A Haploid Expressed Gene Cluster Exists as a Single Chromatin Domain in Human Sperm

Abstract

Mammalian spermiogenesis is marked by the initial disruption of the nuclear-histone-DNA complex by the transition proteins for ultimate replacement with protamines. The genes for three of these low molecular weight basic nuclear proteins exist as a single linear array of PRM1, PRM2, and TNP2 on human chromosome 16p13.2. To begin to address the mechanism governing their transcriptional potentiation, a region of ∼40 kilobases of the human genome encompassing these genes was introduced into the germ line of mice. Fluorescence in situ hybridization and Southern analysis showed that this segment of the human genome integrated into independent chromosomal sites while maintaining its fidelity. Transcript analysis demonstrated that the expression of the endogenous mouse protamine Prm1 and Prm2 genes as well as the mouse transition protein Tnp2 gene were expressed along with their human transgene counterparts. The pattern of expression of these transgenic human genes within this multigenic cluster faithfully represented that observed in vivo. In addition, all members of this transgenic gene cluster were expressed in proportions similar to those in human testis. Copy number-dependent and position-independent expression of the transgenic construct demonstrated that the corresponding biological locus was contained within this segment of the human genome. Furthermore, DNase I sensitivity established that in sperm the human PRM1→ PRM2→ TNP2 genic domain was contained as an ∼28.5-kilobase contiguous segment bounded by an array of nuclear matrix associated topoisomerase II consensus sites. This is the first description of a multigenic male gamete-specific domain as a fundamental gene regulatory unit. A model of haploid-specific gene determination is presented. Mammalian spermiogenesis is marked by the initial disruption of the nuclear-histone-DNA complex by the transition proteins for ultimate replacement with protamines. The genes for three of these low molecular weight basic nuclear proteins exist as a single linear array of PRM1, PRM2, and TNP2 on human chromosome 16p13.2. To begin to address the mechanism governing their transcriptional potentiation, a region of ∼40 kilobases of the human genome encompassing these genes was introduced into the germ line of mice. Fluorescence in situ hybridization and Southern analysis showed that this segment of the human genome integrated into independent chromosomal sites while maintaining its fidelity. Transcript analysis demonstrated that the expression of the endogenous mouse protamine Prm1 and Prm2 genes as well as the mouse transition protein Tnp2 gene were expressed along with their human transgene counterparts. The pattern of expression of these transgenic human genes within this multigenic cluster faithfully represented that observed in vivo. In addition, all members of this transgenic gene cluster were expressed in proportions similar to those in human testis. Copy number-dependent and position-independent expression of the transgenic construct demonstrated that the corresponding biological locus was contained within this segment of the human genome. Furthermore, DNase I sensitivity established that in sperm the human PRM1→ PRM2→ TNP2 genic domain was contained as an ∼28.5-kilobase contiguous segment bounded by an array of nuclear matrix associated topoisomerase II consensus sites. This is the first description of a multigenic male gamete-specific domain as a fundamental gene regulatory unit. A model of haploid-specific gene determination is presented. Protamines are highly basic, sperm-specific nuclear proteins that are characterized by an arginine rich core segment (cf. Refs. 1Dixon G.H. Aiken J.M. Jankowski J.M. McKenzie D.I. Moir R. States J.C. Reeck G.R. Goodwin G.A. Puigdomenech P. Chromosomal Proteins and Gene Expression. Plenum Publishing Corp., New York1986: 287-314Google Scholar and 2Krawetz S.A. Dixon G.H. J. Mol. Evol. 1988; 27: 291-297Crossref PubMed Scopus (35) Google Scholar). In mammals there are two classes of protamine proteins, PRM1 and PRM2, and at least three distinct classes of transition proteins, TNP1, TNP2, and TNP4 (3Grimes S.R. Platz R.D. Meistrich M.L. Hnilica L.S. Biochem. Biophys. Res. Commun. 1975; 67: 182-189Crossref PubMed Scopus (50) Google Scholar, 4Unni E. Meistrich M.L. J. Biol. Chem. 1992; 267: 25359-25363Abstract Full Text PDF PubMed Google Scholar). During the terminal phase of spermiogenesis, the nucleohistone array is replaced with a nucleoprotamine complex. A compact and essentially quiescent genome results. This process is mediated by the transition proteins (cf. Ref. 5Meistrich M.L. Hnilica L.S. Stein G.S. Stein J.L. Histonesand Other Basic Nuclear Proteins. CRC Press, Boca Raton, FL1989: 165-182Google Scholar), which in humans leads to the final replacement of ∼85% of the nuclear histone complement with protamines (6Gatewood J.M. Cook G.R. Balhorn R. Bradbury E.M. Schmid C.W. Science. 1987; 236: 962-964Crossref PubMed Scopus (294) Google Scholar). The functionally related PRM1, PRM2, and TNP2 genes exist as an ordered linear array on human chromosome 16p13.2 (7Callen D.F. Doggett N.A. Stallings R.L. Chen L.Z. Whitemore S.A. Lane S.A. Nancarrow J.K. Aposolou S. Thompson A.D. Lapsys N.M. Eyre H.J. Baker E.G. Shen Y. Holman K. Phillips H. Richards R.I. Sutherland G.R. Genomics. 1992; 13: 1178-1185Crossref PubMed Scopus (68) Google Scholar, 8Nelson J.E. Krawez S.A. J. Biol. Chem. 1993; 268: 2932-2936Abstract Full Text PDF PubMed Google Scholar). The complete primary structure of the 40,573-bp 1The abbreviations used are:bpbase pair(s)kbkilobase(s)LCRlocus control regionFISHfluorescence in situ hybridizationSARscaffold attachment regionMARmatrix attachment regionhhuman. region that encompasses these genes has been determined (9Nelson J.E. Krawetz S.A. J. Biol. Chem. 1994; 269: 31067-31073Abstract Full Text PDF PubMed Google Scholar). Extensive sequence analysis of this chromosomal segment has provided many insights into the biology and evolution of the various members of the cluster and this region of the human genome. base pair(s) kilobase(s) locus control region fluorescence in situ hybridization scaffold attachment region matrix attachment region human. Phylogenetic footprinting (2Krawetz S.A. Dixon G.H. J. Mol. Evol. 1988; 27: 291-297Crossref PubMed Scopus (35) Google Scholar, 10Olivia R. Dixon G.H. J. Mol. Evol. 1990; 30: 333-346Crossref PubMed Scopus (54) Google Scholar, 11Queralt R. Oliva R. Gene (Amst.). 1993; 133: 197-204Crossref PubMed Scopus (32) Google Scholar) and transgenic strategies have been used to identify a series of sequences flanking the 5′ region of the protamine genes that confer tissue specificity. An 859-bp sequence, 5′ of the initiator codon of the mouse Prm2 gene (12Stewart T.A. Hecht N.B. Hollingshead P.G. Johnson P.A. Leong J.A.C. Pitts S.L. Mol. Cell. Biol. 1988; 8: 1748-1755Crossref PubMed Scopus (40) Google Scholar) and a 465-bp sequence, 5′ of the initiator codon of the mouse Prm1 gene (13Peschon J.J. Behinger R.R. Brinster R.L. Palmiter R.D. Proc. Natl. Acad. Sci. U. S. A. 1987; 84: 5316-5319Crossref PubMed Scopus (176) Google Scholar, 14Peschon J.J. Behinger R.R. Brinster R.L. Palmiter R.D. N. Y. Acad. Sci. 1989; 564: 186-197Crossref PubMed Scopus (18) Google Scholar) can direct tissue-specific expression in a homologous system. Deletion analysis has shown that the tissue-specific element(s) resides within a 113-bp segment, 5′ of the initiator codon of the Prm1 gene (15Zambrowicz B.P. Harendza C.J. Zimmerman J.W. Brinster R.L. Palmiter R.D. Proc. Natl. Acad. Sci. U. S. A. 1993; 90: 5071-5075Crossref PubMed Scopus (92) Google Scholar) encompassing nucleotides −77 to −37 (16Zambrowicz B.P. Palmiter R.D. Biol. Reprod. 1994; 50: 65-72Crossref PubMed Scopus (34) Google Scholar). Their significance remains to be assessed, since both testis-specific and ubiquitous protein factors bind to this segment. Although this directly addressed transcriptional regulation, the issue of how these linked genes are selected from the genome for expression remains. Our understanding of how a multigenic locus assumes a potentiated conformation, i.e. readied for expression, or how this state is maintained for transcription is limited to somatic cells. The concept that transcriptionally competent chromatin assumed an altered nuclear structure was validated when it was shown that the β-globin gene acquired a DNase I-sensitive conformation within cells in which it was expressed (17Weintraub H. Goudine M. Science. 1976; 193: 848-856Crossref PubMed Scopus (1203) Google Scholar). These studies were extended to the chicken ovalbumin, x, y locus, which was one of the first chromatin domains containing a multigenic loci to be defined (18Lawson G.M. Knoll B.J. March C.J. Woo S.L.C. Tsai M.J. O'Malley B.W. J. Biol. Chem. 1982; 257: 1501-1507Abstract Full Text PDF PubMed Google Scholar). The complete characterization of the β-globin locus has extended the concept of the genic domain as the primary regulatory unit to mammalian systems (19Smith R.D. Yu J. Annunziato A. Seale R.L. Biochemistry. 1984; 23: 2970-2976Crossref PubMed Scopus (20) Google Scholar, 20Forrester W.C. Thompson C. Elder J.T. Groudine M. Proc. Natl. Acad. Sci. U. S. A. 1986; 83: 1359-1363Crossref PubMed Scopus (270) Google Scholar). The subsequent identification of the LCR, i.e. locus control region (21Grosveld F. van Assendelft G.B. Greaves D.R. Kollias Cell. 1987; Full Text PDF PubMed Scopus Google Scholar), provided a to understanding the mechanism of gene In to the well characterized in and in β-globin locus, understanding of expressed genic domains and how these i.e. PRM1→ PRM2→ TNP2, a potentiated conformation is understanding the mechanism governing male germ gene expression has been by the of a line and the of primary cells in A cells has been M. J. M. K. van P. F. Cell. 1993; Full Text PDF PubMed Scopus Google Scholar), it has to be This has by or transgenic analysis as the that can be used to address these within the sperm To the of the molecular mechanism that the expression of the male a transgenic mouse model was A segment of human chromosome the PRM1→ PRM2→ TNP2 gene cluster J.E. Krawez S.A. J. Biol. Chem. 1993; 268: 2932-2936Abstract Full Text PDF PubMed Google Scholar, J.E. Krawetz S.A. J. Biol. Chem. 1994; 269: 31067-31073Abstract Full Text PDF PubMed Google Scholar) was introduced into mouse This contained to faithfully all of the members of this in a position-independent in the system. A DNase I-sensitive domain of encompassing the TNP2 gene cluster is bounded by an array of topoisomerase II consensus sites. These that somatic the genes by the male are at loci as The of the male in this as a of first control male is in the model of haploid-specific gene and the were from were from and the and were from were from DNase I was from The mouse was from were of or molecular biology and from T.A. C.J. Krawetz S.A. J. Biol. Chem. 1992; 267: Full Text PDF PubMed Google Scholar). A 40,573-bp segment encompassing the human protamine PRM1 and the transition protein TNP2 gene cluster from chromosome 16p13.2 was and characterized J.E. Krawetz S.A. PubMed Scopus Google Scholar). The corresponding containing this human gene cluster was by of the by The was into the of mouse were by analysis of mouse the TNP2 Their was by gene was determined mouse essentially as F. E. the A Scholar). and were to with of the various transgenic mouse and were The were in the hybridization of containing for at least at The were at in hybridization that contained of The human PRM1 and the TNP2 were from of and the were for at in with for at in with to for at Copy was determined as the as Krawetz S.A. Dixon G.H. J. Biochem. 1990; PubMed Scopus (20) Google Scholar). were established as from mouse tissue or tissue R.I. of A of Basic R. New Scholar). The cells were in with and in were and chromosomal J. F. M. The Press, New Scholar). To chromosomal of the integrated human loci and the transgenic mouse were van H. D.R. J. 1975; PubMed Scopus Google Scholar) to of the from was with by the as by the for at the was nucleotides were from the The was to a final of with for at The containing the various of transgenic were in at for the were in a of containing at for were a series of of of the hybridization containing of the containing of sequence mouse was to The were and with at in a to the were in at for by a in at were and the were with were with an fluorescence a and of to its chromosome was determined as a of was the of and P. N. Biochem. 1987; PubMed Scopus Google Scholar), and of were on a H. J. H. Biochemistry. PubMed Scopus Google Scholar). were The corresponding human sequences and the mouse sequence were a series of The PRM1, PRM2, and TNP2 were as The mouse was a of the The were in the hybridization for at least at were at in hybridization that contained of corresponding to the various members of the gene hybridization the were at in a of containing for at in containing for The were for various to the human PRM1, PRM2, TNP2, and mouse The corresponding mouse and Tnp2 were a series of The sequences of the were as and The corresponding were for at in the hybridization containing The were at in hybridization that contained a single hybridization the were at in a of containing for at in containing for The were at for to the mouse that were within the linear of the were the system. were as a of was as a of the of the hybridization of the mouse the when to the human sperm were with of DNase I at and at and The were with the of a containing was to the DNase to a of and the at for The was by with were in at for to The were at for to from sequence of human essentially as of these as the region of human chromosome 16p13.2 that the PRM1→ PRM2→ TNP2 A the region from and The was and the β-globin sequences on and were used as to the were at for in that contained The were at were the analysis system. The of hybridization of DNase was expressed to that of the hybridization as a of was The DNase I i.e. of was defined as the to a in the of hybridization A of human from was into the of of mice. This contained the linear human PRM1, protamine genes and the TNP2 transition protein This gene cluster is by of sequence 5′ to the PRM1 gene and of sequence of the TNP2 gene transgenic two were Gene analysis from of this cluster or its members within the members or The of this segment of the human genome containing these was when introduced into and in the transgenic showed germ line of the integrated The of loci integrated within the from to in that and integrated the TNP2 gene cluster into mouse chromosome and integrated the TNP2 gene cluster into mouse and from and were line was from since the human PRM1→ PRM2→ TNP2 locus integrated into the mouse chromosome and was to the TNP2 locus the and was from analysis was to this of the human genome from contained the to the human PRM1, PRM2, and TNP2 in the shown in the was the of the PRM1, PRM2, and TNP2 in human testis. PRM1 was at a of of that of PRM2, and TNP2 was at a of of that of The proportions of of the human PRM2, PRM1, and TNP2 in transgenic mouse shown in were similar to that observed in human shown in was of or of the mouse or human Furthermore, transgenic and to of the human or mouse The expression of of the human PRM1→ PRM2→ TNP2 transgenic gene cluster was independent of the mouse can that this segment of the human genome the for tissue-specific and expression of of its members in a similar to that observed in vivo. The that this human of ∼40 containing the TNP2 gene cluster as a single locus, i.e. similar to that of the β-globin locus (21Grosveld F. van Assendelft G.B. Greaves D.R. Kollias Cell. 1987; Full Text PDF PubMed Scopus Google Scholar) was by three analysis demonstrated that transgenic mouse the TNP2 locus integrated this multigenic human domain into the mouse genome in a position-independent transgenic expressed the PRM1, PRM2, and TNP2 genes in the The of of the transgenic TNP2 cluster was similar to that observed in vivo. this the of a genic locus, i.e. number-dependent expression of the integrated To address this the expression of of the cluster as a of the of of the integrated locus was containing of the human PRM1→ PRM2→ TNP2 locus showed an of the corresponding human which contained The expression of of the human PRM1→ PRM2→ TNP2 locus within of the transgenic is shown in from and of the locus, the of of the three gene of the human PRM1→ PRM2→ TNP2 locus was directly i.e. to the of i.e. of the integrated Furthermore, within the of gene was similar to that observed in human testis. their expression the of of the integrated The of this series of transgenic that position-independent number-dependent expression of of the members of the human PRM1→ PRM2→ TNP2 that the biological locus is contained within this of the human genome. this region all that locus in to and This region the of the genic expression of the human PRM1, PRM2, and TNP2 genes in transgenic was from transgenic and mouse by that to the and mouse was the system. The integrated for of the transgenic was to the mouse The expression of the cluster in the three was was from transgenic and mouse by that to the and mouse was the system. The integrated for of the transgenic was to the mouse The expression of the cluster in the three was To this defined locus the of human sperm were with DNase I for various of The were by analysis as that the TNP2 transgenic locus as shown in hybridization to human which is provided a DNase I-sensitive control a human which is expressed in cells of as a DNase a in human shown in of DNase I Their were and similar to that of the DNase I-sensitive In A and were similar to that of the β-globin DNase The DNase of that this region of human chromosome 16p13.2 the of the DNase I-sensitive shown in and A and the of an region of DNase these at and are associated with of domains J. 1988; PubMed Scopus Google Scholar, J. Cell. Biochem. 1993; PubMed Scopus Google Scholar, J. Cell. Biochem. 1993; PubMed Scopus Google Scholar). shown in a cluster of topoisomerase II sites within region are the of the These sites are at and at least to the topoisomerase II consensus sequence of the chicken gene J. 1988; PubMed Scopus Google Scholar). are at similar with to the DNase of the This the that these as attachment to the nuclear In and in strategies have a basic understanding of the mechanism the and expression of is that of chromatin encompassing a single gene locus or multigenic locus a DNase I-sensitive conformation within somatic cells to that gene (17Weintraub H. Goudine M. Science. 1976; 193: 848-856Crossref PubMed Scopus (1203) Google Scholar, G.M. Knoll B.J. March C.J. Woo S.L.C. Tsai M.J. O'Malley B.W. J. Biol. Chem. 1982; 257: 1501-1507Abstract Full Text PDF PubMed Google Scholar, R.D. Yu J. Annunziato A. Seale R.L. Biochemistry. 1984; 23: 2970-2976Crossref PubMed Scopus (20) Google Scholar). the this have are i.e. locus control (cf. Ref. Cell. 1990; Full Text PDF PubMed Scopus Google Scholar). These are contained as of the of single independent regulatory W.C. S. Groudine M. Res. 1987; PubMed Scopus (294) Google Scholar), as DNase I-sensitive chromatin that in somatic in Cell. Full Text PDF PubMed Scopus Google Scholar). Our understanding of these within the genome of the male is it is that chromatin domains of an of are within the human sperm J. J. 1994; Google Scholar). a domain that number-dependent and position-independent expression a transgene that has integrated into the genome (21Grosveld F. van Assendelft G.B. Greaves D.R. Kollias Cell. 1987; Full Text PDF PubMed Scopus Google Scholar). To begin to those that control haploid-specific gene expression, transgenic gene of of the mouse Prm1 and mouse Prm2 genes linked to various genes have been (12Stewart T.A. Hecht N.B. Hollingshead P.G. Johnson P.A. Leong J.A.C. Pitts S.L. Mol. Cell. Biol. 1988; 8: 1748-1755Crossref PubMed Scopus (40) Google Scholar, J.J. Behinger R.R. Brinster R.L. Palmiter R.D. Proc. Natl. Acad. Sci. U. S. A. 1987; 84: 5316-5319Crossref PubMed Scopus (176) Google Scholar, 14Peschon J.J. Behinger R.R. Brinster R.L. Palmiter R.D. N. Y. Acad. Sci. 1989; 564: 186-197Crossref PubMed Scopus (18) Google Scholar, B.P. Harendza C.J. Zimmerman J.W. Brinster R.L. Palmiter R.D. Proc. Natl. Acad. Sci. U. S. A. 1993; 90: 5071-5075Crossref PubMed Scopus (92) Google Scholar, R.R. J.J. A. Palmiter R.D. Brinster R.L. Proc. Natl. Acad. Sci. U. S. A. 1988; PubMed Scopus Google Scholar, J.J. R.R. Brinster R.L. Palmiter R.D. 1989; PubMed Scopus Google Scholar). This that within regulatory unit assumed a fundamental the and expression of these the PRM1, PRM2, and TNP2 genes exist as a as regulatory is that this multigenic sperm-specific locus is of similar as the chromatin domains observed in was to this gene cluster as a domain within the sperm the mechanism that the expression of these three genes be of expression of the transgenic PRM1, and TNP2 genes that this segment of human chromosome 16p13.2 all the to of this of was maintained and the of of the transgenic TNP2 was similar to that observed in vivo. Transcript analysis that the of the mouse Tnp2 from that of human in the or transgenic the mouse and human were the of the three gene In the of TNP2 was by when to the corresponding mouse Tnp2 This is to the of a region the of which the of the human M. Biochem. Biophys. Commun. 1989; Scopus Google Scholar). The transgenic analysis in this which low of the corresponding human the that the of this the of the This is with the mechanism governing sperm The of the mouse Prm1 the human PRM1 The biological significance of this is since both The of the PRM1 and TNP2 The from human chromosome 16p13.2 as a position-independent regulatory unit. The biological expression of the transgenic TNP2 gene cluster demonstrated that this segment of the human genome all to this The of the endogenous mouse and Tnp2 in of the transgenic was This was within the that to of the TNP2 can be that factors are within the to the on the transgenic transcription by these human the first this the of a multigenic The corresponding genic domain resides within this segment of human chromosome 16p13.2 as an DNase I-sensitive region from in human This is with this segment contained within the of the human sperm genome that remains histone associated (6Gatewood J.M. Cook G.R. Balhorn R. Bradbury E.M. Schmid C.W. Science. 1987; 236: 962-964Crossref PubMed Scopus (294) Google Scholar). The issue of the of the DNase I-sensitive domain sequence with those to to the nuclear matrix was Extensive analysis to extended the sequences at the of the DNase I-sensitive domain and of those associated with or J. Cell. Biochem. 1993; PubMed Scopus Google Scholar, J. Cell. Biochem. 1993; PubMed Scopus Google Scholar). and both are by a cluster of topoisomerase II sites that are by from topoisomerase II In these are by a series of homologous Their significance is since this region of chromosome 16p13.2 at a as as (9Nelson J.E. Krawetz S.A. J. Biol. Chem. 1994; 269: 31067-31073Abstract Full Text PDF PubMed Google Scholar). by H. 1990; PubMed Scopus Google Scholar, J. J. R. M. Mol. Cell. Biol. 1990; Scholar), these as of gene To this is the compact of multigenic domain and is of similar as the single chicken gene K. Res. 1986; PubMed Scopus Google Scholar). is to that the locus of the β-globin gene cluster or the locus of the single the TNP2 locus that all to expression, is within this DNase I-sensitive The complex pattern of transcription and by of the TNP2 locus a to the mechanism that the expression of this multigenic This the process that is have to address this issue by the mechanism of gene as it This has the of the model in by and Science. PubMed Scopus Google Scholar), the of a of associated genes be determined by a of factors and factors be to the testis-specific factors be to genes that the In be mediated the of the of that from chromatin (17Weintraub H. Goudine M. Science. 1976; 193: 848-856Crossref PubMed Scopus (1203) Google Scholar), multigenic domains the ovalbumin, x, y (18Lawson G.M. Knoll B.J. March C.J. Woo S.L.C. Tsai M.J. O'Malley B.W. J. Biol. Chem. 1982; 257: 1501-1507Abstract Full Text PDF PubMed Google Scholar), and the β-globin (19Smith R.D. Yu J. Annunziato A. Seale R.L. Biochemistry. 1984; 23: 2970-2976Crossref PubMed Scopus (20) Google Scholar) locus were is that and their corresponding factors that are associated with the of DNase I-sensitive domains C. J. Biol. Chem. 1989; Full Text PDF PubMed Google Scholar, D.R. J. 1988; PubMed Scopus Google Scholar, H. Hecht A. C. Biol. 1993; PubMed Scopus Google Scholar) at least of the to the i.e. readied for expression genic shown in the model of the determination of haploid-specific gene expression, domains that of by the have assumed a potentiated state to the first the by the is first in the J.J. Biol. 1992; PubMed Scopus Google Scholar), is as by the the with A. S. M. J. E. and S. A. and in with the the and PRM1→ PRM2→ TNP2 have assumed its potentiated by the have demonstrated and these domains are maintained within the subsequent to of the genome. In the β-globin by the is and The issue when is the male genome to its be the of the potentiated This be within the of A or the of for the be from the system. In this the is marked by a which is by the DNase The of the to its is mediated by the of the that are to be maintained in the conformation M. R. Proc. Natl. Acad. Sci. U. S. A. PubMed Scopus Google Scholar, Greaves Proc. Natl. Acad. Sci. U. S. A. 1992; PubMed Scopus Google Scholar). be nuclear that the and which can to the of their In this genes that are for transcription from the chromatin to chromatin W.C. S. Groudine M. Res. 1987; PubMed Scopus (294) Google Scholar). This one of the of the The of the the that haploid-specific gene expression is for the of the of and of the of and for the human of the of and provided the and