Most of the DNA of eukaryotic cells is organized as arrays of nucleosomes. Nucleosomal DNA can be divided into two structurally distinct regions, core and linker. Core DNA is about 146 bp ()in length and is wrapped around the core histone octamer in about 1.75 turns of left-handed superhelix(1) . Linker DNA, which connects adjacent core regions, is variable in length (ranging from about 15 to 100 bp), and its conformation is not known (2) .

In the presence of histone H1, the nucleosomal arrays fold into higher order structures that can be visualized as fibers approximately 30 nm wide(3) . The organization of nucleosomes in these fibers remains unknown. There are two major classes of model, solenoid models(3, 4, 5) and models based on a zigzag chain of nucleosomes(6, 7, 8, 9, 10, 11, 12, 13) . Solenoid models consist of a helical arrangement of nucleosomes in which the linker DNA passes from one nucleosome to its immediate neighbor along the helical path of the fiber. In these models, the linker DNA is curved to a degree that is similar to or greater than that of core DNA. Models based on a zigzag chain of nucleosomes all have the linker DNA relatively straight. Therefore, information about the curvature of linker DNA should allow the differentiation of these two classes of model from each other. However, information about the structure of linker DNA is sparse and comes mainly from physical measurements of purified chromatin fragments (14) or dinucleosomes(15, 16) . Because higher order chromatin structure and the conformation of linker DNA may be perturbed by chromatin isolation(17) , analysis of linker conformation is best done on chromatin in a minimally altered state, preferably in isolated nuclei or living cells.

Photo-induced formation of cis-syn cyclobutane ring pyrimidine dimers (PDs) has been used to probe the structure of nucleosomal DNA. PD formation in the core region occurs with peaks at approximately 10-nucleotide intervals(18, 19) . Based on a model by Pearlman et al.(20) of the distortion of DNA produced by a thymine dimer, it was postulated that the periodic PD peaks were caused by the curvature of core DNA, with PD formation being favored in regions where the DNA is bent toward the major groove and maximal at the positions just 3` of the bends(19) . Consistent with this postulate, PD formation was also found to occur with peaks at approximately 10-nucleotide intervals in a DNA loop that was free of protein contacts(21) . The relationship of the PD peaks to the curvature of the loop was essentially identical to that observed in the nucleosome, indicating that the pattern of PD formation in the nucleosome is primarily due to DNA conformation rather than histone-DNA contacts. Two recent studies provide additional support for this view. One study found that conditions that unfold nucleosomal core DNA to a more extended conformation without removing core histones abolished the periodicity of PD formation(22) . In another study, nucleosome cores were reconstituted onto DNA already containing PDs(23) . In this experiment, one expects the nucleosomes to take up positions that accommodate the distortions in DNA conformation produced by the PDs. The distribution of PDs in the reconstituted nucleosomes was very similar to that seen in nucleosomes from irradiated chromatin, indicating that DNA conformation was the main determinant in both cases.

This considerable body of evidence that DNA conformation is the main determinant of the pattern of PD formation in the nucleosome supports the idea that PD formation can be used to probe the conformation of nucleosomal DNA. Because this method can be used on isolated nuclei or even living cells, it offers the potential to obtain information about the conformation of linker DNA in situ. A preliminary analysis of PD formation in linker DNA found a nearly uniform distribution of PDs, indicating that linker DNA is relatively straight compared with core DNA(19) . However, results from only a single linker length were obtained, and because there is significant variability in linker length, it was not clear if this result was representative of most chromatin. In the present work, the distribution of PDs in linker regions of several different lengths is reported.

EXPERIMENTAL PROCEDURES

Irradiation of Nuclei and Trimming of Nucleosomal Fragments

Mapping of PDs in Trimmed Dinucleosomal DNA

RESULTS

Nuclei isolated from rat livers were photoirradiated with 313 nm light in the presence of the photosensitizing agent AcD. This method of PD formation produces fewer non-PD photoproducts than irradiation with 254 nm light(27) . The second most common photoproduct produced by irradiation at 254 nm is the pyrimidine-pyrimidone [6-4] photoproduct, which has been shown to have a nearly random distribution in the core region of the nucleosome(28) . Although the contribution of photoproducts to the overall pattern of photoproducts in core DNA is relatively small with 254 nm irradiation, their relative contribution to photoproducts in linker DNA may be greater(28) , and therefore, could partially mask PD patterns in this region. Another benefit of photosensitization is the low absorption of 313 nm light by DNA and most proteins, which should minimize photo-induced damage of the nuclei and shielding of internal components from irradiation.

An outline of the method used to map the distribution of the PDs in nucleosomal linker DNA is shown in Fig. 1A. The irradiated nuclei were digested with micrococcal nuclease and an S2 chromatin fraction was prepared; this fraction contains most of the chromatin but is largely depleted of transcriptionally active regions (24) . The S2 chromatin fragments were trimmed with exonuclease III to remove linker remnants from their ends(25) . DNA from the trimmed chromatin fragments was labeled on the 5` end and separated by electrophoresis in a denaturing polyacrylamide gel. The region of the gel containing dinucleosomal DNA was cut into 0.8-mm slices, and each slice was counted with a scintillation counter. A broad peak of radioactivity was observed in the region corresponding in size to dinucleosomal DNA (Fig. 2). This peak was centered at a DNA length that corresponded to a nucleosome repeat length of about 194 nucleotides, very similar to the average repeat length of 195 nucleotides reported for rat liver(29) . Several smaller peaks were present within the dinucleosomal DNA peak. It is not clear whether these peaks were caused by variability in the size of the gel slices or a nonrandom distribution of linker lengths. The average length of the DNA fragments present in consecutive slices differed by about 1 nucleotide, indicating that a narrow distribution of fragment lengths was present in each slice. The distribution of PDs in the dinucleosomal DNA fragments obtained from individual slices was determined by digesting them with the 3` exonuclease of T4 DNA polymerase. Because this enzyme cannot proceed past a PD, the distribution of fragment lengths produced reflects the distribution of PDs relative to the labeled 5` end(26) . Digests of DNA obtained from seven nonconsecutive slices across the dinucleosome peak were run in a denaturing acrylamide gel to reveal the distributions of PDs in different sized dinucleosomal DNAs (Fig. 3). The nucleosome repeat lengths represented in this experiment ranged from 184 to 206 nucleotides.

Figure 1: Method for examining PD formation in nucleosomal DNA in situ. A, outline of method. S1 and P are the supernatant and pellet produced when micrococcal nuclease digested nuclei were collected by centrifugation. S2 is the supernatant that contains the chromatin fragments solubilized by suspending the digested nuclear pellet with 1 mM EDTA(24) . TT indicates a PD. See ``Experimental Procedures'' and (19) for details. B, diagram of trimmed dinucleosomal DNA fragments used to analyze the distribution of PDs in linker DNA. The size of the linker region is variable and directly related to the size of the dinucleosomal DNA; in the fragments used in this study the linker region varied from about 38 to 60 nucleotides.

Figure 2: Distribution of trimmed dinucleosomal DNA lengths. DNA from the trimmed chromatin fragments was labeled on the 5` end and run in a denaturing polyacrylamide gel. The region of the gel that contained dinucleosomal DNA was cut into 0.8-mm slices, and each slice was counted with a scintillation counter. The asterisks indicate slices used for the PD analysis shown in Fig. 3. The direction of electrophoresis was from left to right.

Figure 3: Distribution of PDs in dinucleosomal DNAs of different lengths isolated from irradiated rat liver nuclei. The distribution of PDs was determined as illustrated in Fig. 1A. DNA from nonconsecutive gel slices of the dinucleosomal DNA region were analyzed (see Fig. 2). The nucleosome repeat lengths represented in these lanes were, from left to right: 206, 201, 196, 194, 190, 186, and 184 bp. The repeat lengths were calculated by determining the difference in length between homologous PD peaks in the 3` and 5` core patterns.

As illustrated in Fig. 1B, the DNA from a trimmed dinucleosome should contain a segment of linker DNA in the middle flanked on each side by 146 nucleotides of core DNA. As expected from previous studies(18, 19) , PD formation through the core regions occurred as a series of peaks separated by approximately 10 nucleotides (Fig. 3). The series of PD peaks seen toward the top of each lane occurred in the core region toward the 3` end of the dinucleosomal DNAs, whereas the peaks toward the bottom of each lane occurred in the 5` core region; only the top half of the 5` core region is shown. The relative positions and intensities of the PD peaks in both core regions are consistent with the PD pattern previously obtained by analysis of DNA isolated from H1-containing mononucleosomes prepared from irradiated nuclei(19) . In contrast to the obvious peaks of PD formation seen in the core region, PD formation in the linker region was nearly uniform.

The positions of the PD peaks in the 5` core pattern were essentially independent of the length of the dinucleosomal DNA being analyzed, whereas the peaks in the 3` core pattern were displaced upward with the longer dinucleosomal DNAs (Fig. 3). This upward displacement is due to an increase in the size of the linker region. Thus, as expected, the linker region was the only part of the pattern that increased in length as the dinucleosome fragment length increased. This shows that the difference in the length of the original dinucleosomal DNA fragments used in these analyses were due to a difference in the length of their linker regions, not to differential trimming of their ends.

The PD analysis shown in Fig. 3used nonconsecutive gel slices of dinucleosomal DNA in order to cover a relatively large range of linker lengths. To minimize the possibility that this analysis missed linker lengths that would show PD peaks in the linker region (see ``Discussion''), PD patterns produced from DNA isolated from twelve consecutive gel slices of the dinucleosome region were examined (Fig. 4A). In this experiment, the difference in the length of the dinucleosomal DNA fragments analyzed in adjacent lanes was only about 1 nucleotide. The PD patterns in all lanes are very similar to those in Fig. 3.

Figure 4: Distribution of PDs in dinucleosomal DNAs. A, PD patterns obtained from DNA isolated from consecutive gel slices of the dinucleosomal DNA region. The range of repeat lengths in this experiment was calculated to be 196-186 bp. B, effect of gel loading on PD patterns in the linker region. Alternate lanes were overloaded to produce a relatively dark exposure of the PD pattern in the linker region.

Because PD formation is lower in the linker region than the core region, it seemed possible that peaks of PD formation in the linker DNA might be too small to see at autoradiographic exposures optimized for the peaks in the core region. Fig. 4B shows an example where lanes were overloaded relative to neighboring lanes. Despite the relatively dark exposure of the overloaded lanes, significant peaks of PD formation were not observed in the linker region.

DISCUSSION

These studies show that PD formation in the nucleosomal linker region is nearly uniform, in contrast to PD formation in the core region, which occurs with obvious peaks at approximately 10-nucleotide intervals. A photosensitizing agent, AcD, was used to produce PDs in this study in order to minimize the formation of non-PD photoproducts that have a different distribution in nucleosomal DNA than PDs. Because linker DNA may have an internal location(30, 31, 32) , the possibility that inaccessibility of linker DNA to AcD affected the pattern must be considered. However, the lower rate of PD formation in linker DNA need not reflect its inaccessibility to AcD, because a previous study, in which PDs were induced with 254 nm light, also found preferential PD formation in the core region(33) . Moreover, it has been shown that linker DNA is preferred over core DNA as a target for a variety of chemicals the size of AcD and even larger(32, 34, 35, 36, 37) . For instance, methidiumpropyl-EDTA-Fe(II), a compound larger than AcD, cleaves linker DNA at an identical rate whether the chromatin is in an open conformation or highly compacted(32) . This indicates that linker DNA in compact chromatin is fully accessible to compounds of this size. AcD also appears to freely penetrate the interior of the fiber because the pattern of PD formation in the core region of nucleosomes in situ was essentially the same whether the PDs were formed by photosensitization with AcD or irradiation with 254 nm light (19) . This shows that the side of the nucleosome core region that faces the interior of the fiber is as accessible to AcD as the sides that face the outside of fiber. Thus, the difference in the pattern of PD formation in linker and core DNA appears to reflect a structural difference between these two regions rather than their different location in the fiber.

One interpretation of the nearly uniform distribution of PDs in the linker region is that this DNA is relatively straight compared with core DNA. Evidence for linker curvature would be obscured if an equal mixture of linkers curved in left-handed and right-handed superhelices were present in the chromatin. In this case the PD peaks from the two types of linker would be out of phase and cancel each other out. Models with left-handed (3, 4) or right-handed (5) linker superhelices have been proposed. In one model it was proposed that the curvature would be left-handed for short linkers and right-handed for longer ones(5) . There is no evidence for such a transition within the range of linker lengths examined in the present study. Indeed, this study indicates that if there is a mixture of linkers curved in opposite directions, then both types of linker are present in similar amounts across the range of linker lengths examined.

Another way that linker curvature could be missed is if the curvature were in a direction that did not effect the rate of PD formation; peaks of PD formation occur where the curvature is toward the major groove. The analysis of 12 consecutive slices of dinucleosomal DNA shown in Fig. 4A covered a range of linker lengths of about 11 bases, or just slightly greater than one helical turn of DNA. If the linker DNA were organized in a fashion that resembled any of the existing solenoid models (i.e. smoothly curved in a left-handed or right-handed superhelix), evidence for such curvature should have been detected in this analysis. Because no evidence of significant linker curvature was seen, this analysis appears to be inconsistent with the existing solenoid models. The PD data cannot completely rule out the possibility that most of the linker curvature occurs as a few sharp kinks, with relatively straight DNA between. However, there is no direct evidence of such kinks in the PD data or in any other study of which I am aware.

Thus, the most plausible interpretation of the nearly uniform distribution PDs in the linker region is that this DNA is relatively straight. This interpretation is consistent with the idea that higher order chromatin structure consists of a zigzag chain of nucleosomes. A zigzag arrangement of nucleosomes has been seen by electron microscopic examination of isolated chromatin partially unfolded at low ionic strength(6, 38) . However, it has not been clear whether the zigzag arrangement is also present in fully folded chromatin. The PD data indicate that a zigzag arrangement exists in situ. This interpretation is consistent with a recent study involving three-dimensional reconstructions of sections of hypotonically swollen echinoderm sperm and chicken erythrocyte nuclei. That study found that the chromatin fiber consists of a relatively open and irregular zigzag ribbon, with the linker DNA being straight and having a generally central location(13) . The interpretation of the PD data as evidence for a relatively straight linker also appears to be consistent with photochemical dichroism studies of isolated chromatin fragments, which indicate that linker DNA does not continue the superhelical path of the DNA in the core region(14) . In contrast to the present results, recent studies of purified dinucleosomes found evidence for an ionic strength-dependent bending of linker DNA(15) , even in the absence of histone H1(16) . However, those recent studies appear to contradict three previous studies that found no evidence for an ionic strength dependent change in the conformation of dinucleosomes (39, 40, 41) . In addition, it is not clear whether dinucleosomes are an adequate model for the chromatin fiber because the formation of the chromatin fiber may involve interactions that are not possible in isolated dinucleosomes.

There are several models of higher order chromatin structure based on a zigzag arrangement of nucleosomes. The PD data alone do not differentiate most of these models. Some zigzag models can be distinguished from others by the location of linker DNA and H1, and recent studies indicate that both have an internal location in isolated chromatin fibers(30, 31, 32) . Of the published zigzag models, those of Williams et. al.(9) , Bordas et. al.(12) , and Staynov (7) have H1 and linker DNA located internally. In these models the linker DNA crisscrosses through the middle of the fiber linking nucleosome cores that are on nearly opposite sides of the fiber. However, these models have highly regular structures, and microscopic examination of chromatin indicates that its structure is not highly regular(13, 32, 42) . Such irregularity is not surprising in view of the heterogeneity in linker length (25) and histone subtypes(43, 44) . A model of the chromatin fiber that incorporates heterogeneous linker lengths and is based on a zigzag chain of nucleosomes has been proposed (45) . This model consists of a relatively open zigzag ribbon, similar to that seen in the recent study cited above(13) . The openness of the structures observed in that study may be due to a partial unfolding of the fibers by the hypotonic swelling that was needed to visualize individual fibers. If this is the case, the observed structures would not seem incompatible with an irregular version of the crossed linker models mentioned above.

The sensitivity of PD formation to DNA curvature and its applicability to the study of intact chromatin without introducing DNA breaks make it a valuable tool for examining chromatin structure in situ. However, some limitations of this approach should be noted. First, the method works best with mixed populations of DNA sequences, where adjacent pyrimidines occur with nearly equal frequency at all positions. Information about nucleosomes on specific DNA sequences are limited to sites of adjacent pyrimidines. Also, it cannot always be assumed that DNA curvature will be the primary factor that determines the pattern of PD formation. For example, proteins that interact directly with DNA bases, such as transcription factors, would be expected to affect PD formation in other ways(46, 47) . Even with these limitations, studies utilizing PD formation have provided valuable information about the conformation of core (18, 19) and linker DNA (19 and present study), and the location of H1 (19) in intact chromatin. Ultimately, comparison of results obtained by several independent methods will be required to understand the structure of chromatin in its diverse functional states.

FOOTNOTES

*

This work was supported by Grants GM-24019, GM-49351, and CA 06927 from the National Institutes of Health. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore by hereby marked ``advertisement'' in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

§

To whom correspondence should be addressed. Tel.: 215-898-0454; Fax: 215-898-9923.

()

The abbreviations used are: bp, base pair(s); PD, pyrimidine dimer; AcD, N-(m-acetylbenzyl)-N,N-dimethylethylenediammonium dichloride.

ACKNOWLEDGEMENTS

REFERENCES

1. Richmond, T. J., Finch, J. T., Rushton, B., Rhodes, D., and Klug, A. (1984) Nature 311,532-537 [CrossRef][Medline] [Order article via Infotrieve]
2. van Holde, K. E. (1988) Chromatin , pp. 291-305 and 334-341, Springer-Verlag New York Inc., New York
3. Finch, J. T., and Klug, A. (1976) Proc. Natl. Acad. Sci. U. S. A. 73,1897-1901 [Abstract/Free Full Text]
4. McGhee, J. D., Nickol, J. M., Felsenfeld, G., and Rao, D. C. (1983) Cell 33,831-841 [CrossRef][Medline] [Order article via Infotrieve]
5. Butler, P. J. G. (1984) EMBO J. 3,2599-2604 [Medline] [Order article via Infotrieve]
6. Worcel, A., Strogatz, S., and Riley, D. (1981) Proc. Natl. Acad. Sci. U. S. A. 78,1461-1465 [Abstract/Free Full Text]
7. Staynov, D. Z. (1983) Int. J. Biol. Macromol. 5,3-9
8. Woodcock, C. L. F., Frado, L.-L. Y., and Rattner, J. B. (1984) J. Cell Biol. 99,42-52 [Abstract/Free Full Text]
9. Williams, S. P., Athey, B. D., Muglia, L. J., Schappe, R. S., Gough, A. H., and Langmore, J. P. (1986) Biophys. J. 49,233-248 [Medline] [Order article via Infotrieve]
10. Subirana, J. A., Muñoz-Guerra, S., Aymami, J., Radermacher, M., and Frank, J. (1985) Chromosoma (Berl.) 91,377-390 [CrossRef][Medline] [Order article via Infotrieve]
11. Burgoyne, L. A. (1985) Cytobios 43,141-147 [Medline] [Order article via Infotrieve]
12. Bordas, J., Perez-Grau, L., Koch, M. H. J., Vega, M. C., and Nave, C. (1986) Eur. Biophys J. 13,175-185 [CrossRef][Medline] [Order article via Infotrieve]
13. Horowitz, R. A., Agard, D. A., Sedat, J. W., and Woodcock, C. L. (1994) J. Cell Biol. 125,1-10 [Abstract/Free Full Text]
14. Sen, D., Mitra, S., and Crothers, D. M. (1986) Biochemistry 25,3441-3447 [CrossRef][Medline] [Order article via Infotrieve]
15. Yao, J., Lowary, P. T., and Widom, J. (1990) Proc. Natl. Acad. Sci. U. S. A. 87,7603-7607 [Abstract/Free Full Text]
16. Yao, J., Lowary, P. T., and Widom, J. (1991) Biochemistry 30,8408-8414 [CrossRef][Medline] [Order article via Infotrieve]
17. Woodcock, C. L. (1994) J. Cell Biol. 125,11-19 [Abstract/Free Full Text]
18. Gale, J. M., Nissen, K. A., and Smerdon, M. J. (1987) Proc. Natl. Acad. Sci. U. S. A. 84,6644-6648 [Abstract/Free Full Text]
19. Pehrson, J. R. (1989) Proc. Natl. Acad. Sci. U. S. A. 86,9149-9153 [Abstract/Free Full Text]
20. Pearlman, D. A., Holbrook, S. R., Pirkle, D. H., and Kim, S. (1985) Science 227,1304-1308 [Abstract/Free Full Text]
21. Pehrson, J. R., and Cohen, L. H. (1992) Nucleic Acids Res. 20,1321-1324 [Abstract/Free Full Text]
22. Brown, D. W., Libertini, L. J., Suquet, C., Small, E. W., and Smerdon, M. J. (1993) Biochemistry 32,10527-10531 [CrossRef][Medline] [Order article via Infotrieve]
23. Suquet, C., and Smerdon, M. J. (1993) J. Biol. Chem. 268,23755-23757 [Abstract/Free Full Text]
24. Rose, S. M., and Garrard, W. T. (1984) J. Biol. Chem. 259,8534-8544 [Abstract/Free Full Text]
25. Strauss, F., and Prunell, A. (1982) Nucleic Acids Res. 10,2275-2292 [Abstract/Free Full Text]
26. Doetsch, P. W., Chan, G. L., and Haseltine, W. A. (1985) Nucleic Acids Res. 13,3285-3304 [Abstract/Free Full Text]
27. Meistrich, M. L., and Lamola, A. A. (1972) J. Mol. Biol. 66,83-95 [CrossRef][Medline] [Order article via Infotrieve]
28. Gale, J. M., and Smerdon, M. J. (1990) Photochem. Photobiol. 51,411-417 [CrossRef][Medline] [Order article via Infotrieve]
29. Thomas, J. O., and Thompson, R. J. (1977) Cell 10,633-640 [CrossRef][Medline] [Order article via Infotrieve]
30. Smith, M. F., Athey, B. D., Williams, S. P., and Langmore, J. P. (1990) J. Cell Biol. 110,245-254 [Abstract/Free Full Text]
31. Graziano, V., Gerchman, S. E., Schneider, D. K., and Ramakrishnan, V. (1994) Nature 368,351-354 [CrossRef][Medline] [Order article via Infotrieve]
32. Zlatanova, J., Leuba, S. H., Yang, G., Bustamante, C., and van Holde, K. (1994) Proc. Natl. Acad. Sci. U. S. A. 91,5277-5280 [Abstract/Free Full Text]
33. Snapka, R. M., and Linn, S. (1981) Biochemistry 20,68-72 [CrossRef][Medline] [Order article via Infotrieve]
34. Cech, T., and Pardue, M. L. (1977) Cell 11,631-640 [CrossRef][Medline] [Order article via Infotrieve]
35. Jahn, C. L., and Litman, G. W. (1979) Biochemistry 18,1442-1449 [CrossRef][Medline] [Order article via Infotrieve]
36. Metzger, G., Wilhelm, F., and Wilhelm, M. (1977) Biochem. Biophys. Res. Commun. 75,703-710 [CrossRef][Medline] [Order article via Infotrieve]
37. Cartwright, I. L., and Elgin, S. C. R. (1989) Methods Enzymol. 170,359-369 [Medline] [Order article via Infotrieve]
38. Thoma, F., Koller, T., and Klug, A. (1979) J. Cell Biol. 83,403-427 [Abstract/Free Full Text]
39. Butler, P. J. G., and Thomas, J. O. (1980) J. Mol. Biol. 140,505-529 [CrossRef][Medline] [Order article via Infotrieve]
40. Osipova, T. N., Pospelov, V. A., Svetlikova, S. B., and Vorob'ev, V. I. (1980) Eur. J. Biochem. 113,183-188 [CrossRef][Medline] [Order article via Infotrieve]
41. Marion, C., Bezot, P., Hesse-Bezot, C., Roux, B., and Bernengo, J. C. (1981) Eur. J. Biochem. 120,169-176 [Medline] [Order article via Infotrieve]
42. Rattner, J. B., and Hamkalo, B. A. (1979) J. Cell Biol. 81,453-457 [Abstract/Free Full Text]
43. Cohen, L. H., Newrock, K. M., and Zweidler, A. (1975) Science 190,994-997 [Abstract/Free Full Text]
44. Cole, R. D. (1984) Anal. Biochem. 136,24-30 [CrossRef][Medline] [Order article via Infotrieve]
45. Woodcock, C. L., Grigoryev, S. A., Horowitz, R. A., and Whitaker, N. (1993) Proc. Natl. Acad. Sci. U. S. A. 90,9021-9025 [Abstract/Free Full Text]
46. Becker, M. M., and Wang, J. C. (1984) Nature 309,682-687 [CrossRef][Medline] [Order article via Infotrieve]
47. Selleck, S. B., and Majors, J. (1987) Nature 325,173-177 [CrossRef][Medline] [Order article via Infotrieve]

CiteULike

Complore

Connotea

Del.icio.us

Digg

Reddit

Technorati

This article has been cited by other articles:

				M. Engelhardt Choreography for nucleosomes: the conformational freedom of the nucleosomal filament and its limitations Nucleic Acids Res., August 17, 2007; (2007) gkm560v1. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) Submit a Letter to Editor Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Pehrson, J. R. Search for Related Content PubMed PubMed Citation Articles by Pehrson, J. R. Social Bookmarking                      What's this?