One- and two-dimensional 1H NMR, fluorescence, and molecular modeling studies on the tomaymycin-d(ATGCAT)2 adduct. Evidence for two covalent adducts with opposite orientations and stereochemistries at the covalent linkage site.

Tomaymycin is a member of the pyrrolo[1,4]benzodiazepine antitumor-antibiotic group that binds covalently to the exocyclic 2-amino group of guanine in DNA. Previous correlation of fluorescence and NMR data suggested that the 11R,11aS and the 11S,11aS diastereomers of tomaymycin could bind to DNA in two orientations relative to the covalently modified guanine (Barkley, M. D.; Cheatham, S.; Thurston, D. E.; Hurley, L. H. Biochemistry 1986, 25, 3021-3031). We now report on fluorescence, one- and two-dimensional proton NMR, and molecular modeling studies of the tomaymycin-d(ATGCAT)2 adduct, which corroborate these earlier observations. Fluorescence measurements show that there are two species of tomaymycin bound to d(ATGCAT)2, which are tentatively identified as the 11R,11aS and 11S,11aS diastereomers. Two distinct sets of signals for the tomaymycin molecule are present in the proton NMR spectrum of the tomaymycin-d(ATGCAT)2 duplex adduct. Two-dimensional correlation spectroscopy (2D-COSY) studies also show connectivities for four cytosine H5-H6 and eight thymine methyl-H6 protons and thus clearly establish the presence of two distinct species of tomaymycin-d(ATGCAT)2 adducts in solution. A single scalar 11-11a 1H NMR coupling in the 2D-COSY spectrum is indicative of an adduct species that has an S configuration at the C-11 position. Two-dimensional nuclear Overhauser effect (NOESY) spectra of the tomaymycin-d(ATGCAT)2 duplex adduct show that the adducts are relatively nondistortive. In a NOESY experiment, cross-peaks were identified between both the aromatic H9 proton and the ethylidine methyl protons of tomaymycin and two different adenine H2 protons of d(ATGCAT)2. Molecular mechanics calculations with AMBER show that the two species with the thermodynamically most favorable binding energies are the 11R,11aS and 11S,11aS isomers with their aromatic rings to the 5' and 3' sides of the covalently bound guanine, respectively. The NOEs observed between tomaymycin protons and adenine H2 protons are in accord with molecular modeling studies. Taken together, these results strongly suggest that the two forms of tomaymycin bound to d(ATGCAT)2 are the 11S,11aS and 11R,11aS species, oriented with their aromatic rings to the 3' and 5' sides, respectively, of the covalently modified guanines.

Salt dependence of the kinetics of the lac repressor-operator interaction: role of nonoperator deoxyribonucleic acid in the association reaction.

The kinetics of binding of lac repressor protein and operator deoxyribonucleic acid (DNA) have been studied as a function of monovalent and divalent cation concentration. The salt dependence of the association and dissociation rate constants has been interpreted in light of recent theoretical analyses based on Manning's counterion condensation model. The bell-shaped dependence of the association rate constant on salt concentration evidences a role for nonoperator DNA binding in the repressor's search for the operator site on a large DNA molecule. At intermediate mono- or divalent cation concentrations, the association rate goes through a maximum. At lower cation concentrations, it decreases and becomes dependent on DNA concentration; the high affinity of repressor for nonoperator DNA confines the protein to the DNA. At higher cation concentrations, the association rate decreases and becomes dependent on the weak affinity of repressor for nonoperator DNA. The kinetic data are fit to the theory of Berg and Blomberg [Berg, O. G., & Blomberg, C. (1978) Biophys. Chem. 8, 271] for the salt dependence of association kinetics with coupled diffusion, using published values of the affinity for nonoperator DNA. From this fit, one-dimensional diffusion of repressor along the DNA chain is estimated to be about 4 times faster on MgDNA than on NaDNA. At higher cation concentrations, the salt dependence of the association and dissociation rate constants is consistent with a preequilibrium mechanism for the association reaction [Lohman, T. M., deHaseth, P. L., & Record, M. T., Jr. (1978) Biophys. Chem. 8, 281]. Agreement between literature values (corrected for the presence of Mg2+) and experimental values of the rate constants in the presence of monovalent salt is quite good.

Toward understanding tryptophan fluorescence in proteins.

A general approach to dissecting the complex photophysics of tryptophan is presented and used to elucidate the effects of amino acid functional groups on tryptophan fluorescence. We have definitively identified the amino acid side chains that quench tryptophan fluorescence and delineated the respective quenching mechanisms in a simple model system. Steady-state and time-resolved fluorescence techniques, photochemical H-D exchange experiments, and transient absorption techniques were used to measure individual contributions to the total nonradiative rate for deactivation of the excited state, including intersystem crossing, solvent quenching, and excited-state proton and electron transfer rates. Eight amino acid side chains representing six functional groups quench 3-methylindole fluorescence with a 100-fold range in quenching rate constant. Lysine and tyrosine side chains quench by excited-state proton transfer; glutamine, asparagine, glutamic and aspartic acid, cysteine, and histidine side chains quench by excited-state electron transfer. These studies provide a framework for deriving detailed structural and dynamical information from tryptophan fluorescence intensity and lifetime data in peptides and proteins.

Self-assembly of designed antimicrobial peptides in solution and micelles.

Hydrophobic interactions are responsible for stabilizing leucine zippers in peptides containing heptad repeats. The effects of substituting leucine by phenylalanine and alanine by glycine on the self-assembly of coiled-coils were examined in minimalist antimicrobial peptides designed to form amphipathic alpha-helices. The secondary structure of these peptides was monitored in solution and in diphosphocholine (DPC) micelles using circular dichroism spectroscopy. The leucine peptides (KLAKLAK)3 and (KLAKKLA)n (n = 3, 4) become alpha-helical with increasing concentrations of salt, peptide, and DPC. The aggregation state and equilibrium constant for self-association of the peptides were measured by sedimentation equilibrium. The glycine peptide (KLGKKLG)3 does not self-associate. The leucine peptides and phenylalanine peptides (KFAKFAK)3 and (KFAKKFA)n (n = 3, 4) are in a monomer-tetramer equilibrium in solution, with the phenylalanine zippers being 2-4 kcal/mol less stable than the equivalent leucine zippers. Thermodynamic parameters for the association reaction were calculated from the temperature dependence of the association constants. Leucine zipper formation has DeltaCp = 0, whereas phenylalanine zipper formation has a small negative DeltaCp, presumably due to the removal of the larger surface area of phenylalanine from water. Self-association of the peptides is coupled to formation of a hydrophobic core as detected using 1-anilino-naphthalene-8-sulfonate fluorescence. Carboxyfluorescein-labeled peptides were used to determine the aggregation state of (KLAKKLA)3 and (KLGKKLG)3 in DPC micelles. (KLAKKLA)3 forms dimers, and (KLGKKLG)3 is a monomer. Aggregation appears to correlate with the cytotoxicity of these peptides.

Comparison of approaches to the instrumental response function in fluorescence decay measurements.

Deconvolution of pulse fluorometry data requires knowledge of the instrumental response, which is not directly observable in some circumstances. Various procedures for approaching the instrumental response function were evaluated for nanosecond fluorescence decay data analyzed by nonlinear least squares, including the commonly used time shift correction and several reference fluorophore methods. A new reference fluorophore technique using a Monte Carlo convolution is introduced and tested. The correction for scattered light in several reference techniques is also presented. The random convolution and one other reference fluorophore method consistently gave superior results over a wide range of experimental conditions.

Interaction of terminal transferase with single-stranded DNA.

A 58-kDa monomer of terminal transferase was isolated from calf thymus using a monoclonal antibody affinity column. The enzymatic activity was comparable to that of the 44-kDa alpha beta dimer isolated by conventional methods. Binding of the two enzyme forms to single-stranded DNA was monitored by fluorescence. The site size of both forms was approximately 11 +/- 2 nucleotides. Binding of the 44-kDa alpha beta dimer to polydeoxyadenosine was examined under several conditions. The cooperativity parameter increased from about 90 in the presence of Mg2+ to 300-400 in the absence of Mg2+. The observed dissociation constant of 3-5 microM was essentially independent of salt concentration, whereas the intrinsic dissociation constant decreased about 5-fold in the presence of Mg2+. The binding parameters of the 58-kDa monomer were independent of buffer composition and were similar to those of the 44-kDa alpha beta dimer in the presence of Mg2+. These results indicate that the additional 14-kDa peptide sequences present in the high molecular mass monomer form are not part of the DNA-binding site of terminal transferase.

Ion effects on the lac repressor--operator equilibrium.

The effects of ions on the interaction of lac repressor protein and operator DNA have been studied by the membrane filter technique. The equilibrium association constant was determined as a function of monovalent and divalent cation concentrations, anions, and pH. The binding of repressor and operator is extremely sensitive to the ionic environment. The dependence of the observed equilibrium constant on salt concentration is analyzed according to the binding theory of Record et al. [Record, M. T., Jr., Lohman, T. M., & deHaseth, P. L. (1976) J. Mol. Biol. 107, 145]. The number of ionic interactions in repressor--operator complex is deduced from the slopes of the linear log-log plots. About 11 ionic interactions are formed between repressor and DNA phosphates at pH 7.4 and about 9 ionic interactions at pH 8.0, in reasonable agreement with previous estimates. A favorable nonelectrostatic binding free energy of about 9-12 kcal/mol is estimated from the extrapolated equilibrium constants at the 1 M standard state. The values are in good accord with recent results for the salt-independent binding of repressor core and operator DNA. The effects of pH on the repressor--operator interaction are small, and probably result from titration of functional groups in the DNA-binding site of the protein. For monovalent salts, the equilibrium constant is slightly dependent on cation type and highly dependent on anion type. At constant salt concentration, the equilibrium constant decreases about 10000-fold in the order CH3CO2- greater than or equal to F- greater than Cl- greater than Br- greater than NO3- greater than SCN- greater than I-. The wide range of accessible equilibrium constants provides a useful tool for in vitro studies of the repressor--operator interaction.

Tryptophan fluorescence of terminal deoxynucleotidyl transferase: effects of quenchers on time-resolved emission spectra.

Terminal deoxynucleotidyl transferase (EC 2.7.7.31) is a eucaryotic DNA polymerase that does not require a template. The tryptophan environments in calf thymus terminal transferase were investigated by fluorescence. The heterogeneous emission from this multitryptophan enzyme was separated by time-resolved emission spectroscopy. Nanosecond fluorescence decays at 296-nm excitation and various emission wavelengths were deconvolved by global analysis, assuming that the lifetimes but not the relative weighting factors were independent of emission wavelength. The data were fit to three exponentials of lifetimes tau 1 = 1.4 ns, tau 2 = 4.5 ns, and tau 3 = 7.7 ns. The corresponding decay-associated emission spectra of the three components had maxima at about 328, 335, and 345 nm. The accessibility of individual tryptophan environments to polar and nonpolar fluorescence quenchers was examined in steady-state and time-resolved experiments. In the presence of iodide and acrylamide, the steady-state emission spectra shift to the blue. However, at low quencher concentrations, the emission from the 7.7-ns component (maximum 345 nm) is hardly affected, suggesting that this hydrophilic tryptophan environment is buried within the protein. On the other hand, the red shift in the steady-state emission spectrum in the presence of trichloroethanol indicates that the 1.4-ns component (maximum 328 nm) is an exposed hydrophobic tryptophan environment. The results are consistent with an inside-out model for terminal transferase protein, with the more hydrophobic tryptophan(s) near the surface and the most hydrophilic tryptophan(s) in the core.

Fluorimetric assay for terminal deoxynucleotidyl transferase activity.

A fluorimetric assay for measuring terminal deoxynucleotidyl transferase activity in purified and crude enzyme preparations has been developed. Etheno-substituted deoxynucleotides are shown to be substrates of the enzyme. The assay involves polymerization of the fluorescent nucleotide 1,N6-ethenodeoxyadenosine triphosphate (epsilon dATP) on an oligodeoxynucleotide initiator, [poly(deoxyadenylic acid) with an average chain length of 50 residues] under the reaction conditions used in the standard radiometric assay. The incorporation of epsilon dATP into polymer is quantitated by fluorescence after isolation and nuclease digestion of the product. The enzymological properties of etheno substrates were also determined. Epsilon dATP binds about twofold tighter than dATP to terminal transferase, but has a twofold-lower catalytic rate. However etheno substitution does not affect initiator binding. The fluorimetric assay is suitable for clinical analysis of terminal transferase in human leukemias, and may be a useful adjunct to recently developed immunochemical methods which detect protein, not activity.

Intrinsic tryptophan fluorescence measurements suggest that polylactosaminyl glycosylation affects the protein conformation of the gelatin-binding domain from human placental fibronectin.

Glycosylation can affect the physical and biochemical properties of the polypeptide chain in glycoproteins. Asparagine-N-linked polylactosaminyl glycosylation of the chymotryptic 44-kDa gelatin-binding domain from human placental fibronectin confers protease resistance [Zhu, B. C. R., Fisher, S. F., Panda, H., Calaycay, J., Shively, J. E. & Laine, R. A. (1984) J. Biol. Chem. 259, 3962-3970] and weaken the binding to gelatin [Zhu, B. C. R. & Laine, R. A. (1985) J. Biol. Chem. 260, 4041-4045]. Intrinsic tryptophan fluorescence of the gelatin-binding domain was used to probe glycosylation-dependent protein conformation changes. In gelatin-binding fragments containing incrementally smaller polylactosamine oligosaccharides, the fluorescence intensity progressively decreased and the emission spectrum shifted about 7 nm to the blue. Removal of the polylactosamine chains from a highly glycosylated fragment with endo-beta-galactosidase from Escherichia freundii also quenched the protein fluorescence. The fluorescence lifetimes did not appear to be affected by the extent of glycosylation, suggesting static quenching of the tryptophan emission in the low glycosylated fragments. Acrylamide quenching studies showed that the accessibility of the tryptophans to small solutes was not altered by glycosylation. The steady-state emission anisotropy increased with decreasing polylactosamine chain length. The results indicate that the polylactosamine chains alter the tryptophan environments in the gelatin-binding domain, probably by changing the polypeptide conformation. These putative protein conformation changes may be partially responsible for the altered gelatin binding, protease resistance, and cell adhesion functions of fetal tissue fibronectin.

Characterization of a unique tomaymycin-d(CICGAATTCICG)2 adduct containing two drug molecules per duplex by NMR, fluorescence, and molecular modeling studies.

Tomaymycin is a member of the pyrrolo[1,4]benzodiazepine [P(1,4)B] antitumor antibiotic group. This antibiotic is proposed to react with the exocyclic 2-amino group (N2) of guanine to form a covalent adduct that lies snugly within the minor groove of DNA. While DNA-footprinting experiments using methidiumpropyl-EDTA have revealed the favored bonding sequences for tomaymycin and related drugs on DNA, the stereochemistry at the covalent bonding site (C-11) and orientation in the minor groove were not established by these experiments. In previous studies using a combined fluorescence, high-field NMR, and molecular modeling approach, we have shown that for tomaymycin there are two diastereomeric species (11R and 11S) on both calf thymus DNA and d(ATGCAT)2. Although we were able to infer the identity (stereochemistry at C-11 and orientation in the minor groove) of the two species on d(ATGCAT)2 by high-field NMR and fluorescence studies, in combination with molecular mechanics calculations, definitive experimental evidence was lacking. We have designed and synthesized a self-complementary 12-mer [d(CICGAATTCICG)2] based on the Dickerson dodecamer [d(CGCGAATTCGCG)2] that bonds identically two tomaymycin molecules, each having a defined orientation and stereochemistry. Thus the bis(tomaymycin)-12-mer adduct maintains the self-complementarity of the original duplex molecule. Two-dimensional proton J-correlated spectroscopy (COSY) of the bis(tomaymycin)-d(CICGAATTCICG)2 adduct (I = inosine) unequivocally shows that C-11 of tomaymycin covalently bonds through N2 of guanine with an 11S stereochemistry in the sequence 5'-CGA-3'.(ABSTRACT TRUNCATED AT 250 WORDS)

Pyrrolo[1,4]benzodiazepine antitumor antibiotics: evidence for two forms of tomaymycin bound to DNA.

Tomaymycin is an antibiotic belonging to the pyrrolo[1,4]benzodiazepine group of antitumor compounds. Previous studies have shown that tomaymycin and other members of this group, which include anthramycin, sibiromycin, and the neothramycins, bind covalently through N-2 of guanine and lie within the minor groove of DNA. Two fluorescent ground-state species of tomaymycin were observed in protic solvents and on DNA. 1H NMR studies showed that the two fluorescent species in methanol are the 11R,11aS and 11S,11aS diastereomeric 11-methyl ethers of tomaymycin. On the basis of epimerization experiments and exchange of carbon-13 from 13CH3OH into the C-11 methoxy group of the tomaymycin methyl ether, a mechanism is proposed for their interconversion via 10,11-anhydrotomaymycin. Coupling information revealed that the solution conformations of the two diastereomers differ, with the C-5 carbonyl lying closer to the plane of the aromatic ring in the 11R,11aS diastereomer. The fluorescence excitation and emission spectra of the two emitting species in methanol were separated by time-resolved fluorescence spectroscopy and were associated with the diastereomeric forms identified by 1H NMR. Time-resolved fluorescence studies of tomaymycin in protic solvents and on DNA indicated that the absorption spectrum of the longer lifetime component (11R,11aS form) is red-shifted relative to the absorption spectrum of the shorter lifetime component (11S,11aS form), consistent with more extensive conjugation. The two conformational forms of tomaymycin on DNA were tentatively identified as the 11S,11aS and 11R,11aS diastereomeric adducts, which bind in opposite orientations in the minor groove. This proposal is supported by molecular modeling studies using a 6-mer duplex adduct of d(ATGCAT)2.

Steady-state fluorescence and molecular-modeling studies of tomaymycin-DNA adducts.

The interaction of tomaymycin and 8-O-methyltomaymycin with calf thymus DNA was studied by steady-state fluorescence techniques. The 8-phenolic proton of tomaymycin has a pK = 8.0, and the phenolate anion is essentially nonfluorescent. However, the fluorescence of the DNA adduct does not decrease until pH greater than 10.5, when the DNA double helix denatures. Acrylamide quenches the fluorescence of the free antibiotic with a quenching rate constant kq = 7 x 10(9) M-1 s-1. In DNA adducts, the quenching rate constant is reduced about 50-fold, indicating that the aromatic ring of the drug is shielded from the solvent. The four possible binding modes of the antibiotics were modeled on a 6-mer duplex by molecular mechanics calculations in the absence and presence of water and counterions. The modeling studies show that the antibiotic is buried in the minor groove in all binding modes, with the 8-substituent pointing away from the DNA core. Three or five waters are displaced from the minor groove, depending on the orientation of the drug on the DNA.

Molecular mechanism of sequence-specific termination of lentiviral replication.

The central termination sequence (CTS) terminates (+) strand DNA synthesis in certain lentiviruses. The molecular mechanism underlying this event, catalyzed by equine infectious anemia virus reverse transcriptase (EIAV RT), was evaluated by pre-steady-state kinetic techniques. Time courses in nucleotide incorporation using several DNA substrates were biphasic, consistent with release of enzyme from extended DNA being the rate-limiting step for turnover. While the burst amplitude reflecting the amount of functional RT-DNA complex was sequence-dependent, rate constants for initial product formation were not. Filter binding assays indicate the K(d) for CTS-containing substrate is only 2-fold higher than a random DNA and cannot account entirely for the large diminution in burst amplitudes. Measurements of processive DNA replication on a millisecond time scale indicate that the rate of polymerization is unaffected by the T(6)-tract within the CTS. However, termination products accumulate due to a substantial increase in the rate of nonproductive enzyme-nucleic acid complex formation after incorporation of four to five adenosines of a T(6)-tract within the CTS. During strand displacement synthesis through the CTS, products accumulate after incorporation of three to four adenosines. The rate of polymerization during strand displacement synthesis decreases 2-fold while the rate of nonproductive enzyme-nucleic acid complex formation is identical in the absence or presence of the displacement strand. These results have allowed us to develop a model for CTS-induced termination of (+) strand synthesis.

Time-resolved fluorescence studies of tomaymycin bonding to DNA.

Tomaymycin is an antibiotic that reacts at guanine N2 in the minor groove of the DNA helix. The number and type of tomaymycin-DNA adducts present on natural sequence DNA were identified using time-resolved fluorescence spectroscopy. At low bonding density, only two discrete species were observed with lifetimes of 4.3 and 7.1 ns and relative amplitudes of 40% and 60%. These two lifetime species are proposed to represent either R5' or S5' and S3' binding modes at the preferred bonding sequence 5'-AGA. R and S denote the configuration at C11 of tomaymycin, and 5' and 3' describe the orientation of the aromatic ring on the covalently modified strand. These two species were present over a range of solution conditions, including pH, nucleotide to drug ratio, DNA concentration, and DNA size. They have the same emission spectra, but slightly shifted absorption spectra. The weak temperature dependence of the fluorescence lifetimes presumably is due to the excited-state proton-transfer reaction that quenches tomaymycin fluorescence. The rate of formation of the longer lifetime species of DNA adduct is about twice as fast as that of the shorter lifetime species. Under saturating conditions, the fluorescence decay shows a bimodal lifetime distribution whether analyzed by least-squares assuming a Gaussian distribution model or by the maximum entropy method. The two groups of lifetimes are centered around 2-3 and 6-6.6 ns, reflecting multiple species on different bonding sequences.

Time-resolved fluorescence spectroscopy of human adenosine deaminase: effects of enzyme inhibitors on protein conformation.

Adenosine deaminase, a purine salvage enzyme essential for immune competence, was studied by time-resolved fluorescence spectroscopy. The heterogeneous emission from this four-tryptophan protein was separated into three lifetime components: tau 1 = 1 ns and tau 2 = 2.2 ns an emission maximum at about 330 nm and tau 3 = 6.3 ns with emission maximum at about 340 nm. Solvent accessibility of the tryptophan emission was probed with polar and nonpolar fluorescence quenchers. Acrylamide, iodide, and trichloroethanol quenched emission from all three components. Acrylamide quenching caused a blue shift in the decay-associated spectrum of component 3. The ground-state analogue enzyme inhibitor purine riboside quenched emission associated with component 2 whereas the transition-state analogue inhibitor deoxycoformycin quenched emission from both components 2 and 3. The quenching due to inhibitor binding had no effect on the lifetimes or emission maxima of the decay-associated spectra. These observations can be explained by a simple model of four tryptophan environments. Quenching studies of the enzyme-inhibitor complexes indicate that adenosine deaminase undergoes different protein conformation changes upon binding of ground- and transition-state analogue inhibitors. The results are consistent with localized structural alterations in the enzyme.

Time-resolved fluorescence studies of tomaymycin bonding to synthetic DNAs.

Tomaymycin reacts covalently with guanine in the DNA minor groove, exhibiting considerable specificity for the flanking bases. The sequence dependence of tomaymycin bonding to DNA was investigated in synthetic DNA oligomers and polymers. The maximum extent of bonding to DNA is greater for homopurine and natural DNA sequences than for alternating purine-pyrimidine sequences. Saturation of DNA with tomaymycin has little effect on the melting temperature in the absence of unbound drug. Fluorescence lifetimes were measured for DNA adducts at seven of the ten unique trinucleotide bonding sites. Most of the adducts had two fluorescence lifetimes, representing two of the four possible binding modes. The lifetimes cluster around 2-3 ns and 5-7 ns; the longer lifetime is the major component for most bonding sites. The two lifetime classes were assigned to R and S diastereomeric adducts by comparison with previous NMR results for oligomer adducts. The lifetime difference between binding modes is interpreted in terms of an anomeric effect on the excited-state proton transfer reaction that quenches tomaymycin fluorescence. Bonding kinetics of polymer adducts were monitored by fluorescence lifetime measurements. Rates of adduct formation vary by two orders of magnitude with poly(dA-dG).poly(dC-dT), reacting the fastest at 4 x 10(-2) M-1 s-1. The sequence specificity of tomaymycin is discussed in light of these findings and other reports in the literature.