, a relatively weak interaction, involves portions of the gene V monomer which function in holding together two single strands of gene V protein bound to DNA. ...
Gene V protein exists principally as a dimer in neutral buffers which are 0.15 M in NaCl, NaF, or NaClO4. Higher concentrations of NaClO4 or NaCl disrupt the dimers, but higher concentrations of NaF do not. There is a single sulfhydryl group per gene V protein monomer in native monomers, dimers, and DNA-protein complexes. Disulfide formation leads to loss of protein solubility and DNA binding capacity. The fluorescence of tyrosyl groups is the same for monomers and dimers in NaCl and NaClO4 solutions, but it is extensively quenched on binding to poly(dT) and fd-DNA. Complexes of gene V protein and fd-DNA isolated from lysates of infected cells were found to contain 4.70- +/- 0.13 nucleotides per monomer of gene V protein whereas complexes formed in vitro contain 4.05 +/- 0.17 nucleotides/monomer. It is postulated that tyrosyl groups are not involved in the protein-protein interactions of the monomer-dimer equilibrium, that tyrosyl groups stack with DNA bases in the complexes, and that each subunit of gene V protein in the intracellular complexes with fd-DNA is replaced by exactly two subunits of major coat protein during final assembly of the virus.
... such case.Ikehara et al., 1975; Oey & Knippers, 1972) (Lohman & Kowalczykowski, 1981; Schneider & Wetmur, 1982Park et al., 1982;Winter et al., 1981; Lohman & Kowalczykowski, 1981Dunker, 1975).Figure 8a shows the results obtained from the binding time course for three numbers of binding sites. Since the binding starts with the formationof an isolated complex at a slower rate, there is clearly seen a time lag that is proportional to the-saturationof 100,200, and 800 binding sites are 3.3,2.4, ...
The highly cooperative binding of fd gene 5 to single-stranded DNA was studied kinetically by rapid photo-cross-linking and stopped-flow UV absorption measurements. The observed change in absorbance was shown to be due to the binding by direct evidence of rapid photo-cross-linking of the bound proteins to fd DNA. The bimolecular rate constant obtained for the association was 1.6 X 10(10) M-1 s-1 (in terms of the molecular concentration of DNA), which was concluded to be diffusion controlled. The breakdown of cluster complexes on fd DNA was induced by the addition of large excess amounts of short single-stranded DNA. The breakdown took place in about 1 s. The kinetic process of redistribution of dissociated proteins was monitored by rapid photo-cross-linking and subsequent electrophoresis of the cross-linked complex. The dissociated proteins first formed isolated complexes, but later they were again converted into the cluster. The kinetic results showed that the cooperativity originated from the stabilization of the protein-DNA complex by the cluster formation, not from the accelerated association in the cluster formation. This kind of cooperative binding was shown to perform negative feedback control in the cluster formation. On the basis of the kinetic results obtained, we proposed a model for the regulatory role of the fd gene 5 protein in the synthesis of single-stranded fd DNA.
Publisher Summary This chapter focuses on the proteins that bind preferentially and nonspecifically to single-stranded DNA and have no other (enzymatic) activity. These proteins are essential to many physiological functions, including replication, recombination, and repair, in a host of organisms ranging from bacteriophage to higher eukaryotes. Thus, single-stranded DNA binding proteins represent systems that have evolved substantially beyond primitive precursors, which may only have been capable of direct and uncontrolled nucleic acid binding. The chapter describes molecular aspects of the involvement of DNA binding proteins in entire systems of DNA replication, recombination, and repair. Nature and measurement of DNA–protein interactions along with general purification strategies for single-stranded DNA binding proteins are also presented. In addition, the chapter discusses the ways in which the single-stranded DNA binding proteins have been exploited as tools in molecular biological research, particularly in the electron microscopy of biological macromolecules and in certain biochemical assays. All DNA binding proteins seem to operate stoichiometrically (as opposed to catalytically), in that they are present at intracellular levels sufficient to effectively saturate the single-stranded DNA intermediates produced during replication, recombination, and repair. Binding cooperativity is essential in permitting complete coverage of single-stranded sequences and also in effectively destabilizing the small duplex hairpins formed by intrastrand base pairing in single-stranded DNA.
H1 factor is a heat-stable protein found in large amounts in Escherichia coli. In vitro, this protein has been found to stimulate transcription of λ templates by E. coli DNA-dependent RNA polymerase.
The subunit molecular weight of this factor has been re-estimated and found to be 15500 ± 1000 in the presence of 2-mercaptoethanol. Crosslinking experiments performed with dimethylsuberimidate in the presence or absence of the DNA template indicate the presence of multiples of the 15500-Mr subunit up to the tetramer, the dimeric species being predominant. One cysteinyl residue per 15000-Mr subunit is labeled by 4-chloro-7-nitrobenzofurazan. This residue is not labeled if the factor is exposed to oxidizing conditions. In this case, three lysyl residues are titrated.
H1 factor behaves as a DNA-binding protein. We have detected binding to DNA by two independent methods: displacement of a fluorescently labeled factor by the native protein and retention of radioactive DNA on millipore filters in the presence of the factor. Under our experimental conditions (high ionic strength, absence of magnesium ions), the saturation function of λplac DNA as well as of wild-type λ DNA has been found to be non-cooperative. Saturation is reached when 300 ± 30 molecules of dimeric factor are bound per λ molecule, the average dissociation constant of the complex being 10 nM.
The dissociation time of the H1· DNA complex is less than 5 s at 37°C. The binding of this factor lowers the affinity of native DNA for ethidium bromide. In the presence of this intercalating dye, the solubility of the complex decreases drastically.
A filter assay for Eco helix-destabilizing Protein I is reported and shown by immunological analysis specifically to detect Eco HD Protein I in crude extracts. The assay detects no activity in strains carrying a temperature-sensitive allele for Eco HD Protein I. The assay is linear over a 20-fold range and can reliably detect a nanogram or less of Eco HD Protein I.
A single-stranded DNA-dependent ATP γ-phosphohydrolase of Mr 56000 induced after infection of Escherichia coli cells with bacteriophage T4, probably the ATPase dependent on gene dda of the phage, was isolated. Studies on the enzyme show that in the presence of ATP and Mg2+ ions it is capable of dissociating partially double-stranded fd bacteriophage DNA into the single strands and that some 3000 enzyme copies are required to unwind the 6400-nucleotides-long DNA. Unwinding is inhibited by reducing the length of the single-stranded portion of DNA to two nucleotides. In addition it can be inhibited by sulfhydryl reagents which block the ATPase or by trapping free enzyme molecules in the assay system. The results suggest that unwinding is initiated near the single-stranded portion of the DNA and is driven by the ATPase. It further appears that the enzyme unwinds by adsorbing to the DNA. Affinity of the enzyme for double-stranded DNA is not detectable by DNA binding assay.
The gene-5 protein of the fd filamentous bacterial virus binds to single-stranded DNA over a pH range of 2-10.3. Binding to fd DNA is several hundred-fold stronger than to bacteriophage R17 RNA or to DNA tetranucleotides.
Isolated gene 5 protein from bacteriophage fd-infected Escherichia coli has been shown by sedimentation equilibrium to exist primarily as a dimer under non-denaturing conditions. The dimer was stable under conditions of high ionic strength, extremes in pH, dilution to 0.075 mg/ml, and increased temperature. Gene 5 protein did not undergo the indefinite self-association observed with gene 32 protein.Three lines of evidence for co-operative binding of gene 5 protein to DNA were developed. First, the interaction between gene 5 protein and phage T4 DNA was examined using a nitrocellulose filter assay. Scatchard plots of the binding data indicated that the interaction was co-operative. Similar results were obtained with gene 32 protein. Second, the co-operative binding of both proteins to DNA was shown by the sensitivity of the protein-DNA interaction to increasing ionic strength at various ratios of protein to DNA. Finally, by using the cross-linking agent, dimethyl suberixmidate, oligomeric structures containing at least seven monomers were found when the DNA was less than saturated.The possibility that gene 5 protein dimers undergo indefinite self-association in the presence of oligonucleotides was examined by sedimentation equilibrium. With oligo[d(pT)4], the protein dimer was complexed with this oligonucleotide but no self-association was observed. With oligo[d(pT)8], gene 5 protein formed tetramers, but no significant indefinite association was noted. These results do not suggest a DNA-induced conformational change, which results in indefinite association. A model for the co-operative binding of gene 5 protein to DNA is presented.
Previous studies have shown that the single-stranded DNA binding protein of bacteriophage f1 (gene V protein) represses the translation of the mRNA of the phage-encoded replication protein (gene II protein). We have characterized phage mutations in the repressor and in its target. Using a gene II-lacZ translational fusion, we have defined a 16-nucleotide-long region in the gene II mRNA sequence that is required in vivo for repression by the gene V protein. We have shown that in vitro the binding affinity of the gene V protein is at least 10-fold higher to an RNA carrying this sequence than to an RNA lacking it. We propose that this sequence constitutes the gene II mRNA operator.
The gene V protein of the filamentous bacteriophages fl, fd and M13, and the gene 32 protein of bacteriophage T4 share the property of binding strongly and co-operatively to single-stranded nucleic acids, especially DNA. Moreover, both are capable of repressing the translation of specific mRNAs (gene 32 protein its own, and gene V protein that of the filamentous phage gene II), both in vivo and in vitro. If the mechanism of repression by either of these proteins were based solely on its ability to bind single strands co-operatively, then the other would be expected to mimic or interfere with its effect in vitro. We have found no such mimicry or interference, even at protein concentrations high enough to have substantial non-specific effects on translation. This suggests that the sites of repression on the mRNAs must offer something other than simple "unstructuredness" for binding and repression to occur.
A new type of protein essential for DNA replication and genetic recombination has been isolated from T4 bacteriophage-infected cells of E. coli. This protein binds cooperatively to single-stranded DNA, and it catalyses DNA denaturation and renaturation in physiological conditions in vitro.
Excerpt
The phenomenon of lysogeny was cited by Jacob and Monod in 1961 as an example of gene control by repressors. It was hypothesized that each temperate phage produces a repressor which specifically blocks the functioning of early genes of that phage. Since the late genes function only if provided with certain early gene products, the proposed repressor action would suffice to turn off all the lytic phage genes, thereby permitting lysogenization. According to these ideas, the dormant phage genes would resume functioning upon removal of the repressor, and it was proposed that inducing agents, such as UV light, function by inactivating the repressor.
Several of the basic aspects of this picture may now be formulated in biochemical terms. The λ and 434 phage repressors, the products of their respective CI genes, have been isolated and their activities explored in vitro (Ptashne, 1967a and 1967b; Pirrotta and Ptashne, 1969). Both repressors
The product of gene 5 of filamentous bacteriophages is required for synthesis of the single-stranded progeny DNA. When extracts made from Escherichia coli infected with bacteriophages fd or M13 are chromatographed on single-stranded DNA/cellulose, a DNA-binding protein can be eluted which has a molecular weight of about 10,000 daltons and is made in at least 100,000 copies per cell. This protein is altered in amber mutant 5-H3 and temperature-sensitive mutant 5-HS1 of phage M13, which identifies it as the gene 5 product (Henry & Pratt, 1969). As judged by sucrose-gradient sedimentation at 4 °C, the pure protein binds tightly and co-operatively to single-stranded, but not to double-stranded, DNA's; at saturation, one protein monomer is bound per every 4 DNA nucleotides. This selective DNA-binding enables the gene 5 protein to denature double-stranded DNA's rapidly at physiological temperatures. In these respects, the gene 5 protein resembles the T4 bacteriophage gene 32 protein described previously, a protein which is required for both the replication and recombination of T4 bacteriophage DNA. However, electron microscopy reveals that the structure of the gene 5 protein complex with single-stranded DNA is quite different: whereas gene 32 protein forces the DNA into an extended linear conformation, the gene 5 protein coalesces two protein-covered DNA stands into a helical, rodlike structure. For this and other reasons, these two “DNA unwinding” proteins could have quite different roles in the replication process.
Gene 5 protein of bacteriophage fd was purified from phage infected bacteria. The protein binds to single-stranded DNA but not to double-stranded DNA. The protein inhibits the bacterial exonucleases and, at high concentrations, the cellular DNA polymerases I and II. Gene 5 protein, at low concentrations, increases the template activity of a viral replicative form DNA for polymerase II.
Gene 5 (G5) protein of fd bacteriophage has an unusual circular dichroic (CD) spectrum with a positive band at 228 nm. This band has been assigned to tyrosine according to its pH dependence and data for model compounds. Computer fitting to reference CD spectra for α-helical, β, and random conformations, as well as for tyrosyl contributions, show that the CD of G5 protein is that of a non-α-helical protein. Binding to single-stranded DNA leads to little change in average peptide conformation, although changes in the tyrosyl CD band and in tyrosyl difference absorbance bands are observed. Changes in absorbance intensity over a broad region around 262 nm show that the bases of single-stranded DNA in complexes with G5 protein are unstacked regardless of their prior state in isolated DNA. The various spectral changes have been used to show that each protein monomer of 10,000 daltons binds four nucleotides. The fd DNA-G5 protein complex, having unstacked bases and non-α-helical subunits, is very different from the fd bacteriophage particle, which has stacked bases and highly α-helical subunits of G8 protein. It is suggested that circular single-stranded fd DNA is shifted from an unstacked to a stacked state by means of a subunit exchange reaction during the course of replication and phage assembly.
Lysates of bacteriophage M13-infected cells contain numerous unbranched filamentous structures approximately 1·1 μm long × 160 Å wide, that is, slightly longer and considerably wider than M13 virions. These structures are complexes of viral single-stranded DNA molecules with M13 gene 5 protein, a non-capsid protein required for single-stranded DNA production. All, or nearly all, of the single-stranded DNA from the infected cells and at least half to two-thirds of the gene 5 protein molecules are found as complex in the lysates. The complex contains about 1300 gene 5 protein molecules per DNA molecule but little if any of the two known capsid proteins. The complex is much less stable than virions in the presence of salt or ionic detergent solutions and in electron micrographs it appears to have a much looser and more open structure. If an excess of M13 single-stranded DNA is added to complex in a lysate, the gene 5 protein molecules from the complex redistribute onto all of the added as well as the original DNA, again suggesting a rather loose association of protein and DNA.By electron microscopy, the complex from infected cells appears to differ structurally from complex formed in vitro between purified single-stranded DNA and purified gene 5 protein. Because of this apparent structural difference and because previous experiments suggested the presence of complex in vivo, we presume that the complex which we have found in lysates of infected cells previously did exist as such inside the cells, but we have been unable to exclude that it formed during or after lysis. If it is assumed that complex does occur in vivo, the results of pulse-chase radioactive labeling experiments on infected cells can be interpreted as showing that with time the single-stranded DNA leaves complex, presumably to be matured into virions, while the gene 5 protein molecules are re-used to form more complex.
Initiation of the lethal process usually associated with abortive infection by filamentous virus requires two components: continued replication of viral duplex DNA and an unidentified temperature-sensitive component. We present evidence here that the temperature-sensitive component in the lethal process is viral gene 5 protein, and suggest that the size of the gene 5 protein pool influences duplex DNA replication in some critical way that determines whether or not an abortively infected cell will die.
During the first ten minutes after infection of bacteria with fd, the rate of DNA synthesis in an infected culture becomes several-fold larger than the rate in a parallel uninfected culture. This stimulation of rate is due to the synthesis of 100 to 200 double-stranded forms of viral DNA, superimposed on continuing bacterial DNA synthesis. At the end of the ten-minute period, the rate of viral plus bacterial DNA synthesis stops increasing, and remains constant for the next 50 minutes. The abrupt decrease in acceleration of net DNA synthesis corresponds in time to the onset of synthesis of single-stranded viral DNA.The results are discussed in terms of an inhibitor model for the switch from double-strand replication to single-strand synthesis.
The filamentous bacterial virus is a simple and well-characterized model system for studying how genetic information is transformed
into molecular machines. The viral DNA is a single-stranded circle coding for about 10 proteins. The major viral coat protein
is largely α-helical, with about 46 amino acid residues. Several thousand identical copies of this protein in a helical array
form a hollow cylindrical tube 1–2μ long, of outer diameter 60 Å and inner diameter 20 Å, with the twisted circular DNA extending down the core of the tube.
Before assembly, the viral coat protein spans the cell membrane, and assembly involves extrusion of the coat from the membrane.
X-ray fibre diffraction patterns of the Pf 1 species of virus at 4°C, oriented in a strong magnetic field, give three-dimensional
data to 4 Å resolution. An electron density map calculated from native virus and a single iodine derivative, using the maximum
entropy technique, shows a helix pitch of 5.9 Å. This may indicate a stretched A-helix, or it may indicate a partially 310 helix conformation, resulting from the fact that the coat protein is an integral membrane protein before assembly, and is
still in the hydrophobic environment of other coat proteins after assembly.
Physically homogeneous preparations of DNA can be isolated from fd. This DNA is resistant to hydrolysis by exonuclease I. The kinetics of pancreatic DNase action on the DNA as followed by sedimentation and viscosity in suitable solvents show that a single hit (biological inactivation) leads to an expansion of the DNA structure but no change in molecular weight. The native DNA molecule is resistant to a variety of agents which destroy secondary structure. These results are taken to indicate that the DNA has a stable ring topology.
- L Day
Day, L. (1973) Biochemistry 12, 5329-5339.
- B Alberts
- L Frey
Alberts, B. and Frey, L. (1970) Nature 227, 313-315.
- J L Oey
- R Knippers
Oey, J. L. and Knippers, R. (1972) J. Mol. Biol. 68,
125-138.
- A D Riggs
- S Bourgeois
Riggs, A. D. and Bourgeois, S. (1968) J. Mol. Biol. 34,
361-364.
- M Yarns
- P Berg
Yarns, M. and Berg, P. (1968) J. Mol. Biol. 42, 171-
189.
- K Temmis
- D A Marvin
Temmis, K. and Marvin, D. A. (1974) Virology 59,293-
301.
- D Pratt
- P Laws
- J Griffith
[ll ] Pratt, D., Laws, P. and Griffith, J. (1974) J. Mol. Biol.
82,425-439.
- B Alberts
- L Frey
- H Delins
Alberts, B., Frey, L. and Delins, H. (1972) J. Mol. Biol.
68, 139-152.