Proc. Nati. Acad. Sci. USA
Vol. 73, No. 6, pp. 2038-2042, June 1976
Cell Biology
Locations of chromosomal proteins in polytene chromosomes* (chromosome structure/histone antibodies/immunofluorescence detection of a nonhistone chromosomal protein/Drosophila chromosomes)
CANDIDO RODRIGUEZ ALFAGEME, GEORGE T. RUDKIN, AND LEONARD H. COHEN The Institute for Cancer Research, Fox Chase Cancer Center, Philadelphia, Pennsylvania 19111
Communicated by Thomas F. Anderson, March 23,1976
ABSTRACT DI, a nonhistone chromosomal protein rich in both basic and acidic amino acids, has been localized at a limited number of specific loci in polytene chromosomes of Drosophila melanogaster. H2B, a nucleosomal histone, and Hi, a nonnucleosomal histone, are both found throughout most chromosomal regions. Study of the role of proteins in gene function in eukaryotes has been hampered by the complexity of chromatin. Thus, whereas many proteins have been detected in purified chromatin preparations, none of them has yet been shown to be associated with specific genetic loci. We have now been able to localize by immunofluorescence a nonhistone protein in the polytene chromosomes of Drosophila. A major difficulty has been that conventional methods for spreading polytene chromosomes in acetic acid result in the extensive loss (1-3) and possible translocation of chromosomal proteins, particularly basic proteins. We have devised a method for obtaining, with minimal loss of the associated proteins, chromosome spreads capable of reacting with specificity with antibodies against purified chromosomal proteins. The recent isolation of such proteins in our laboratory (4, t) has permitted us to visualize Di, a Drosophila nonhistone protein, as well as two Drosophila histones, Hi and H2B. DI is a protein, obtained from Drosophila chromatin, that is rich in both acidic and basic amino acids (24% and 21%, respectively)t. It may belong to a class of proteins of general occurrence among eukaryotes (5-8). It shares many properties with a group of proteins of calf-thymus chromatin known as the high mobility group (5), notably, solubility in 5% perchloric acid, extractability from chromatin by 0.35 M NaCl, and high content of basic and acidic amino acids. However, it differs from those proteins of the high mobility group thus far isolated in having a higher molecular weight and in having aspartic acid rather than glutamic acid as its most abundant amino acid. The amount of Di in embryo chromatin is 2-5% that of histone Hi on a molar basis, and the amount in salivary gland chromatin, while not yet accurately determined, is not more than that. The possibility existed, therefore, that Di might be confined to a proportionately small fraction of the genome. MATERIALS AND METHODS Protein Di and histones Hi and H2B were isolated from 4 to This paper is dedicated to the memory of the late Jack Schultz, whose understanding of chromosomes triggered the collaborative effort involved. Abbreviations: D1, Drosophila protein 1; phosphate-buffered saline, 0.14 M NaCI, 10 mM sodium phosphate, pH 7.3. The histone nomenclature used is that adopted at the Ciba Foundation symposium on structure and function of chromatin, 1975, in which H1 = F1, H2A = F2a2, H2B = F2b, H3 = F3, and H4 = F2al. * A preliminary report was presented at the 15th Annual Meeting of the American Society for Cell Biology, November 11-14, 1975, San Juan, Puerto Rico [(1975) J. Cell Biol. 67, 6a]. t Manuscript in preparation.
20-hr embryos of Drosophila melanogaster Oregon R as described elsewhere (4, t). Protein DI was obtained from embryo chromatin together with histone Hi by acid extraction followed by precipitation of other proteins by 5% perchloric acid. It was separated from Hi by preparative gel electrophoresis. This isolation yields a single electrophoretic component in acid urea gels (Fig. 1) and in sodium dodecyl sulfate gelst. Preparation of Sera Against DI, HI, and H2B. Rabbits were immunized (9) with a suspension of protein (83 ,g of Di, or 240 ,gg of HI, or 185 .sg of H2B) and yeast RNA (except in the case of HI) in 1 ml of phosphate-buffered saline (0.14 M NaCl, 10 mM sodium phosphate pH 7.3) and 1.5 ml of complete Freund's adjuvant. Yeast RNA type VI was obtained from Sigma Chemical Co. This inoculum was mixed for 5 min with a Vortex mixer and injected subcutaneously in the back at multiple sites. Booster shots were given 2, 7, and 9 weeks later (the last two injections with half the amount of protein-RNA complexes) and sera obtained 1 week after the last inoculation. Immunodiffusion Analysis of Rabbit Sera. Double immunodiffusion (10) on microscope slides (I1) was performed in 1.5% agarose gels containing 0.14 M NaCl, 2 mM MgCl2, 0.1 mM CaCl2, and sodium 5,5'-diethylbarbiturate at pH 7.3 (12). Agar is unsuitable for this purpose because it contains a component that precipitates histones. The antiserum against Di showed little or no crossreactivity with the histones (Fig. 2). Antisera prepared against Hi and H2B showed no crossreactivity with H2B and HI, respectively, or with DI, and little or no crossreactivity with other histones. Absorption of Immune Sera. Antisera were treated with appropriate histone-RNA complexes to remove crossreacting antibodies. These histone-RNA complexes (1:1) were prepared by precipitating the histone mixtures (dissolved in phosphatebuffered saline) with yeast RNA type VI. The precipitates were recovered by centrifugation (10,000 rpm, 10 min), washed and resuspended in phosphate-buffered saline, and added to whole serum in the ratio of 1 iug of each histone-RNA complex per gl of serum. After incubation at 40 for 18 hr the absorbed sera were centrifuged at 10,000 rpm for 10 min and the supernatants were used for the assays. Preparation of Chromosomes for Antibody Binding. D. melanogaster salivary glands were dissected from giant (gt) larvae obtained as daughters of a cross between females homozygous for the gt mutation and males carrying an X chromosome deficient for the gt locus [Df(1)62g'8] (courtesy of T. C. Kaufman). A D. virilis wild-type stock was obtained from S. K. Majumdar. Salivary glands were treated with 2% formaldehyde (in 50 mM Na glycerophosphate, 25 mM sucrose, 1 mM MgCl2, pH 6.8, for 2 min at 4°-8°) to prevent the loss of chromosomal proteins during subsequent treatment; the glands were then transferred to a drop of 2% formaldehyde in 5% acetic acid, were homogenized lightly either with a Dounce type microhomogenizer or by forcing through a micropipette, and were finally squashed between a microscope slide and a siliconed cover slip. The cover slip was removed by immersing 2038
Cell Biology:
Alfageme et al.
D1 H
FIG. 1 (left). Purity of D. melanogaster D1 protein. Left gel: D1 isolated by preparative gel electrophoresis of 5% perchloric acid-soluble nuclear protein fraction. Right gel: The 5% perchloric acid-soluble fraction, shown for comparison. Electrophoresis was in acidurea-polyacrylamide gels (5% acetic acid, 4.2 M urea). Gels were stained with amido black. FIG. 2 (right). Immunodiffusion analysis of antiserum against D. melanogaster D1. The center well contained 4 jAl of whole, unabsorbed antiserum against D1 prepared as described in Materials and Methods. The other wells contained 3jug in 3,gl of the protein as indicated. A single and strong precipitin line is formed between D1 and antiserum against D1. A weak reaction can be seen with H1. These two precipitin lines cross each other, indicating lack of identity between the antigenic determinants (10). The precipitate around the well containing H4 is caused by a serum component other than antibodies against D1 because it is also formed when this histone is tested with nonimmune rabbit serum.
the slide in phosphate-buffered saline. The conditions chosen for the fixation of chromosomes are based on data of Brutlag et al. (13) and Chalkley and Hunter (14). Binding of Antibodies to Polytene Chromosomes and Visualization of Bound Antibodies. Sera absorbed as described above were diluted with phosphate-buffered saline (ko for DI, '/40 for either HI or H2B) and applied to the squashes, which were then covered with cover slips. Incubation in a humid chamber for I hr was followed by extensive washing in phosphate-buffered saline (1 hr at 220 and 18 hr at 40) to eliminate weakly bound antibodies. The bound rabbit antibodies were detected with fluorescein isothiocyanate conjugated antibody against rabbit IgG prepared in goat (A280/A495 = 1.7, purchased from Miles-Yeda Ltd.) diluted 1ko with phosphatebuffered saline. After incubation for 30 min at 220 in the humid chamber, the chromosome preparations were washed in phosphate-buffered saline for 1 hr at 220 and mounted in buffered glycerol (Becton, Dickinson and Company). A Zeiss fluorescence microscope equipped with epiillumination was used. To avoid fading of fluorescence, we searched and focussed in the phase contrast mode using a tungsten light source. Then a fluorescence photomicrograph was taken by a 15-sec exposure on Kodak Tri-X-pan film and developed in Diafine (Acufine, Inc., Chicago, Ill. 60611) at ASA 1600. Light source: Osram HBO-200 W, primary filters: BG 38, SP500; secondary filter: 50; dichroic mirror: F1 (Zeiss). RESULTS The procedure adopted for visualization of proteins in polytene chromosomes involves three steps. First, the salivary glands are fixed in formaldehyde, homogenized lightly in acetic acidformaldehyde to disperse the cytoplasm, and squashed. The aldehyde fixation is essential since many chromosomal proteins are known to be removed by the acid conditions needed for obtaining cytologically suitable preparations (1-3). Second, the
Proc. Natl. Acad. Sci. USA 73 (1976)
2039
rabbit antiserum is treated with a mixture of appropriate histones, as described in Materials and Methods, and applied to the slide with the chromosomes. Finally, the chromosomal proteins are visualized indirectly (15) by labeling the bound rabbit antibodies with fluorescein-labeled antibody against rabbit IgG prepared in goat.
The patterns of fluorescence obtained with D. melanogaster salivary gland chromosomes indirectly stained with antisera against D1, HI, and H2B are shown in Fig. 3 (left panel). The contrast between the staining by antisera against the histones and by the antiserum against DI is apparent. Antibodies against. Hi and H2B are distributed quite generally throughout all the chromosomes, whereas chromosomes stained with antibody against Dl have a few very intensely fluorescent regions in the vicinity of the chromocenter. In addition, there are other chromosomal regions that fluoresce prominently but much less strongly, such as the distal ends of some chromosomes and a few other regions along every chromosome. The less intensely fluorescing regions have not been studied in detail. The results obtained with D. virilis chromosomes have a bearing on the specificity of staining of the less intensely fluorescing regions. Dl is species-specific in that D. melanogaster Di differs considerably in electrophoretic mobility from D. virilis DIt. It is therefore of interest that the antiserum prepared against D. melanogaster Di failed to bind to D. virilis chromosomes in contrast to the antisera against D. melanogaster Hi and H2B (Fig. 3, right panel). This indicates a high degree of specificity of the antiserum against DI, suggesting that even the low-level staining, at many sites, of D. melanogaster chromosomes is due to the presence at these sites of DI or of proteins very closely related to it. Results of control experiments to assess the immunological specificity of staining are shown in Table 1. The use of antiserum against H2B that had been treated with H2B-RNA complexes (0.5 ,tg of H2B per gl of serum) did not yield fluorescence; in contrast, treatment of antiserum against H2B with histones Hi, H2A, H3, and H4 complexed with RNA (0.5 ,ug of each histone per gl of serum) had no effect on the fluorescence. Similarly, no blocking of antiserum against DI was obtained by treatment with a mixture of all histones, nor of antiserum against HI by treatment with RNA complexed with H2A, H2B, H3, and H4. These results indicate that the fluorescence response in the chromosomes is specific for the antigen. The chromosomal regions stained intensely by antiserum against Di are seen at higher resolution in Fig. 4. In the case of visualization with antiserum against Di, two intensely fluorescent regions can be seen at the base of 3R and two in chromosome 4. This pattern is reproducible and was observed in salivary gland chromosomes from late third instar larvae and from prepupae. In contrast, antisera against HI and H2B stain all the bands resolvable with phase contrast optics. The general distribution of H2B throughout the chromosomes supports the current view that this histone is part of an oligomeric unit associated with all of the DNA (16-18). On the other hand, Hi is not considered to be part of the oligomer and no indication of its distribution has been available. Our method shows HI to be distributed throughout every chromosomal region, although not necessarily in constant proportion to H2B. The ubiquitous staining of chromosomal regions by antibody against H2B is a good test of the accessibility of proteins in these chromosome preparations, since H2B in chromatin is much less accessible to antibodies than is Hi (19) and is also more difficult to extract than either Dit or HI (20-22). It seems likely that the acid conditions used in spreading the chromosomes increases the accessibility of the proteins. The possibility still remains that some sites are not available to antibodies.
2040
Cell Biology: Alfageme et al.
Proc. Natl. Acad. Sci. USA 73 (1976)
'~~~~~~~~~~~~~I
l.l.....=. ...
.-
~
FIG. 3. Polytene chromosomes from D. melanogaster (left) and D. virilis (right) indirectly stained with antisera againstD. melanogaster D1, H1, or H2B. Anti-D1 was absorbed with H1, H2A, H2B, H3, and H4; anti-Hi was absorbed with H2A, H2B, H3, and H4; anti-H2B was absorbed with H1, H2A, H3, and H4 (see Materials and Methods and Table 1). Prior to use, the antisera were diluted with phosphate-buffered saline, 'o for anti-D1 and /40 for either anti-H1 or anti-H2B. The inset is a phase contrast image of the D. virilis nucleus, which was in the microscope field for the upper right fluorescence photomicrograph. The scales are 10 ,jm.
Chromosomes fixed under a variety of conditions (1-5% formaldehyde for 1-20 min at 4°-22°) show essentially identical fluorescent patterns when stained with antibody against D1, thus suggesting that the fixation procedure adopted is sufficient to prevent significant loss of protein Di. DISCUSSION The method presented here for the preparation of polytene chromosome spreads for immunofluorescence is a compromise in that the formaldehyde treatment used to prevent loss of proteins renders the chromosomes difficult to spread and thus hampers identification of some fluorescent regions. Nevertheless, it has permitted us to identify the regions that fluoresce most brightly with antibody against D1 as 81F and 83C-E in chromosome 3R and two regions in chromosome 4 [Bridges standard map (23)]. The marked difference in intensity between those few fluorescent sites near the chromocenter and
the faintly fluorescent sites scattered through the genome is most simply interpreted as due to a difference in D1 content between these two classes of sites. This interpretation is supported by the results reported here, although the accessibility of the protein for reaction with the antibody may still be a factor worthy of investigation. If we interpret subjective fluorescence intensity as an indicator of the relative concentration of the primary antigen, it would appear that a disproportionately large fraction of the D1 is localized in a few short regions of the genome, with the remainder widely distributed over all the chromosomes. The restricted distribution of D1 implies that the Dl-containing regions share structural features capable of recognizing this protein. Common structural features might conceivably arise from any chromosomal macromolecule; if they arise from DNA, one possibility is that they arise from repetitive DNA. This class of DNA has been detected by hybridization in situ
Cell Biology:
Alfageme et al.
Proc. Natl. Acad. Sci. USA 73 (1976)
2041
Table 1. Effect of various serum treatments on the immunofluorescent staining of D. melanogaster polytene chromosomes Chromo-
41. .
j
.#
.
VP
J
Serum
Treatment of serum*
Nonimmune None Anti-H2B Anti-H2B Anti-H2B
None None Yeast RNA H2B and yeast RNA H4, H3, H2A, H1, and yeast RNA H4, H3, H2A, H2B, H1, and yeast RNA H4, H3, H2A, H2B, and yeast RNA
1.
./
Anti-D1 Anti-H1
some fluorescencet
+ +
+ +
* Antisera were absorbed with mixtures of the proteins and RNA listed (see Materials and Methods), diluted 1/20 with phosphatebuffered saline, and then used in the indirect method for fluorescence staining of chromosomes. t A minus score indicates that fluorescence was not detected in the chromosomes or that it was hardly visible and insufficient to produce a photographic image under standard exposjure conditions (see Materials and Methods). B
FIG. 4. Proximal ends of D. melanogaster salivar gland chromosomes 3R indirectly stained with antisera against D1, H1, and H2B. Preparations were stained as described in Materials and Methods with absorbed antisera diluted with phosphate-buffered saline Y40 for either H2B or H1 and 'Ao for D1. Fluorescent granules (A) such as those in the top photograph are occasionally seen. The fluorescence from these granules can be removed by washing the stained preparations with 5% acetic acid. This treatment does not affect the fluorescence intensity of chromosomes. The arrows point to subsections 81F (right) and 83 C-E (left). Chromosome 4 can be clearly seen at the extreme right of the top pair of photomicrographs. The scale is 10 gm.
at chromocentral regions and at many other loci interspersed along the chromosomes (24-30). Precise localization of such DNA sequences might help in determining whether DI is associated with specific nucleotide sequences.
Studies of several investigators on the staining of chromosomes by quinacrine and Hoechst 33258 (31-39) provide clues
to the chemical nature of the principal binding regions for antibody against D1 (Fig. 4). Salivary gland chromosomes of D. melanogaster stained with quinacrine are reported to exhibit brightest fluorescence at subsections 8WF and 83D in chromosome 3, and 101 and 102D in chromosome 4 [these regions, with the exception of 83D, have been shown to be Hoechst bright as well (39)].* The quinacrine bright regions coincide with the principal binding sites for antibody against DI. The fluorescence of the dyes is enhanced by adenylate- and thymidylaterich DNA (35, 36, 38), although, as Holmquist has pointed out, adenylate and thymidylate richness is not alone a sufficient property to account for fluorochrome brightness (39). These observations suggest that nucleotide sequences rich in adenylate and thymidylate might be a property of D1 sites. An answer to this question is important, because identification of proteins associated with repetitive DNA sequences would be valuable for studies of the functional role, or roles, of this class of DNA, a class that may be involved in gene regulation (40). The method described here§ permits localization of proteins at the level of resolution of the chromomere. The application of this method to other nonhistone proteins, to histone variants (4, 41, 42), and to modified histones should make accessible detailed knowledge of the organization of proteins in the eukaryotic genome. We thank Dr. M. Bayer for the use of the fluorescence microscope; Drs. M. Bosma, S. Hausman and I. Millman for their help and advice in the preparation and analysis of antibodies; Drs. B. Mintz, R. Perry and K. Tartof for their helpful review of the manuscript; and Mrs. Mara Bard, Mrs. S. Bianchi-Ellquist, Mr. R. Rudkin, and Miss D. Hazler for * Note Added in Proof. Precise coincidence of the principal DI sites (81, 83, 101, and 102) with the quinacrine bright sites has been ver-
ified by double-label experiments. § After this manuscript was completed, Silver and Elgin reported (43) the use of antibodies against preparations of total nonhistone chromosomal proteins to stain Drosophila polytene chromosomes by
methods similar to those described here.
2042
Cell Biology: Alfageme et al.
excellent technical assistance. This study was supported by U.S. Public Health Service Grants CA-12544, CA-06927, and RR-05539 from the National Institutes of Health, a grant from the National Science Foundation (BO-41476), and an appropriation from the Commonwealth of Pennsylvania.
1. Swift, H. (1964) "The histones of polytene chromosomes," in The Nucleohistones, eds. Bonner, J. & Ts'o, P. (Holden-Day, Inc., San Francisco), pp. 169-183. 2. Murray, K. (1969) "Stepwise removal of histones from native deoxyribonucleoprotein by titration with acid and some properties of the resulting partial nucleoproteins," J. Mol. Biol. 39, 125-144. 3. Brody, T. (1974) "Histones in cytological preparations," Exp. Cell Res. 85,255-263. 4. Alfageme, C. R., Zweidler, A., Mahowald, A. & Cohen, L. H. (1974) "Histones of Drosophila embryos: Isolation and structural studies," J. Biol. Chem. 249,3729-3736. 5. Goodwin, G. H. & Johns, E. W. (1973) "Isolation and characterization of two calf-thymus chromatin non-histone proteins with high contents of acidic and basic amino acids," Eur. J. Biochem. 40,215-219. 6. Marushige, K. & Dixon, G. H. (1971) "Transformation of trout testis chromatin," J. Biol. Chem. 246,5799-5805. 7. Elgin, S. C. R. & Bonner, J. (1972) "Partial characterization of the major nonhistone chromosomal proteins," Biochemistry 11, 772-781. 8. Smith, J. A. & Stocken, L. A. (1973) "The characterization of a non-histone protein isolated from histone F1 preparations," Biochem. J. 131, 859-861. 9. Stollar, D. D. & Ward, M. (1970) "Rabbit antibodies to histone fractions as specific reagents for preparative and comparative studies," J. Biol. Chem. 245, 1261-1266. 10. Ouchterlony, 0. & Nilsson, L. A. (1973) in Handbook of Experimental Immunology, ed. Weir, D. M. (Blackwell Scientific Publications, London, England), pp. 19.1-19.39. 11. Crowle, A. J. (1960) "Four modifications of the microagar diffusion precipitin test," J. Lab. Clin. Med. 55,593-604. 12. Kabat, E. A. & Mayer, M. M. (1961) in Experimental Immunochemistry (C. C Thomas, Springfield, Ill.), p. 149. 13. Brutlag, D., Schlehuber, C. & Bonner, J. (1969) "Properties of formaldehyde treated nucleohistone," Biochemistry 8, 32143218. 14. Chalkley, R. & Hunter, C. (1975) "Histone-histone propinquity by aldehyde fixation of chromatin," Proc. Natl. Acad. Sci. USA 72, 1304-1308. 15. Coons, A. H., Leduc, E. H. & Connolly, J. M. (1955) "Studies on antibody production I. A method for the histochemical demonstration of specific antibody and its application to a study of the hyper-immune rabbit," J. Exp. Med. 102, 49-60. 16. Olins, A. L. & Olins, D. E. (1974) "Spheroid chromatin units (v bodies)," Science 183, 330-332. 17. Kornberg, R. D. & Thomas, J. 0. (1974) "Chromatin structure: Oligomers of the histones," Science 184, 865-868. 18. Kornberg, R. D. (1974) "Chromatin structure: A repeating unit of histones and DNA," Science 184, 868-871. 19. Goldblatt, D. & Bustin, M. (1975) "Exposure of histone antigenic determinants in chromatin," Biochemistry 14, 1689-1695. 20. Ohlenbusch, H. H., Olivera, B. M., Tucin, D. & Davidson, N. (1967) "Selective dissociation of histones from calf thymus nucleoprotein," J. Mol. Biol. 25, 299-315. 21. Wilhelm, X. & Champagne, M. (1969) "Dissociation de la nucleoproteine d'erythrocytes de poulets par le sels," Eur. J. Biochem. 10, 102-109.
Proc. Natl. Acad. Sci. USA 73 (1976) 22. Georgiev, G. P., Ananieva, L. N. & Kozlov, L. N. (1966) "Stepwise removal of protein from a deoxyribonucleoprotein complex and derepression of the genome," J. Mol. Biol. 22, 365-371. 23. Bridges, C. B. (1935) "Salivary gland chromosome maps; with a key to the banding of the chromosomes of Drosophila melanogaster," J. Hered. 26, 60-64. 24. Rae, P.' M. M. (1970) "Chromosomal distribution of rapidly reannealing DNA in Drosophila melanogaster," Proc. Natl. Acad. Sci. USA 67,1018-1025. 25. Gall, J. G., Cohen, E. H. & Polan, M. L. (1971) "Repetitive DNA sequences in Drosophila," Chromosoma 33,319-344. 26. Peacock, W. J., Brutlag, D., Goldring, E., Appels, R., Hinton, C. W. & Lindsley, D. L. (1973) "The organization of highly repeated DNA sequences in Drosophila melanogaster chromosomes," Cold Spring Harbor Symp. Quant. Biol. 38,405-416. 27. Rudkin, G. T. & Tartof, K. D. (1973) "Repetitive DNA in polytene chromosomes of Drosophila melanogaster," Cold Spring Harbor Symp. Quant. Biol. 38,397-403. 28. Manning, J. E., Schmid, C. W. & Davidson, N. (1975) "Interspersion of repetitive and non-repetitive DNA sequences in the Drosophila melanogaster genome," Cell 4, 141-155. 29. Wensink, P. C., Finnegan, D. F., Donelson, J. E. & Hogness, D. S. (1975) "A system for mapping DNA sequences in the chromosome of Drosophila melanogaster," Cell 4,315-325. 30. Sederoff, R., Lowenstein, L. & Birnboin, H. C. (1975) "Polypyrimidine segments in Drosophila melanogaster DNA: II. Chromosome location and nucleotide sequence," Cell 5, 183-194. 31. Vosa, C. G. (1970) "The discriminating fluorescence patterns of the chromosomes of Drosophila melanogaster," Chromosoma 31,446-451. 32. Adkisson, K. P., Perrault, W. J. & Gay, H. (1971) "Differential fluorescent staining of Drosophila chromosomes with quinacrine mustard," Chromosoma 34,190-205. 33. Ellison, J. R. & Barr, H. J. (1971) "Differences in the quinacrine staining of the chromosomes of a pair of sibling species: Drosophila melanogaster and Drosophila simulans," Chromosoma 34,424-435. 34. Ellison, J. R. & Barr, H. J. (1972) "Quinacrine fluorescence of specific chromosome regions," Chromosoma 36,375-390. 35. Barr, H. J. & Ellison, J. R. (1972) "Ectopic pairing of chromosome regions containing chemically similar DNA," Chromosoma 39, 53-61. 36. Weisblum, B. & Haseth, P. de (1972) "Quinacrine, a chromosome stain specific for deoxyadenylate-deoxythymidilate-rich regions in DNA," Proc. Natl. Acad. Sci. USA 69,629-632. 37. Weisblum, B. (1973) "Fluorescent probes of chromosomal DNA structure: three classes of acridines," Cold Spring Harbor Symp. Quant. Biol. 38,441-449. 38. Mayfield, J. E. & Ellison, J. R. (1975) "The organization of interphase chromatin in Drosophilidae. The self adhesion of chromatin containing the same DNA sequence," Chromosoma 52, 37-48. 39. Holmquist, G. (1975) "Hoechst 33258 fluorescent staining of Drosophila chromosomes," Chromosoma 49, 333-356. 40. Britten, R. J. & Davidson, E. H. (1969) "Gene regulation for higher cells: A theory," Science 165, 349-357. 41. Zweidler, A. & Cohen, L. H. (1973) "Histone variants and new histone species isolated from mammalian tissues," J. Cell Biol. 59, 378a. 42. Kinkade, J. M. & Cole, R. D. (1966) "The resolution of four lysine-rich histones derived from calf thymus," J. Biol. Chem. 241, 5790-5797. 43. Silver, L. M. & Elgin, S. C. R. (1976) "A method for determination of the in situ distribution of chromosomal proteins," Proc. Natl. Acad. Sci. USA 73,423-427.
Vol. 73, No. 6, pp. 2038-2042, June 1976
Cell Biology
Locations of chromosomal proteins in polytene chromosomes* (chromosome structure/histone antibodies/immunofluorescence detection of a nonhistone chromosomal protein/Drosophila chromosomes)
CANDIDO RODRIGUEZ ALFAGEME, GEORGE T. RUDKIN, AND LEONARD H. COHEN The Institute for Cancer Research, Fox Chase Cancer Center, Philadelphia, Pennsylvania 19111
Communicated by Thomas F. Anderson, March 23,1976
ABSTRACT DI, a nonhistone chromosomal protein rich in both basic and acidic amino acids, has been localized at a limited number of specific loci in polytene chromosomes of Drosophila melanogaster. H2B, a nucleosomal histone, and Hi, a nonnucleosomal histone, are both found throughout most chromosomal regions. Study of the role of proteins in gene function in eukaryotes has been hampered by the complexity of chromatin. Thus, whereas many proteins have been detected in purified chromatin preparations, none of them has yet been shown to be associated with specific genetic loci. We have now been able to localize by immunofluorescence a nonhistone protein in the polytene chromosomes of Drosophila. A major difficulty has been that conventional methods for spreading polytene chromosomes in acetic acid result in the extensive loss (1-3) and possible translocation of chromosomal proteins, particularly basic proteins. We have devised a method for obtaining, with minimal loss of the associated proteins, chromosome spreads capable of reacting with specificity with antibodies against purified chromosomal proteins. The recent isolation of such proteins in our laboratory (4, t) has permitted us to visualize Di, a Drosophila nonhistone protein, as well as two Drosophila histones, Hi and H2B. DI is a protein, obtained from Drosophila chromatin, that is rich in both acidic and basic amino acids (24% and 21%, respectively)t. It may belong to a class of proteins of general occurrence among eukaryotes (5-8). It shares many properties with a group of proteins of calf-thymus chromatin known as the high mobility group (5), notably, solubility in 5% perchloric acid, extractability from chromatin by 0.35 M NaCl, and high content of basic and acidic amino acids. However, it differs from those proteins of the high mobility group thus far isolated in having a higher molecular weight and in having aspartic acid rather than glutamic acid as its most abundant amino acid. The amount of Di in embryo chromatin is 2-5% that of histone Hi on a molar basis, and the amount in salivary gland chromatin, while not yet accurately determined, is not more than that. The possibility existed, therefore, that Di might be confined to a proportionately small fraction of the genome. MATERIALS AND METHODS Protein Di and histones Hi and H2B were isolated from 4 to This paper is dedicated to the memory of the late Jack Schultz, whose understanding of chromosomes triggered the collaborative effort involved. Abbreviations: D1, Drosophila protein 1; phosphate-buffered saline, 0.14 M NaCI, 10 mM sodium phosphate, pH 7.3. The histone nomenclature used is that adopted at the Ciba Foundation symposium on structure and function of chromatin, 1975, in which H1 = F1, H2A = F2a2, H2B = F2b, H3 = F3, and H4 = F2al. * A preliminary report was presented at the 15th Annual Meeting of the American Society for Cell Biology, November 11-14, 1975, San Juan, Puerto Rico [(1975) J. Cell Biol. 67, 6a]. t Manuscript in preparation.
20-hr embryos of Drosophila melanogaster Oregon R as described elsewhere (4, t). Protein DI was obtained from embryo chromatin together with histone Hi by acid extraction followed by precipitation of other proteins by 5% perchloric acid. It was separated from Hi by preparative gel electrophoresis. This isolation yields a single electrophoretic component in acid urea gels (Fig. 1) and in sodium dodecyl sulfate gelst. Preparation of Sera Against DI, HI, and H2B. Rabbits were immunized (9) with a suspension of protein (83 ,g of Di, or 240 ,gg of HI, or 185 .sg of H2B) and yeast RNA (except in the case of HI) in 1 ml of phosphate-buffered saline (0.14 M NaCl, 10 mM sodium phosphate pH 7.3) and 1.5 ml of complete Freund's adjuvant. Yeast RNA type VI was obtained from Sigma Chemical Co. This inoculum was mixed for 5 min with a Vortex mixer and injected subcutaneously in the back at multiple sites. Booster shots were given 2, 7, and 9 weeks later (the last two injections with half the amount of protein-RNA complexes) and sera obtained 1 week after the last inoculation. Immunodiffusion Analysis of Rabbit Sera. Double immunodiffusion (10) on microscope slides (I1) was performed in 1.5% agarose gels containing 0.14 M NaCl, 2 mM MgCl2, 0.1 mM CaCl2, and sodium 5,5'-diethylbarbiturate at pH 7.3 (12). Agar is unsuitable for this purpose because it contains a component that precipitates histones. The antiserum against Di showed little or no crossreactivity with the histones (Fig. 2). Antisera prepared against Hi and H2B showed no crossreactivity with H2B and HI, respectively, or with DI, and little or no crossreactivity with other histones. Absorption of Immune Sera. Antisera were treated with appropriate histone-RNA complexes to remove crossreacting antibodies. These histone-RNA complexes (1:1) were prepared by precipitating the histone mixtures (dissolved in phosphatebuffered saline) with yeast RNA type VI. The precipitates were recovered by centrifugation (10,000 rpm, 10 min), washed and resuspended in phosphate-buffered saline, and added to whole serum in the ratio of 1 iug of each histone-RNA complex per gl of serum. After incubation at 40 for 18 hr the absorbed sera were centrifuged at 10,000 rpm for 10 min and the supernatants were used for the assays. Preparation of Chromosomes for Antibody Binding. D. melanogaster salivary glands were dissected from giant (gt) larvae obtained as daughters of a cross between females homozygous for the gt mutation and males carrying an X chromosome deficient for the gt locus [Df(1)62g'8] (courtesy of T. C. Kaufman). A D. virilis wild-type stock was obtained from S. K. Majumdar. Salivary glands were treated with 2% formaldehyde (in 50 mM Na glycerophosphate, 25 mM sucrose, 1 mM MgCl2, pH 6.8, for 2 min at 4°-8°) to prevent the loss of chromosomal proteins during subsequent treatment; the glands were then transferred to a drop of 2% formaldehyde in 5% acetic acid, were homogenized lightly either with a Dounce type microhomogenizer or by forcing through a micropipette, and were finally squashed between a microscope slide and a siliconed cover slip. The cover slip was removed by immersing 2038
Cell Biology:
Alfageme et al.
D1 H
FIG. 1 (left). Purity of D. melanogaster D1 protein. Left gel: D1 isolated by preparative gel electrophoresis of 5% perchloric acid-soluble nuclear protein fraction. Right gel: The 5% perchloric acid-soluble fraction, shown for comparison. Electrophoresis was in acidurea-polyacrylamide gels (5% acetic acid, 4.2 M urea). Gels were stained with amido black. FIG. 2 (right). Immunodiffusion analysis of antiserum against D. melanogaster D1. The center well contained 4 jAl of whole, unabsorbed antiserum against D1 prepared as described in Materials and Methods. The other wells contained 3jug in 3,gl of the protein as indicated. A single and strong precipitin line is formed between D1 and antiserum against D1. A weak reaction can be seen with H1. These two precipitin lines cross each other, indicating lack of identity between the antigenic determinants (10). The precipitate around the well containing H4 is caused by a serum component other than antibodies against D1 because it is also formed when this histone is tested with nonimmune rabbit serum.
the slide in phosphate-buffered saline. The conditions chosen for the fixation of chromosomes are based on data of Brutlag et al. (13) and Chalkley and Hunter (14). Binding of Antibodies to Polytene Chromosomes and Visualization of Bound Antibodies. Sera absorbed as described above were diluted with phosphate-buffered saline (ko for DI, '/40 for either HI or H2B) and applied to the squashes, which were then covered with cover slips. Incubation in a humid chamber for I hr was followed by extensive washing in phosphate-buffered saline (1 hr at 220 and 18 hr at 40) to eliminate weakly bound antibodies. The bound rabbit antibodies were detected with fluorescein isothiocyanate conjugated antibody against rabbit IgG prepared in goat (A280/A495 = 1.7, purchased from Miles-Yeda Ltd.) diluted 1ko with phosphatebuffered saline. After incubation for 30 min at 220 in the humid chamber, the chromosome preparations were washed in phosphate-buffered saline for 1 hr at 220 and mounted in buffered glycerol (Becton, Dickinson and Company). A Zeiss fluorescence microscope equipped with epiillumination was used. To avoid fading of fluorescence, we searched and focussed in the phase contrast mode using a tungsten light source. Then a fluorescence photomicrograph was taken by a 15-sec exposure on Kodak Tri-X-pan film and developed in Diafine (Acufine, Inc., Chicago, Ill. 60611) at ASA 1600. Light source: Osram HBO-200 W, primary filters: BG 38, SP500; secondary filter: 50; dichroic mirror: F1 (Zeiss). RESULTS The procedure adopted for visualization of proteins in polytene chromosomes involves three steps. First, the salivary glands are fixed in formaldehyde, homogenized lightly in acetic acidformaldehyde to disperse the cytoplasm, and squashed. The aldehyde fixation is essential since many chromosomal proteins are known to be removed by the acid conditions needed for obtaining cytologically suitable preparations (1-3). Second, the
Proc. Natl. Acad. Sci. USA 73 (1976)
2039
rabbit antiserum is treated with a mixture of appropriate histones, as described in Materials and Methods, and applied to the slide with the chromosomes. Finally, the chromosomal proteins are visualized indirectly (15) by labeling the bound rabbit antibodies with fluorescein-labeled antibody against rabbit IgG prepared in goat.
The patterns of fluorescence obtained with D. melanogaster salivary gland chromosomes indirectly stained with antisera against D1, HI, and H2B are shown in Fig. 3 (left panel). The contrast between the staining by antisera against the histones and by the antiserum against DI is apparent. Antibodies against. Hi and H2B are distributed quite generally throughout all the chromosomes, whereas chromosomes stained with antibody against Dl have a few very intensely fluorescent regions in the vicinity of the chromocenter. In addition, there are other chromosomal regions that fluoresce prominently but much less strongly, such as the distal ends of some chromosomes and a few other regions along every chromosome. The less intensely fluorescing regions have not been studied in detail. The results obtained with D. virilis chromosomes have a bearing on the specificity of staining of the less intensely fluorescing regions. Dl is species-specific in that D. melanogaster Di differs considerably in electrophoretic mobility from D. virilis DIt. It is therefore of interest that the antiserum prepared against D. melanogaster Di failed to bind to D. virilis chromosomes in contrast to the antisera against D. melanogaster Hi and H2B (Fig. 3, right panel). This indicates a high degree of specificity of the antiserum against DI, suggesting that even the low-level staining, at many sites, of D. melanogaster chromosomes is due to the presence at these sites of DI or of proteins very closely related to it. Results of control experiments to assess the immunological specificity of staining are shown in Table 1. The use of antiserum against H2B that had been treated with H2B-RNA complexes (0.5 ,tg of H2B per gl of serum) did not yield fluorescence; in contrast, treatment of antiserum against H2B with histones Hi, H2A, H3, and H4 complexed with RNA (0.5 ,ug of each histone per gl of serum) had no effect on the fluorescence. Similarly, no blocking of antiserum against DI was obtained by treatment with a mixture of all histones, nor of antiserum against HI by treatment with RNA complexed with H2A, H2B, H3, and H4. These results indicate that the fluorescence response in the chromosomes is specific for the antigen. The chromosomal regions stained intensely by antiserum against Di are seen at higher resolution in Fig. 4. In the case of visualization with antiserum against Di, two intensely fluorescent regions can be seen at the base of 3R and two in chromosome 4. This pattern is reproducible and was observed in salivary gland chromosomes from late third instar larvae and from prepupae. In contrast, antisera against HI and H2B stain all the bands resolvable with phase contrast optics. The general distribution of H2B throughout the chromosomes supports the current view that this histone is part of an oligomeric unit associated with all of the DNA (16-18). On the other hand, Hi is not considered to be part of the oligomer and no indication of its distribution has been available. Our method shows HI to be distributed throughout every chromosomal region, although not necessarily in constant proportion to H2B. The ubiquitous staining of chromosomal regions by antibody against H2B is a good test of the accessibility of proteins in these chromosome preparations, since H2B in chromatin is much less accessible to antibodies than is Hi (19) and is also more difficult to extract than either Dit or HI (20-22). It seems likely that the acid conditions used in spreading the chromosomes increases the accessibility of the proteins. The possibility still remains that some sites are not available to antibodies.
2040
Cell Biology: Alfageme et al.
Proc. Natl. Acad. Sci. USA 73 (1976)
'~~~~~~~~~~~~~I
l.l.....=. ...
.-
~
FIG. 3. Polytene chromosomes from D. melanogaster (left) and D. virilis (right) indirectly stained with antisera againstD. melanogaster D1, H1, or H2B. Anti-D1 was absorbed with H1, H2A, H2B, H3, and H4; anti-Hi was absorbed with H2A, H2B, H3, and H4; anti-H2B was absorbed with H1, H2A, H3, and H4 (see Materials and Methods and Table 1). Prior to use, the antisera were diluted with phosphate-buffered saline, 'o for anti-D1 and /40 for either anti-H1 or anti-H2B. The inset is a phase contrast image of the D. virilis nucleus, which was in the microscope field for the upper right fluorescence photomicrograph. The scales are 10 ,jm.
Chromosomes fixed under a variety of conditions (1-5% formaldehyde for 1-20 min at 4°-22°) show essentially identical fluorescent patterns when stained with antibody against D1, thus suggesting that the fixation procedure adopted is sufficient to prevent significant loss of protein Di. DISCUSSION The method presented here for the preparation of polytene chromosome spreads for immunofluorescence is a compromise in that the formaldehyde treatment used to prevent loss of proteins renders the chromosomes difficult to spread and thus hampers identification of some fluorescent regions. Nevertheless, it has permitted us to identify the regions that fluoresce most brightly with antibody against D1 as 81F and 83C-E in chromosome 3R and two regions in chromosome 4 [Bridges standard map (23)]. The marked difference in intensity between those few fluorescent sites near the chromocenter and
the faintly fluorescent sites scattered through the genome is most simply interpreted as due to a difference in D1 content between these two classes of sites. This interpretation is supported by the results reported here, although the accessibility of the protein for reaction with the antibody may still be a factor worthy of investigation. If we interpret subjective fluorescence intensity as an indicator of the relative concentration of the primary antigen, it would appear that a disproportionately large fraction of the D1 is localized in a few short regions of the genome, with the remainder widely distributed over all the chromosomes. The restricted distribution of D1 implies that the Dl-containing regions share structural features capable of recognizing this protein. Common structural features might conceivably arise from any chromosomal macromolecule; if they arise from DNA, one possibility is that they arise from repetitive DNA. This class of DNA has been detected by hybridization in situ
Cell Biology:
Alfageme et al.
Proc. Natl. Acad. Sci. USA 73 (1976)
2041
Table 1. Effect of various serum treatments on the immunofluorescent staining of D. melanogaster polytene chromosomes Chromo-
41. .
j
.#
.
VP
J
Serum
Treatment of serum*
Nonimmune None Anti-H2B Anti-H2B Anti-H2B
None None Yeast RNA H2B and yeast RNA H4, H3, H2A, H1, and yeast RNA H4, H3, H2A, H2B, H1, and yeast RNA H4, H3, H2A, H2B, and yeast RNA
1.
./
Anti-D1 Anti-H1
some fluorescencet
+ +
+ +
* Antisera were absorbed with mixtures of the proteins and RNA listed (see Materials and Methods), diluted 1/20 with phosphatebuffered saline, and then used in the indirect method for fluorescence staining of chromosomes. t A minus score indicates that fluorescence was not detected in the chromosomes or that it was hardly visible and insufficient to produce a photographic image under standard exposjure conditions (see Materials and Methods). B
FIG. 4. Proximal ends of D. melanogaster salivar gland chromosomes 3R indirectly stained with antisera against D1, H1, and H2B. Preparations were stained as described in Materials and Methods with absorbed antisera diluted with phosphate-buffered saline Y40 for either H2B or H1 and 'Ao for D1. Fluorescent granules (A) such as those in the top photograph are occasionally seen. The fluorescence from these granules can be removed by washing the stained preparations with 5% acetic acid. This treatment does not affect the fluorescence intensity of chromosomes. The arrows point to subsections 81F (right) and 83 C-E (left). Chromosome 4 can be clearly seen at the extreme right of the top pair of photomicrographs. The scale is 10 gm.
at chromocentral regions and at many other loci interspersed along the chromosomes (24-30). Precise localization of such DNA sequences might help in determining whether DI is associated with specific nucleotide sequences.
Studies of several investigators on the staining of chromosomes by quinacrine and Hoechst 33258 (31-39) provide clues
to the chemical nature of the principal binding regions for antibody against D1 (Fig. 4). Salivary gland chromosomes of D. melanogaster stained with quinacrine are reported to exhibit brightest fluorescence at subsections 8WF and 83D in chromosome 3, and 101 and 102D in chromosome 4 [these regions, with the exception of 83D, have been shown to be Hoechst bright as well (39)].* The quinacrine bright regions coincide with the principal binding sites for antibody against DI. The fluorescence of the dyes is enhanced by adenylate- and thymidylaterich DNA (35, 36, 38), although, as Holmquist has pointed out, adenylate and thymidylate richness is not alone a sufficient property to account for fluorochrome brightness (39). These observations suggest that nucleotide sequences rich in adenylate and thymidylate might be a property of D1 sites. An answer to this question is important, because identification of proteins associated with repetitive DNA sequences would be valuable for studies of the functional role, or roles, of this class of DNA, a class that may be involved in gene regulation (40). The method described here§ permits localization of proteins at the level of resolution of the chromomere. The application of this method to other nonhistone proteins, to histone variants (4, 41, 42), and to modified histones should make accessible detailed knowledge of the organization of proteins in the eukaryotic genome. We thank Dr. M. Bayer for the use of the fluorescence microscope; Drs. M. Bosma, S. Hausman and I. Millman for their help and advice in the preparation and analysis of antibodies; Drs. B. Mintz, R. Perry and K. Tartof for their helpful review of the manuscript; and Mrs. Mara Bard, Mrs. S. Bianchi-Ellquist, Mr. R. Rudkin, and Miss D. Hazler for * Note Added in Proof. Precise coincidence of the principal DI sites (81, 83, 101, and 102) with the quinacrine bright sites has been ver-
ified by double-label experiments. § After this manuscript was completed, Silver and Elgin reported (43) the use of antibodies against preparations of total nonhistone chromosomal proteins to stain Drosophila polytene chromosomes by
methods similar to those described here.
2042
Cell Biology: Alfageme et al.
excellent technical assistance. This study was supported by U.S. Public Health Service Grants CA-12544, CA-06927, and RR-05539 from the National Institutes of Health, a grant from the National Science Foundation (BO-41476), and an appropriation from the Commonwealth of Pennsylvania.
1. Swift, H. (1964) "The histones of polytene chromosomes," in The Nucleohistones, eds. Bonner, J. & Ts'o, P. (Holden-Day, Inc., San Francisco), pp. 169-183. 2. Murray, K. (1969) "Stepwise removal of histones from native deoxyribonucleoprotein by titration with acid and some properties of the resulting partial nucleoproteins," J. Mol. Biol. 39, 125-144. 3. Brody, T. (1974) "Histones in cytological preparations," Exp. Cell Res. 85,255-263. 4. Alfageme, C. R., Zweidler, A., Mahowald, A. & Cohen, L. H. (1974) "Histones of Drosophila embryos: Isolation and structural studies," J. Biol. Chem. 249,3729-3736. 5. Goodwin, G. H. & Johns, E. W. (1973) "Isolation and characterization of two calf-thymus chromatin non-histone proteins with high contents of acidic and basic amino acids," Eur. J. Biochem. 40,215-219. 6. Marushige, K. & Dixon, G. H. (1971) "Transformation of trout testis chromatin," J. Biol. Chem. 246,5799-5805. 7. Elgin, S. C. R. & Bonner, J. (1972) "Partial characterization of the major nonhistone chromosomal proteins," Biochemistry 11, 772-781. 8. Smith, J. A. & Stocken, L. A. (1973) "The characterization of a non-histone protein isolated from histone F1 preparations," Biochem. J. 131, 859-861. 9. Stollar, D. D. & Ward, M. (1970) "Rabbit antibodies to histone fractions as specific reagents for preparative and comparative studies," J. Biol. Chem. 245, 1261-1266. 10. Ouchterlony, 0. & Nilsson, L. A. (1973) in Handbook of Experimental Immunology, ed. Weir, D. M. (Blackwell Scientific Publications, London, England), pp. 19.1-19.39. 11. Crowle, A. J. (1960) "Four modifications of the microagar diffusion precipitin test," J. Lab. Clin. Med. 55,593-604. 12. Kabat, E. A. & Mayer, M. M. (1961) in Experimental Immunochemistry (C. C Thomas, Springfield, Ill.), p. 149. 13. Brutlag, D., Schlehuber, C. & Bonner, J. (1969) "Properties of formaldehyde treated nucleohistone," Biochemistry 8, 32143218. 14. Chalkley, R. & Hunter, C. (1975) "Histone-histone propinquity by aldehyde fixation of chromatin," Proc. Natl. Acad. Sci. USA 72, 1304-1308. 15. Coons, A. H., Leduc, E. H. & Connolly, J. M. (1955) "Studies on antibody production I. A method for the histochemical demonstration of specific antibody and its application to a study of the hyper-immune rabbit," J. Exp. Med. 102, 49-60. 16. Olins, A. L. & Olins, D. E. (1974) "Spheroid chromatin units (v bodies)," Science 183, 330-332. 17. Kornberg, R. D. & Thomas, J. 0. (1974) "Chromatin structure: Oligomers of the histones," Science 184, 865-868. 18. Kornberg, R. D. (1974) "Chromatin structure: A repeating unit of histones and DNA," Science 184, 868-871. 19. Goldblatt, D. & Bustin, M. (1975) "Exposure of histone antigenic determinants in chromatin," Biochemistry 14, 1689-1695. 20. Ohlenbusch, H. H., Olivera, B. M., Tucin, D. & Davidson, N. (1967) "Selective dissociation of histones from calf thymus nucleoprotein," J. Mol. Biol. 25, 299-315. 21. Wilhelm, X. & Champagne, M. (1969) "Dissociation de la nucleoproteine d'erythrocytes de poulets par le sels," Eur. J. Biochem. 10, 102-109.
Proc. Natl. Acad. Sci. USA 73 (1976) 22. Georgiev, G. P., Ananieva, L. N. & Kozlov, L. N. (1966) "Stepwise removal of protein from a deoxyribonucleoprotein complex and derepression of the genome," J. Mol. Biol. 22, 365-371. 23. Bridges, C. B. (1935) "Salivary gland chromosome maps; with a key to the banding of the chromosomes of Drosophila melanogaster," J. Hered. 26, 60-64. 24. Rae, P.' M. M. (1970) "Chromosomal distribution of rapidly reannealing DNA in Drosophila melanogaster," Proc. Natl. Acad. Sci. USA 67,1018-1025. 25. Gall, J. G., Cohen, E. H. & Polan, M. L. (1971) "Repetitive DNA sequences in Drosophila," Chromosoma 33,319-344. 26. Peacock, W. J., Brutlag, D., Goldring, E., Appels, R., Hinton, C. W. & Lindsley, D. L. (1973) "The organization of highly repeated DNA sequences in Drosophila melanogaster chromosomes," Cold Spring Harbor Symp. Quant. Biol. 38,405-416. 27. Rudkin, G. T. & Tartof, K. D. (1973) "Repetitive DNA in polytene chromosomes of Drosophila melanogaster," Cold Spring Harbor Symp. Quant. Biol. 38,397-403. 28. Manning, J. E., Schmid, C. W. & Davidson, N. (1975) "Interspersion of repetitive and non-repetitive DNA sequences in the Drosophila melanogaster genome," Cell 4, 141-155. 29. Wensink, P. C., Finnegan, D. F., Donelson, J. E. & Hogness, D. S. (1975) "A system for mapping DNA sequences in the chromosome of Drosophila melanogaster," Cell 4,315-325. 30. Sederoff, R., Lowenstein, L. & Birnboin, H. C. (1975) "Polypyrimidine segments in Drosophila melanogaster DNA: II. Chromosome location and nucleotide sequence," Cell 5, 183-194. 31. Vosa, C. G. (1970) "The discriminating fluorescence patterns of the chromosomes of Drosophila melanogaster," Chromosoma 31,446-451. 32. Adkisson, K. P., Perrault, W. J. & Gay, H. (1971) "Differential fluorescent staining of Drosophila chromosomes with quinacrine mustard," Chromosoma 34,190-205. 33. Ellison, J. R. & Barr, H. J. (1971) "Differences in the quinacrine staining of the chromosomes of a pair of sibling species: Drosophila melanogaster and Drosophila simulans," Chromosoma 34,424-435. 34. Ellison, J. R. & Barr, H. J. (1972) "Quinacrine fluorescence of specific chromosome regions," Chromosoma 36,375-390. 35. Barr, H. J. & Ellison, J. R. (1972) "Ectopic pairing of chromosome regions containing chemically similar DNA," Chromosoma 39, 53-61. 36. Weisblum, B. & Haseth, P. de (1972) "Quinacrine, a chromosome stain specific for deoxyadenylate-deoxythymidilate-rich regions in DNA," Proc. Natl. Acad. Sci. USA 69,629-632. 37. Weisblum, B. (1973) "Fluorescent probes of chromosomal DNA structure: three classes of acridines," Cold Spring Harbor Symp. Quant. Biol. 38,441-449. 38. Mayfield, J. E. & Ellison, J. R. (1975) "The organization of interphase chromatin in Drosophilidae. The self adhesion of chromatin containing the same DNA sequence," Chromosoma 52, 37-48. 39. Holmquist, G. (1975) "Hoechst 33258 fluorescent staining of Drosophila chromosomes," Chromosoma 49, 333-356. 40. Britten, R. J. & Davidson, E. H. (1969) "Gene regulation for higher cells: A theory," Science 165, 349-357. 41. Zweidler, A. & Cohen, L. H. (1973) "Histone variants and new histone species isolated from mammalian tissues," J. Cell Biol. 59, 378a. 42. Kinkade, J. M. & Cole, R. D. (1966) "The resolution of four lysine-rich histones derived from calf thymus," J. Biol. Chem. 241, 5790-5797. 43. Silver, L. M. & Elgin, S. C. R. (1976) "A method for determination of the in situ distribution of chromosomal proteins," Proc. Natl. Acad. Sci. USA 73,423-427.
