- Email: [email protected]
Temporal expression of the transgenic human protamine gene cluster
FERTILITY AND STERILITYt VOL. 71, NO. 4, APRIL 1999 Copyright ©1999 American Society for Reproductive Medicine Published by Elsevier Science Inc. Printed on acid-free paper in U.S.A.
Temporal expression of the transgenic human protamine gene cluster Kathy S. Stewart, M.D.,*† Jeffrey A. Kramer, Ph.D.,*‡ Mark I. Evans, M.D.,*‡ and Stephen A. Krawetz, Ph.D.*‡ Wayne State University, Detroit, Michigan
Received September 11, 1998; revised and accepted November 25, 1998. Supported in part by the Fund for Medical Education and Research and by a Research Faculty Award from Wayne State University, Detroit, Michigan. Jeffrey A. Kramer was supported in part by a Lalor Foundation (Providence, Rhode Island) postdoctoral fellowship. Presented in part at the 18th Annual Meeting of the Society of Perinatal Obstetricians, Miami Beach, Florida, February 2–7, 1998. Reprint requests: Stephen A. Krawetz, Ph.D., Department of Obstetrics and Gynecology and Center for Molecular Medicine and Genetics, Wayne State University School of Medicine, 275 East Hancock, Detroit, Michigan 48201 (FAX: 313577-8854; E-mail: steve @compbio.med.wayne .edu). * Department of Obstetrics and Gynecology. † Present address: Department of Obstetrics and Gynecology, University of Wisconsin, Madison, Wisconsin. ‡ Center for Molecular Medicine and Genetics. 0015-0282/99/$20.00 PII S0015-0282(98)00548-2
Objective: To ascertain the fidelity of expression of the genes from the transgenic human sperm-specific nuclear packaging protamine-13protamine-23transition protein-2 (PRM13 PRM23 TNP2) locus. Design: Controlled human transgene study. Setting: Basic science laboratory. Animal(s): Age-matched transgenic and nontransgenic mice. Intervention(s): Transgenic mice containing the human protamine locus were mated. One testis from each offspring was frozen at 280°C and the other was preserved in formalin. Main Outcome Measure(s): The temporal expression of the human and mouse protamines was evaluated by Northern blot analysis. Orientation of the transgenic locus was determined by Southern blot analysis. Tissue morphology was assessed histologically. Result(s): Conservation of transgenic morphology was confirmed. Head-to-tail integration of the PRM13 PRM23TNP2 locus was shown. Temporal expression of the mouse and human protamine genes was maintained in the transgenic state. Conclusion(s): These results show that the head-to-tail concatomer of the PRM13 PRM23 TNP2 locus contains all the necessary elements for appropriate temporal expression while maintaining testicular structure and function. (Fertil Sterilt 1999;71:739 – 45. ©1999 by American Society for Reproductive Medicine.) Key Words: Protamine, transgenic, infertility, spermatogenesis, Mus musculus, locus
Protamines are low-molecular-weight, malespecific, arginine-rich proteins that replace histones during sperm maturation. This replacement mediates the restructuring of the chromatin, condensing the genetic material into a speciesspecific shaped nucleus that is approximately one-sixth the size of a somatic nucleus (1). The human protamine locus resides within an approximately 30-kilobase (kb) region along human chromosome 16p13.13. Its chromatin domain encompasses a linear array of three genes, protamine 1 (PRM1), protamine 2 (PRM2), and transition protein 2 (TNP2) (2, 3). This cluster of genes is defined as the PRM13 PRM23TNP2 locus. Disturbances in nuclear condensation that result from abnormalities in the sperm nuclear proteins have been implicated in male factor infertility. Abnormal protamine production may manifest in several ways, including reduced sperm number and abnormal sperm morphology (4 – 8). Protamines have also been im-
plicated in certain disturbances of sperm nuclear condensation that lead to male factor infertility (5– 8). Although the protamine gene cluster is carried on an autosome found in all somatic cells, the genes are expressed only in the male gonad. The protamine locus is an ideal model for studying gene regulation, in that expression is specific for both developmental stage and site. Spermatogenesis can be divided into three major stages: mitosis, meiosis, and spermiogenesis. In the first stage, the stem cells, primitive type A spermatogonia, either differentiate into premeiotic 4N spermatocytes or replicate to replenish the stem cell pool (9, 10). During meiosis, chromosomal synapsis and genetic recombination occur. After two reduction divisions, meiosis I and meiosis II, the haploid spermatid begins the process of spermiogenesis, a phase of morphologic maturation. During spermiogenesis, the genome is restructured and condensed as a result of the replacement of the 739
FIGURE 1 Conservation of gonadal development and morphologic structure of transgenic mice of various ages. (A), Twelve-day-old mouse testis. Original magnification, 3200. (B), Fourteen-day-old mouse testis. Original magnification, 3300. (C), Twentyone– day-old mouse testis. Original magnification, 3250. (D), Six-week-old mouse testis. Original magnification, 3200. Open arrows indicate spermatogonia, solid arrows indicate pachytene spermatocytes, open arrowheads indicate round spermatids, and solid arrowheads indicate mature spermatozoon.
nucleohistones by nucleoprotamines. The spermatids become elongated and develop a tail, the chromatin becomes tightly condensed, and the acrosome develops. The mature spermatozoon adopts a species-specific shape. Spermatogenesis yields an essentially transcriptionally inert mature spermatozoon (11), although RNA is still present (12). Lack of normal chromatin condensation, as a function of proper protamine expression, is indicative of developmental abnormalities that have been associated with certain cases of infertility. This has been supported in part by the observation of infertile individuals who exhibit decreased or absent protamine expression, as a result of a mutation in a region of locus control (8). 740
Stewart et al.
Transgenic human protamine gene cluster
To begin to study the role of protamines in human gametogenesis, a transgenic animal model using the approximately 30-kb human PRM13 PRM23 TNP2 locus was created in C57BL/6 mice. Initial studies demonstrated conservation of spatial expression (i.e., human transcripts were observed only in transgenic testicular tissue) (2). Further, this locus was shown to display copy number– dependent expression (i.e., the amount of product identified was dependent on the number of copies of the transgene integrated into the host genome) (2). Site of integration–independent expression also was identified. This demonstrated that the protamine locus contained all the elements necessary for expression. Vol. 71, No. 4, April 1999
FIGURE 2 Comparison of testicular development between mature, age-matched transgenic and nontransgenic animals. Cellular morphology and testicular development are conserved. (A), Mature, 6-month-old transgenic mouse. Original magnification, 3200. (B), Age-matched, 6-month-old nontransgenic mouse. Original magnification, 3200. Solid arrows indicate pachytene spermatocytes, open arrowheads indicate round spermatids, and solid arrowheads indicate mature spermatozoon.
To complete the characterization of the transgenic PRM13 PRM23 TNP2 locus, our objectives were as follows: [1] to confirm conservation of testicular morphology and function in the transgenic mice; [2] to determine the orientation of the integrated locus; and [3] to define the temporal expression of the transgenic locus.
MATERIALS AND METHODS Serial Matings Heterozygous 12 locus copy transgenic offspring were created by mating F8 homozygous line 882 transgenic male and nontransgenic female mice, Mus musculus. Alternatively, homozygous 9 locus copy offspring were created by mating F7 homozygous line 892 males and females. The transgenic status of each mouse was verified using the tailblood polymerase chain reaction-genotyping protocol as previously described (13). Mice were euthanized by carbon dioxide inhalation followed by cervical dislocation in accordance with the protocol approved by the Department of Laboratory Animals Research at Wayne State University (Institutional Review Board no. A 05-08-96-(02)). Testes were obtained at 12, 17, 28, and 80 days of life. Agematched nontransgenic mice were used as controls.
Testicular Morphology Testicular tissue samples were collected over the course of 2 months. As they were obtained, one testicle from each pair was placed in 10% buffered formalin for fixation and the other was flash frozen and stored at 270°C. Once tissue procurement was complete, the formalin-fixed, gross speciFERTILITY & STERILITYt
mens were embedded in paraffin, sectioned, and stained with hematoxylin and eosin. This tissue collection and processing strategy provided the most efficient means to assemble the required samples from animals of various ages that were collected at different times. As expected, some sample variation was observed (i.e., the sporadic appearance of variously sized vacuole-like spaces), because the size of the tissue and the total time in tissue fixative varied. Slides were viewed and morphology was compared between age-matched transgenic and nontransgenic mice.
Orientation of the Integrated Transgene The orientation of the transgenes was determined by Southern blot hybridization. Total genomic DNA was prepared from the liver of transgenic founders or members of the F1 generation of each transgenic line created essentially as described (14). Approximately 10 mg of genomic DNA was digested to completion with the restriction endonuclease BamHI and then resolved by field inversion gel electrophoresis (MJ Research, Cambridge, MA) on a 0.75% agarose gel. DNA was transferred to positively charged nylon membranes and hybridized with [a32P]-radiolabeled subclone 8e4 (base pairs 38,765–39,464) or 5f1 (base pairs 0 –546) from cosmid clone hP3.1 (accession no. HSU15422) as previously described (15). After hybridization, the membranes were washed for 30 minutes at ambient temperature in a solution of 23 SSPE with 0.1% sodium dodecyl sulfate (SDS) and then at 42°–50°C in 0.13 SSPE with 0.1% SDS for an additional 15–30 minutes. The membranes then were autoradiographed at 270°C for 5–10 days. 741
resolved on 1.5% agarose–formaldehyde gel and transferred to nylon membranes for hybridization.
FIGURE 3 Orientation of the integrated transgene. Genomic DNA was isolated from the livers of transgenic mice from several independent lines, digested to completion with BamHI, and resolved by field inversion gel electrophoresis on a 0.75% agarose gel. Cosmid clone hP3.1 digested with BamHI and lambda phage DNA digested with HindIII provided molecular weight markers as indicated in kilobases. The nucleic acids were transferred to positively charged nylon membranes and then hybridized with radiolabeled subclones 8e4 and 5f1 (data not shown). The membranes were washed at different levels of stringency to remove background hybridization and then autoradiographed at 270°C. A map of the approximately 40-kilobase construct (derived from hP3.1) used in the creation of the transgenic animals is indicated at the bottom. The genes are depicted as arrows, indicating their direction of transcription; BamHI restriction sites are shown as barbells; and the positions of the subclones used as probes are indicated by black boxes. kbp 5 kilobase pair.
Random primed [a32P]-radiolabeled probes specific to the human PRM1, PRM2, and TNP2, and to the mouse b-actin, as well as to the mouse PRM1, PRM2, and TNP2 messenger RNAs were generated as previously described from corresponding specific polymerase chain reaction products (16). (Polymerase chain reaction primers and amplification conditions have been deposited on the Internet at the URL http://compbio.med.wayne.edu/.) The b-actin control probe also hybridizes to the round spermatid-expressed smooth muscle g-actin at approximately 1.5 kb (17). After hybridization, the membranes were washed for 30 minutes at ambient temperature in a solution of 23 SSPE with 0.1% SDS and then at 45°C in 0.13 SSPE with 0.1% SDS for an additional 5–30 minutes to remove nonspecific background hybridization. The membranes then were autoradiographed at 270°C for 6 –24 hours.
RESULTS Testicular Morphology The formalin-preserved testes were stained with hematoxylin and eosin, and morphology was compared between the transgenic and nontransgenic mice. Tissue architecture and sperm maturation were conserved throughout development. As shown in Figure 1, “normal” testicular development as well as cellular morphology was observed in the transgenic mice of various ages. In addition, conservation of testicular structures and cellular development was maintained in transgenic compared with nontransgenic animals (Fig. 2).
Orientation of the Integrated Transgene
Temporal Expression Appropriate temporal expression of the human genes in homozygous mice harboring nine copies of the human transgene was verified using Northern blot analysis, as previously described (2). Briefly, total RNA was isolated from the testis of transgenic mice at 12, 17, 28, and 80 days of age and from age-matched nontransgenic mice (15). Four testes were used for RNA isolation from 12-day-old mice, three from 17-dayold mice, two from 28-day-old mice, and one from 80-dayold mice. The total RNA prepared in this manner then was 742
Stewart et al.
Transgenic human protamine gene cluster
The orientation of the integrated transgene was established by Southern blot analysis. Hybridization using the 39 end–specific subclone 8e4 probe revealed an approximately 20-kb fragment. This indicates that the transgene was inserted in the head-to-tail orientation (Fig. 3). This orientation is independent of the site of integration and the number of copies integrated. Orientation was confirmed using the 59 end-specific subclone 5f1 probe (data not shown). As expected, the terminal end fragments varied by size depending on the site of integration for each transgenic line. The varied end fragments reflected the position of the next endogenous BamHI restriction site.
Temporal Expression Temporal expression of the transgenes was assessed using Northern blot analysis of the messenger RNA isolated from the testes of 12-, 17-, 28-, and 80-day-old mice. The relative expression of both the human transgenic and endogenous mouse messages is summarized in Table 1. As expected and confirmed in Figure 4, the testis-specific transgenes were not expressed in the 12- or 17-day-old mice, consistent with the absence of the mouse homologues (Table 1). However, the Vol. 71, No. 4, April 1999
TABLE 1 Temporal expression of the human and mouse PRM1, PRM2, and TNP2 genes in transgenic and nontransgenic mice as determined by Northern blot analysis. Age (d)
Predominant cell types*
12
Spermatogonia and Sertoli cells
17
Leptotene and zygotene spc
28
Pachytene spc and round spt
80
Elongating spt and mature spz
Transgenic status
hPRM1
mPrm1
hPRM2
mPrm2
hTNP2
mTnp2
Actin†
Positive Negative Positive Negative Positive Negative Positive Negative
2‡ 2 2 2 11 2 11 2
2 2 2 2 11 11 11 11
2 2 2 2 11 2 11 2
2 2 2 2 11 11 11 11
2 2 2 2 1 2 1 2
2 2 2 2 1 1 1 1
11 11 11 11 111 111 111 111
Note: Spc 5 spermatocytes; Spt 5 spermatids; Spz 5 spermatozoon. * Predominant cell types in testes of mice of the given age. † The b-actin probe identifies both the ubiquitously expressed b-actin messenger RNA as well as the postmeiotically expressed smooth muscle g-actin messenger RNA. ‡ The relative level of each transcript is indicated as 1/2.
b-actin message was present, confirming that each sample contained RNA and that approximately the same amount of RNA was loaded in each lane. With the appearance of round spermatids, developmentally appropriate for 28-day-old mice, transgenic messenger RNA was identified for the human PRM1, PRM2, and TNP2 genes along with the endogenous smooth muscle g-actin (Fig. 4). Transcripts from the endogenous mouse locus also were observed at this stage of development (Table 1). As expected, sexually mature and fertile transgenic 80-day-old males also showed expression of both human and mouse messages.
DISCUSSION Morphology Histologic evaluation of testicular morphology showed conservation of structure and spermatogenic development when the human PRM13 PRM23 TNP2 locus was incorporated into the mouse genome and expressed. Spermatogenic arrest would have presented as disordered testicular morphology (18). Fertility was not compromised by the expression of the transgene, as evidenced by the multiple serial matings and the multiple generations of transgenic animals that have been obtained. Accordingly, these animals provide a useful model for evaluating human spermatogenesis, especially in relation to the role of chromatin structure in fertility.
Orientation of the Integrated Transgene The orientation of the various copies of the transgenic locus was established to distinguish whether they were integrated in the head-to-tail or head-to-head manner (19). BamHI digestion of the 40-kb PRM13 PRM23 TNP2 insert from cosmid hP3.1 yields a 39 tail fragment of approxFERTILITY & STERILITYt
imately 13 kb and a 59 head fragment of approximately 7 kb (Fig. 5). On one hand, if head-to-tail integration had occurred, Southern blot analysis using the 39-specific 8e4 probe would have identified an approximately 20-kb fragment. This fragment would have also contained a portion of the head sequence (Fig. 5). Thus, the same band would have been identified when the 59-specific 5f1 head probe was used. On the other hand, if the transgene had been integrated in a head-to-head manner, an approximately 26-kb fragment containing the 39 end of the sequence would have been observed when the 39-specific 8e4 probe was used. Further, an additional approximately 14-kb fragment containing the 59 end of the sequence would have been identified using the 59-specific 5f1 probe. Southern blot analysis (Fig. 3) showed an approximately 20-kb fragment when the 39-specific 8e4 fragment was used as a probe. Similar results were observed when the 59specific 5f1 probe was used. This confirms head-to-tail insertion orientation of the integrated transgene. The terminal end fragment of the insert could be visualized as a faint band. The size of each of the terminal fragments varied between mouse lines depending on the site of integration of the transgene and the next adjacent BamHI restriction site found in the mouse genome.
Temporal Expression Having begun spermatogenesis, the 12-day-old mouse will generate cells up to the zygotene spermatocyte stage of differentiation (10). In comparison, germ cells in every stage of meiosis up to diakinesis can be isolated from the gonads of a 17-day-old mouse. Starting at 21 days of age, the mouse begins to produce round spermatids. As in humans, the expression of the PRM1, PRM2, and TNP2 genes is initiated 743
FIGURE 4
FIGURE 5
Temporal analysis of human transgene expression in mouse testes. Total RNA was isolated from testis of the nine copy homozygous line 892 (nine copy) mice at 12, 17, 28, and 80 days of age, resolved on 1.5% agarose–formaldehyde gel, and transferred to nylon membranes. Membranes were hybridized using probes specific to human PRM1, PRM2, TNP2, and mouse b-actin. The membranes were washed at different levels of stringency to remove nonspecific background hybridization and then autoradiographed at 270°C. The age of the mice used for testicular RNA isolation is indicated above each lane and the size of each message in nucleotides (nt) or kilobases (kb) is indicated along the side. Note that the b-actin probe also identifies the round spermatid– expressed smooth muscle g-actin of approximately 1.5 kb.
Schematic representation of the possible orientation of the inserted transgenes. Representations of the two most common orientations of the inserted transgene are shown. Three copies of the gene cluster are indicated in either a head-totail (top) or a head-to-head/tail-to-tail (bottom) orientation. The arrowheads correspond to the 39 tail end of the transgene. The genes are indicated by lines with filled circles beneath the map and the terminal BamHI restriction sites are indicated by lines with open circles above the map. Top: Both subclone probes 8e4 and 5f1 annealed to an identical approximately 20-kilobase BamHI restriction fragment in a head-to-tail orientation. Bottom: In a head-to-head orientation, 8e4 recognized two sites in an approximately 26-kilobase BamHI restriction fragment, whereas 5f1 annealed to sites within a separate approximately 14-kilobase fragment.
centage increases (10). Accordingly, 28-day-old mice were selected for analysis because of the proportionately higher number of round spermatids present in their gonads. As expected, both the endogenous and transgenic haploid-expressed messages were present in normally developing transgenic mice at 28 days of age. This demonstrated conservation of temporal expression of the human transgene.
at this stage (20). The continuous nature of the spermatogenic wave enables the isolation of cells from all stages of spermatogenesis from the sexually mature male mouse testis. Only transgenic animals that are .21 days of age should express the human message if the elements controlling temporal expression are contained within the transgenic locus. Twelve- and 17-day-old mice were analyzed initially because the endogenous protamine and transition protein message is absent at this stage. In accord with this fact, the transgenic human messages were not observed at this stage of development, indicating appropriate temporal expression. At 21 days of age, approximately 4% of all testicular cells are round spermatids in which the PRM1, PRM2, and TNP2 genes are transcribed (20). As the animal matures, this per744
Stewart et al.
Transgenic human protamine gene cluster
The 80-day-old mice were all of proven fertility. As in the 28-day-old mice, the transgenic products as well as the endogenous products were present (Table 1). This confirmed conservation of temporal expression for both the human transgene and endogenous mouse genes. Thus, all elements necessary for the appropriate temporal expression of the human PRM13 PRM23 TNP2 transgene were contained within the transgenic locus. Characterization of the transgenic human PRM13 PRM23 TNP2 locus has shown that integration of the locus occurs in a head-to-tail manner. Once integrated, expression of the various members is independent of host determinants as copy number– dependent, site of integration–independent expression was observed. Appropriate temporal expression of each member of the PRM13PRM23TNP2 locus now has been shown. The incorporation of the transgene does not affect host development or fertility. As we previously have shown (2), the transcription of each member of the human locus does not appear to affect the transcription of the various members Vol. 71, No. 4, April 1999
of the endogenous mouse locus. Although transcribed, it is not yet known whether these transgenic human messenger RNAs are translated and/or the resulting human protein products are used to package the mouse genome. Nevertheless, each member of this transgenic human locus is replicated and expressed at constant and proportional levels through many generations. The proven stability of this transgenic animal provides an ideal model system for studying factors that affect the transcriptional control of human spermatogenesis. We recently completed a gene mutation scan of the PRM13 PRM23 TNP2 locus in five men with oligospermia (8). Two of these men were found to carry a mutation in a region of locus control analogous to that of gdb-thalassemia (21). The technology that we developed has provided a means to screen men seeking evaluation for their infertility and to initiate appropriate therapy, such as intracytoplasmic sperm injection. Once the regulatory elements are defined, this information may be used to design discrete functioning genic loci that can be engineered for gene therapy. Thus, the analysis of the protamine locus in the transgenic state is now focused on determining the role of the region of locus control that regulates the expression of this gene cluster. The results of our efforts to define this regulatory mechanism have been described recently (22). The clinical implications of this project extend beyond the study of male factor infertility. References 1. Kramer JA, Krawetz SA. RNA in sperm: implications for the alternative genome. Human Molecular Reproduction 1997;3:473– 8. 2. Choudhary SK, Wykes SM, Kramer JA, Mohamed AN, Koppitch F, Nelson JE, et al. A haploid expressed gene cluster exists as a single chromatin domain in human sperm. J Biol Chem 1995;270:8755– 62. 3. Nelson JE, Krawetz SA. Characterization of a human locus in transition. J Biol Chem 1994;269:31067–73.
FERTILITY & STERILITYt
4. Jager S. Sperm nuclear stability and male infertility. Arch Androl 1990;25:253–9. 5. Balhorn R, Reed S, Tanphaichitr N. Aberrant protamine 1/protamine 2 ratios in sperm of infertile human males. Experientia 1988;44:52–5. 6. Belokopytova IA, Kostyleva EI, Tomlin AI, Vorob’ev VI. Human male infertility may be due to a decrease in the protamine P2 content in sperm chromatin. Mol Reprod Dev 1993;34:53–7. 7. De Yerba L, Ballesca JL, Vanrell JA, Bassas L, Oliva R. Complete selective absence of protamine P2 in humans. J Biol Chem 1993;268: 10553–7. 8. Kramer JA, Zhang S, Yaron Y, Zhao Y, Krawetz SA. Genetic testing for male infertility: a postulated role for mutations in sperm nuclear matrix attachment regions. Genetic Testing 1997;1:125–9. 9. Dym M. Spermatogonial stem cells of the testis. Proc Natl Acad Sci USA 1994;91:11287–9. 10. Belve´ AR. Purification, culture, and fractionation of spermatogenic cells. Methods Enzymol 1993;225:84 –113. 11. Kierszenbaum AL, Tres LL. Structural and transcriptional features of the mouse spermatid genome. J Cell Biol 1975;65:258 –70. 12. Wykes SM, Visscher DW, Krawetz SA. Haploid transcripts persist in mature human spermatozoa. Mol Hum Reprod 1997;3:15–9. 13. Kramer JA, Krawetz SA. Analysis of transgenes by PCR: establishing and maintaining the PRM13 PRM23 TNP2 line. Anal Biochem 1996; 235:110 –1. 14. Hogan B, Costantini F, Lacy E. Manipulating the mouse embryo: a laboratory manual. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory USA, 1986:174 –5. 15. Chomczynski P, Sacchi N. Single-step method of RNA isolation by acid guanidinium thiocyanate-phenol-chloroform extraction. Anal Biochem 1987;162:156 –9. 16. Kramer JA, Krawetz SA. Nuclear matrix interactions within the sperm genome. J Biol Chem 1996;271:11619 –22. 17. Kim E, Waters SH, Hake LE, Hecht NB. Identification and developmental expression of a smooth-muscle gamma-actin in postmeiotic male germ cells of mice. Mol Cell Biol 1989;9:1875– 81. 18. Okabe M, Mashahito I, Ashkenas J. Gametogenesis ’98: male infertility and the genetics of spermatogenesis. Am J Hum Genet 1998;62:1274 – 81. 19. Palmiter RD, Brinster RL. Germ-line transformation of mice. Annu Rev Genet 1986;20:465–99. 20. Wykes SM, Nelson JE, Visscher DW, Djakiew D, Krawetz SA. Coordinate expression of the PRM1, PRM2, and TNP2 multigene locus in human testis. DNA Cell Biol 1995;14:155– 61. 21. Forrester WC, Epner E, Driscoll C, Enver T, Brice M, Papayannopoulou T, et al. A deletion of the human b-globin locus activation region causes a major alteration in chromatin structure and replication across the entire b-globin locus. Genes Dev 1990;4:1637– 49. 22. Kramer JA, McCarrey JR, Djakiew D, Krawetz SA. Differentiation: the selective potentiation of chromatin domains. Development 1998;125: 4749 –55.
745
Temporal expression of the transgenic human protamine gene cluster Kathy S. Stewart, M.D.,*† Jeffrey A. Kramer, Ph.D.,*‡ Mark I. Evans, M.D.,*‡ and Stephen A. Krawetz, Ph.D.*‡ Wayne State University, Detroit, Michigan
Received September 11, 1998; revised and accepted November 25, 1998. Supported in part by the Fund for Medical Education and Research and by a Research Faculty Award from Wayne State University, Detroit, Michigan. Jeffrey A. Kramer was supported in part by a Lalor Foundation (Providence, Rhode Island) postdoctoral fellowship. Presented in part at the 18th Annual Meeting of the Society of Perinatal Obstetricians, Miami Beach, Florida, February 2–7, 1998. Reprint requests: Stephen A. Krawetz, Ph.D., Department of Obstetrics and Gynecology and Center for Molecular Medicine and Genetics, Wayne State University School of Medicine, 275 East Hancock, Detroit, Michigan 48201 (FAX: 313577-8854; E-mail: steve @compbio.med.wayne .edu). * Department of Obstetrics and Gynecology. † Present address: Department of Obstetrics and Gynecology, University of Wisconsin, Madison, Wisconsin. ‡ Center for Molecular Medicine and Genetics. 0015-0282/99/$20.00 PII S0015-0282(98)00548-2
Objective: To ascertain the fidelity of expression of the genes from the transgenic human sperm-specific nuclear packaging protamine-13protamine-23transition protein-2 (PRM13 PRM23 TNP2) locus. Design: Controlled human transgene study. Setting: Basic science laboratory. Animal(s): Age-matched transgenic and nontransgenic mice. Intervention(s): Transgenic mice containing the human protamine locus were mated. One testis from each offspring was frozen at 280°C and the other was preserved in formalin. Main Outcome Measure(s): The temporal expression of the human and mouse protamines was evaluated by Northern blot analysis. Orientation of the transgenic locus was determined by Southern blot analysis. Tissue morphology was assessed histologically. Result(s): Conservation of transgenic morphology was confirmed. Head-to-tail integration of the PRM13 PRM23TNP2 locus was shown. Temporal expression of the mouse and human protamine genes was maintained in the transgenic state. Conclusion(s): These results show that the head-to-tail concatomer of the PRM13 PRM23 TNP2 locus contains all the necessary elements for appropriate temporal expression while maintaining testicular structure and function. (Fertil Sterilt 1999;71:739 – 45. ©1999 by American Society for Reproductive Medicine.) Key Words: Protamine, transgenic, infertility, spermatogenesis, Mus musculus, locus
Protamines are low-molecular-weight, malespecific, arginine-rich proteins that replace histones during sperm maturation. This replacement mediates the restructuring of the chromatin, condensing the genetic material into a speciesspecific shaped nucleus that is approximately one-sixth the size of a somatic nucleus (1). The human protamine locus resides within an approximately 30-kilobase (kb) region along human chromosome 16p13.13. Its chromatin domain encompasses a linear array of three genes, protamine 1 (PRM1), protamine 2 (PRM2), and transition protein 2 (TNP2) (2, 3). This cluster of genes is defined as the PRM13 PRM23TNP2 locus. Disturbances in nuclear condensation that result from abnormalities in the sperm nuclear proteins have been implicated in male factor infertility. Abnormal protamine production may manifest in several ways, including reduced sperm number and abnormal sperm morphology (4 – 8). Protamines have also been im-
plicated in certain disturbances of sperm nuclear condensation that lead to male factor infertility (5– 8). Although the protamine gene cluster is carried on an autosome found in all somatic cells, the genes are expressed only in the male gonad. The protamine locus is an ideal model for studying gene regulation, in that expression is specific for both developmental stage and site. Spermatogenesis can be divided into three major stages: mitosis, meiosis, and spermiogenesis. In the first stage, the stem cells, primitive type A spermatogonia, either differentiate into premeiotic 4N spermatocytes or replicate to replenish the stem cell pool (9, 10). During meiosis, chromosomal synapsis and genetic recombination occur. After two reduction divisions, meiosis I and meiosis II, the haploid spermatid begins the process of spermiogenesis, a phase of morphologic maturation. During spermiogenesis, the genome is restructured and condensed as a result of the replacement of the 739
FIGURE 1 Conservation of gonadal development and morphologic structure of transgenic mice of various ages. (A), Twelve-day-old mouse testis. Original magnification, 3200. (B), Fourteen-day-old mouse testis. Original magnification, 3300. (C), Twentyone– day-old mouse testis. Original magnification, 3250. (D), Six-week-old mouse testis. Original magnification, 3200. Open arrows indicate spermatogonia, solid arrows indicate pachytene spermatocytes, open arrowheads indicate round spermatids, and solid arrowheads indicate mature spermatozoon.
nucleohistones by nucleoprotamines. The spermatids become elongated and develop a tail, the chromatin becomes tightly condensed, and the acrosome develops. The mature spermatozoon adopts a species-specific shape. Spermatogenesis yields an essentially transcriptionally inert mature spermatozoon (11), although RNA is still present (12). Lack of normal chromatin condensation, as a function of proper protamine expression, is indicative of developmental abnormalities that have been associated with certain cases of infertility. This has been supported in part by the observation of infertile individuals who exhibit decreased or absent protamine expression, as a result of a mutation in a region of locus control (8). 740
Stewart et al.
Transgenic human protamine gene cluster
To begin to study the role of protamines in human gametogenesis, a transgenic animal model using the approximately 30-kb human PRM13 PRM23 TNP2 locus was created in C57BL/6 mice. Initial studies demonstrated conservation of spatial expression (i.e., human transcripts were observed only in transgenic testicular tissue) (2). Further, this locus was shown to display copy number– dependent expression (i.e., the amount of product identified was dependent on the number of copies of the transgene integrated into the host genome) (2). Site of integration–independent expression also was identified. This demonstrated that the protamine locus contained all the elements necessary for expression. Vol. 71, No. 4, April 1999
FIGURE 2 Comparison of testicular development between mature, age-matched transgenic and nontransgenic animals. Cellular morphology and testicular development are conserved. (A), Mature, 6-month-old transgenic mouse. Original magnification, 3200. (B), Age-matched, 6-month-old nontransgenic mouse. Original magnification, 3200. Solid arrows indicate pachytene spermatocytes, open arrowheads indicate round spermatids, and solid arrowheads indicate mature spermatozoon.
To complete the characterization of the transgenic PRM13 PRM23 TNP2 locus, our objectives were as follows: [1] to confirm conservation of testicular morphology and function in the transgenic mice; [2] to determine the orientation of the integrated locus; and [3] to define the temporal expression of the transgenic locus.
MATERIALS AND METHODS Serial Matings Heterozygous 12 locus copy transgenic offspring were created by mating F8 homozygous line 882 transgenic male and nontransgenic female mice, Mus musculus. Alternatively, homozygous 9 locus copy offspring were created by mating F7 homozygous line 892 males and females. The transgenic status of each mouse was verified using the tailblood polymerase chain reaction-genotyping protocol as previously described (13). Mice were euthanized by carbon dioxide inhalation followed by cervical dislocation in accordance with the protocol approved by the Department of Laboratory Animals Research at Wayne State University (Institutional Review Board no. A 05-08-96-(02)). Testes were obtained at 12, 17, 28, and 80 days of life. Agematched nontransgenic mice were used as controls.
Testicular Morphology Testicular tissue samples were collected over the course of 2 months. As they were obtained, one testicle from each pair was placed in 10% buffered formalin for fixation and the other was flash frozen and stored at 270°C. Once tissue procurement was complete, the formalin-fixed, gross speciFERTILITY & STERILITYt
mens were embedded in paraffin, sectioned, and stained with hematoxylin and eosin. This tissue collection and processing strategy provided the most efficient means to assemble the required samples from animals of various ages that were collected at different times. As expected, some sample variation was observed (i.e., the sporadic appearance of variously sized vacuole-like spaces), because the size of the tissue and the total time in tissue fixative varied. Slides were viewed and morphology was compared between age-matched transgenic and nontransgenic mice.
Orientation of the Integrated Transgene The orientation of the transgenes was determined by Southern blot hybridization. Total genomic DNA was prepared from the liver of transgenic founders or members of the F1 generation of each transgenic line created essentially as described (14). Approximately 10 mg of genomic DNA was digested to completion with the restriction endonuclease BamHI and then resolved by field inversion gel electrophoresis (MJ Research, Cambridge, MA) on a 0.75% agarose gel. DNA was transferred to positively charged nylon membranes and hybridized with [a32P]-radiolabeled subclone 8e4 (base pairs 38,765–39,464) or 5f1 (base pairs 0 –546) from cosmid clone hP3.1 (accession no. HSU15422) as previously described (15). After hybridization, the membranes were washed for 30 minutes at ambient temperature in a solution of 23 SSPE with 0.1% sodium dodecyl sulfate (SDS) and then at 42°–50°C in 0.13 SSPE with 0.1% SDS for an additional 15–30 minutes. The membranes then were autoradiographed at 270°C for 5–10 days. 741
resolved on 1.5% agarose–formaldehyde gel and transferred to nylon membranes for hybridization.
FIGURE 3 Orientation of the integrated transgene. Genomic DNA was isolated from the livers of transgenic mice from several independent lines, digested to completion with BamHI, and resolved by field inversion gel electrophoresis on a 0.75% agarose gel. Cosmid clone hP3.1 digested with BamHI and lambda phage DNA digested with HindIII provided molecular weight markers as indicated in kilobases. The nucleic acids were transferred to positively charged nylon membranes and then hybridized with radiolabeled subclones 8e4 and 5f1 (data not shown). The membranes were washed at different levels of stringency to remove background hybridization and then autoradiographed at 270°C. A map of the approximately 40-kilobase construct (derived from hP3.1) used in the creation of the transgenic animals is indicated at the bottom. The genes are depicted as arrows, indicating their direction of transcription; BamHI restriction sites are shown as barbells; and the positions of the subclones used as probes are indicated by black boxes. kbp 5 kilobase pair.
Random primed [a32P]-radiolabeled probes specific to the human PRM1, PRM2, and TNP2, and to the mouse b-actin, as well as to the mouse PRM1, PRM2, and TNP2 messenger RNAs were generated as previously described from corresponding specific polymerase chain reaction products (16). (Polymerase chain reaction primers and amplification conditions have been deposited on the Internet at the URL http://compbio.med.wayne.edu/.) The b-actin control probe also hybridizes to the round spermatid-expressed smooth muscle g-actin at approximately 1.5 kb (17). After hybridization, the membranes were washed for 30 minutes at ambient temperature in a solution of 23 SSPE with 0.1% SDS and then at 45°C in 0.13 SSPE with 0.1% SDS for an additional 5–30 minutes to remove nonspecific background hybridization. The membranes then were autoradiographed at 270°C for 6 –24 hours.
RESULTS Testicular Morphology The formalin-preserved testes were stained with hematoxylin and eosin, and morphology was compared between the transgenic and nontransgenic mice. Tissue architecture and sperm maturation were conserved throughout development. As shown in Figure 1, “normal” testicular development as well as cellular morphology was observed in the transgenic mice of various ages. In addition, conservation of testicular structures and cellular development was maintained in transgenic compared with nontransgenic animals (Fig. 2).
Orientation of the Integrated Transgene
Temporal Expression Appropriate temporal expression of the human genes in homozygous mice harboring nine copies of the human transgene was verified using Northern blot analysis, as previously described (2). Briefly, total RNA was isolated from the testis of transgenic mice at 12, 17, 28, and 80 days of age and from age-matched nontransgenic mice (15). Four testes were used for RNA isolation from 12-day-old mice, three from 17-dayold mice, two from 28-day-old mice, and one from 80-dayold mice. The total RNA prepared in this manner then was 742
Stewart et al.
Transgenic human protamine gene cluster
The orientation of the integrated transgene was established by Southern blot analysis. Hybridization using the 39 end–specific subclone 8e4 probe revealed an approximately 20-kb fragment. This indicates that the transgene was inserted in the head-to-tail orientation (Fig. 3). This orientation is independent of the site of integration and the number of copies integrated. Orientation was confirmed using the 59 end-specific subclone 5f1 probe (data not shown). As expected, the terminal end fragments varied by size depending on the site of integration for each transgenic line. The varied end fragments reflected the position of the next endogenous BamHI restriction site.
Temporal Expression Temporal expression of the transgenes was assessed using Northern blot analysis of the messenger RNA isolated from the testes of 12-, 17-, 28-, and 80-day-old mice. The relative expression of both the human transgenic and endogenous mouse messages is summarized in Table 1. As expected and confirmed in Figure 4, the testis-specific transgenes were not expressed in the 12- or 17-day-old mice, consistent with the absence of the mouse homologues (Table 1). However, the Vol. 71, No. 4, April 1999
TABLE 1 Temporal expression of the human and mouse PRM1, PRM2, and TNP2 genes in transgenic and nontransgenic mice as determined by Northern blot analysis. Age (d)
Predominant cell types*
12
Spermatogonia and Sertoli cells
17
Leptotene and zygotene spc
28
Pachytene spc and round spt
80
Elongating spt and mature spz
Transgenic status
hPRM1
mPrm1
hPRM2
mPrm2
hTNP2
mTnp2
Actin†
Positive Negative Positive Negative Positive Negative Positive Negative
2‡ 2 2 2 11 2 11 2
2 2 2 2 11 11 11 11
2 2 2 2 11 2 11 2
2 2 2 2 11 11 11 11
2 2 2 2 1 2 1 2
2 2 2 2 1 1 1 1
11 11 11 11 111 111 111 111
Note: Spc 5 spermatocytes; Spt 5 spermatids; Spz 5 spermatozoon. * Predominant cell types in testes of mice of the given age. † The b-actin probe identifies both the ubiquitously expressed b-actin messenger RNA as well as the postmeiotically expressed smooth muscle g-actin messenger RNA. ‡ The relative level of each transcript is indicated as 1/2.
b-actin message was present, confirming that each sample contained RNA and that approximately the same amount of RNA was loaded in each lane. With the appearance of round spermatids, developmentally appropriate for 28-day-old mice, transgenic messenger RNA was identified for the human PRM1, PRM2, and TNP2 genes along with the endogenous smooth muscle g-actin (Fig. 4). Transcripts from the endogenous mouse locus also were observed at this stage of development (Table 1). As expected, sexually mature and fertile transgenic 80-day-old males also showed expression of both human and mouse messages.
DISCUSSION Morphology Histologic evaluation of testicular morphology showed conservation of structure and spermatogenic development when the human PRM13 PRM23 TNP2 locus was incorporated into the mouse genome and expressed. Spermatogenic arrest would have presented as disordered testicular morphology (18). Fertility was not compromised by the expression of the transgene, as evidenced by the multiple serial matings and the multiple generations of transgenic animals that have been obtained. Accordingly, these animals provide a useful model for evaluating human spermatogenesis, especially in relation to the role of chromatin structure in fertility.
Orientation of the Integrated Transgene The orientation of the various copies of the transgenic locus was established to distinguish whether they were integrated in the head-to-tail or head-to-head manner (19). BamHI digestion of the 40-kb PRM13 PRM23 TNP2 insert from cosmid hP3.1 yields a 39 tail fragment of approxFERTILITY & STERILITYt
imately 13 kb and a 59 head fragment of approximately 7 kb (Fig. 5). On one hand, if head-to-tail integration had occurred, Southern blot analysis using the 39-specific 8e4 probe would have identified an approximately 20-kb fragment. This fragment would have also contained a portion of the head sequence (Fig. 5). Thus, the same band would have been identified when the 59-specific 5f1 head probe was used. On the other hand, if the transgene had been integrated in a head-to-head manner, an approximately 26-kb fragment containing the 39 end of the sequence would have been observed when the 39-specific 8e4 probe was used. Further, an additional approximately 14-kb fragment containing the 59 end of the sequence would have been identified using the 59-specific 5f1 probe. Southern blot analysis (Fig. 3) showed an approximately 20-kb fragment when the 39-specific 8e4 fragment was used as a probe. Similar results were observed when the 59specific 5f1 probe was used. This confirms head-to-tail insertion orientation of the integrated transgene. The terminal end fragment of the insert could be visualized as a faint band. The size of each of the terminal fragments varied between mouse lines depending on the site of integration of the transgene and the next adjacent BamHI restriction site found in the mouse genome.
Temporal Expression Having begun spermatogenesis, the 12-day-old mouse will generate cells up to the zygotene spermatocyte stage of differentiation (10). In comparison, germ cells in every stage of meiosis up to diakinesis can be isolated from the gonads of a 17-day-old mouse. Starting at 21 days of age, the mouse begins to produce round spermatids. As in humans, the expression of the PRM1, PRM2, and TNP2 genes is initiated 743
FIGURE 4
FIGURE 5
Temporal analysis of human transgene expression in mouse testes. Total RNA was isolated from testis of the nine copy homozygous line 892 (nine copy) mice at 12, 17, 28, and 80 days of age, resolved on 1.5% agarose–formaldehyde gel, and transferred to nylon membranes. Membranes were hybridized using probes specific to human PRM1, PRM2, TNP2, and mouse b-actin. The membranes were washed at different levels of stringency to remove nonspecific background hybridization and then autoradiographed at 270°C. The age of the mice used for testicular RNA isolation is indicated above each lane and the size of each message in nucleotides (nt) or kilobases (kb) is indicated along the side. Note that the b-actin probe also identifies the round spermatid– expressed smooth muscle g-actin of approximately 1.5 kb.
Schematic representation of the possible orientation of the inserted transgenes. Representations of the two most common orientations of the inserted transgene are shown. Three copies of the gene cluster are indicated in either a head-totail (top) or a head-to-head/tail-to-tail (bottom) orientation. The arrowheads correspond to the 39 tail end of the transgene. The genes are indicated by lines with filled circles beneath the map and the terminal BamHI restriction sites are indicated by lines with open circles above the map. Top: Both subclone probes 8e4 and 5f1 annealed to an identical approximately 20-kilobase BamHI restriction fragment in a head-to-tail orientation. Bottom: In a head-to-head orientation, 8e4 recognized two sites in an approximately 26-kilobase BamHI restriction fragment, whereas 5f1 annealed to sites within a separate approximately 14-kilobase fragment.
centage increases (10). Accordingly, 28-day-old mice were selected for analysis because of the proportionately higher number of round spermatids present in their gonads. As expected, both the endogenous and transgenic haploid-expressed messages were present in normally developing transgenic mice at 28 days of age. This demonstrated conservation of temporal expression of the human transgene.
at this stage (20). The continuous nature of the spermatogenic wave enables the isolation of cells from all stages of spermatogenesis from the sexually mature male mouse testis. Only transgenic animals that are .21 days of age should express the human message if the elements controlling temporal expression are contained within the transgenic locus. Twelve- and 17-day-old mice were analyzed initially because the endogenous protamine and transition protein message is absent at this stage. In accord with this fact, the transgenic human messages were not observed at this stage of development, indicating appropriate temporal expression. At 21 days of age, approximately 4% of all testicular cells are round spermatids in which the PRM1, PRM2, and TNP2 genes are transcribed (20). As the animal matures, this per744
Stewart et al.
Transgenic human protamine gene cluster
The 80-day-old mice were all of proven fertility. As in the 28-day-old mice, the transgenic products as well as the endogenous products were present (Table 1). This confirmed conservation of temporal expression for both the human transgene and endogenous mouse genes. Thus, all elements necessary for the appropriate temporal expression of the human PRM13 PRM23 TNP2 transgene were contained within the transgenic locus. Characterization of the transgenic human PRM13 PRM23 TNP2 locus has shown that integration of the locus occurs in a head-to-tail manner. Once integrated, expression of the various members is independent of host determinants as copy number– dependent, site of integration–independent expression was observed. Appropriate temporal expression of each member of the PRM13PRM23TNP2 locus now has been shown. The incorporation of the transgene does not affect host development or fertility. As we previously have shown (2), the transcription of each member of the human locus does not appear to affect the transcription of the various members Vol. 71, No. 4, April 1999
of the endogenous mouse locus. Although transcribed, it is not yet known whether these transgenic human messenger RNAs are translated and/or the resulting human protein products are used to package the mouse genome. Nevertheless, each member of this transgenic human locus is replicated and expressed at constant and proportional levels through many generations. The proven stability of this transgenic animal provides an ideal model system for studying factors that affect the transcriptional control of human spermatogenesis. We recently completed a gene mutation scan of the PRM13 PRM23 TNP2 locus in five men with oligospermia (8). Two of these men were found to carry a mutation in a region of locus control analogous to that of gdb-thalassemia (21). The technology that we developed has provided a means to screen men seeking evaluation for their infertility and to initiate appropriate therapy, such as intracytoplasmic sperm injection. Once the regulatory elements are defined, this information may be used to design discrete functioning genic loci that can be engineered for gene therapy. Thus, the analysis of the protamine locus in the transgenic state is now focused on determining the role of the region of locus control that regulates the expression of this gene cluster. The results of our efforts to define this regulatory mechanism have been described recently (22). The clinical implications of this project extend beyond the study of male factor infertility. References 1. Kramer JA, Krawetz SA. RNA in sperm: implications for the alternative genome. Human Molecular Reproduction 1997;3:473– 8. 2. Choudhary SK, Wykes SM, Kramer JA, Mohamed AN, Koppitch F, Nelson JE, et al. A haploid expressed gene cluster exists as a single chromatin domain in human sperm. J Biol Chem 1995;270:8755– 62. 3. Nelson JE, Krawetz SA. Characterization of a human locus in transition. J Biol Chem 1994;269:31067–73.
FERTILITY & STERILITYt
4. Jager S. Sperm nuclear stability and male infertility. Arch Androl 1990;25:253–9. 5. Balhorn R, Reed S, Tanphaichitr N. Aberrant protamine 1/protamine 2 ratios in sperm of infertile human males. Experientia 1988;44:52–5. 6. Belokopytova IA, Kostyleva EI, Tomlin AI, Vorob’ev VI. Human male infertility may be due to a decrease in the protamine P2 content in sperm chromatin. Mol Reprod Dev 1993;34:53–7. 7. De Yerba L, Ballesca JL, Vanrell JA, Bassas L, Oliva R. Complete selective absence of protamine P2 in humans. J Biol Chem 1993;268: 10553–7. 8. Kramer JA, Zhang S, Yaron Y, Zhao Y, Krawetz SA. Genetic testing for male infertility: a postulated role for mutations in sperm nuclear matrix attachment regions. Genetic Testing 1997;1:125–9. 9. Dym M. Spermatogonial stem cells of the testis. Proc Natl Acad Sci USA 1994;91:11287–9. 10. Belve´ AR. Purification, culture, and fractionation of spermatogenic cells. Methods Enzymol 1993;225:84 –113. 11. Kierszenbaum AL, Tres LL. Structural and transcriptional features of the mouse spermatid genome. J Cell Biol 1975;65:258 –70. 12. Wykes SM, Visscher DW, Krawetz SA. Haploid transcripts persist in mature human spermatozoa. Mol Hum Reprod 1997;3:15–9. 13. Kramer JA, Krawetz SA. Analysis of transgenes by PCR: establishing and maintaining the PRM13 PRM23 TNP2 line. Anal Biochem 1996; 235:110 –1. 14. Hogan B, Costantini F, Lacy E. Manipulating the mouse embryo: a laboratory manual. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory USA, 1986:174 –5. 15. Chomczynski P, Sacchi N. Single-step method of RNA isolation by acid guanidinium thiocyanate-phenol-chloroform extraction. Anal Biochem 1987;162:156 –9. 16. Kramer JA, Krawetz SA. Nuclear matrix interactions within the sperm genome. J Biol Chem 1996;271:11619 –22. 17. Kim E, Waters SH, Hake LE, Hecht NB. Identification and developmental expression of a smooth-muscle gamma-actin in postmeiotic male germ cells of mice. Mol Cell Biol 1989;9:1875– 81. 18. Okabe M, Mashahito I, Ashkenas J. Gametogenesis ’98: male infertility and the genetics of spermatogenesis. Am J Hum Genet 1998;62:1274 – 81. 19. Palmiter RD, Brinster RL. Germ-line transformation of mice. Annu Rev Genet 1986;20:465–99. 20. Wykes SM, Nelson JE, Visscher DW, Djakiew D, Krawetz SA. Coordinate expression of the PRM1, PRM2, and TNP2 multigene locus in human testis. DNA Cell Biol 1995;14:155– 61. 21. Forrester WC, Epner E, Driscoll C, Enver T, Brice M, Papayannopoulou T, et al. A deletion of the human b-globin locus activation region causes a major alteration in chromatin structure and replication across the entire b-globin locus. Genes Dev 1990;4:1637– 49. 22. Kramer JA, McCarrey JR, Djakiew D, Krawetz SA. Differentiation: the selective potentiation of chromatin domains. Development 1998;125: 4749 –55.
745