Chromosoma (1990) 100:32-36

CHROMOSOMA 9 Springer-Verlag 1990

The centromere specific histone CENP-A is selectively retained in discrete foci in mammalian sperm nuclei Douglas K. Palmer, Kathleen O'Day, and Robert L. Margolis Fred Hutchinson Cancer Research Center, 1124 Columbia Street, Seattle, WA 98104, USA Received June 15, 1990 Accepted July 10, 1990 by W.C. Earnshaw

Abstract. The 17 kDa human autoantigen designated CENP-A is a centromere specific histone. We show here that CENP-A is present in tissue of bovine origin, and that it is quantitatively retained in mature spermatozoa. This result is striking, as a prominent feature of spermatogenesis in mammals is the replacement of most somatic and testes specific histones with protamines. Indirect immunofluorescence studies further show that CENP-A is retained in sperm nuclei in discrete foci, rather than being dispersed throughout the sperm head. These observations suggest that CENP-A is a functionally important component of centromeres, and that pre-existing C E N P - A : D N A interactions are likely to be important in organizing the centromeres of the paternal genome during early embryogenesis.

Introduction During spermatogenesis in mammals, somatic histones are ultimately replaced by cysteine- and arginine-rich sperm specific peptides called protamines, a process resulting in loss of the nucleosomal organization of chromatin and its replacement by a highly compacted structure characteristic of mature sperm. Details of this process vary depending on the histone considered, and on the organism (reviewed by Grimes 1986). In mice and rams replacement of histones is nearly complete, although a small subset of histones remains (O'Brien and Bellv~ 1980; Moss etal. 1989; Uschewa etal. 1982), whereas in humans approximately 15% of the somatic and testes specific histones are retained (Tamphaichitr et al. 1978; Gatewood et al. 1987). In humans, regions of D N A retained as nucleohistone are distributed nonrandomly with respect to D N A sequence (Gatewood et al. 1987), suggesting that the nucleohistone-nucleoprotamine distinction may have important consequences Offprint requests to: D.K. Palmer

for expression of the paternal genome in early embryogenesis. Anticentromere antibodies present in sera from patients with CREST scleroderma have enabled the identification of several centromere specific proteins in mammalian cells (Guldner et al. 1984; Earnshaw et al. 1984; Earnshaw and Rothfield 1985; Valdivia and Brinkley 1985; Kingwell and Rattner 1987; Hadlaczky etal. 1989). Human CREST autoantigens with Mrs as determined by SDS-polyacrylamide gel electrophoresis of 17000, 80000, and 140000 have been designated CENPA, CENP-B, and CENP-C, respectively (Earnshaw and Rothfield 1985). Centromere specific human autoantigen(s) of Mr 50000-60000 have also been reported (Kingwell and Rattner 1987; Hadlaczky et al. 1989). We have demonstrated that the Mr 17000 CREST autoantigen CENP-A is an acid soluble component of highly purified nucleosome core particles prepared from HeLa cells (Palmer et al. 1987). Further, CENP-A elutes with H3 and H4 from both ion exchange (Palmer et al. 1987) and gel exclusion columns (D.K. Palmer, K. O'Day, and R.L. Margolis, unpublished observations) under non-denaturing conditions. CENP-A is also present in Indian muntjac cells (Hadlaczky etal. 1986; Kingwell and Rattner 1987). Antibodies affinity purified from protein blots of the Indian muntjac CENP-A bind specifically to the kinetochore regions of Indian muntjac chromosomes (Kingwell and Rattner 1987). Therefore, CENP-A appears to function as a centromere specific, and possibly a kinetochore specific, core histone. Here, we demonstrate that CENP-A is quantitatively retained during spermatogenesis in bulls. Indirect immunofluorescence studies further show that CENP-A is retained in sperm nuclei in discrete foci, rather than being dispersed throughout the sperm head. These observations suggest that CENP-A is a functionally important component of centromeres, and that pre-existing CENPA: D N A interactions are likely to be important in organizing the centromeres of the paternal genome during early embryogenesis.

33

Materials and methods Cells, tissues, and purification of nuclei. HeLa cells were cultured as monolayers in Dulbecco's Modified Eagle's medium (Gibco Laboratories). Human cells designated "ManCa", derived from a patient with non-Hodgkin's lymphoma (Nishikori et al. 1984), were cultured in spinner flasks in RPMI 1640 medium (Whittaker Bioproducts). Both media contained 5% bovine calf serum (Hyclone Laboratories). Indian muntjac cells, kindly provided by Dr. J.B. Rattner, were grown as monolayers in Dulbecco's Modified Eagle's medium containing 10 % fetal calf serum (Hyclone Laboratories). Cells were collected by scraping or by centrifugation as appropriate, and were washed with PBS (136 mM NaC1, 2 mM KC1, 10.6 mM NazPO4, 1.5 mM KH2PO4, pH 7.4). Nuclei were prepared as described previously for HeLa cells (Palmer et al. 1987), using 0.2% Triton X-100 in buffer A [60 mM KC1, 15 mM NaC1, 0.15raM spermine, 0.5 mM spermidine, 1.0 mM EDTA, 0.2 mM EGTA, 30% (w/v) glycerol, 15 mM fi-mercaptoethanol, 20 gM phenylmethylsulfonylfluoride, 15 mM Tris-HC1, pH 7.4] to lyse cells, and buffer A for subsequent washes. Calf thymus nuclei were purified from frozen tissue, which was thawed, minced, and homogenized (Waring blendor) in buffer A. Crude nuclei were collected by centrifugation, and further purified using 0.2% Triton X-100 in buffer A as for human tissue culture cell nuclei. Bull spermatozoa were obtained by perfusion of fresh epididymides as described by Gagnon (1986), except that 10 mM Tris-HC1, 150 mM NaC1, pH 7.4 was substituted for mineral oil. Sperm nuclei were purified using alkylammonium bromides as described by Balhorn et al. (1977) and Gatwood et al. (1987). Purified nuclei were stored in buffer A at - 7 0 ~ C. The DNA content of the nuclear preparations was estimated from the absorbance at 260 nm of acid hydrolysates (1 M NaC1, 1 M perchloric acid; 20 rain, 100~ C). Indirect immunofluorescence. Bull sperm nuclei were incubated for 60 min at 22~ and centrifuged onto glass coverslips in 10 mM Tris-HC1, pH 8 containing 0.2 M NaC1, 2.5 M urea, and 10 mM dithiothreitol. Coverslips were washed with PBS, fixed in PBS containing 2% paraformaldehyde for 20 min at 22~ C, washed with PBS, and probed with whole anticentromere serum GD diluted 500-fold in PBS containing 0.05% Tween 20 and 0.05% sodium azide (PBS/Tween/azide), which was also used as a wash buffer. The secondary probe was affinity purified fluorescein isothiocyanate (FITC)-conjugated goat anti-human immunoglobulin (Tago). DNA was stained with 0.25 pg/ml propidium iodide in PBS. Coverslips were mounted using a medium containing 1,4-diazabicyclo(2.2.2)octane (Johnson et al. 1982), and were observed using an MRC-500 Laser Scanning Confocal Microscope (BioRad Microscience, Cambridge, Mass.) with a Nikon Optiphot (Nikon, Torrance, Calif.). Photographs were taken on TMAX 100 film (Kodak) at f5.6 for 1 s.

added to stop the digestions, resulting in final concentrations of SDS and urea of 3% and 2 M, respectively. Following sonication and heating at 100~ for 3 rain, samples were electrophoresed in 75 mm thick, 10% polyacrylamide-SDS gels (Laemmli 1970). Proteins were visualized by staining with Coomassie blue, or by immunoblotting.

Immunoblotting. Proteins resolved in Triton-acid-urea polyacrylamide gels were electrophoretically transferred to polyvinylidene difluoride (PVDF) membrane (Immobilon P, Millipore), using 0.7% acetic acid as the transfer buffer. Proteins resolved by SDSpolyacrylamide gel electrophoresis were transferred to nitrocellulose (BA 83, Schleicher and Schuell), using Tris-gtycine-methanol (Towbin et al. 1979). In both cases transfer was for 90 rain at 60 V. The protein blots were blocked with 2% non-fat dry milk in PBS/ Tween/azide, and were probed using anticentromere serum GD, and [lZS]protein A (NEN) as the secondary reagent. To remove unbound primary antibodies or protein A, blots were washed sequentially with PBS/Tween/azide, PBS/2 M urea, PBS/2 M NaC1, and PBS/Tween/azide. Autoradiography was performed using Kodak XAR-5 film and intensifying screens.

Results C E N P - A is p r e s e n t w i t h b u l k h i s t o n e s in a c i d extracts o f c a l f t h y m u s nuclei. T h e b o v i n e p r o t e i n m i g r a t e s to the s a m e p o s i t i o n as h u m a n a n d I n d i a n m u n t j a c C E N P A in T r i t o n - a c i d - u r e a p o l y a c r y l a m i d e gels, a n d is recognized b y a n t i b o d i e s p r e s e n t in h u m a n a n t i c e n t r o m e r e s e r u m (Fig. 1). A p r o t e i n w h i c h b i n d s a n t i b o d i e s b u t w h i c h m i g r a t e s m o r e slowly t h a n the b u l k o f the C E N P A is also p r e s e n t in the c a l f t h y m u s a c i d extract. This p r o t e i n r e p r e s e n t s d i m e r i c C E N P - A ( D . K . P a l m e r , K. O ' D a y , H. L e T r o n g , H. C h a r b o n n e a u , R . L . M a r g o l i s , m a n u s c r i p t in p r e p a r a t i o n ) . B o v i n e C E N P - A also m i -

Acid extraction of proteins and Triton-acid-urea polyacrylamide gel electrophoresis. Nuclei were dissociated with 1 M NaC1, which was then made 0.25 M in HC1 and centrifuged to remove insoluble material. Extracted proteins were precipitated with trichloroacetic acid, dissolved in Triton-acid-urea sample buffer, and resolved in gels containing 12% acrylamide (140:1 acrylamide:bis-acrylamide), 6.5 M urea, 5% acetic acid, and 0.38% (w/v) Triton X-100 (Zweidler 1978). Proteins were visualized by staining with Coomassic Brilliant Blue R, or by immunoblotting. In one experiment, a commercially available acid extract of calf thymus nuclei (Histone, type II, Sigma) was used as the sample.

SDS and urea extraction of proteins, and SDS-polyacrylamide gel electrophoresis. Calf thymus nuclei and bull sperm nuclei in 10 mM Tris-HC1, 1 mM CaC1;, 15 mM fl-mercaptoethanol, pH 7.5, were digested with 200 units/ml micrococcal nuclease for 30 min at 37~ C. An equal volume of 2 x SDS gel application buffer was

Fig. 1. Triton-acid-urea polyacrylamide gel electrophoresis and immunoblotting comparison of acid soluble human, Indian muntjac, and calf thymus nuclear proteins. Samples containing approximately 50 gg of acid extracted nuclear protein were electrophoresed in Triton-acid-urea polyacrylamide gels. For immunoblotting, proteins were transferred to polyvinylidene difluoride membrane. M ManCa cells (human lymphoma); H HeLa cells; I Indian muntjac; C calf thymus (Histone, type IIA, Sigma). Left-hand panel Coomassie blue stained gel. Right-hand panel Autoradiogram of the corresponding immunoblot

34

Fig. 2. Coomassie blue stained gel (left-hand panel), and the corresponding immunoblot (right-handpanel) of acid soluble calf thymus nuclear proteins (73, and bull sperm nuclear proteins (S) resolved by Triton-acid-urea polyacrylamide gel electrophoresis. All loadings represented 0.5 260 nm absorbance units of nuclei grates similarly to human and muntjac C E N P - A in SDSpolyacrylamide gels (Fig. 4, and data not shown), and like human C E N P - A (Palmer et al. 1989) elutes at 60% acetonitrile from a Cls reverse phase column, when a gradient o f acetonitrile in 0.3% trifluoroacetic acid is used to elute bound proteins (D.K. Palmer et al., manuscript in preparation). A comparison by Triton-acid-urea polyacrylamide gel electrophoresis and immunoblotting of acid soluble proteins from calf thymus nuclei and bull sperm nuclei reveals that C E N P - A is present in both types of nuclei in comparable amounts relative to D N A (Fig. 2). This is not true for the other histones, which are not detectable on the Coomassie blue stained gels of acid extracted bull sperm nuclear protein. Using reverse phase chromatography to compare acid soluble proteins from calf thymus and bull sperm nuclei, we estimate that no more than 1% o f somatic histones are retained in the bull sperm heads (D.K. Palmer et al., manuscript in preparation). Using similar methods, we have also found quantitative retention of C E N P - A in human sperm nuclei, where approximately 85% of the somatic and testes specific histones are replaced with protamines (data not shown). Indirect immunofluorescence techniques reveal punctate foci in bull sperm nuclei when anticentromere serum

Fig. 3A, B. Indirect immunofluorescence visualization of centromere foci only A, and centromere foci plus DNA B in purified bull sperm nuclei. Bar represents 5 gm

Fig. 4A, B. Immunoblotting analysis of proteins solubilized with SDS and urea from (/4) human tissue culture cell nuclei (ManCa cell nuclei), (73 calf thymus nuclei, (S) bull sperm nuclei. The lefthand panel shows the Coomassie blue stained gel, the right-hand panel shows the corresponding protein blot probed with serum GD and [lZSI]protein A. All loadings represent 0.5 260 nm absorbance units of nuclei is used as the primary detection reagent (Fig. 3), but no such foci are observed when a variety of nonimmune or irrelevant autoimmune sera are substituted for anticentromere serum. Depending on the conditions of salt and urea used to treat the nuclei prior to fixation, we

35 detect beaded substructures within the punctate loci in some images (Fig. 3), and occasionally, centromeric chromatin has dispersed slightly beyond the nuclear periphery and become adsorbed to the coverslip (not shown). As judged by immunoblotting of nuclear proteins resolved on SDS-polyacrylamide gels, the whole C R E S T serum we have used for the work presented here contains antibodies that bind to the human centromere specific protein CENP-B, in addition to CENP-A. However, using this serum in immunoblotting procedures we detect no proteins other than C E N P - A among whole calf thymus or bull sperm nuclear proteins solubilized with urea and SDS (Fig. 4). We conclude the C E N P - A is retained in discrete foci in bull sperm nuclei.

Discussion

C E N P - A appears to be a centromere specific core histone. Thus, C E N P - A is extracted from chromatin by histone extraction procedures, and it elutes with H 3 and H 4 in apparent tetrameric complexes during ion exchange or sizing column chromatography o f histones under nondenaturing conditions (Palmer et al. 1987 and our unpublished observations). Further, it is a component of highly purified nucleosome core particles prepared from H e L a cell nuclei (Palmer et al. 1987). We have also recently found that the N-terminal half of a 12 amino acid chrymotryptic fragment of C E N P - A is identical to a sequence near the carboxy-terminus of bovine H 3 (D.K. Palmer et al., manuscript in preparation). Spermatogenesis in placental mammals is characterized by the replacement of most histones with protamines (Grimes 1986). Although the degree of replacement varies among different species, in most mammals the portion of the genome that remains as nucleohistone is < 2 % (O'Brian and Bellive 1980; Balhorn 1982; Uschewa et al. 1982; Moss et al. 1989). It is therefore striking that C E N P - A is quantitatively retained during spermatogenesis in bulls, where replacement of somatic histones is essentially complete (Fig. 2). Further, our indirect immunotluorescence studies indicate that CENPA is retained in sperm nuclei in discrete foci, suggesting its continued association with centromere specific D N A sequences. If the presence of C E N P - A in foci represents its retention in centromeric heterochromatin domains, centromeric satellite D N A should also be retained in foci. The results o f Moens and Pearlman (1989), which demonstrate the presence of centromeric satellite D N A in foci in rat sperm, are consistent with this prediction. Others have demonstrated the retention of centromeric elements in sperm nuclei by indirect immunofluorescence (Brinkley et al. 1986; Sumner 1987; H a a f et al. 1990) and immunoblotting techniques (Brinkley et al. 1986). To date, these elements have not been identified with proteins o f known function. Our data identify a centromere specific histone as a retained element, in striking contrast to the near total replacement of the major somatic cell histories. These results suggest that C E N P - A is a functionally important component of cen-

tromeres, and that pre-existing C E N P - A : D N A interactions are likely to be important in maintaining the organization o f the centromeres of the paternal genome for early embryogenesis. The findings presented here offer the possibility of exploring questions o f substantial interest. One consequence has already reached fruition. The absence o f somatic histones from bull sperm nuclei has enabled us to purify C E N P - A to homogeneity (D.K. Palmer et al., manuscript in preparation), permitting further analysis of its primary structure and role(s) in chromatin organization. Further, whereas protamine associated D N A is resistant to nuclease digestion, D N A associated with histone is extractable from sperm (Tamphaichitr et al. 1982; Gatewood et al. 1987). Using nuclease or footprinting cleavage techniques, we should now be able selectively to excise C E N P - A and its associated D N A from sperm. This tissue may therefore be uniquely well suited as starting material for isolation of C E N P - A associated D N A sequences and analysis of the molecular organization o f the centromere. The findings presented here also raise new questions. It will be of interest to determine, for instance, the mechanisms which are used to exclude C E N P - A from replacement with protamines in sperm, whether other core histones are retained with C E N P - A in sperm centromeric chromatin, and if C E N P - A containing chromatin in sperm has a typical nucleosomal organization. References

Balhorn R, Gledhill BL, Wyrobek AJ (i 977) Mouse sperm chromatin proteins: quantitative isolation and partial characterization. Biochemistry 16: 4074-4080 Balhorn R (1982) A model for the structure of chromatin in mammalian sperm. J Cell Biol 93 : 298-305 Brinkley BR, Brenner SL, Hall JM, Tousson A, Balczon RD, Valdivia MM (1986) Arrangements of kinetochores in mouse cells during meiosis and spermiogenesis. Chromosoma (Berl) 94:309-317 Earnshaw WC, Rothfield N (1985) Identification of a family of human centromere proteins using autoimmune sera from patients with scleroderma. Chromosoma (Berl) 91 : 313-321 Earnshaw WC, Halligan N, Cooke C, Rothfield N (1984) The kinetochore is part of the metaphase chromosome scaffold. J Cell Biol 98:352-357 Gagnon C (1986) Extraction and properties of dyneins from bull spermatozoa. Methods Enzymol 134: 318-324 Gatewood JM, Cook GR, Balhorn R, Bradbury EM, Schmid CW (1987) Sequence-specific packaging of DNA in human sperm chromatin. Science 236: 962-964 Grimes SR Jr (1986) Nuclear proteins in spermatogenesis. Comp Biochem Physiol 83B :495-500 Guldner H-H, Lakomek HJ, Bautz FA (1984) Human anti-centromere sera recognize a 19.5 kD non-histone chromosomal protein in HeLa cells. Clin Exp Immunol 58:13-20 Haaf T, Grunenberg H, Schmid M (1990) Paired arrangement of non-homologous centromeres during vertebrate spermiogenesis. Exp Cell Res 187:157-161 Hadlaczky GY, Went M, Rugertz NR (1986) Direct evidence for the non-random localization of mammalian chromosomes in the interphase nucleus. Exp Cell Res 167:1-15 Hadlaczky G, Praznovszky T, Rasko I, Kereso J (1989) Centromere proteins I. Mitosis specific centromere antigen recognized by anti-centromere antibodies. Chromosoma 97:282-288 Johnson GD, Davidson RS, McNamee KC, Russel G, Goodwin

36 D, Holoborow EJ (1982) Fading of immunofluorescenceduring microscopy: a study of the phenomenon and its remedy. J Immunol Methods 55 : 231-242 Kingwell B, Rattner JB (1987) Mammalian kinetochore/centromere composition: A 50 kDa antigen is present in the mammalian kinetochore/centromere. Chromosoma 95 : 403-407 Laemmli UK (1970) Cleavage of structural proteins during the assembly of the head of bacteriophage T 4. Nature 227:149-154 Moens PB, Pearlman RE (1989) Satellite DNA I in chromatin loops of rat pachytene chromosomes and in spermatids. Chromosoma 98: 287-294 Moss SB, Challoner PB, Groudine M (1989) Expression of a novel histone 2B during mouse spermiogenesis. Dev Biol 133:83-92 Nishikori M, Hansen H, Jhanwar S, Fried J, Sordillo P, Koziner B, Lloyd K, Clarkson B (1984') Establishment of a B-cell lymphoma line with duplication of the 8;14 translocation. Cancer Genet Cytogenet 12:39-50 O'Brien DA, Bellv+ AR (1980) Protein constituents of the mouse spermatozoan. Dev Biol 75:386-404 Palmer DK, O'Day K, Wener MH, Andrews BS, Margolis RL (1987) A 17-kD centromere protein (CENP-A) copurifies with nucleosome core particles and with histones. J Cell Biol 104:805-815 Palmer DK, O'Day KO, Margolis RL (1989) Biochemical analysis

of CENP-A, a centromeric protein with histone-like properties. In: Resnick M (ed) Mechanisms of aneuploidy. Liss, New York, pp 61-72 Sumner AT (1987) Immunocytochemical demonstration of kinetochores in human sperm heads. Exp Cell Res 171:250.253 Tamphaichitr N, Sobhon P, Taluppeth N, Chalermisarachai P (1978) Basic nuclear proteins in testicular cells and ejaculated spermatozoa in man. Exp Cell Res 117:347-356 Tamphaichitr N, Sobhon P, Chalermisarachai P, Chutatape C (1982) Biochemical and ultrastructural characterization of nucleoprotamine in human sperm heads treated with micrococcal nuclease and salt. Gamete Res 6:235-255 Towbin H, Staehelin T, Gordon J (1979) Electrophoretic transfer of proteins from polyacrylamide gels to nitrocellulose sheets: procedure and some applications. Proc Natl Acad Sci USA 76:4350M354 Uschewa A, Auramova Z, Tsanev R (1982) Tightly bound somatic histones in mature ram sperm nuclei. FEBS Lett 138:50-54 Valdivia MM, Brinkley BR (1985) Fractionation and initial characterization of the kinetochore from mammalian metaphase chromosomes. J Cell Biol 101:1124-1134 Zweidler A (1978) Resolution of histones by polyacylamide gel electrophoresis in presence of monionic detergents. Methods Cell Biol 17: 223-233