Original article

Immunoelectron microscopic visualizationof intermediate basic proteins HPI1 and HPI2 in

human spermatids and spermatozoa

Yann Prigenta Frédéric Troalenb Jean-Pierre Dadoune

a Groupe d’étude de la formation et de la maturation du gamète mâle, Laboratoire d’histologie, JEMENESR 349, UFR Biomédicale des Saints-Pères, 75280 Paris cedex 06, Franceb Unité d’immunochimie, Institut Gustave-Roussy, 39, rue Camille-Desmoulins,

94805 Villejuif, France

(Received 2 March 1998; accepted 15 June 1998)

Abstract - In humans, intermediate basic proteins HPII and HPL are considered as common precursorsof the P2 protamine family, according to data provided by structural studies of these proteins. The occur-rence and fate of proteins HPI, and HP12 were investigated in nuclei of human spermatids and sper-matozoa by means of immunoelectron microscopy. A specific polyclonal antibody against a syntheticpeptide overlapping the N-terminus of HPI1 and HPL was prepared and used to detect these proteinson sections of testis and ejaculated sperm. A quantitative analysis of labelling density was performedon micrographs using an interactive image analysis system. The first signs of labelling of intermediatebasic proteins appeared in spermatid nuclei at steps 4-5 of spermiogenesis, i.e. during the chroma-tin condensation process. The nuclear labelling density strongly increased in elongating spermatids(steps 5 and 6) and then sharply decreased from step 6 to step 8 of spermiogenesis. However, weaklabelling persisted in the nuclei of mature spermatids and ejaculated spermatozoa. The present resultsshow that the intermediate basic proteins HPI! and HPL are synthesized in large amounts in humanspermatids during elongation phase and disappear almost totally in mature spermatids when depositionof protamines is completed in condensed nuclei. © Inra/Elsevier, Paris

intermediate basic proteins / protamines / male gamete / spermiogenesis / immunocytochemistry

Résumé - Mise en évidence par immunocytochimie ultrastructurale des protéines basiquesintermédiaires HPh et HP12 dans les spermatides et les spermatozoïdes humains. Chez l’homme,les protéines basiques intermédiaires HPII et HPL sont considérées comme les précurseurs com-muns des protamines P2, selon les données fournies par les études structurales de ces protéines. La

1 mise en place et le devenir des protéines HPII et HP12 ont été étudiées dans les noyaux des sperma-

* Correspondence and reprintsE-mail: Jean-Pierre Dadoune@tnn.ap-hop-paris.fr

tides et des spermatozoïdes humains par immunocytochimie ultrastructurale. Un anticorps polyclo-nal spécifique dirigé contre un peptide de synthèse recouvrant le domaine N-terminal des protéinesHPII et HP12 a été préparé et utilisé pour détecter ces protéines sur des coupes de testicule et despermatozoïdes éjaculés. Une analyse quantitative de la densité du marquage a été effectuée sur lesmicrographies à l’aide d’un système d’analyse d’image interactif. Le marquage des protéines inter-médiaires basiques se manifeste dans les noyaux des spermatides aux stades 4-5 de la spermiogenèse,durant la condensation de la chromatine. La densité du marquage nucléaire augmente fortementdans les spermatides en élongation (stades 5 et 6) et décroît ensuite brusquement du stade 6 au stade8 de la spermiogenèse. Cependant, un faible marquage persiste dans les noyaux des spermatidesmatures et des spermatozoïdes éjaculés. Ces résultats montrent que les protéines basiques intermé-diaires HPh et HP12 sont synthétisées en grande quantité dans les spermatides humaines durant la phased’élongation et disparaissent presque totalement dans les spermatides matures, au terme de la mise enplace des protamines dans les noyaux condensés. © Inra/Elsevier, Paris

protéines basiques intermédiaires / protamines / gamète mâle / spermiogenèse / immunocyto-chimie

1. INTRODUCTION

In mammals, the spermatid nucleus iscompletely reorganized during mid-sper-miogenesis. Concomitantly with structuralchanges in the spermatid chromatin, his-tones are replaced by protamines [21]. Inman, histone-to-protamine replacementoccurs at the beginning of the spermatidmaturation phase [22]. Human sperm nucleicontain two families of protamines, P, andP2, which differ in their molecular mass,amino acid composition and sequence. ThePi family is represented by HP, (50 resi-dues) with different degrees of phosphory-lation, and is related to other mammalian

P, protamines. Protamines HPZ (57 resi-dues), HP3 (54 residues) and HP4 (58 resi-dues) belong to the PZ family. All three pro-teins contain high amounts of arginine,cysteine and histidine, and differ only intheir N-terminal sequence [3, 4, 10, 13, 20].Protamines Pj and PZ appear in human sper-matid nuclei at steps 4-5 of spermiogenesis.[17, 18, 22]. However, it can be inferredfrom biochemical findings in the mouse [6,9, 12, 33] and the description of genomicsequences of both protamine genes in man[10, 11] that protamine P, is synthesized asa mature protein, whereas protamine PZ 2derives from a precursor form.

Four basic proteins of intermediate sizebetween histones and protamines (HPILHP12, HPS1, HPS2 ) are also present in lowamounts in testis and human sperm nuclei,and they are immunologically related to PZ 2protamines [23]. Intermediate basic proteinsthe sequences of which are now knownshare a common C-terminal domain iden-tical to the amino acid sequence of prota-mines HP2, HP3 and HP4 [1, 19, 25]. Fromthese structural studies, it can be assumedthat protein HPI1, the largest of the inter-mediate basic proteins (101 residues), mayrepresent a common precursor of humansperm PZ protamines. However, the locali-zation and expression pattern of putative Pz 2family precursors during spermiogenesishave not been described.

The aim of the present work was to inves-tigate the occurrence of intermediate basicproteins HPI1 and HP12 during human sper-miogenesis and to follow their fine structu-ral distribution in spermatids and sperma-tozoa by means of a specific purified peptideantibody.

2. MATERIALS AND METHODS

2.1. Peptide synthesis and purification

A cysteinyl peptide (HD3) was synthesized.Its amino acid sequence corresponded to the

sequence of 15 residues located at the N-terminaldomain of the human sperm intermediate pro-tein HPI1 (sequence 17-31). Acysteine residuewas added at the C-terminus of the peptidesequence for coupling to a carrier protein (figure I ).

The 16 amino acid HD3 peptide was preparedaccording to standard Fmoc procedures on anautomatic solid-phase synthesizer and cleavedfrom the solid support by trifluoroacetic acidtreatment. The crude peptide was purified byhigh performance liquid chromatography. Peptidepurity was estimated at higher than 90 % by ana-lytical HPLC and by amino acid analysis.

2.2. Peptide-carrier conjugation

The carrier was keyhole limpet hemocyanin(KLH: Pierce product, Ref.: 77100) because of itslarge molecular mass (MW: 6.5 x 106). A hete-robifunctional cross-linker, (sulfo-MBS: m-maleimidobenzoyl-N-hydroxysulfosuccinimideester, Pierce product, Ref.: 22312) was chosento react toward carrier primary amines and pep-tide sulfhydryls. The carrier was first activated byconjugating to the active ester end of the cross-linker sulfo-MBS via lysine amino groups (mole-cular ratio: linker/carrier = 4 000, reaction buffer:

potassium phosphate 20 mM pH: 7.0, total reac-tion volume: 1.5 mL, reaction time: 30 min atroom temperature with constant stirring). Thisactivated carrier was then isolated by gel filtrationon a PD1 pre-packed column (Pharmacia Bio-tech, Ref.: 17-085 1 -0 1 ) to remove excess rea-gents and possessed, via cross-linker molecules,a large number of reactive maleimide groups onits surface (equilibration and elution buffer:sodium, phosphate 50 mM/NaCl 0.15 M pH:6.0). The reaction product was an opalescentsolution (2 mL). The HD3 peptide, which wassynthesized with a C-terminal cysteinyl residue,had a sulfhydryl group that provided a highlyspecific conjugation site for reacting with thecross-linker sulfo-MBS maleimide group to forma stable thio-ether bond (molecular ratio: pep-tide/activated carrier = 2 000, peptide solubili-zation and reaction buffer: sodium phosphate100 mM/NaCl 0.15 M, pH: 7.5, total reactionvolume: 1.5 mL, reaction time: 3 h at room tem-perature with constant stirring).

This peptide/carrier conjugate was then puri-fied by gel filtration on a PD10 column (equili-bration and elution buffer: sodium phosphate100 mM/NaCl 0.15 M pH: 7.5) The conjugateproduct was always an opalescent solution(2 mL). After hydrolysis of 100 ilL of conjugatesolution with 6N hydrochloride acid ( h,

150 °C), amino acid analysis was performed toestimate the number of haptens coupled to thecarrier prior to immunization [26]. A couplingmolecular ratio was calculated at 1 800.

2.3. Immunizations and immunoassays

Two male rabbits aged 3-5 months wereimmunized by intradermal injection of I50 pgequivalent peptide in complete Freund’s adju-vant on day 0. Boots (intradermal injections)were performed with 25 gg equivalent peptidein incomplete Freund’s adjuvant on days 21, 42and 63. Bleedings were performed 7 days aftereach injection. Sera were tested at various dilu-tions in immunoassays (ELISA procedure) totest their capacity to react with the HD3-peptide.

2.4. Electrophoresis and immunoblotting

The specificity of the antibody was assessedby western blotting of basic nuclear proteinsextracted from human sperm nuclei as reportedpreviously [13]. Total basic nuclear proteins wereseparated on 17 % acid-urea polyacrylamide gelelectrophoresis. Proteins were transferred onto0.45 !m nitrocellulose sheets (Schleicher andSchuell) at 2 mA/cm2 for 30 min at room tem-perature using 1 % acetic acid buffer. The blotswere blocked for 2 h at room temperature in tris-buffered saline (TBS), pH 7.2, containing 10 %low fat milk (w/v). The sheets were incubatedwith the anti-peptide antibody (1:200 dilution)in 0.1 M TBS, pH 7.2, containing 0.1 % Tween20 (v/v) for 2 h at room temperature followed,after washing, by a peroxidase-conjugated anti-rabbit antibody (ECL, Amersham Inc.) (1:5 00)for 1 h at room temperature. Visualization wascarried out using the luminescent substrate ofthe ECL system (ECL, Amersham Inc.). ).

2.5. Ultrastructural immunocytochemistry

Human testicular biopsies were obtained fromfour healthy 20-25-year-old men with provenbrain death. Ejaculated spermatozoa were col-lected from four voluntary donors assessed asnormal after spermogram and spermocytogramaccording to WHO criteria [32]. Pieces of testesand the pellets of ejaculated spermatozoa werefixed by immersion for 1 h in 0.1 M Sorensenbuffer, pH 7.4, containing 2 % paraformalde-hyde and 0.1 % glutaraldehyde. Some sperm pel-

lets were immersed for 60 min at 37 °C in 0.05 Mborate buffer, pH 9.0, containing 1 % SDS and6 mM EDTA to modify the chromatin structureand then were washed twice in the same bufferbefore fixation. Samples were dehydrated sequen-tially in 50, 75 and 90 % dimethylformamide,and then processed for embedding in LowicrylK4M medium (Polysciences Paris, France) [2].Ultrathin sections (80 nm) collected on uncoatednickel grids were incubated successively withthe anti-HD3 antibody (1:500) for 2 h, followedby a gold-labelled (15 nm particle size) goat anti-rabbit antibody (1:100) (Biocell, Cardiff, UK). Ascontrol, some sections of ejaculated spermato-zoa were incubated in the same conditions withanother monoclonal antibody recognizing the PZ 2protamine family (h BNP C4 p) which was pre-viously obtained in the laboratory [ 17]. Sectionswere counterstained with aqueous uranyl ace-tate. Controls of the immunostaining specificityincluded incubations with pre-immune serum,increasing dilutions of first and second antibodyand omission of first antibody. Ultrathin sectionswere examined under a JEOL 120 CX electron

microscope at 60 kV (Jeol Ltd., Tokyo, Japan).

2.6. Quantitative evaluationof gold particle distribution

Quantitative evaluation was carried out onfour tissue blocks per subject. The spermatidswere divided into eight steps according to theclassification of Holstein and Roosen-Riinge[16]. Three categories of spermatids could bedistinguished: 1) young spermatids with roundnuclei (steps 1 and 2); 2) intermediate spermatidswith elongating nuclei (steps 3-5); and 3) maturespermatids with condensed nuclei (steps 6-8).To avoid bias in subsequent analysis, only nucleiof spermatids exhibiting visible anterio-poste-rior structures (cap, acrosomal vesicle, matureacrosome sliced longitudinally, neck structures)were selected. Gold particles were recorded overthe spermatid nuclei at the different steps of sper-miogenesis and the ejaculated sperm nuclei. Foreach cell type, 50 sections selected at randomwere counted. The EM micrographs were alltaken at standard magnification X 8000. Goldgrain counting was performed on negative printsusing an interactive image analysis system(Samba, Alcatel, France). The nuclear samplesconsisted of the number of gold particles perJlm2. Background subtraction was realized takinginto consideration adjacent tissue-free resinregions.

2.7. Statistical analysis

Statistical analysis of the data was carried outby one-way analysis of variance (ANOVA) todetermine significant labelling between the dif-ferent steps of spermiogenesis (Macintosh Stat-view II Program). The Scheff6 multiple rangetest was used to study specific comparisons.

3. RESULTS

The presence of antibodies against HD3peptide in sera of the two rabbits immunized(7375 and 7376) was shown by the ELISAprocedure on the 7th day after the third boos-ter injection (figure 2). Only serum 7376was used for the study. Following acid-ureaPAGE of proteins extracted from humansperm nuclei, the immunoblot assays revea-

led that antiserum bound to proteins cor-responding to the molecular weight of inter-mediate basic proteins HPI, and HP12. Bycontrast, neither histones nor protaminesshowed any immunoreactivity ( fi’gure 3).

In immunoelectron microscopy, no label-ling was found in the nucleus or cytoplasmof young spermatids. The labelling firstappeared in nuclei of spermatids at steps4-5 (figure 4), but only a few grains weredistributed in these cells. It then increasedand was much more intense in step 6 sper-matids (figure 5). Colloidal gold particlesprogressively became scarce in mature sper-matids with condensed nuclei (steps 6―8)and a few gold particles were still presentin the nuclei of mature spermatids (figure 6)and ejaculated spermatozoa (figure 7). Com-paratively, a strong labelling of ejaculated

sperm nuclei was obtained with the anti-PZ 2

protamine family antibody (figure 8) whilethe immunolabelling with the anti-HD3 anti-body was not modified after chromatindecondensation by SDS-EDTA treatment(figure 9). The nucleus and the cytoplasmof spermatogonia and spermatocytes as wellas of testicular somatic cells were all unla-belled. Control grids incubated with preim-mune serum or in the absence of the pri-mary antibody did not show any labelling(data not shown).

The pattern of nuclear labelling at thedifferent steps of spermiogenesis is shown infigure 10. No significant differences werefound between the mean grain counts obtain-ed from each subject. The differences in thenumber of gold particles per nucleus areawere statistically significant between thedifferent steps of spermiogenesis (P < 0.001 ).

The nuclear labelling density of proteinsHPI1 and HP12 strongly increased in inter-mediate spermatids (steps 5 and 6) (Scheff6test, P < 0.01) and then sharply decreasedfrom stage 6 to stage 8 of spermiogenesis(Scheff6 test, P < 0.01). However, it remai-ned detectable in nuclei of mature sperma-tids and ejaculated spermatozoa.

4. DISCUSSION

In mouse, as in man, protamines Pi and

P2 are encoded by single-copy genes (PRM,and PRM2) closely linked on chromosome16 [15, 31]. However, mouse protamine P2(MP2; 63 residues) is derived from a pre-cursor form, pmP2 (106 residues), whereasmouse protamine P, is synthesized as amature protamine [33]. In the later steps ofspermiogenesis, the N-terminal part of pmP! 2is processed by limited proteolysis genera-ting a series of polypeptides (pmP2/5,pmP2/1 l, pmP2/16, pmP2/20, pmP2/26 andpmP2/32) which differ in the length of theirN-terminal extension [6, 12]. With theexception of the largest precursors, pmP2 2and pmP2/5, the intermediates of proteoly-sis generated from pmP2 inside spermatid

’

nuclei persist in mature sperm [9]. PZ pro-tamines from mouse (mP2) and man (HP2,HP3, HP4), as well as pmP2 and the inter-mediate basic protein HPI1, share strongstructural homology [4, 7]. In humans, aunique start codon (ATG) which is localized110 nucleotides downstream from the trans-

cription initiation site has been determinedfrom the nucleotide sequence of the PZ gene[11]. Therefore, all proteins related to humanP2 protamines must arise from HPI) whichcan be considered as their common precur-sor. The latter seems to be post-translatio-nally processed in several steps (BP12, HPS1,HPS2, HP4, HP2 and HP31 successively) viaproteolytic cleavages at specific sites lea-ding from lipli to protamine HP3 [ 1, 4, 19,24].

The specificity of the polyclonal anti-HD3 antibody used in this study was asses-

sed in ELISA and western blot: it recognizedonly the two intermediate basic proteinsHPII and HPI2. Ultrastructural controls sho-wed that the accessibility of chromatin tothe antibody was not affected by the com-paction state of nuclei. The deposition ofproteins HPI1 and HPI2 inside human sper-matid nuclei appeared to occur at the endof elongation phase (steps 4-5). At that time,as shown by previous quantitative immu-nocytochemical data [22], the amount ofsomatic-type histones H!B and H3 wasdecreasing, whereas the first signs of label-ling of protamines were detectable in the

nucleus. It is likely that, in elongating sper-matids, histones are displaced by sperma-tid-specific transition proteins (TP1 and

TP 2)’ as suggested by prior cytochemicalfindings [8]. Transition proteins, in turn, aresubsequently removed from the condensingchromatin at the end of the spermatid elon-gation phase and are replaced by protamines(reviewed by Balhom [5] and Hecht [14]).

From step 5 to step 6 of spermiogenesis,the sharp increase in the nuclear labellingdensity of HPI, and HP12 paralleled thelabelling which was observed for protamines

localized by means of a specific monoclonalantibody (hBNP 12 C3) against both P! andPZ families. However, the labelling of HPI¡and HP12 sharply decreased in step 6 sper-matid nuclei, in contrast with the protaminerate that was found to remain high and stablein mature spermatids and ejaculated sper-matozoa [22]. We previously showed thatmonoclonal antibody hBNP 12 C3 and ano-

ther monoclonal antibody recognizing theP2 protamine family (hBNP C4 p) gave thesame results in immunoelectron microscopy[17]. This finding was not surprising, since

antigenic sites common to the different pro-tamines had been detected by polyclonalantisera against purified protamine HPjb[23] and by monoclonal antibodies [27].Elsewhere, on western blots of human inter-mediate basic proteins separated by acid-urea gel electrophoresis, both hBNP 12 C3and hBNP C4P cross-reacted with the four

proteins HPIt, HP12, HPSI and HPS2 [17].Therefore, the present results, together withpreceding data on immunolabelling of pro-tamines, are in good agreement with struc-tural studies showing that the intermediate

basic proteins HPII and HP12 represent pre-cursors (pro-protamines) of the Pz prota-mine family.

The lack of cytoplasmic labelling sug-gests that intermediate proteins do not accu-mulate inside the spermatid cytoplasm. Thisobservation reinforces the hypothesis thatthe proteases implicated in the maturationof pro-protamines are closely associatedwith nuclei [9].

The persistence of low amounts of pro-teins HPI, and HPI2 in human sperm nuclei,as revealed by immunoelectron microscopy,is in accordance with earlier biochemicaldata indicating that the four basic proteins ofintermediate size between histones and pro-tamines (HPI1, HPI2, HPSI and HPS2) repre-sent about 10 % of the amount of basic pro-teins isolated from sperm nuclei [13, 28-30].

ACKNOWLEDGEMENT

The authors would like to thank Mrs S.

Reposo for her skillful technical assistance.

REFERENCES

[1] ] Alimi E., Martinage A., Arkhis A., Belaiche D.,SautiBre P., Chevaillier Ph., Amino acidsequence of the human intermediate basic pro-tein 2 (HP12) from sperm nuclei, Eur. J. Bio-chem. 214 (1993) 445-450.

[2] Altman L.G., Schneider B.G., Papermeister D.S., .,

Rapid embedding of tissues in Lowicryl K4Mfor immunoelectron microscopy, J. Histochem.Cytochem. 32 (1984) 1217-1224.

[3] Ammer H., Henschen A., Lee C.H., Isolationand amino acid sequence analysis of humansperm protamines PI and P2, Biol. Chem. 367(1986) 515-522.

[4] Arkhis A., Martinage A., Sauti6re P., ChevaillierPh., Molecular structure of human protamineP4 (HP4) a minor basic protein of human spermnuclei, Eur. J. Biochem. 200 (1991) 387-392.

[5] Balhom R., Mammalian protamines: Structureand molecular interactions, in: Adolph K.W.(Ed.), Molecular Biology of Chromosome Func-tion, Springer Verlag, New York, 1989, pp366-395.

[6] Chauvi6re M., Martinage A., Debar]6 M., Sau-tiere P., Chevallier Ph., Molecular characteri-zation of six intermediate proteins in the pro-cessing of mouse protamine P2 precursor, Eur.J. Biochem. 204 (1992) 759-765.

[7] Chevaillier Ph., Martinage A., Arkhis A., Sau-ti6re P., Les pr6curseurs des protamines dans lespermatozdfde humain, Reprod. Nutr. Dev. 30( 1990) 343-347.

[8] Dadoune J.P., Alfonsi M.F., Ultrastructural andcytochemical changes of the head components ofhuman spermatids and spermatozoa, GameteRes. 14 (1986) 33-46.

[9] Debarlé M., Martinage A., Sautière P., Che-vaillier Ph., Persistence of protamine precursorsin mature sperm nuclei of the mouse, Mol.

Reprod. Dev. 40 ( 1995) 84-90.

[10] Domenjoud L., Fronia C., Uhde F., Engel W.,Sequence of human protamine 2 cDNA, Nucl.Acid Res. 16 (1988) 7733.

[11] Domenjoud L., Nussbaum G., Adham LM.,Greeske G., Engel W., Genomic sequences ofhuman protamines whose genes, PRM1 andPRM2 are clustered, Genomics 8 (1990)127-133.

[12] Elsevier S.M., Noirau J., Carr6-Eus6be D., Pro-cessing of the precursor of protamine P2 inmouse. Identification of intermediates by theirinsolubility in the presence of sodium dodecylsulfate, Eur. J. Biochem. 196 (1991) 167-175.

[13] Gusse M., SautiBre P., Belafche D., MartinageA., Rouch Ch., Dadoune J.P., Chevaillier Ph.,Purification and characterization of nuclear basic

proteins of human sperm, Biochem. Biophys.Acta 884 (1986) 224-234.

[ 14J Hecht N.B., Mammalian protamines and theirexpression, in: Hnilica L., Stein G., Stein J.(Eds.), Histones and Other Basic Nuclear Pro-teins, CRC Press, Boca Raton FL, USA, 1989,pp 347-373.

[15] Hecht N.B., Kleene K.C., Yelick P.C., John-son P.A., Pratcheva D.D., Ruddle F.H., Haploidgene mapping: the genes for both mouse prota-mines are located or chromosome 16, Som. CellMol. Genet. 12 ( 1986) 191-196.

[ 16] Holstein A.F., Roosen-Runge E.C., (Eds.), Atlasof Human Spermatogenesis, Grosse Verlag, Ber-lin, 1981. 1 .

[17] Le Lannic G., Arkhis A., Vendrely E., Che-vaillier Ph., Dadoune J.P., Production, charac-terization and immunocytochemical applica-tions of monoclonal antibodies to human spermprotamines, Mol. Reprod. Dev. 36 (1993)106-112.

[18] Lescoat D., Blanchard Y., Lavault M.T., Quer-n6e D., Le Lannou D., Ultrastructural andimmunocytochemical study of PI protaminelocalization in human testis, Andrologia 25(1993)93-99.

[19] Martinage A., Arkhis A., Alimi E., Sautière P.,Chevaillier Ph., Molecular characterization ofnuclear basic protein HPI1, a putative precur-sor of human sperm protamines HP2 and HP3,Eur. J. Biochem. 191 (1990) 449-451. 1.

[20] McKay D.J., Renaux B.S., Dixon G.H., Humansperm protamines. Amino-acid sequence of twoforms of protamine P2, Eur. J. Biochem. 156(1986)5-8.

[21] Oliva R.F., Dixon G.H., Vertebrate protaminegenes and the histone-to-protamine replacementreaction, Prog. Nucl. Acids Res. Mol. Biol. 40(1991)25-94.

[22] Prigent Y., Muller S., Dadoune J.P., Immunoe-lectron microscopical distribution of histonesH2B and H3 and protamines during humanspermiogenesis, Mol. Human Reprod. 2 (1996)929-935.

[23] Roux C., Mathieson J., Dadoune J.P., Locali-sation immunocytologique des protamines dugroupe HP1 dans le testicule humain et le sper-matozdide 6jacul6, Bull. Assoc. Anat. 71 (1987)65-69.

[24] Roux C., Arkhis A., Dadoune J.P., Chevaillier Ph.,Intermediate basic proteins of human spermnuclei are structurally related to protamine HP2:immunological evidence, Mol. Androl. 3 (1990)227-232.

[25] Sautière P., Martinage A., Belafche D., ArkhisA., Chevaillier Ph., Comparison of the aminoacid sequences of human protamine HP2 andHP3 and of intermediate basic nuclear proteinsHPS 1 and HPS2: structure evidence that HPS 1and HPS2 are pro-protamines, J. Biol. Chem.263 (1988) 11.059-11.062.

[26] Simpson R.J., Neuberger M.R., Liu T.Y., Com-plete amino acid analysis of proteins from asingle hydrolysate, J. Biol. Chem. 251 (1976)1936-1940.

[27] Stanker L.H., Wyrobek A., Balhom R., Mono-clonal antibodies to human protamines, Hybri-doma 6 (1987) 293-303.

[28] Svasti J., Taluppeth N., Improvement in theresolution of human sperm protamines by useof iodoacetamide as alkylating agent, Biochim.Biophys. Acta 577 (1979) 221-225.

[29] Tanphaichitr N., Sohbon P., Taluppeth N., Cha-lermisarachai P., Basic nuclear proteins in tes-ticular cells and ejaculated spermatozoa in man,Exp. Cell Res. 117 (1978) 347-350.

[30] Tanphaichitr N., Sohbon P., Chalermisarachai P.,Patilantakarnkool M., Acid extracted nuclearproteins and ultrastructure of human sperm chro-matin as revealed by differential extraction withurea, mercaptoethanol and salt, Gamete Res. 4(1981)297-315.

[31 ] Viguie F., Domenjoud L., Rousseau-Merck F.M.,Dadoune J.P., Chevaillier Ph., Chromosomallocalisation of the human protamine genes,PRM 1 and PRM2, to 16p 13.3, by in situ hybri-dization, Hum. Genet. 85 (1990) 171-174.

[32] World Health Organization, WHO LaboratoryManual for the Examination of Human Semenand Sperm-Cervical Mucus Interaction, 3rd ed.,Cambridge University Press, Cambridge, UK,1992.

[33] Yelick P.C., Balhom R., Johnson P.A., Corzett M.,Mazrimas J.A., Kleene K.C., Hecht N.B., Mouseprotamine 2 is synthesized as a precursor, whe-reas mouse protamine 1 is not, Mol. Cell Biol. 7(1987)2173-2179.