Domain effects on the DNA-interactive properties of bacteriophage T4 gene 32 protein.

Bacteriophage T4 gene 32 protein, a model for single-strand specific nucleic acid-binding proteins, consists of three structurally and functionally distinct domains. We have studied the effects of the N and C domains on the protein structure and its nucleic acid-interactive properties. Although the presence of the C domain decreases the proteolytic susceptibility of the core (central) domain, quenching of the core tryptophan fluorescence by iodide is unaltered by the presence of the terminal domains. These results suggest that the overall conformation of the core domain remains largely independent of the flanking domains. Removal of the N or the C terminus does not abolish the DNA renaturation activity of the protein. However, intact protein and its three truncated forms differ in DNA helix-destabilizing activity. The C domain alone is responsible for the kinetic barrier to natural DNA helix destabilization seen with intact protein. Intact protein and core domain potentiate the DNA helix-destabilizing activity of truncated protein lacking only the C domain (*I), enhancing the observed hyperchromicity while increasing the melting temperature. Proteolysis experiments suggest that the affinity of core domain for single-stranded DNA is increased in the presence of *I. We propose that *I can "mingle" with intact protein or core domain while bound to single-stranded DNA.

Details of the nucleic acid binding site of T4 gene 32 protein revealed by proteolysis and DNA Tm depression methods.

The affinities and location of oligonucleotides bound to intact and truncated bacteriophage T4 gene 32 protein have been elucidated by two independent and sensitive methods. The nucleic acid binding site is located within the core domain of 32 protein, residues 22-253. Oligonucleotides protect the core domain against proteolysis catalyzed by mammalian endoproteinase Arg-C. Of the three cleavage sites, Arg111, within the internal "LAST" ((Lys/Arg)3(Ser/Thr)2) motif, is selectively protected. We have previously suggested that these LAST residues, Lys-Arg-Lys-Thr-Ser, residues 110-114, are involved in nucleic acid binding, and our results are also consistent with crystallographic studies. The inhibitory effects of oligonucleotides on the kinetics of core domain proteolysis were used to quantify binding affinities. In addition, affinities of oligonucleotides for both core domain and intact protein were obtained from their effect on the Tm-depressing activities of these proteins. For both core and intact protein, the degree of affinity increases with oligonucleotide length. The presence of a 5' terminal phosphate increases the affinity two- to fourfold. Placement of methylphosphonodiester (uncharged) linkages at alternating linkages vastly lowers binding affinity for the intact protein and core domain. We conclude that at least two and likely three adjacent phosphodiester linkages are a minimal requirement for binding, further defining the electrostatic component of the interaction. The length-dependence of binding affinity suggests that additional interactions, both ionic and non-ionic, likely occur with longer oligonucleotides.

Monitoring metal ion flux in reactions of metallothionein and drug-modified metallothionein by electrospray mass spectrometry.

The capabilities of electrospray ionization mass spectrometry are demonstrated for monitoring the flux of metal ions out of and into the metalloprotein rabbit liver metallothionein and, in one example, chlorambucil-alkylated metallothionein. Metal ion transfers may be followed as the reactions proceed in situ to provide kinetic information. More uniquely to this technique, metal ion stoichiometries may be determined for reaction intermediates and products. Partners used in these studies include EDTA, carbonic anhydrase, a zinc-bound hexamer of insulin, and the core domain of bacteriophage T4 gene 32 protein, a binding protein for single-stranded DNA.

An RNA chaperone activity of non-specific RNA binding proteins in hammerhead ribozyme catalysis.

We have previously shown that a protein derived from the p7 nucleocapsid (NC) protein of HIV type-1 increases kcat/Km and kcat for cleavage of a cognate substrate by a hammerhead ribozyme. Here we show directly that the increase in kcat/Km arises from catalysis of the annealing of the RNA substrate to the ribozyme and the increase in kcat arises from catalysis of dissociation of the RNA products from the ribozyme. A peptide polymer derived from the consensus sequence of the C-terminal domain of the hnRNP A1 protein (A1 CTD) provides similar enhancements. Although these effects apparently arise from non-specific interactions, not all non-specific binding interactions led to these enhancements. NC and A1 CTD exert their effects by accelerating attainment of the thermodynamically most stable species throughout the ribozyme catalytic cycle. In addition, NC protein is shown to resolve a misfolded ribozyme-RNA complex that is otherwise long lived. These in vitro results suggest that non-specific RNA binding proteins such as NC and hnRNP proteins may have a biological role as RNA chaperones that prevent misfolding of RNAs and resolve RNAs that have misfolded, thereby ensuring that RNA is accessible for its biological functions.

Studies on primer binding of HIV-1 reverse transcriptase using a fluorescent probe.

The fluorescent nucleotide analog, 2',3'-trinitrophenyladenosine-5'-triphosphate (TNP-ATP), was utilized to quantify the affinities of human immunodeficiency virus-1 reverse transcriptase (HIV-1 RT) for its substrates. Interaction of this probe with the enzyme brings about a twofold increase in the magnitude of fluorescence emission from the probe, and a blue-shift in wavelength maximum, from 561 to 553 nm. TNP-ATP binds HIV-1 RT with a dissociation constant of 21 microM. The presence of millimolar levels of deoxynucleoside triphosphates or micromolar levels of an oligonucleotide primer analogue, p(dT)12-18, suppressed this enhancement of fluorescence. The fact that inhibition was achieved with much lower levels of primer than of dNTPs suggests that TNP-ATP is a probe for the binding site of primer on the enzyme, rather than that of deoxynucleoside triphosphate. In support of this, the effect of TNP-ATP on the kinetics of DNA synthesis catalyzed by the enzyme indicated that the probe is a competitive inhibitor with respect to template-primer. The ability of primers and primer analogs to reverse the fluorescence enhancement was determined, and the corresponding affinities of these compounds for reverse transcriptase were calculated. The affinity increased with primer length, increasing more than 50-fold from a span of 5 to 15 nucleotide residues. The interaction of polydeoxynucleotides was consistent with a model in which the enzyme bound at adjacent internal sites of about 15 residues in length. Several mammalian and bacterial transfer RNA primers were tested, including the natural primer, tRNA(3Lys). The affinities were found to be between 0.55 and 1.2 microM, with no obvious selectivity for the natural primer, which had a Kd of 0.79 microM. These results are discussed within the context of data for HIV-1 RT obtained by other methodologies.

Bacteriophage T4 gene 32 protein: modulation of protein-nucleic acid and protein-protein association by structural domains.

The cooperative binding of bacteriophage T4 gene 32 protein to single-stranded nucleic acids is dependent on homotypic protein-protein interactions between the N-terminus of a protein monomer with the core domain of an adjacent protein. In a previous report [Casas-Finet et al. (1992) Proc. Natl. Acad. Sci. U.S.A. 89, 1050-1054], we demonstrated that synthetic peptides corresponding to various portions of the N-terminal B-domain (residues 1-21) formed a 1:1 complex with core domain and identified a sequence, residues 3-5, Lys-Arg-Lys-Ser-Thr (the LAST motif) strongly homologous to a sequence within the central portion of protein (core domain) that was likely to function in nucleic acid binding. On the basis of these observations, we proposed a model where cooperative binding involves an exchange of intramolecular protein-protein interactions involving the internal LAST sequence for intermolecular protein-protein interactions utilizing the N-terminal LAST sequence. In this paper, we have tested various predictions of the model, and utilizing several proteases, further have defined the domain structure of 32 protein. The interaction of peptides containing LAST sequences with 32 protein qualitatively reduces its binding cooperativity, indicating that the peptides bind at the same site within the core domain as the N-terminus of an adjacent intact protein bound to the polynucleotide lattice. As expected, these peptides bind to nucleic acids. The N-terminus of 32 protein is predicted to be largely alpha-helical, and the circular dichroism spectrum of a peptide corresponding to residues 1-17 is consistent with this prediction. On the basis of the magnitude of protein tryptophan fluorescence quenching, the conformational change in 32 protein brought about by LAST peptides may be similar to that effected by oligonucleotides. As predicted by our model, in the presence of interacting peptide, the binding of 32 protein to oligonucleotide becomes salt-dependent. Arg-C endoproteolysis of intact 32 protein indicates that the loss of as few as three or four amino acids from the N-terminus appears to eliminate binding cooperativity, although the remainder of the N-terminal B-domain appears to protect the core from proteolysis. In contrast, this enzyme will catalyze the breakdown of trypsin-generated core domain, which lacks the first 21 residues of the protein. Thus, the presence of residues 4/5-21 attached to core alters its conformation and/or accessibility to protease. Poly(dT) inhibits this digestion, whereas the presence of N-terminal peptide accelerates proteolysis, in agreement with our model.(ABSTRACT TRUNCATED AT 400 WORDS)

Mammalian heterogeneous ribonucleoprotein A1 and its constituent domains. Nucleic acid interaction, structural stability and self-association.

With a view toward further understanding the structure-function relationships of the eukaryotic heterogeneous ribonucleoprotein (hnRNP) A1, and in particular its multiplicity of nucleic acid-interactive domains, we have studied the nucleic acid binding properties of the globular N-domain (UP1) and sequence-repetitive, flexible C-domain, the thermal denaturation of UP1 and the concomitant effects of binding polynucleotide, and the self-associative properties of the full-length protein. Utilizing protein tryptophan fluorescence as a probe, polynucleotide binding was shown to stabilize UP1 against thermal unfolding. The denaturation profile of UP1-poly(thymidylic acid) complexes was biphasic, suggesting that unfolding of the two subdomains of UP1 can occur independently. This is in agreement with a previously proposed structure in which only one of the two UP1 subdomains binds the nucleic acid. The subdomains of UP1 can be prepared by controlled proteolysis of A1, further indicating that these two globular segments within A1 are connected by an exposed, flexible linkage. Circular dichroism measurements on UP1 confirm previous data that this portion of A1 binds single-stranded nucleic acids non-co-operatively. UP1 clearly shows a preference for single-stranded nucleic acids with a 2'-OH, since its affinity for poly(U) is three times higher than for poly(dU), and five times higher than its affinity for poly(2'-OCH3U). The nucleic acid-interactive properties of the C-domain were further examined by preparing a synthetic peptide polymer (M(r) approximately 12,000) containing about seven repeats of a 16-residue sequence, GNFGGGRGGNYGGSRG, which in turn comprises two copies of the C-terminal consensus, GN(F/Y)GG(G/S)RG. The polymer of this sequence exhibited significant affinity for the fluorescent polyribonucleotide, poly(ethenoadenylic acid), binding stoichiometrically at < or = 0.2 M-Na+. Complex formation was accompanied by an increase in aggregate formation, as indicated by the appearance of scattering. For purposes of comparison, the data were analyzed via the linear co-operative model of McGhee and von Hippel, though this model may not be fully descriptive of the protein-nucleic acid complex(es) formed in this case. In contrast to the non-co-operative binding mode of the UP1 domain, the C-polymer exhibited moderate co-operativity, comparable to that seen with full-length A1. Although addition of sufficient NaCl reversed the interaction, a sigmoidal binding isotherm could still be observed (with sufficient added polymer) at 0.8 M-NaCl. This suggests that non-electrostatic interactions contribute significantly to the free energy of binding.(ABSTRACT TRUNCATED AT 400 WORDS)

Mammalian DNA polymerase beta: characterization of a 16-kDa transdomain fragment containing the nucleic acid-binding activities of the native enzyme.

The 39-kDa DNA polymerase beta (beta-Pol) molecule can be readily converted into two constituent domains by mild proteolysis; these domains are represented in an 8-kDa N-terminal fragment and a 31-kDa C-terminal fragment [Kumar et al. (1990a) J. Biol. Chem. 265, 2124-2131]. Intact beta-Pol is a sequence-nonspecific nucleic acid-interactive protein that binds both double-stranded (ds) and single-stranded (ss) polynucleotides. These two activities appear to be contributed by separate portions of the enzyme, since the 31-kDa domain binds ds DNA but not ss DNA, and conversely, the 8-kDa domain binds ss DNA but not ds DNA [Casas-Finet et al. (1991) J. Biol. Chem. 266, 19618-19625]. Truncation of the 31-kDa domain at the N-terminus with chymotrypsin, to produce a 27-kDa fragment (residues 140-334), eliminated all DNA-binding activity. This suggested that the ds DNA-binding capacity of the 31-kDa domain may be carried in the N-terminal segment of the 31-kDa domain. We used CNBr to prepare a 16-kDa fragment (residues 18-154) that spans the ss DNA-binding region of the 8-kDa domain along with the N-terminal portion of the 31-kDa domain. The purified 16-kDa fragment was found to have both ss and ds polynucleotide-binding capacity. Thermodynamic binding properties for these activities are similar to those of the intact enzyme.(ABSTRACT TRUNCATED AT 250 WORDS)

Selective photochemical modification by trichloroethanol of tryptophan residues in proteins with a high tyrosine-to-tryptophan ratio.

We present an improved procedure for the selective modification of tryptophan residues in proteins. A simple, low-cost set-up allows rapid tryptophan photoreaction upon ultraviolet irradiation in the presence of 2,2,2-trichloroethanol. This photochemical reaction is carried out under native conditions, occurs only in the excited state of tryptophan, and yields a single, as yet unidentified, photoproduct. Except for tyrosine, no reaction with other amino acid side chains are known. Stringent photoselection of tryptophan, ensuring that tyrosine residues are not affected, is achieved in situ without the need for an elaborate system of optical filters or lenses. Illumination with a medium-wave uv lamp of samples placed in disposable, dual pathlength, polystyrene fluorescence cuvettes allows treatment of small sample volumes (greater than or equal to 100 microliters) of various optical density. Chromophore accessibility in oligomeric assemblies or protein-nucleic acid complexes can be assessed by this reaction since the integrity of these structures is preserved. Moreover, this technique can be used to evaluate the involvement of tryptophan residues in catalytic or ligand binding processes.

Nucleic acid interactive properties of a peptide corresponding to the N-terminal zinc finger domain of HIV-1 nucleocapsid protein.

An 18-residue peptide (NC-F1) with an amino acid sequence corresponding to the N-terminal zinc finger of human immunodeficiency virus-1 nucleocapsid protein has been shown to bind to nucleic acids by fluorescence and NMR methods. Previously, this peptide has been shown to fold into a defined structure when bound to zinc (Summers et al., 1990). We have used a fluorescent polynucleotide, poly(ethenoadenylic acid), to monitor binding of this peptide to nucleic acids. In the presence of zinc, the peptide had a smaller site size (1.75 nucleotide residues/peptide) than in the absence of the metal ion (2.75). The salt sensitivity of the interaction indicated that two ion pairs are involved in the association of Zn2+ (NC-F1) with polynucleotide, whereas one ion pair is found in the metal-free peptide-nucleic acid complex. Competition experiments with single-stranded DNA (ss DNA) in either the presence or absence of Zn2+ showed that the peptide bound to ss DNA. Using NMR methods, we monitored the binding of a synthetic oligonucleotide, d(TTTGGTTT), to Zn(NC-F1). The hydrophobic residues F2 and I10, which are on the surface of the peptide and have been implicated in viral RNA recognition, were shown to interact with the oligomer. In accord with this observation, analysis of the salt dependence of the polynucleotide-peptide interaction indicates a nonelectrostatic component of about -6 kcal/mol, a value consistent with theoretical estimates of stacking energies of phenylalanine with nucleic acid bases.

Characterization of the copper- and silver-thiolate clusters in N-terminal fragments of the yeast ACE1 transcription factor capable of binding to its specific DNA recognition sequence.

N-terminal fragments of ACE1 protein spanning residues 1-122 or 1-110, termed ACE1(122*) and ACE1(110*), respectively, were investigated in regard to their metal- and double-stranded DNA-binding properties. Band mobility shift assays showed that binding to a specific oligonucleotide (termed UASc), containing two ACE1(122*) binding sites, requires the presence of Cu(I) or Ag(I) but does not occur in the presence of divalent metal ions. Both the Ag(I) and the Cu(I) forms of ACE1(122*) were characterized spectroscopically. The Tyr and metal cluster luminescence emission of Cu-ACE1(122*) was specifically quenched by the oligonucleotide UAScL, but not by an oligonucleotide of the same length and base composition but scrambled sequence. The room-temperature luminescence of Cu(I)-ACE1(122*) was assigned to a phosphorescence emission, on the basis of its long-lived luminescence of approximately 3.5 microseconds. We report the first observation of a Ag(I) metal cluster in solution for Ag(I)-ACE1(122*), which was found to exhibit a quantum yield and average luminescence lifetime that are ca. 6% of that of Cu(I)-ACE1(122*). The three-dimensional structure brought about by the binding of either metal ion appears to be very similar, since dynamic tyrosine fluorescence lifetime measurements, as well as circular dichroism spectra, were nearly identical for Cu- and Ag-ACE1(122*). Based on these results, we present a hypothetical model for the structure of the metal cluster in this class of proteins.

Structural basis for the nucleic acid binding cooperativity of bacteriophage T4 gene 32 protein: the (Lys/Arg)3(Ser/Thr)2 (LAST) motif.

To identify the functional residues of the N-terminal B region of bacteriophage T4 gene 32 protein involved in its cooperative binding to single-stranded nucleic acids, a process dependent on homotypic protein-protein interaction, we have studied the interaction of the protein with synthetic peptides containing different portions of this domain. Gel-permeation chromatography showed that a 6-acryloyl-2-dimethylaminonaphthalene (acrylodan)-labeled fluorescent peptide corresponding to the first 17 residues of gene 32 protein formed a complex with whole protein. The fluorescence was blue-shifted 14 nm upon interaction with intact protein, and somewhat less so (7-11 nm) with cleavage products of the protein lacking B domains. The intrinsic tryptophan fluorescence of whole and truncated protein was quenched by this peptide and by the nonderivatized peptide. The peptide bound tightly to truncated protein at both 0.015 and 0.44 M Na+, with a stoichiometry of 1:1. Similar tryptophan quenching or acrylodan blue shifts were obtained with peptides corresponding to residues 1-9 and 3-8, but not residues 1-4, 5-9, or 5-17, indicating that the essential amino acids are contained within positions 3-8, Lys-Arg-Lys-Ser-Thr-Ala. Several other DNA binding proteins contain a LAST motif with documented involvement of these residues in nucleic acid interaction. The amino acid and coding sequence of residues 110-114, a region proposed to be involved in nucleic acid binding, is virtually identical to that of residues 3-7. Based on these observations, we have formulated a model for the cooperative interactions of gene 32 protein with single-stranded nucleic acids.

Spectroscopic studies of the structural domains of mammalian DNA beta-polymerase.

The 8- and 31-kDa fragments of beta-polymerase, prepared by controlled proteolysis as described (Kumar, A., Widen, S. G., Williams, K. R., Kedar, P., Karpel, R. L., and Wilson, S. H. (1990) J. Biol. Chem. 265, 2124-2131), constitute domains that are structurally and functionally dissimilar. There is little disruption of secondary structure upon proteolysis of the intact enzyme, as suggested from CD spectra of the fragments. beta-Polymerase is capable of binding both single- and double-stranded nucleic acids: the 8-kDa fragment binds specifically to single-stranded lattices, whereas the 31-kDa domain displays affinity exclusively for double-stranded polynucleotides. These domains are connected by a highly flexible protease-hypersensitive segment that may allow the coordinate functioning of the two binding activities in the intact protein. beta-Polymerase binds to poly(ethenoadenylic acid) with higher affinity, similar cooperativity, but lesser salt dependence than the 8-kDa fragment. Under physiological conditions, the intact enzyme displays greater binding free energy for single-stranded polynucleotides than the 8-kDa fragment, suggesting that the latter may carry a truncated binding site. Binding of double-stranded calf thymus DNA brings about a moderate quenching of the Tyr and Trp fluorescence emission of both the 31-kDa fragment and beta-polymerase and induces a 6-nm blue shift in the Trp emission maximum of the intact enzyme, but not in the fragment. This latter result is likely due to a change in the relative orientation of the 8- and 31-kDa domains in the intact protein upon interaction with double-stranded DNA; alternatively, the binding mode of intact protein may differ from that of the fragment. Simultaneous interaction of both domains with polynucleotides most likely does not occur since double-stranded DNA binding to the 31-kDa domain of intact beta-polymerase induces the displacement of single-stranded polynucleotides from the 8-kDa domain. These results are evaluated in light of the role of beta-polymerase in DNA repair.

Physical studies of tyrosine and tryptophan residues in mammalian A1 heterogeneous nuclear ribonucleoprotein. Support for a segmented structure.

The mammalian heterogeneous ribonucleoprotein (hnRNP) A1 and its constituent N-terminal domain, termed UP1, have been studied by steady-state and dynamic fluorimetry, as well as phosphorescence and optically detected magnetic resonance (ODMR) spectroscopy at cryogenic temperatures. The results of these diverse techniques coincide in assigning the site of the single tryptophan residue of A1, located in the UP1 domain, to a partially solvent-exposed site distal to the protein's nucleic acid binding surface. In contrast, tyrosine fluorescence is significantly perturbed when either protein associates with single-stranded polynucleotides. Tyr to Trp energy transfer at the singlet level is found for both UP1 and A1 proteins. Single-stranded polynucleotide binding induces a quenching of their intrinsic fluorescence emission, which can be attributed to a significant reduction (greater than 50%) of the Tyr contribution, while Trp emission is only quenched by approximately 15%. Tyrosine quenching effects of similar magnitude are seen upon polynucleotide binding by either UP1 (1 Trp, 4 Tyr) or A1 (1 Trp, 12 Tyr), strongly suggesting that Tyr residues in both the N-terminal and C-terminal domain of A1 are involved in the binding process. Tyr phosphorescence emission was strongly quenched in the complexes of UP1 with various polynucleotides, and was attributed to triplet state energy transfer to nucleic acid bases located in the close vicinity of the fluorophore. These results are consistent with stacking of the tyrosine residues with the nucleic acid bases. While the UP1 Tyr phosphorescence lifetime is drastically shortened in the polynucleotide complex, no change of phosphorescence emission maximum, phosphorescence decay lifetime or ODMR transition frequencies were observed for the single Trp residue. The results of dynamic anisotropy measurements of the Trp fluorescence have been interpreted as indicative of significant internal flexibility in both UP1 and A1, suggesting a flexible linkage connecting the two sub-domains in UP1. Theoretical calculations based on amino acid sequence for chain flexibility and other secondary structural parameters are consistent with this observation, and suggest that flexible linkages between sub-domains may exist in other RNA binding proteins. While the dynamic anisotropy data are consistent with simultaneous binding of both the C-terminal and the N-terminal domains to the nucleic acid lattice, no evidence for simultaneous binding of both UP1 sub-domains was found.

Spectroscopic characterization of the copper(I)-thiolate cluster in the DNA-binding domain of yeast ACE1 transcription factor.

A polypeptide containing the amino-terminal region of ACE1 (residues 1-122; 122*), the activator of yeast Cu-metallothionein gene transcription, shows charge-transfer and metal-centered UV absorption bands, and orange luminescence which are characteristic of Cu-cysteinyl thiolate cluster structures. These spectral features are abolished by the Cu(I) complexing agents CN- and diethyldithiocarbamate or exposure to acid, but not by the Cu(II) chelator, EDTA. Binding of the polypeptide to its specific DNA recognition site, but not to calf-thymus double-stranded DNA, induces quenching of its Tyr and Cu-S cluster luminescence emission. The CD spectrum is characteristic of a tightly folded structure that may be organized around the Cu cluster.

Mammalian heterogeneous nuclear ribonucleoprotein A1. Nucleic acid binding properties of the COOH-terminal domain.

A1 is a core protein of the eukaryotic heterogeneous nuclear ribonucleoprotein complex and is under study here as a prototype single-stranded nucleic acid-binding protein. A1 is a two-domain protein, NH2-terminal and COOH-terminal, with highly conserved primary structure among vertebrate homologues sequenced to date. It is well documented that the NH2-terminal domain has single-stranded DNA and RNA binding activity. We prepared a proteolytic fragment of rat A1 representing the COOH-terminal one-third of the intact protein, the region previously termed COOH-terminal domain. This purified fragment of 133 amino acids binds to DNA and also binds tightly to the fluorescent reporter poly(ethenoadenylate), which is used to access binding parameters. In solution with 0.41 M NaCl, the equilibrium constant is similar to that observed with A1 itself, and binding is cooperative. The purified COOH-terminal fragment can be photochemically cross-linked to bound nucleic acid, confirming that COOH-terminal fragment residues are in close contact with the polynucleotide lattice. These binding results with isolated COOH-terminal fragment indicate that the COOH-terminal domain in intact A1 can contribute directly to binding properties. Contact between both COOH-terminal domain and NH2-terminal domain residues in an intact A1:poly(8-azidoadenylate) complex was confirmed by photochemical cross-linking.

A single-stranded nucleic acid binding sequence common to the heterogeneous nuclear ribonucleoparticle protein A1 and murine recombinant virus gp70.

We have used an antisynthetic peptide antiserum to a murine recombinant virus gp70 to probe normal mouse tissues for immunologically related proteins. In addition to cognate gp70s, this antiserum reacts with the heterogeneous nuclear ribonucleoparticle protein A1 by virtue of a 5-amino acid epitope, PRNQG. Further structural similarity is evident both 5' and 3' of this epitope. Since the function of the heterogeneous nuclear ribonucleoprotein particles in the cell is to aid in the stabilization and processing of newly synthesized RNA, we have investigated whether this retroviral sequence exhibits any nucleic acid-binding properties by the same criteria established for the identification of heterogeneous nuclear ribonucleoprotein particles. Analysis of the peptide in a poly(eA) binding assay shows this retroviral sequence to bind with high affinity to single-stranded nucleic acid. This binding occurs in a salt-sensitive manner characteristic of single-stranded nucleic acid-binding proteins. Flanking peptides not containing this sequence generated from either the A1 or gp70 show no ability to bind single-stranded nucleic acids by this assay.

Studies of the domain structure of mammalian DNA polymerase beta. Identification of a discrete template binding domain.

Characterization of the domain structure of DNA polymerase beta is reported. Large scale overproduction of the rat protein in Escherichia coli was achieved, and the purified recombinant protein was verified by sequencing tryptic peptides. This protein is both a single-stranded DNA binding protein and a DNA polymerase consisting of one polypeptide chain of 334 amino acids. As revealed by controlled proteolysis experiments, the protein is organized in two relatively protease-resistant segments linked by a short protease-sensitive region. One of these protease-resistant segments represents the NH2-terminal 20% of the protein. This NH2-terminal domain (of about 75 residues) has strong affinity for single-stranded nucleic acids. The other protease-resistant segment, representing the COOH-terminal domain of approximately 250 residues, does not bind to nucleic acids. Neither domain, tested as purified proteins, has substantial DNA polymerase activity. The results suggest that the NH2-terminal domain is principally responsible for the template binding activity of the intact protein.

Mammalian heterogeneous nuclear ribonucleoprotein complex protein A1. Large-scale overproduction in Escherichia coli and cooperative binding to single-stranded nucleic acids.

Characterization of mammalian heterogeneous nuclear ribonucleoprotein complex protein A1 is reported after large-scale overproduction of the protein in Escherichia coli and purification to homogeneity. A1 is a single-stranded nucleic acid binding protein of 320 amino acids and 34,214 Da. The protein has two domains. The NH2-terminal domain is globular, whereas the COOH-terminal domain of about 120 amino acids has low probability of alpha-helix structure and is glycinerich. Nucleic acid binding properties of recombinant A1 were compared with those of recombinant and natural proteins corresponding to the NH2-terminal domain. A1 bound to single-stranded DNA-cellulose with higher affinity than the NH2-terminal domain peptides. Protein-induced fluorescence enhancement was used to measure equilibrium binding properties of the proteins. A1 binding to poly (ethenoadenylate) was cooperative with the intrinsic association constant of 1.5 X 10(5) M-1 at 0.4 M NaCl and a cooperativity parameter of 30. The NH2-terminal domain peptides bound noncooperatively and with a much lower association constant. With these peptides and with intact A1, binding was fully reversed by increasing [NaCl]; yet. A1 binding was much less salt-sensitive than binding by the NH2-terminal domain peptides. A synthetic polypeptide analog of the COOH-terminal domain was prepared and was found to bind tightly to poly-(ethenoadenylate). The results are consistent with the idea that the COOH-terminal domain contributes to A1 binding through both cooperative protein-protein interaction and direct interaction with the nucleic acid.

Photoaffinity labeling of T4 bacteriophage 32 protein.

With a view toward the determination of nucleic acid binding domains and sites on nucleic acid helix-destabilizing (single strand-specific) proteins (HDPs), we have studied the interactions of the copolymer polynucleotide photoaffinity label, poly(adenylic, 8-azidoadenylic acid), (poly(A,8-N3A] with the T4 bacteriophage HDP, 32 protein. Poly(A,8-N3A) quenched the intrinsic tryptophan fluorescence of 32 protein in a manner similar to that observed with other polynucleotides, and the effect could be reversed by addition of sufficient NaCl. The binding affinity and site size of this noncovalent interaction of poly(A,8-N3A) with 32 protein are similar to the values obtained for poly(A) and this protein. When [3H]poly(A,8-N3A)/32 protein mixtures were irradiated at 254 nm, fluorescence quenching was not reversed by NaCl, suggesting that the label was covalently bound to the protein. Mixtures of photolabel and protein subjected to short periods of irradiation (generally 1 min, 2000 erg mm-2) formed high molecular weight complexes, which when electrophoresed on sodium dodecyl sulfate (SDS)-polyacrylamide gels were radioactive and stained with Coomassie Blue R. Under the same conditions, [3H]poly(A) failed to label 32 protein. The radioactivity of [3H]poly(A,8-N3A)-labeled complexes subjected to micrococcal nuclease after irradiation was seen to migrate just behind the free 32 protein monomer on SDS-polyacrylamide gels, indicating that portions of the photolabel not in direct contact with protein were accessible to this enzyme. By several criteria, we conclude that 32 protein was photolabeled specifically at its single-stranded nucleic acid binding site. Single-stranded nucleic acids with affinities for protein greater than that of poly(A,8-N3A) effectively inhibited photolabeling. The [NaCl] dependence of photolabeling monitored on SDS gels paralleled the NaCl reversal of (noncovalent) poly(A,8-N3A)-32 protein binding. Photolabeling reached a plateau after 1-2 min. The formation of high molecular weight complexes with increasing [poly(A,8-N3A)] paralleled the disappearance of free protein on SDS gels, and reached a saturation level of about 75% labeling. Several chromatographic procedures appear to be useful for the separation of the photolabeled complexes from free protein and photolabel. Limited trypsin hydrolysis of photolabeled 32 protein indicated that all the label was within the central ("III") portion of the protein. This approach should have general applicability to the identification of nucleic acid binding sites on helix-destabilizing proteins.