nation for the reported blocking effect of GnRH on various estrogen-dependent processes, including ovulation, ovum transport, ovum implantation, and mam- mary tumorigenesis. Moreover, the in- hibitory effect of GnRH and agonists on FSH-induced progesterone production in cultured granulosa cells could offer a possible explanation for the interference by GnRH and its agonists in such a pro- gesterone-dependent process as preg- nancy. Our results, however, do not ex- clude the possibility that administration in vivo of pharmacological doses of GnRH in intact animals may also cause imbalances in pituitary gonadotropin production that also result in the inhibi- tion of various reproductive functions (7). The mechanisms by which GnRH ex- erts inhibition on ovarian granulosa cells are not known. Although GnRH recep- tors have been identified in the anterior pituitary by radiotracer binding tech- nique and by immunocytochemical methods (8), their tissue specificity has not been demonstrated. Bernardo et al. (9) have reported putative GnRH binding sites in the mouse adrenal cells. It is pos- sible that ovarian granulosa cells may have GnRH receptors and that the inhib- itory effect of GnRH that we observed is mediated through hormone-receptor in- teractions. A. J. W. HSUEH G. F. ERICKSON Department ofReproductive Medicine, University of California, San Diego, La Jolla 92093 References and Notes 1. A. V. Schaily, Science 202, 18 (1978); R. Guile- min, ibid., p. 390; S. S. C. Yen, B. L. Lasley, C. F. Wang, H. Leblanc, T. M. Siler, Recent Prog. Horm. Res. 31, 321 (1975). 2. A. Corbin, C. W. Beattie, J. Tracy, R. Jones, T. J. Foell, J. Yardley, R. W. A. Rees, Int. J. Fer- til. 23, 81 (1978); Y. C. Lin and K. Yoshinaga, Program, 58th annual meeting of the Endocrine Society, San Francisco, Calif. (1976), p. 143; E. Johnson, R. Gendrich, W. White, Fertil. Steril. 27, 853 (1976); E. S. Johnson, J. H. Seely, W. F. White, E. R. DeSombre, Science 194, 329 (1976). 3. J. H. Dorrington, Y. S. Moon, D. T. Armstrong, Endocrinology 97, 1328 (1975); G. F. Erickson and A. J. W. Hsueh, ibid. 102, 1275 (1978); A. Nimrod and H. R. Lindner, Mol. Cell. Endo- crinol. 5, 315 (1976); D. W. Schomberg, R. L. Stouffer, L. Tyrey, Biochem. Biophys. Res. Commun. 68, 77 (1976); D. T. Armstrong and J. H. Dorrington, Endocrinology 99, 1411 (1976); A. W. Lucky, J. R. Schreiber, S. G. Hiflier, J. D. Schulman, G. T. Ross, ibid. 100, 128 (1977). 4. G. F. Erickson and A. J. W. Hsueh, Endocrinol- ogy 102, 1275 (1978). 5. Abbreviations are Leu, leucine; Gly, glycine; Try, tryptophan; Pro, proline; NHEt, ethyl- amide; and Me, methyl. 6. The potencies of the GnRH analog I and analog 2 relative to native GnRH, as determined by their ability to cause luteinizing hormone release in cultured pituitary cells, are approximately 6- and 144-fold, respectively. 7. P. E. Belchetz, T. M. Plant, Y. Nakai, E. J. Keogh, E. Knobil, Science 202, 631 (1978). 8. G. Grant, W. Vale, J. Rivier, Biochem. Biophys. Res. Commun. 50, 771 (1973); L. A. Sternberger SCIENCE, VOL. 204, 25 MAY 1979 and J. P. Petrali, Cell Tissue Res. 162, 141 technical assistance; Dr. S. S. C. Yen for his in- (1975); J. Spona, FEBS Lett. 34, 24 (1973); E. terest; Drs. N. Ling and J. Rivier for providing Pendoze, J. A. Vilchez-Martinez, J. Fishback, the GnRH analogs; and Dr. H. R. Papkoff and A. Arimura, A. V. Schally, Biochem. Biophys. the NIAMDD Pituitary Hormone Distribution Res. Commun. 79, 234 (1977). Program for providing FSH preparations. Sup- 9. L. A. Bernardo, J. P. Petrali, L. P. Weiss, L. A. ported by NIH grant 1-ROI-CA 21867 and Rock- Sternberger, J. Histochem. Cytochem. 26, 613 efeller Foundation grant RF-75029. (1978). 10. We thank L. Tucker and C. Fabics for their 14 December 1978; revised 22 February 1979 Periodicity of Deoxyribonuclease I Digestion of Chromatin Abstract. Two methods have been used to measure the single-strand lengths of the DNA fragments produced by deoxyribonuclease I digestion of chromatin. The aver- age lengths obtained are multiples of about 10.4 bases, significantly different from the value of 10 previously reported. This periodicity in fragment lengths is closely related to the periodicity of the DNA double helix in chromatin, but the two values need not be exactly the same. The first level of condensation of DNA in chromatin is brought about by its in- teraction with histones to form nucle- osomes, the elementary subunits of the structure. For a detailed understanding of how the DNA is folded, it is necessary to know both the path of the double helix and its periodicity, or the number of base pairs per turn, in the nucleosome. X-ray crystallographic studies indicate a path in which the DNA is wrapped twice around the histones (1). This appears to Bases -95 -64 Bases 95 conflict with measurements on closed, circular DNA extracted from SV 40 chromatin showing nearer one super- helical turn (2). The conflict can be re- solved by postulating a change in the pe- riodicity of the DNA as it is folded into a nucleosome (1). A decrease of only about 5 percent is required, and so be- fore firm conclusions can be drawn it is necessary to know the relevant parame- ters as accurately as possible. Noll (3) has suggested that the perio- 42 40 100 - 31- 31 27~~~~~~~2 21 2 -17 1 .0 DNase I DNase I Markers DNase I DNase I Markers *t 80 + + (/. Markers Markers Fig. 1 (left). Comparison of DNase I and marker fragments in polyacrylamide gels. 70 DNase I fragments were prepared by DNase I digestion of rat liver nuclei and extraction of l the DNA (3). Marker fragments were pre- pared as described in Table 1. All fragments 60 / were labeled with 32P as follows. DNA (40 ,ug/ ml) in 50 mM sodium acetate (pH 4.6) and I E 1 mM EDTA was boiled for I minute and treated with spleen acid phosphomonoester- Migration ase B (16) (0.1 unit per milliliter) for 2 hours at 37°C. The pH was then raised to 8.0 by the addition of tris base to 85 mM and the mixture was boiled for 1 minute, supplemented with MgCl2 (10 mM), 2-mercaptoethanol (15 miM), and y-32P-labeled ATP (tenfold excess over 5'-OH termini), and treated with polynucleotide kinase (20 unit/ml) (New England Biolabs) for 1 hour at 37°C. The labeled DNA was purified by filtration through Sephadex G-25 in a mixture of 0. IM NaCl, 10 mM tris-HCl (pH 7.5), and I mM EDTA and subjected to electrophoresis in 12 per- cent (left panel) and 20 percent (right panel) polyacrylamide-98 percent formamide gels (17). Negatives of autoradiograms of the gels are shown. Fig. 2 (right). Size determination of DNase I bands. Densitometer traces of lanes in left panel of Fig. 1 containing DNase I (bottom) and marker fragments were aligned. The sizes and distances of migration of the marker frag- ments (filled circles) were used to construct a calibration curve from which the sizes of the DNase I fragments were derived. The distances of migration of the DNase I fragments (solid vertical lines) are contrasted with those expected if the fragments were multiples of 10.0 bases (dashed vertical lines). 0036-8075/79/0525-0855$00.50/0 Copyright © 1979 AAAS 855