Sequence dependence of the circular dichroism of synthetic double-stranded RNAs

We have synthesized and studied the CD spectra of five new double-stranded RNA polymers: poly[r(A-G)·r(C-U)], poly[r(A-U-C)·r(G-A-U)], poly[r(A-C-U)·r(A-G-U)], poly[r(A-A-C)·r(G-U-U)], and poly[r(A-C-C)·r(G-G-U)]. Together with previously published spectra of seven other RNA sequences, the spectra of these new sequences provide a library sufficient to approximate the spectra of all other RNA sequences by first-neighbor formulas and, in addition, give four spectra with which we may test the validity of first-neighbor approximations. (1) We find that the spectra of RNA sequence isomers are very different, but that the spectra essentially do obey first-neighbor relationships. (2) We have derived tentative first-neighbor assignments of negative bands at about 295 and 210 nm in the CD spectra. (3) A test of spectral independence shows that among the 12 polymer spectra there are at least seven significant independent spectral shapes, one less than the eight needed to give the most accurate spectral analysis of an unknown RNA sequence for its first-neighbor frequencies. (4) Spectra are calculated for RNAs of random base composition, approximating natural RNAs having complex sequences. (5) A T-matrix of spectral components assigned to the first-neighbor base pairs is derived from 10 of the spectra. This matrix allows an estimation of the CD spectrum of any other known RNA sequence or an analysis of the spectrum of an unknown sequence for its distribution of first-neighbor base-pair frequencies. (6) Test analyses of two of the synthetic polymers and of two natural RNAs set a probable limit on the accuracy of first-neighbor frequency determinations using this T-matrix. (7) Finally, we summarize in an appendix the melting temperatures for all the RNA and corresponding DNA sequences; it appears that the T_m values of both DNAs and RNAs approximately obey first-neighbor relationships.     

A comparison of the circular dichroism spectra of synthetic DNA sequences of the homopurine . homopyrimidine and mixed purine- pyrimidine types.

We have obtained the ultraviolet circular dichroism spectra of two repeating trinucleotide DNAs, poly [d(A-G-G).d(C-C-T)] and poly[d(A-A-G).d(C-T-T)], that have all purines on one strand and all pyrimidines on the other. These spectra, together with spectra of other synthetic polymers, can be combined to give 3 first-neighbor calculations of the spectrum of poly[d(A).d(T)] and 2 first-neighbor calculations of the spectrum of poly [d(G).d(C)]. The results show (1) that first-neighbor calculations utilizing only spectra of homopurine.homopyrimidine DNA sequences are no more accurate than are similar calculations that involve spectra of mixed purine-pyrimidine sequences, demonstrating that double-stranded homopurine.homopyrimidine sequences do not obviously belong to a special class of secondary conformations, and (2) that the wavelength region above 250 nm in the CD spectra of synthetic DNAs is least predictable from first-neighbor equations, probably because this region is especially sensitive to sequence-dependent conformational differences.     

The analysis of circular dichroism spectra of natural DNAs using spectral components from synthetic DNAs

We have tested 21 different basis sets of synthetic DNA circular dichroism spectra and have slected one for use in spectral analyses of natural DNAs. This “standard” set consists of spectra of eight polymers: poly[d(A-A-T)·d(A-T-T)], poly[d(A-G-G)·d(C-C-T)], poly[d(A-T)·d(A-T)], poly[d(G-C)·d(G-C)], poly[d(A-G)·d(C-T)], poly[d(A-C)·d(G-T)], poly[d(A-T-C)·d(G-A-T)], and poly[d(A-G-C)·d(G-C-T)]. This basis set, applied according to the first-neighbor polymer procedure of Gray and Tinoco, allows a more uniformly accurate spectral analysis of six natural complex DNAs and eight (A+T)-rich satellite DNAs for base composition and first-neighbor frequencies than was previously possible. We find that spectra of poly[d(A)·d(T)] and/or poly[d(A-C-T-)·d(A-G-T)] are not generally required for good analysis results but we show in this and the following paper that these spectra are needed for the most accurate analyses of some satellite DNAs.     

The circular dichroism of poly d(AAT): d(ATT) and poly r(AAU):r(AUU). A test of the first-neighbor hypothesis

We present absorption and circular dichroism (CD) spectra for the synthetic polymers poly d(AAT):d(AAT) and poly r(AAU):r(AAU), in both native and heat-denatured forms. As a means of evaluating the first-neighbor hypothesis, the CD spectra are compared with approximations derived from spectra of other synthetic polymers containing the same first-neighbor sequences. This is the first instance where such a comparison has been possible using spectra of double-stranded RNA sequences, and the agreement between the measured and approximated spectra for poly r(AAU):r(AUU) is surprisingly good. We have also subjected the CD spectrum of poly d(AAT):d(AAT) to a previously published analytical procedure for obtaining estimates of first-neighbor frequencies. In this first independent test of the procedure, we find that the analysis does infer the existence of a majority (86%) of AA, TT, AT, and TA first neighbors but does not precisely indicate their relative proportions.     

On the first-neighbor analysis of nucleic acid CD spectra: The definitive T Matrix and considerations of various methods

The first-neighbor approximation for the representation of double-stranded polynucleotide CD has been widely applied. Two apparently different formalisms have appeared. We discuss and compare those formalisms and show them to be identical. We present an updated optimum basis set of polydeoxynucleotide CD spectra and the resulting first-neighbor unit contribution matrix derived from these data.     

A simple method for the computation of first neighbour frequencies of DNAs from CD spectra

A procedure for the computation of the first neighbour frequencies of DNA's is presented. This procedure is based on the first neighbour approximation of Gray and Tinoco. We show that the knowledge of all the ten elementary CD signals attached to the ten double stranded first neighbour configurations is not necessary. One can obtain the ten frequencies of an unknown DNA with the use of eight elementary CD signals corresponding to eight linearly independent polymer sequences. These signals can be extracted very simply from any eight or more CD spectra of double stranded DNA's of known frequencies. The ten frequencies of a DNA are obtained by least square fit of its CD spectrum with these elementary signals. One advantage of this procedure is that it does not necessitate linear programming, it can be used with CD data digitalized using a large number of wavelengths, thus permitting an accurate resolution of the CD spectra. Under favorable case, the ten frequencies of a DNA (not used as input data) can be determined with an average absolute error < 2%. We have also observed that certain satellite DNA's, those of Drosophila virilis and Callinectes sapidus have CD spectra compatible with those of DNA's of quasi random sequence; these satellite DNA's should adopt also the B-form in solution.     

Absolute configuration of tropane alkaloids bearing two alpha,beta-unsaturated ester functions using electronic CD spectroscopy: application to (R,R)-trans-3-hydroxysenecioyloxy-6-senecioyloxytropane

The absolute configuration of heterocyclic natural products substituted with two mobile alpha,beta-unsaturated esters was studied using electronic circular dichroism (CD) spectroscopy. The conformational flexibility of the side chains imposed the use of density functional theory calculation to determine the set of the most probable conformations in solution. The electronic CD and UV spectra were calculated by Boltzmann-weighted average of the simulated spectra using the results of the excited states calculation of a set of simplified structures. Comparison with the experimental CD spectrum allowed to determine whether the calculations were made with the right enantiomer. The method was applied to the determination of the absolute configuration of (R,R)-trans-3-hydroxysenecioyloxy-6-senecioyloxytropane.     

Predicting ultraviolet spectrum of single stranded and double stranded deoxyribonucleic acids

Synthetic oligodeoxynucleotides are widely used in many biological, biochemical and biophysical applications. The concentration, composition and structure of DNA are often determined from its ultraviolet spectrum. Although parameters for use with the nearest-neighbor model for prediction of extinction coefficients of single stranded DNAs at 260 nm were published some time ago, similar parameters for other wavelengths or for use with DNA duplexes have not been reported. Practical formulae and parameters for prediction of UV spectra, hypochromism and peak wavelengths were experimentally determined for both single stranded and double stranded oligodeoxynucleotides in the range from 215 to 310 nm. The accuracy of predictions made using the nearest-neighbor model and the base composition model was determined and compared. The spectrum of any DNA oligomer can be calculated using a Microsoft Excel application that is available in the Appendix A.     

Analysis of circular dichroism spectra of homopurine: homopyrimidine DNA's for sequence information and comparison of results with heterobase DNA

Procedures are developed which make a first neighbor frequency analysis possible from a CD spectrum of homopyrimidine: homopurine DNA's. The contribution to the CD spectrum from the various first neighbor frequencies present in homopyrimidine: homopurine-type DNA's has been determined, and hence the CD spectrum for any DNA of this type with known first neighbor frequencies can easily be calculated. An identical analysis is presented for the determination of extinction coefficients. It is further shown that unlike the more usual heteropurine–pyrimidine DNA's a random sequence does not lead to a simplified formalism. Finally, it is concluded that the homopyrimidine: homopurine DNA's have a structure that is different from that of the more usual heterobase DNA's. A procedure capable of determining first neighbor frequencies from a CD spectrum for heterobase and/or homopurine: homopyrimidine DNA's is described. This procedure is used to determine that there is only minimal interference between these two types of DNA in the first neighbor analysis.     

An ochre suppressor of bacteriophage T4 that is associated with a transfer RNA

Ultraviolet circular dichroism spectra have been obtained for native and heat-denatured Drosophila virilis satellite DNAs I, II and III. Gall &; Atherton (1974) have found that these DNAs have simple, unique sequences. We compare here the circular dichroism spectra of these satellite sequences with the circular dichroism spectra of synthetic DNAs of simple sequences which are combined in first-neighbor calculations. We also apply an analytical procedure for determining nearest-neighbor frequencies from the DNA spectra (Allen et al., 1972). The results are an indication of the potential usefulness and present limitations of circular dichroism measurements in confirming or determining the nearestneighbor frequencies of satellite DNAs of simple sequences.     

Circular dichroism spectra of twelve short DNA restriction fragments of known sequence: a comparison of measured and calculated spectra.

The CD spectra of twelve DNA restriction fragments ranging in size from 12 to 360 base pairs are reported. Since the sequences of these fragments are known, it is possible to calculate their CD spectra from a set of nearest neighbor contributions derived from a combination of synthetic polydeoxyribonucleotides. While the calculations lead to good agreement in the negative band at approximately 245 nm, they generally reproduce the positive band at approximately 270 nm only poorly. The experimentally observed positive band consists of two peaks centered around 270 and 285 nm. The comparison of calculated and measured spectra reveals that end effects lead to increased disagreement for fragments smaller than approximately 40 base pairs. The disagreement between calculated and measured spectra can be partially attributed to the fraction of next nearest neighbors in the DNAs, which are also in the spectral components. Thus, the sequence specific CD contributions in the long wavelength region of the spectra extend at least to next nearest neighbor nucleotides and may extend beyond.     

The Resolution of nucleic acid-ligand binding sites by CD first-neighbor analysis

The first-neighbor approximation for representation of double-stranded polynucleotide CD has been used to determine which first-neighbor units are favorable binding sites for a ligand. Here, we refine the process to yield a more definitive answer. The binding at the first-neighbor units, ApA:TpT and CpC:GpG, may be unequivocally resolved, and when fewer than six of the remaining eight base-paired first-neighbor units are bound by the ligand, they too can be resolved. If this condition is not met, then the binding can be characterized as nonspecific.     

Conformation and circular dichroism of DNA.

CD spectra of calf thymus, C. perfringens, E. coli, and M. luteus DNA have been measured in the vacuum-uv region to about 168 nm for the A-, B-, and C-forms. The positive band at about 187 nm and the negative band at about 170 nm found for each type and form of DNA are sensitive to the source of the DNA and the base–base interactions of the double-stranded helix. The A-form spectra confirm that these bands are indeed sensitive to secondary structure. In the near-uv, the CD of B-form DNA is well analyzed as a linear combination of 27% A-form and 78% C-form. However, an analysis of the extended spectrum demonstrates that the near-uv analysis is not correct. The extended analysis shows that the base–base interactions are similar for B- and C-forms in solution, which implies that these two forms have nearly the same number of base pairs per turn. Various types of CD difference spectra are also discussed.     

Spectroscopic analysis of the interaction of Escherichia coli DNA-dependent RNA polymerase with T7 DNA and synthetic polynucleotides.

We have studied the circular dichroism and ultraviolet difference spectra of T7 bacteriophage DNA and various synthetic polynucleotides upon addition of Escherichia coli RNA polymerase. When RNA polymerase binds nonspecifically to T7 DNA, the CD spectrum shows a decrease in the maximum at 272 but no detectable changes in other regions of the spectrum. This CD change can be compared with those associated with known conformational changes in DNA. Nonspecific binding to RNA polymerase leads to an increase in the winding angle, theta, in T7 DNA. The CD and UV difference spectra for poly[d(A-T)] at 4 degrees C show similar effects. At 25 degrees C, binding of RNA polymerase to poly[d(A-T)] leads to hyperchromicity at 263 nm and to significant changes in CD. These effects are consistent with an opening of the double helix, i.e. melting of a short region of the DNA. The hyperchromicity observed at 263 nm for poly[d(A-T)] is used to determine the number of base pairs disrupted in the binding of RNA polymerase holoenzyme. The melting effect involves about 10 base pairs/RNA polymerase molecule. Changes in the CD of poly(dT) and poly(dA) on binding to RNA polymerase suggest an unstacking of the bases with a change in the backbone conformation. This is further confirmed by the UV difference spectra. We also show direct evidence for differences in the template binding site between holo- and core enzyme, presumably induced by the sigma subunit. By titration of the enzyme with poly(dT) the physical site size of RNA polymerase on single-stranded DNA is approximately equal to 30 bases for both holo- and core enzyme. Titration of poly[d(A-T)] with polymerase places the figure at approximately equal to 28 base pairs for double-stranded DNA.     

First-neighbor specificities of actinomycin-DNA bindings by circular dichroism.

The circular dichroism spectra of eleven double-stranded DNAs, five natural with known nearest neighbor frequencies and six synthetic polydimers and polytrimers, were measured from 210 to 310 nm in the absence and presence of increasing amounts of actinomycin up to saturation. Based on the fact that the circular dichroism of nucleic acids is a nearest-neighbor frequency-dependent property, matrix analysis of the problem revealed which neighbor sets were perturbed by actinomycin, presumably by intercalation of the planar moiety of the molecule. The intercalation sites can be separated into three families. The first-neighbor units GpC and CpG are very favorable binding sites for actinomycin. ApG, CpC, ApC, TpC, and TpG appear to be less attractive sites, while ApT, TpA, and ApA are unfavorable sites.     

Raman spectra of the model B-DNA oligomer d(CGCGAATTCGCG)2 and of the DNA in living salmon sperm show that both have very similar B-type conformations

Raman spectra were obtained from aqueous solutions of the deoxyoligonucleotide d(CGCGAATTCGCG)2 (I), which has been suggested as a model for B-type DNA conformation. These spectra were compared with the Raman spectra of the aqueous solutions of several DNAs of natural origin taken under identical solution conditions. Since the model sequence has a high percent GC (66%), the Raman spectrum was compared with the Raman spectrum of the DNA from Micrococcus lysodeikticus (72% GC), and the spectra of the two different DNAs were found to be rather similar in both 50 mM salt and 6 M salt solutions. Computer-aided band-shape analysis of the backbone vibrational region of the Raman spectra shows the existence of several bands corresponding to different furanose ring puckers. This appears to indicate a heterogeneity of furanose ring pucker in both the model dodecamer and the native DNA. Significant differences were found in the intensity of the conformational marker band at 810 cm-1, which indicates corresponding differences in furanose ring pucker heterogeneities in these two high GC content DNAs. The Raman spectrum of the dodecamer (I) was used to analyze the Raman spectrum of the DNA inside the head of living intact salmon sperm. Sperm spectra were taken with both our conventional Raman spectrograph and a newly developed intracavity laser Raman microscope system. Although the DNA in the sperm head is required by packing considerations to be in a highly compact and condensed state, the Raman spectra of the intact sperm are almost identical with that of the model dodecamer (I) if the difference in base composition is taken into account.(ABSTRACT TRUNCATED AT 250 WORDS)     

Sequence-dependence of the CD of synthetic double-stranded RNAs containing inosinate instead of guanylate subunits

The CD spectra and melting profiles have been measured for nine synthetic double-stranded RNAs containing I · C instead of G · C base pairs: poly[r(I) · r(C)], poly[r(I-C) · r(I-C)], poly[r(A-I-C) · r(I-C-U)], poly[r(A-C) · r(I-U)], poly[r(A-I) · r(C-U)], poly[r(A-C-C) · r(I-I-U)], poly[r(A-A-C) · r(I-U-U)], poly[r(A-C-U) · r(A-I-U)], and poly[r(A-U-C) · r(I-A-U)]. CD spectra have not previously been reported for the latter six of these polymers. The substitution of inosinate for guanylate led to recognizable CD differences, with all but two of the polymers having two resolved positive bands above 230 nm. Also, the I-containing RNAs differed from their G-containing counterparts in the almost complete absence of negative CD bands at long wavelengths and in the reduction of negative CD bands near 210 nm. First-neighbor comparisons showed that the CD spectra of the I-containing RNAs were consistent with the nearest-neighbor sequences of the polymers, as previously shown for G-containing RNAs (D. M. Gray, J.-J. Liu, R. L. Ratliff, and F. S. Allen, Biopolymers (1981) 20 , 1337–1382). Moreover, two of the first-neighbor comparisons involved spectra of poly[r(A) · r(U)] and poly[r(I) · r(C)], polymers known to be in the A family of conformations in fibers (S. Arnott, D. W. L. Hukins, S. D. Dover, W. Fuller, and A. Hodgson, (1973) J. Mol. Biol. 81 , 107–122). Thus, differences in the CD spectra of I- and G-containing RNAs could be simply explained as resulting from differences in the hypoxanthine and guanine chromophores, without invoking differences in conformation. Finally, melting temperatures of the I-containing RNAs were found to vary much less with base composition than do the melting temperatures of G-containing RNAs, since A · U base pairs are closer to I · C than to G · C base pairs in stability.     

Changes in DNA conformation induced by gamma irradiation in the presence of copper

Gamma irradiation of DNA solutions containing copper causes changes in DNA conformation in oligonucleotides and in natural and synthetic DNAs. Diagnostic for these conformational changes is a ubiquitous 187-nm peak in the circular dichroism (CD) difference spectrum that has been predicted for a transformation from a right-handed to a left-handed helical DNA conformation. Changes in CD are correlated with changes in the UV spectrum. Reduction of DNA-bound Cu(II) to Cu(I) with ascorbic acid produces similar changes in CD spectra. These changes can be produced by the peroxy radical anion (O2*-) and the OH radical in the presence of copper. O2*- is approximately twice as efficient as *OH in initiating these changes in natural DNA. The changes in DNA conformation induced by ionizing radiation are remarkable in that they are dependent on the copper-ion concentration in a highly nonlinear manner at low copper concentrations and are not observed in the absence of copper ions. Possible implications of our results for radiobiological and oxidative damage in the cell nucleus are discussed.