Biochemical and ultrastructural study of the sperm chromatin from Mytilus galloprovincialis

Biochemical and ultrastructural study of the sperm chromatin from Mytilus galloprovincialis

Experimental Biochemical and Ultrastructural from Mytilus ZOYA AVRAMOVA,’ ‘Institute Cell Research ANDRE1 of Molecular Biology, and 2Znstitute o...

6MB Sizes 0 Downloads 26 Views

Experimental

Biochemical

and Ultrastructural from Mytilus

ZOYA AVRAMOVA,’ ‘Institute

Cell Research

ANDRE1

of Molecular Biology, and 2Znstitute of Cytology,

152 (1984) 231-239

Study of the Sperm galloprovincialis

ZALENSKY*

and ROUMEN

Chromatin

TSANEV’,*

Bulgarian Academy of Sciences, 1113 Sofia, Bulgaria, USSR Academy of Sciences, Leningrad, USSR

Protein composition and ultrastructure of the mature spermatozoa of the mussel Myths were studied upon gradual decondensation of the nuclei with increasing NaCl concentration. Three types of protein were found, associated with the sperm DNA: (1) the sperm-specific proteins Sl, S2 and S3 (80% of the acid-soluble proteins); (2) the four core histones (20%); (3) three non-histone proteins tightly bound to DNA (about 4 ug protein per 100 ug DNA). The sperm-specific protein S3 was the first to dissociate at about 0.5 M NaCl and electron micrographs of spread nuclei indicated its participation in the final compaction of the nucleus. Hypotonically treated sperm nuclei revealed the presence of 21-25 nm large granules irregularly scattered along some of the DNA fibers. These granules correspond to the ‘superbeads’ of histone-containing chromatins. The tightly bound non-histone proteins were represented by a triplet in the range 60-80 kD. They formed 30-60 nm large annular bodies holding DNA fibers and resisting high salt-detergent treatment. galloprovincialis

The sperm nuclei of various Bivalvia molluscs are characterized by a considerable variety in their DNA-associated proteins. Although the chromatin composition of these species has been studied [l-4], little is known about the ultrastructural organization of their mature sperm chromatin. The Mytilus class of DNA-associated proteins are interesting in view of the simultaneous presence of both somatic-type histones and sperm-specific proteins, some of which are considered to be intermediate between histones and protamines as regards their size and amino acid composition [l-4]. In an attempt to reveal the common principles of chromatin organization of the mature sperm nuclei and the possible structural importance of stable DNA-protein complexes, we were interested to study the sperm chromatin of evolutionary distinct species. In our previous studies we have revealed some common features in the organization of the sperm chromatin of a mammal [5, 61 and a fish [7, 81. Now we have chosen the mature sperm nucleus of a Bivalvia mollusc, Mytilus galloprouincialis, as an invertebrate species, where the protein composition of the sperm chromatin is expected to be the same as that of Mytilus edulis. The sperm protein composition of the latter is well studied, but there are no data on its structural organization, except for an old work on ultrathin sections of compact nuclei, where 40-70 nm large granules of unknown nature were observed [91. *To whom offprint requests should be sent. Copyright @ 1984 by Academic Press, Inc. All rights of reproduction in any form reserved 0014-4827/84 $03.00

232

Avramova,

Zalensky

and

Tsanev

In all previous studies, partial removal of the chromosomal basic proteins has been achieved by their preferential extraction with acids [I], partial solubilization in Hðanol, or selective precipitation with H2S04-acetone solutions [l-4]. The structures remaining after such extractions are highly denatured and no conclusions regarding their fine organization could be drawn. To avoid this we have used another approach [6] consisting in a gradual salt-induced decondensation of the mature sperm nuclei. This procedure permits the analysis of both the proteins extracted at different ionic strengths and the ultrastructure of the remaining DNA-protein complexes.

MATERIALS

AND

METHODS

Preparation of Mature Sperm Nuclei Excised mature male gonads of the mussel Myths gailoprwincialis (collected in the Black Sea) were placed in a bag of nylon gauze immersed in boiled and faltered sea water. The mature spermatozoa were spontaneously shed in the water and were centrifuged for 5 min at 800 g. Under these conditions the moving mature sperm cells remained floating in the supematant. They were pelleted at 3 800 g for 30 min, suspended in a buffer containing 0.25 M sucrose, 10 mM ‘Iris-HCl, pH 7.6, 0.1 M NaCl, 1 mM PMSF and sonicated twice for 2 min in a cell disruptor W 375 (Ultrasonics, Inc., USA). The suspension was layered over 1 M sucrose in the same buffer and pelleted for 15 min at 3800 g. The nuclei resistant to the ultrasonic disintegration were suspended in the above buffer containing 1% ‘Biton X-100. After 30 min of incubation the suspension was centrifuged several times through 1 M sucrose. The final pellet showed a homogeneous population of compact demembranized sperm nuclei when visualized under both phase contrast and electron microscope. Contaminating nuclei were less than 2 %.

Differential Extraction

of Nuclear Proteins

Unless otherwise indicated, the extraction procedures were performed at room temperature in the presence of 1 mM PMSF. The nuclear pellets were dispersed in NaCl solutions of the following concentrations: 0.2; 0.4; 0.5; 0.65; 0.8; 1.O; 2.0 and 2.5 M. After 30 min incubation the suspension was centrifuged for 18 h at 190000 g in the Ti 60 rotor of a Beckman ultracentrifuge. The supematants and the pellets were collected and analysed for their protein composition. The pellet obtained after extraction of the nuclei with 2 M NaCl was re-extracted with 2 M NaCl and then dispersed in 1% Mton X-100. After 60 min at 0°C the sample was centrifuged as above and the DNA pellet containing the tightly bound proteins was dispersed in 20 mM Tris-HCl, pH 7.6, and dialysed extensively against the same buffer. The tightly. bound proteins were obtained as a precipitate after an extensive digestion of DNA with DNase I (5, 71. The precipitate contained only traces of DNase I (less than 10 % of its protein content as judged by gel electrophoresis) and could be dissolved in SDS solution. Acid-soluble proteins were extracted with 0.25 N H$S04 at @‘C for 45 mm from the corresponding supematants or pellets, followed by a 30 min centrifugation at 25000 g to remove insoluble material. The supematant was dialysed against water and the proteins were precipitated with 6 vol of acidified acetone. Protein concentrations were determined ad modum Lowry et al. [lo] using protamine sulfate (Merck) and bovine serum albumine (Sigma) as standards for protamine-like and non-histone proteins respectively.

Electrophoresis Proteins were fractionated by electrophoresis in acetic acid : urea polyacrylamide gels [I l] and by a two-dimensional electrophoresis using acetic acid : urea gels in the first direction and SDS slab gels [12] in the second. To this end, the gel from the first direction was washed at 4°C with several changes Exp Cell Res I52 (1984)

Sperm chromatin

structure

of a mollusc

233

ABCDE’ Fig. 1. Acetic acid : urea gel electrophoresis [l 11 of nuclear sperm proteins of Myths. A, total acidsoluble proteins; B, proteins extracted with 0.65 M NaCl; C, proteins remaining in the pellet after 0.65 M NaCI; D, proteins soluble in 5 % HClO.,; E, proteins precipitated by 5 % HClO+ Fig. 2. Two-dimensional electrophoresis of the total acid-soluble proteins of Myfilus sperm chromatin: A, first direction in acetic acid : urea polyacrylamide gels [ll]; B, second direction in SDScontaining polyacrylamide gels [ 121.

of a solution containing 4% SDS, 20% ethanol, 50 mM *is-HCl, pH 8.0, and was run in the second direction for 2 h at 4V/cm followed by 12 h at 8V/cm. The proteins were stained with Coomassie Brilliant Blue R 250 (Serva). Approximate molecular masses of the non-histone proteins were estimated using a standard mobility curve of proteins of known molecular mass.

Decondensation of the Sperm Nuclei for Electron Microscopy Demembranized sperm nuclei were dispersed in NaCl of different concentrations and incubated for 15 min at room temperature in the presence of 1 mM PMSF. They were centrifuged on freshly glowdischarged carbon-coated grids for 30 min at 0°C and 3 800 g in the microcentrifugation chamber [ 131 through 0.1 M sucrose containing the corresponding salt concentration. For each salt concentration one of the grids with the corresponding material was floated for 20 min at 37°C on a drop containing 50 EUlml of DNase I (Sigma) in 2 mM ‘Bis-HCl buffer, pH 7.6, 1 mM Mg*+, and then washed with distilled water. After staining with alcoholic uranyl acetate, the grids were rotary shadowed with W02. The exact magnification was determined with a Polaron grating replica.

RESULTS As seen in figs. 1 and 2 the acid-extractable proteins of the sperm chromatin of Mytilus galloprouincialis consist of four somatic-type core histones, three already Exp Cell

Res 152 (1984)

234 Avramova,

Zalensky

and Tsanev

106 60 . 60 . 40 20 0

3 1

2

MNaCl

Fig. 3. Dependence of the dissociation of the proteins of Mytilus sperm chromatin on the salt concentration. The amount of proteins extracted with 2 M NaCl is taken as 100%. Fig. 4. Electron micrograph of a Mytilus sperm nucleus spread in 0.4 M NaCl and treated on the grid with DNase I. Bar, 0.9 urn.

described basic proteins Sl, S2 and S3, and a non-identified protein X. As estimated from the area under the densitometric peaks of stained gels, their relative amount is roughly 50 % for S2, 20 % for S3, 6 % for Sl and 20% for the somatic-type histones. Due to the low solubility of Sl, S2 and S3 in SDS solutions, their electrophoretic patterns in SDS gels (fig. 2) do not reflect their real proportions. When demembranized sperm nuclei (1 AZ&ml) were treated with increasing concentrations of NaCl and the amount of dissociated protein plotted against salt concentration, the curve shown in fig. 3 was obtained. In 0.2 M NaCl practically no proteins were dissociated and the nuclei preserved their compact state. Raising the salt concentration to 0.4 M NaCl leads to a small but detectable dissociation of proteins (fig. 3) paralleled by a substantial increase (about 60 %) of the nuclear diameter, although as a whole the nucleus preserves its shape. Digestion of such partially decondensed nuclei with DNase I results in a further Fig. 5. Electron micrograph showing highly decondensed sperm chromatin spread in 0.65 M NaCI. Bar, 0.5 pm. Fig. 6. (A) Electron micrograph of spread residual structures remaining after extraction of sperm nuclei with 2 M NaCI-1 % Mton X-100; (B) SDS-polyacrylamide gel electrophoresis of the proteins obtained after digestion of the residual material from (A) with DNase I. The figures show approximate molecular masses in kD. (A) Bar, 0.22 urn. Exp

Cell

Res IS2 11984)

Sperm

chromatin

structure

of a mollusc

Exp Cell

235

Res 152 (1984)

236

Avramova,

Zalensky

and Tsanev

Fig. 7. Electron micrograph of the periphery of a sperm nucleus spread in distilled water. Inset shows an enlarged picture of the granules. Bar, 0.21 pm; inset, 55 nm.

increase of the nuclear volume, and a highly reproducible picture is obtained showing a fibrous DNase-resistant network (fig. 4). When the salt concentration is increased to 0.5-0.65 M NaCl the main protein which is extracted is S3 together with a small amount of S2 (fig. 1 B), while all core histones, the bulk of S2 and Sl remain insoluble (fig. 1 C). The electron microscopic picture of sperm nuclei treated with 0.65 M NaCl shows that at this salt concentration they lose their contours and the chromatin appears decondensed to a different degree, thus revealing some heterogeneity of the nuclei. The highly decondensed (0.65-I M NaCl) chromatin appears as a network of smooth fibers attached to compact electron-dense material (fig. 5). At still higher salt concentrations (l-2.5 M NaCl) and 1% Triton X-100 this material disaggregates into electron-dense, annular bodies of varying size, usually from 30 to 60 nm (fig. 6A). Repeated extractions of the sperm nuclei with 2 M NaCl and 1% Triton X100, both in the absence and in the presence of 20 mM 2-mercaptoethanol, do not destroy these structures. When after such a treatment the remaining material is extensively digested with DNase I, some non-histone proteins are released as an insoluble precipitate. On SDS electrophoresis they show three bands, with molecular masses in the range 60-80 kD (fig. 6B). These proteins are insoluble in 0.25 N HZS04 and their amount is 3-5 ug per 100 ug DNA. Exp Ceil Res 152 (1984)

Sperm chromatin

structure

of a mollusc

237

If the residual structures obtained by high salt-detergent treatment are digested with pronase all DNA remains in the supematant after the low-speed centrifugation used for spreading. When sperm nuclei are subjected to a hypotonic shock and spread in distilled water, an increase of the nuclar volume and a ‘blowing up’ of the DNA fibers takes place, although no detectable dissociation of protein is observed. Electron microscopic pictures of such nuclei reveal at their periphery radially dispersed fibers containing 20-25 nm large granules irregularly scattered (fig. 7).

DISCUSSION The high homogeneity of the material used in our experiments (more than 98 % mature spermatozoa) when compared with the rather high amount of somatictype histones (about 20%) practically excludes the possibility that they might originate from contaminating nuclei. This confirms literature data [ 1, 3, 41 that in the mature spermatozoa of Mytilus somatic histones ‘do coexist with the spermspecific proteins Sl, S2 and S3 which correspond to the proteins found in M. edulis and denoted as $3 [2], $1 [ 1, 31 and #2b [14] respectively. Our unpublished experiments have shown that M. galloprouincialis has sperm-chromosomal proteins practically identical with those of M. edulis and Crenomytilus grayanus and thus confirm the view [l-4, 141 that S 1, S2 and S3 can be considered as spermspecific for the Mytilus family. The two proteins S 1 and S2 have been well characterized [l-4]. Conclusions on the nature of S3 are still contradictory. On the basis of its total amino acid composition it has been related to the somatic H2b (and hence, its detonation $2b [14]). However, there are strong indications that S3 is a Hl-type protein. It is soluble in 5 % HC104 (fig. 1 D) and is preferentially dissociated from chromatin at about 0.5 M NaCl. In addition, bromosuccinimide and trypsin degradation of the corresponding protein isolated from C. grayanus has shown a similarity with Hl [15]. In agreement with these results the two-dimensional electrophoresis (fig. 2) clearly shows that S3 is a protein different from a core histone. The curve reflecting the salt-induced dissociation of total protein from Mytilus sperm chromatin is intermediate between that of the dissociation of histones [ 161 and that of protamines [7, 171. This suggests that some of the sperm-specific proteins interact with DNA in the mussel sperm nuclei in a manner analogous to the interaction of protamines with DNA in the trout sperm nucleus. Such a suggestion is supported by the following findings: (1) the #l protein (corresponding to S2) conformationally resembles a protamine [18]; (2) its complexes with DNA behave as typical nucleoprotamines [193; (3) the residual structures after DNase I treatment of partially decondensed nuclei (fig. 4) are very similar to the corresponding structures of trout sperm nuclei 181. The initially highly condensed chromatin is rapidly decondensed when the Exp Cell Res 152 (1984)

238

Avramova,

Zalensky

and Tsanev

sperm-specific proteins are dissociated (fig. 5). This is especially true for protein S3 which is the first to dissociate in the interval of 0.4-0.65 M NaCl (fig. I B). On the basis of all these data it may be suggested that Sl and S2 probably play a protamine-like role of complexing with DNA for its protection and partial condensation, while S3 is involved in the final high compaction of the Myrilus sperm nucleus. When some sperm-specific proteins are dissociated at about 0.65 M NaCl, the core histones still remain (fig. 1 C) but no nucleosomes could be seen in the spread nuclei (fig. 6). The same has been observed when mitotic chromosomes have been extracted with 0.6 M NaCl [20]. A possible explanation may be that nucleosomes are easily destroyed when unfixed material is spread for electron microscopy [2 11. Unlike the salt-treated material, many discrete granules larger than nucleosomes are seen dispersed along the DNA fibers when demembranized sperm nuclei are spread for electron microscopy in distilled water (fig. 7). These large granules closely resemble the superbeads consisting of several nucleosomes which have been observed earlier in somatic nuclei [22-241, in sea urchin and in H. tubulosa sperm nuclei [25, 261 after hypotonic treatments. The possibility that the granules seen in water-spread Mytilus sperm nuclei consist of nucleosomes is supported by preliminary results showing that digestion of water-dispersed sperm chromatin with micrococcal nuclease gives mono-, di- and trisomes. An important finding in this study is that upon extraction of all sperm-specific proteins and core histones with high-salt and detergent treatment, the residual material still contains three proteins with molecular masses of 60, 67, and 77 kD (fig. 6B). The strong attachment of DNA to these proteins leads to the formation of a network which makes possible the sedimentation of this material under lowspeed centrifugation. When sedimented, this network represents fibers attached to electron-dense material (fig. 5) of proteinaceous nature as indicated by its digestion with pronase. The formation of these large masses of condensed material is most probably an artifact of precipitation and aggregation. Higher salt concentrations in the presence of Triton X-100 cause the disaggregation of this material into annular bodies of varying size, usually between 30 and 60 nm (fig. 6A), to which several DNA fibers are attached. These structures are very stable, resisting further high salt-detergent treatment and are very similar to those observed in ram [6] and in trout [8] sperm chromatin. In all these three species the residual nuclear bodies in the sperm have the same characteristics: (1) they are resistant to high salt and detergent treatment (even in the presence of 2-mercaptoethanol); (2) several DNA fibers are radially attached to them; (3) they have an annular granular structure varying in size between 30 and 70 nm; (4) they contain three proteins with molecular masses in the range 6O-gO kD; (5) in response to increased salt concentration they show a tendency to aggregate into dense masses from which DNA fibers emerge. It is interesting to note that under the same high salt-detergent-DNase I Exp CeNRes

152 (1984)

Sperm

chromatin

structure

of a mollusc

239

treatment, DNA-associated granules (25-30 nm large) have been isolated from nuclei and chromosomes of somatic cells [27]. When bound together, these granules formed larger structures and were considered to be a basic element of the nuclear matrix, chromosome scaffold, and nuclear envelope. The granular structure and the variable size of the annular bodies observed by us in the sperm nuclei of ram [6], trout [8] and mussel suggest that these structures may be formed by the same granules. The finding of similar structures in the sperm nuclei of species so widely separated in evolution indicates that they may be important universal elements which can serve the role of attachment sites organizing DNA into separate domains and thus transmitting structural information to the progeny. The authors are grateful to Drs Konsulovs from the Institute of Marine Products and Resources in Vama for their help with the supply of mussels.

REFERENCES 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27.

Subirana, J A, Cozcolluela, C, Palau, J & Unzeta, M, Biochim biophys acta 317 (1973) 364. Phelan, J J, Colom, J, Cozcolluela, C, Subirana, J A & Cole, R D, J biol them 249 (1974) 1099. Colom, J & Subirana, J A, Biochim biophys acta 581 (1979) 217. Zalensky, A 0 & Zalenskaya, I A, Comp biochem physiol66B (1980) 415. Avramova, 2, Dessev, G & Tsanev, R, FEBS lett 118 (1980) 58. Tsanev, R & Avramova, Z, Eur j cell biol 24 (1981) 139. Avramova, Z, Uschewa, A, Stephanova, E & Tsanev, R, Eur j cell biol31 (1983) 137. Tsanev, R & Avramova, Z, Eur j cell biol 31 (1983) 143. Longo, F J & Domfeld, E J, J ultrastruct res 20 (1967) 462. Lowry, 0 H, Rosebrough, N J, Farr, A L & Randall, R J, J biol them 193 (1951) 265. Panyim, S & Chalkley, R, Arch biochem biophys 130 (1969) 337. Laemmli, U K, Nature 227 (1970) 680. Miller, 0 L, & Bakken, A H, Acta endocrinol, suppl. 168 (1972) 155. Ausi6, J & Subirana, J A, Exp cell res 141 (1982) 39. Odintsova, T I, Ermochina, T M & Krasheninnikov, I A, Biokhimya 47 (1982) 1532. Vassilev, L, Russev, G & Tsanev, R, Int j biochem 13 (1981) 1247. Marushige, K & Dixon, G H, J biol them 246 (1971) 5799. Puigdomenech, P, Cabre, 0, Palau, J, Bradbury, E M & Crane-Robinson, C, Eur j biochem 59 (1975) 237. Subirana, J A & Puigianer, L C, Conformation of biological molecules and polymers (ed E D Bergan & B Pullman) p. 645. The Jerusalem symposia on quantum chemistry and biochemistry. V. The Israel Academy of Sciences and Humanities, Jerusalem (1973). Hadlaczky, G, Sumner, A T & Ross, A, Chromosoma 81 (1981) 537. Thoma, F, Koller, Th & Klug, A, J cell biol 83 (1979) 403. Renz, M, Nehls, P & Hozier, J, Cold Spring Harbor symp quant bio142 (1978) 245. Kiryanov, G I, Manamshjan, T A, Polyakov, V Yu, Fais, D & Chentsov, Ju S, FEBS lett 67 (1976) 323. Olins, A L, Cold Spring Harbor symp quant bio142 (1978) 325. Zentgraf, H, Miiller, U & Franke, W W, Eur j cell biol 20 (1980) 254. Subirana, J A, Mutioz-Guerra, S, Martinez, A B, Perez-Grau, L, Marcet, X & Fita, I, Chromosoma 83 (1981) 455. Engelhardt, P, Plagens, U, Zbarsky, I B & Filatova, L S, Proc natl acad sci US 79 (1982) 6937.

Received November 29, 1983

16-848335 Printed

in Sweden

Exp Cell

Res 152 (1984)