Skip Navigation

This Article
Right arrow Abstract Freely available
Right arrow Print PDF (120K) Freely available
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrow Search for citing articles in:
ISI Web of Science (18)
Right arrowRequest Permissions
Right arrow Commercial Re-use Guidelines
for Open Access NAR Content
Google Scholar
Right arrow Articles by Dlakic, M.
Right arrow Articles by Harrington, R. E.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Dlakic, M.
Right arrow Articles by Harrington, R. E.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?

Nucleic Acids Research Pages 4274-4279  

Unconventional helical phasing of repetitive DNA motifs reveals their relative bending contributions
Introduction
Materials And Methods
   Purification, labeling and ligation of oligonucleotides and markers
   Two-dimensional gel electrophoresis
   Computer modeling
Results And Discussion
Conclusions
Acknowledgements
References

Unconventional helical phasing of repetitive DNA motifs reveals their relative bending contributions

Unconventional helical phasing of repetitive DNA motifs reveals their relative bending contributions

Mensur Dlakic* and Rodney E. Harrington+

Department of Biochemistry 330, University of Nevada at Reno, Reno, NV 89557-0014, USA

Received April 15, 1998; Revised June 29, 1998; Accepted July 26, 1998

ABSTRACT

A novel, multiple DNA phasing analysis is described in which three sequence motifs associated with bent DNA are clustered together in oligomers of identical base composition, but with different phasing relationships of these motifs to each other. Synthetic oligonucleotides containing different combinations of AAAAA(A), GGGCCC and GAGAG sequence motifs were ligated and analyzed by gel mobility and cyclization experiments to determine their global curvature. These assays were used to obtain relative bending contributions of the analyzed sequence motifs. The experimental results also provide a rigorous test of predictive models for DNA bending. We report, using molecular modeling, that none of the most widely used dinucleotide (nearest neighbor) models can accurately describe the conformational properties of these DNA sequences and that more complex models, at least at the trinucleotide level, are required.

INTRODUCTION

DNA bending was initially noticed with repeated adenine tracts (A-tracts) because such molecules migrate more slowly on polyacrylamide gels than does DNA of the same length but composed of random sequence (1,2). This observation provided the experimental support for an earlier hypothesis stating that local deflections within DNA, when repeated in phase with the helical turn, can add up and lead to macroscopic DNA curvature (3). Further attempts to characterize DNA curvature involved gel mobility experiments on repetitive synthetic DNA sequences (4-6). The idea behind this approach was that local structural variations of DNA (bends), if separated by an integral number of helical turns, will retain the same global direction and add coherently into a defined macroscopic shape (curve). Conversely, local DNA bends with the same direction will cancel each other when separated by half a helical turn (6). Although anomalous gel mobility of curved DNA is not yet fully understood theoretically (7), this method has often provided unique information about the direction and magnitude of intrinsic and protein-induced DNA bending (2,8). It was also used in developing an important predictive model for DNA structure (9). However, a serious problem with gel mobility studies was that they were usually conducted in low ionic strength buffers, which are not necessarily representative of in vivo conditions. Another weakness in this approach has been its relative inability to differentiate flexible versus bent DNA sequences (deformable versus deformed) (10-12). In addition, it has been shown that physiological ionic conditions give rise to DNA curvature in a very different way from that deduced by gel experiments without divalent cations (11-13). Finally, the sequence context in which DNA sequences of interest are located is also a major determinant of DNA structural properties (12). To incorporate these phenomena fully, higher order predictive models are required.

One of the most serious problems in defining the origin of DNA curvature has not been the limitations of gel mobility studies but rather the fact that results from crystal and solution studies disagree on the question of A-tract-associated DNA bending (14-16). Crystallographic results to date uniformly show essentially straight A-tracts (summarized in 17), whereas earlier studies in solution strongly suggest that these sequence elements are bent (5,18,19). To reconcile this dichotomy, it has been proposed that dehydrating agents used to facilitate DNA crystallization might induce structural changes in DNA (20), either directly or through their effect on water activity. This hypothesis has been investigated with gel mobility experiments and shown to be feasible (21). It is also supported by many independent experiments using different dehydrating agents and their effects on DNA structure (22-28). Although the dilemma concerning the origin of DNA bending is not yet fully resolved (12,17,29), several possible explanations have been offered (30,31). Thus, evidence currently available strongly suggests that A-tracts may be polymorphic and their specific structural details may depend on their local hydration environment.

In the present work, we have developed a multiple phasing analysis using AAAAA(A), GGGCCC and GAGAG motifs, which have been previously characterized by us in various combinations (12,31). These sequence motifs were assembled differently into several oligomers of identical base composition (Fig. 1). In each case, one of these motifs is separated from an identical repeat by 15-16 bp (~1.5 helical turns), which effectively cancels its bending contribution (6,19), while the other two sequence motifs remain in phase, separated by either one or two DNA helical turns. By comparing the electrophoretic mobilities and cyclization properties of ligated oligomers, we deduce their relative bending contributions. In addition, we employ computer modeling to show that dinucleotide models in general are not successful in predicting the curvature of these sequences and that trinucleotide models can provide a much improved description of their relative structural characteristics.

MATERIALS AND METHODS

Purification, labeling and ligation of oligonucleotides and markers

Oligonucleotides were obtained from commercial sources. Their purification, labeling and ligation have been described elsewhere (11). A BamHI linker and a HpaII digest of pBR322 were purchased from New England Biolabs and used as molecular weight markers. The BamHI linker was kinased at its 5[prime]-ends with [[gamma]-32P]ATP. Single strands were then annealed and ligated at room temperature for 4 h. The HpaII digest of pBR322 was dephosphorylated using alkaline phosphatase and subsequently labeled with [[gamma]-32P]ATP. Gel mobility experiments, both with and without 10 mM MgCl2, were performed according to a published protocol (12).

Two-dimensional gel electrophoresis

Products of ligation were analyzed by two-dimensional (2D) gel electrophoresis using methods described elsewhere (12,31). Briefly, the first dimension was on a 5% polyacrylamide gel with 25% glycerol. The gel was run at 10 V/cm until xylene cyanol migrated 9 cm on a 14 cm long gel. The gel lane was excised and placed horizontally between the slab gel plates to be used for the second dimension separation. Chloroquine phosphate was added to a final concentration of 50 µg/ml to both the running buffer and the 8% polyacrylamide mixture used for the second dimension gel. Glycerol was added to a final concentration of 10%. After running for 12-18 h at 7 V/cm, X-ray film was placed over the wet gel at room temperature and kept overnight. Sizes of linear and circular DNA fragments were determined by comparison with the original ligation products on 8% denaturing polyacrylamide gels. BamHI linker ligation products were used as additional markers.

Computer modeling

Spatial trajectories in Figures 5 and 6 were generated using DIAMOD, a PC-based program written for this purpose (32). We used four dinucleotide (9,33-35) and two trinucleotide (15,36) models. Angles for the Bolshoy model were used in the form of roll and tilt instead of wedge and direction (the conversion from one to another is described in 9). Angular parameters from trinucleotide models were applied to the 5[prime] dinucleotide within a trinucleotide (15). The statistical parameters in Table 1 were also calculated in DIAMOD following the published algorithm (37).

RESULTS AND DISCUSSION

Figure 1 shows the design of the multiple phasing experiment. Three basic sequence motifs were used: AAAAA(A), GAGAG and GGGCCC. In each oligomer, one of these motifs is out of phase, which cancels out its contribution in terms of bending, while the other two are in phase. For example, in AA-out the distance between the centers of AAAAA(A) motifs is ~1.5 helical turns, which means that they are out of phase. At the same time, GAGAG and GGGCCC motifs are in phase, because their centers are separated by one and/or two helical turns. In GC-out the GGGCCC motif is out of phase and in GA-out the GAGAG motif is out of phase. The concept behind this approach is to determine the relative bending contributions of these three motifs, since phasing out one of them at a time while phasing in the other two should have different effects on DNA bending.


Figure 1. The sequences were designed so that one of the three used sequence motifs [AAAAA(A), GAGAG and GGGCCC] is out of phase, while the other two are in phase. For example, in AA-out the distance between centers of AAAAA(A) motifs is ~1.5 helical turns, which means that their bending contributions cancel each other out. At the same time, the GAGAG and GGGCCC motifs are in phase because their centers are separated by one or two helical turns. Similar logic applies to the other two sequences. In GC-out, the GGGCCC motif is out of phase, while the GAGAG and AAAAA(A) motifs are separated by either one or two helical turns. In GA-out, the GAGAG motif is out of phase, while both the AAAAA(A) and GGGCCC motifs are in phase. All sequences have identical base composition. Complementary strands were designed to produce 3 bp 5[prime]-end overhangs after annealing. Numbers indicate the interval between centers of identical motifs in terms of helical turns.

Gel mobility experiments under standard conditions (low ionic strength buffer, no divalent cations) were used initially to determine bending properties of the oligomers from Figure 1. Results show that gel retardation monotonically increases from AA-out to GA-out (Fig. 2A). In other words, in a sequence where A-tracts cancel each other (with the other two motifs remaining in phase), the global curvature is insignificant. On the other hand, when the GAGAG motif is out of phase, that has the least influence on DNA curvature.


Figure 2. (A) Gel retardation is AA-out < GC-out < GA-out without Mg2+ in the gel and buffer. (B) The general trend of retardation values remains, but retardation increases in the presence of 10 mM Mg2+.

As noted earlier (38), the addition of divalent cations usually increases DNA curvature. However, this increase in retardation anomaly was generally smaller in sequences with A-tracts and larger for GC-rich sequences when only one of these motifs was present in the sequence (11-13). We observed an increase in retardation when 10 mM Mg2+ was added to the gel (Fig. 3), while the relative ratio of these three sequence motifs remained approximately the same (Fig. 2B). This was not unexpected, as all the sequences have GC-rich motifs, which are the main targets for Mg2+ effects (11-13). The fact that the relative order of bending magnitudes remains the same indicates that the main determinant of bending in these sequences is not the presence of GC-rich motifs, but rather the differential phasing between individual bending elements.


Figure 3. Gel mobility of ligated oligomers (in 10 mM MgCl2) against the ligated BamHI linker (M1) and HpaII digest of pBR322 (M2). Lanes T1, T2 and T3 correspond to multimers of AA-out, GC-out and GA-out, respectively.

It was shown earlier that cyclization experiments sometimes produce very different results from gel mobility studies, attributed to the presence of Mg2+ in cyclization experiments and its absence in gel electrophoresis experiments (11,12). The results of cyclization and gel experiments without Mg2+ are usually more similar in a qualitative sense for DNA sequences containing A-tracts (11). The pattern observed here in cyclization experiments is consistent with that seen in gel retardation analysis. Sequence AA-out failed to produce circles after ligation (Fig. 4A), while GA-out yielded small circles due to its bent character (Fig. 4B). As in the gel mobility experiments, GC-out was intermediate (data not shown).


Figure 4. (A) 2D cyclization gel of AA-out. (B) 2D cyclization gel of GA-out. With regard to the orientation of the picture, the first dimension is from left to right and the second is from top to bottom. Linear fragments are in the lower diagonal, as they migrate faster. Circles, when present, are located above linear molecules. Numbers show how many original oligomers were ligated in each DNA fragment. Cyclization parallels gel mobility experiments, as sequences with low retardation anomalies, e.g. AA-out, do not produce small circles, which in this assay is a hallmark of bent DNA sequences. Sequence GA-out shows the highest gel mobility anomaly (Fig. 2) and forms the smallest circles.


Figure 5. Modeling five repeats of sequences from Figure 1 using the dinucleotide models of: (A) Bolshoy et al. (9); (B) Gorin et al. (34); (C) Ulyanov and James (35). Sequences AA-out, GC-out and GA-out correspond to rows I, II and III, respectively. Plots are shown in the plane of largest curvature.


Figure 6. Modeling five repeats of sequences from Figure 1 using the trinucleotide models of: (A) Goodsell and Dickerson (15) based on experimental results by Satchwell et al. (43) and (B) Brukner et al. (36). Sequences AA-out, GC-out and GA-out correspond to rows I, II and III, respectively. Plots are shown in the plane of largest curvature. Figures 5 and 6 were produced using DIAMOD (32).

Comparative gel mobility experiments and cyclization assays register only global DNA structural features (15,16). For that reason, our results can be explained in either of two ways: (i) the GAGAG motif is not bent, while A-tracts and the GGGCCC motifs produce bends of similar magnitude but of opposite direction; (ii) A-tracts are not bent, while GAGAG and GGGCCC motifs both bend in the same direction, with the magnitude of the latter being roughly twice that of the former (11,12). The experiments presented here cannot clearly distinguish between these possibilities. However, we can determine with certainty the relative bending contributions of sequence motifs used in this study. We will make the well-justified assumption, based on both theory and experiments, that only roll contributes significantly to bending (14,39,40). In this case, the bending can be described as an absolute difference between the roll contributions in the sequence motifs used here (16,41). For example, the GAGAG motifs cancel in GA-out and, for that reason, bending in this sequence can be described by means of a difference in the roll angles only between the AAAAA(A) and GGGCCC motifs. Since GA-out shows the largest bending, it can be inferred that roll angles within the AAAAA(A) and GGGCCC motifs have the greatest net difference (31,42). This is consistent with the notion that the roll angles of AAA and GGC are on completely opposite sides of the bending scale (15,36,43). Applying the same argument to the other two sequences, we observe that the roll-dominated bending in the GAGAG motif is between AAAAA(A) and GGGCCC. However, it must be closer to the latter because the net difference between bends in the GGGCCC and GAGAG motifs in AA-out is smaller than the difference between the AAAAA(A) and GAGAG motifs in GC-out. This result is slightly different from our earlier results showing that bending in the GAGAG motif is central between the AAAAA(A) and GGGCCC motifs (12). One explanation for this discrepancy comes from the fact that stable DNA bending seems to require phased A-tracts (11), regardless of whether A-tracts are themselves bent or straight. Since AA-out has A-tracts out of phase, it is feasible that bending of this sequence might be underestimated. It is also possible that different sequence contexts of the GAGAG motif in this analysis, compared with that in the previous study (12), contribute partially to the observed variance.

As remarked above, different experimental approaches (e.g. gel mobility versus cyclization assays) and different conditions (e.g. low ionic strength buffers versus buffers with divalent cations) often produce variable results even on the same DNA sequence (11,31). In that regard, the results are not always easy to bring under a common denominator and analyze by objective theoretical methods. The sequences used in the present study overcome that problem as they show consistent results by several techniques. In addition, the monotonic increase in both gel retardation and cyclization from AA-out through GC-out to GA-out offers a good way of objectively testing the predictive models for DNA bending. To carry out this test, we generated pentamer oligos for each sequence and determined their bending properties. This size was chosen because pentamers of GA-out already produce circles (Fig. 4B). We made no attempt here to furnish a full quantitative evaluation of all predictive models and determine the best one, as that problem was independently addressed on a larger set of sequences (M.Dlakic and R.E.Harrington, in preparation). Instead, we addressed only the ability of predictive models to match the experimental curvature profiles of our three sequences. According to that rule, any predictive model will be proclaimed good if it can predict the largest curvature in GA-out and smallest in AA-out, with GC-out being intermediate.

Several statistical parameters are presented in Table 1. The ratio of end-to-end distance (EED) to the contour length of a DNA molecule is one indicator of bending. For straight DNA this ratio equals one and it approaches zero as bending increases (bending brings the ends of the molecule closer together). The ratio of largest to smallest principle moments of inertia (RMI) is roughly comparable with the length/diameter ratio of the molecule. This value is large for an extended molecule, as its length is always bigger than any other dimension, but it converges to one as the molecule becomes more tightly packed by bending. Finally, the radius of gyration (RG) indicates the compactness of the molecule. When comparing DNA molecules of identical sizes, it will be smaller for curved and larger for extended molecules. For the purposes of our evaluation, a DNA molecule is straight when EED is close to 1 and RMI assumes a large double digit value (e.g. the Brukner model prediction for AA-out in Table 1). Conversely, DNA is bent when EED is close to 0 and RMI approaches 1 (e.g. the Bolshoy model prediction for GC-out in Table 1).

Almost all models are successful in predicting that AA-out is essentially straight (EED close to 1 and large RMI), which is reflected in Figures 5 and 6. However, dinucleotide models (9,33-35) do not show the required gradual increase in DNA bending (Table 1). For example, middle sequence GC-out has a larger curvature than the other two with the Bolshoy and De Santis models and a smaller one than the others with the Gorin and Ulyanov models (Table 1 and Fig. 5; data obtained using the De Santis model are not shown in Fig. 5). Neither of these two trends is consistent with the experimental results. On the other hand, the trinucleotide models (15,36,43) produce bending patterns that qualitatively match the experimental results (Fig. 6), which is reflected in a gradual decrease in EED, RMI and RG from AA-out to GA-out (Table 1). A similar conclusion about the superiority of the trinucleotide models was reached by others (15,44) and is additionally strengthened by an independent analysis on a larger set of sequences (M.Dlakic and R.E.Harrington, in preparation).


Table 1. The statistics for dinucleotide predictive models by Bolshoy et al. (9), De Santis et al. (33), Gorin et al. (34) and Ulyanov and James (35) and the models by Goodsell and Dickerson (15) and Brukner et al. (36) based on trinucleotides
The numbers in each cell were calculated for DNA multimers of five repeats using DIAMOD (32): the ratio of EED over contour length is at the top, the ratio of the longest to the smallest RMI is in the middle and the RG is at the bottom.

Having said that the trinucleotide models exhibit better predictive power, we should point out that they cannot be used to distinguish between models for DNA curvature. Although the trinucleotide models reported to date all have roll values scaled from 0-10° (15,36), these models cannot support straight A-tracts and bent general sequence DNA per se, since the choice of scaling was completely arbitrary. For example, the experimental data used for these models (43,45) could have been scaled from -5° to 5° (the absolute difference between the smallest and largest roll values would also be 10°), in which case they would support bent A-tracts while still producing identical global DNA trajectories (M.Dlakic and R.E.Harrington, in preparation).

CONCLUSIONS

We have reported a novel, multiple phasing analysis to facilitate interpretation of DNA curvature studies. By combining three sequence motifs at a time and placing them into different phasing arrangements, we can obtain new information about their relative bending contributions. Extending these studies to a larger number of sequences is straightforward and could be used to construct a scale of relative bending propensities, similar to those already described (15,36).

The design of our sequences and the nature of their bending have provided a way of testing several predictive models of DNA structure. Confirming earlier results (15,44), we show that trinucleotide models are more successful in explaining our data than dinucleotide models. A potential problem with trinucleotide models is that they utilize sequence-independent values of twist, which is not true from a structural point of view. However, this incorrect assumption seems to be offset by three correct postulates that are embedded in trinucleotide models: (i) roll contributes mostly to DNA bending; (ii) tilt is insignificant compared with roll; (iii) sequence-dependent differential bendability, as opposed to fixed angular parameters used in dinucleotide models, can be used to capture structural features of DNA (all these points are reviewed in 46).

ACKNOWLEDGEMENTS

We thank Professor Steve Harvey for many constructive discussions. This work was supported by grants CA70274 and GM53517 from the NIH and from USDA Hatch Project NEV032D (R.E.H.) and by a supplementary grant from the Open Society Institute (M.D.).

REFERENCES

1. Marini,J.C., Levene,S.D., Crothers,D.M. and Englund,P.T. (1982)Proc. Natl Acad. Sci. USA, 79, 7664-7668.

2. Wu,H.M. and Crothers,D.M. (1984) Nature (Lond.), 308, 509-513.

3. Trifonov,E.N. and Sussman,J.L. (1980) Proc. Natl Acad. Sci. USA, 77, 3816-3820. MEDLINE Abstract

4. Hagerman,P.J. (1985) Biochemistry, 24, 7033-7037. MEDLINE Abstract

5. Hagerman,P.J. (1986) Nature (Lond.), 321, 449-450.

6. Koo,H.S., Wu,H.M. and Crothers,D.M. (1986) Nature (Lond.), 320, 501-506.

7. Zimm,B.H. (1993) Curr. Opin. Struct. Biol., 3, 373-376.

8. Zinkel,S.S. and Crothers,D.M. (1987) Nature (Lond.), 328, 178-181.

9. Bolshoy,A., McNamara,P., Harrington,R.E. and Trifonov,E.N. (1991) Proc. Natl Acad. Sci. USA, 88, 2312-2316. MEDLINE Abstract

10. Kahn,J.D., Yun,E. and Crothers,D.M. (1994) Nature (Lond.), 368, 163-166.

11. Dlakic,M. and Harrington,R.E. (1995) J. Biol. Chem., 270, 29945-29952. MEDLINE Abstract

12. Dlakic,M. and Harrington,R.E. (1996) Proc. Natl Acad. Sci. USA, 93, 3847-3852. MEDLINE Abstract

13. Brukner,I., Susic,S., Dlakic,M., Savic,A. and Pongor,S. (1994) J. Mol. Biol., 236, 26-32. MEDLINE Abstract

14. Goodsell,D.S., Kopka,M.L., Cascio,D. and Dickerson,R.E. (1993)Proc. Natl Acad. Sci. USA, 90, 2930-2934. MEDLINE Abstract

15. Goodsell,D.S. and Dickerson,R.E. (1994) Nucleic Acids Res., 22, 5497-5503. MEDLINE Abstract

16. Haran,T.E., Kahn,J.D. and Crothers,D.M. (1994) J. Mol. Biol., 244, 135-143. MEDLINE Abstract

17. Dickerson,R.E., Goodsell,D. and Kopka,M.L. (1996) J. Mol. Biol., 256, 108-125. MEDLINE Abstract

18. Ulanovsky,L.E. and Trifonov,E.N. (1987) Nature (Lond.), 326, 720-722.

19. Koo,H.S. and Crothers,D.M. (1988) Proc. Natl Acad. Sci. USA, 85, 1763-1767. MEDLINE Abstract

20. DiGabriele,A.D. and Steitz,T.A. (1993) J. Mol. Biol., 231, 1024-1039. MEDLINE Abstract

21. Sprous,D., Zacharias,W., Wood,Z.A. and Harvey,S.C. (1995)Nucleic Acids Res., 23, 1816-1821. MEDLINE Abstract

22. Drew,H.R. and Travers,A.A. (1984) Cell, 37, 491-502. MEDLINE Abstract

23. Buckin,V.A., Kankiya,B.I., Bulichov,N.V., Lebedev,A.V., Gukovsky,I.Y., Chuprina,V.P., Sarvazyan,A.P. and Williams,A.R. (1989) Nature (Lond.), 340, 321-322.

24. Herrera,J.E. and Chaires,J.B. (1989) Biochemistry, 28, 1993-2000. MEDLINE Abstract

25. Chan,S.S., Breslauer,K.J., Hogan,M.E., Kessler,D.J., Austin,R.H., Ojemann,J., Passner,J.M. and Wiles,N.C. (1990) Biochemistry, 29, 6161-6171. MEDLINE Abstract

26. Breslauer,K.J. (1991) Curr. Biol., 1, 416-422.

27. Park,Y.W. and Breslauer,K.J. (1991) Proc. Natl Acad. Sci. USA, 88, 1551-1555. MEDLINE Abstract

28. Ganunis,R.M., Guo,H. and Tullius,T.D. (1996) Biochemistry, 35, 13729-13732. MEDLINE Abstract

29. Han,G.W., Kopka,M.L., Cascio,D., Grzeskowiak,K. and Dickerson,R.E. (1997) J. Mol. Biol., 269, 811-826. MEDLINE Abstract

30. Harvey,S.C., Dlakic,M., Griffith,J., Harrington,R., Park,K., Sprous,D. and Zacharias,W. (1995) J. Biomol. Struct. Dyn., 13, 301-307. MEDLINE Abstract

31. Dlakic,M., Park,K., Griffith,J.D., Harvey,S.C. and Harrington,R.E. (1996) J. Biol. Chem., 271, 17911-17919. MEDLINE Abstract

32. Dlakic,M. and Harrington,R.E. (1998) Bioinformatics, 14, 326-331. MEDLINE Abstract

33. De Santis,P., Palleschi,A., Savino,M. and Scipioni,A. (1990) Biochemistry, 29, 9269-9273. MEDLINE Abstract

34. Gorin,A.A., Zhurkin,V.B. and Olson,W.K. (1995) J. Mol. Biol., 247, 34-48. MEDLINE Abstract

35. Ulyanov,N.B. and James,T.L. (1995) Methods Enzymol., 261, 90-120. MEDLINE Abstract

36. Brukner,I., Sanchez,R., Suck,D. and Pongor,S. (1995) J. Biomol. Struct. Dyn., 13, 309- 317. MEDLINE Abstract

37. Bansal,M., Bhattacharyya,D. and Ravi,B. (1995) Comput. Appl. Biosci., 11, 281-287. MEDLINE Abstract

38. Diekmann,S. (1987) Nucleic Acids Res., 15, 247-265. MEDLINE Abstract

39. Zhurkin,V.B., Lysov,Y.P. and Ivanov,V.I. (1979) Nucleic Acids Res., 6, 1081-1096. MEDLINE Abstract

40. Dickerson,R.E., Goodsell,D.S. and Neidle,S. (1994) Proc. Natl Acad. Sci. USA, 91, 3579-3583. MEDLINE Abstract

41. Calladine,C.R. and Drew,H.R. (1992) Understanding DNA: The Molecule and How It Works. Academic Press, London, UK.

42. Han,W., Dlakic,M., Zhu,Y.J., Lindsay,S.M. and Harrington,R.E. (1997) Proc. Natl Acad. Sci. USA, 94, 10565-10570. MEDLINE Abstract

43. Satchwell,S.C., Drew,H.R. and Travers,A.A. (1986) J. Mol. Biol., 191, 659-675. MEDLINE Abstract

44. Gabrielian,A. and Pongor,S. (1996) FEBS Lett., 393, 65-68. MEDLINE Abstract

45. Brukner,I., Sanchez,R., Suck,D. and Pongor,S. (1995) EMBO J., 14, 1812-1818. MEDLINE Abstract

46. Dickerson,R.E. (1998) Nucleic Acids Res., 26, 1906-1926. MEDLINE Abstract

*To whom correspondence should be addressed at present address: Howard Hughes Medical Institute, 4570 MSRB II, 1150 West Medical Center Drive, Ann Arbor, MI 48109-0650, USA. Tel: +1 734 764 3554; Fax: +1 734 763 9323; Email: mensur@umich.edu
+Present address: Department of Microbiology, Arizona State University, Tempe, AZ 85287-2701, USA

This page is run by Oxford University Press, Great Clarendon Street, Oxford OX2 6DP, as part of the OUP Journals
Comments and feedback: www-admin{at}oup.co.uk
Last modification: 29 Aug 1998
Copyright©Oxford University Press, 1998.
Add to CiteULike CiteULike   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us    What's this?


This article has been cited by other articles:


Home page
 Nucleic Acids ResHome page
A. Kanhere and M. Bansal
An assessment of three dinucleotide parameters to predict DNA curvature by quantitative comparison with experimental data
Nucleic Acids Res., May 15, 2003; 31(10): 2647 - 2658.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Print PDF (120K) Freely available
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrow Search for citing articles in:
ISI Web of Science (18)
Right arrowRequest Permissions
Right arrow Commercial Re-use Guidelines
for Open Access NAR Content
Google Scholar
Right arrow Articles by Dlakic, M.
Right arrow Articles by Harrington, R. E.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Dlakic, M.
Right arrow Articles by Harrington, R. E.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?