| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Articles |
a Department of Biology, Brandeis University, Waltham, Massachusetts 02454
b Department of Biology, Boston College, Chestnut Hill, Massachusetts 02167
c Boston Fertility Laboratories, Brookline, Massachusetts 02146
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Given these considerations, we have built an in vitro system for efficient human pronucleus formation and replication. The process of human sperm pronucleus formation has previously been analyzed in intact hamster eggs [5] and in several in vitro systems using Xenopus egg extracts [6–9]. The system we describe here also utilizes Xenopus egg extracts, but it is more efficient and synchronous than previous systems. The key difference is that our system uses two different Xenopus egg extracts: one prepared from unactivated Xenopus eggs arrested in meiotic metaphase II (MII-Extract), and a second prepared from activated eggs at the peak of their DNA synthetic capacity for human sperm pronuclei (Interphase-Extract). This two-cytoplasm approach mimics the events of fertilization. During normal fertilization, the sperm nucleus with its surrounding cytoskeleton contacts the cytoplasm of the egg while the egg is still arrested in meiotic metaphase II. The egg then reenters the cell cycle, and its cytoplasm advances into the interphase state. Earlier studies from our laboratory have demonstrated that the two-cytoplasm approach results in efficient plasmid DNA synthesis in intact Xenopus eggs [10] and egg extracts [11], as well as efficient reactivation and replication of Xenopus red blood cell nuclei, which, like sperm nuclei, are quiescent and highly condensed [12, 13]. The two-cytoplasm approach is also the basis for several recent successes in mammalian cloning [14–16], in accord with our predictions [13].
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Female frogs, obtained from Nasco (Fort Atkinson, WI), were cared for and ovulated as described previously, and the resulting eggs were hardened and used to prepare both high-speed MII-Extract (previously called CSF-Extract) and low-speed Interphase-Extract (previously called Activated-Extract) as previously described [13, 17]. The eggs used for preparation of Interphase-Extract were activated by calcium ionophore treatment and incubated at 20°C for a total of 22 min, rather than 28 min, before being centrifuged (see Results). The kinetics of cell cycle progression at 20°C has been previously established [10, 13].
Preparation of Human Sperm
Frozen samples of sperm from fertile donors and subfertile patients were provided by Boston Fertility Laboratories, Inc. (Brookline, MA). Samples of sperm were obtained from fertile sperm donors and from subfertile infertility patients. The sperm donors were contributors to the Boston Fertility Laboratory sperm bank whose sperm had been used to achieve ongoing pregnancies through in vitro fertilization or intrauterine insemination. These samples had prefreezing sperm counts between 60 and 70 million/ml, a sperm morphology range of 6–14% normal forms, and at least 50% motility using strict criteria developed at Boston Fertility Laboratory. The subfertile males were from couples seeking infertility treatment due to failure to achieve pregnancy after unprotected intercourse for a least one year. The primary diagnosis in each case was male factor infertility, and all samples showed less than 10% normal morphology readings. All donors and patients gave informed consent for use of their sperm according to the guidelines approved by the Committee on Clinical Investigations at the Beth Israel Deaconess Hospital, Boston, MA. Sperm morphology was assessed using the strict criteria implemented by Boston Fertility Laboratories, Inc.
All samples were first allowed to liquefy and were then mixed with an equal volume of TEST Yolk buffer (Irvine Scientific, Santa Ana, CA) before being frozen by exposing aliquots to liquid nitrogen vapor for 2 h followed by immersion in liquid nitrogen. As needed, sperm samples were thawed on ice and washed twice by suspension and centrifugation in cold nuclear isolation buffer, NIB (250 mM sucrose, 25 mM NaCl, 10 mM PIPES, 1.5 mM MgCl2, 0.5 mM spermidine, 0.15 mM spermine, pH 7.0) for 15 min at 1200 x g at 4°C. The resulting pellet was resuspended in NIB, and aliquots containing 3 x 107 cells in 200 µl were then refrozen in liquid N2 in siliconized tubes.
In Vitro Reactivation of Human Sperm Nuclei
Chemical pretreatment. Individual aliquots of human sperm were thawed on ice and were permeabilized by incubation in 100 µg/ml of lysolecithin in NIB for 5 min at 25°C in a final volume of 1 ml. Lysolecithin treatment was then stopped by addition of BSA and soybean trypsin inhibitor to final concentrations of 0.4% (w:v) and 30 µg/ml, respectively, and samples were centrifuged at 1200 x g for 20 min at 2°C. The pellet was washed once in 1 ml of 0.4% BSA in NIB. After removal of the supernatant, the pellet was incubated for 20 min at 25°C in 400 µl NIB containing 5 mM dithiothreitol (DTT). DTT reduction was stopped by addition of N-ethylmaleimide to a 1 mM concentration and incubated for 10 min at 25°C. Samples were centrifuged at 1200 x g for 20 min at 2°C, and each pellet was resuspended in NIB to a final concentration of 2 x 104 sperm/µl.
Further pretreatment in MII-Extract. Frozen aliquots of MII-Extract were thawed on ice and were supplemented with 1/10th volume of a 10-strength ATP-regenerating mix (single-strength: 0.4 mM creatine phosphate, 0.4 µg/ml creatine phosphokinase, 0.1 mM CaCl2 in sterile water). Pretreated sperm were then diluted 1:5 into the extract to a final concentration of 4000 sperm/µl. The reaction was incubated at 25°C for 10–18 min, depending on the MII-Extract, until the sperm had undergone maximum swelling as observed by fluorescent staining with Hoechst 33342. The sample was then placed on ice for 60 min.
Nuclear activation and replication in Interphase-Extract. The MII-Extract was triggered to enter interphase by addition of 1.2 mM CaCl2 and incubation at 25°C for 10 min; it was then diluted into 9 volumes of Interphase-Extract supplemented with 0.4 mM creatine phosphate and 0.4 µg/ml creatine phosphokinase. The Interphase-Extract was incubated at 25°C for 180 min and throughout this time was sampled at regular intervals by withdrawing 5-µl aliquots. The kinetics of DNA synthesis were determined by incorporation of [
-32P]dCTP followed by gel electrophoresis of the labeled product; DNA replication on a per nucleus level was determined by incorporation of biotinylated dUTP followed by Texas Red/streptavidin staining; quantitation of the extent of DNA synthesis was determined by incorporation of bromo-dUTP (Br-dUTP) followed by CsCl density gradient centrifugation, as previously described [10, 13].
By using Gaussian curve analysis of the peaks resolved in the CsCl density gradient, we determined the percentages of replicated and unreplicated molecules present at the beginning and end of S-phase and used these values to deduce the efficiency of replication [10]. The formula used to quantitate the process of DNA replication was:
HL ;eq once-replicated heavy-light DNA; LL ;eq unreplicated light-light DNA; LLH ;eq partially replicated DNA.
Fluorescent Imaging and Deconvolution
Fluorescent images of nuclei stained with Hoechst 33342 were acquired and analyzed using the CELLscan deconvolution system as described previously [18]. In brief, this system collects a series of optical sections at focal planes 0.25 µm apart using a 100x/1.4 N.A. Olympus (Tokyo, Japan) objective connected to a piezoelectric Z-axis focusing device and a computer-controlled excitation light shutter. The set of images is then deconvolved by application of the Exhaustive Photon Reassignment algorithm built into the CELLscan software. This algorithm vectorially reassigns the light haze contributed by fluorescent structures located above and below the plane of optimal focus to its proper places of origin after accurate characterization of the blurring function of the optical system. Measurements of distances in the XY-axis were performed after calibration of the CELLscan software using a CELLscan system stage micrometer as an external metric standard.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
We set out to develop a 4-step protocol for in vitro reactivation and replication of human sperm that recapitulates the process of pronucleus development in fertilized human eggs. In this protocol, intact sperm were first treated with lysolecithin to permeabilize the plasma membrane and then with DTT to disrupt the disulfide bonds between the sperm protamines. Chemically pretreated sperm were then incubated in MII-Extract and next diluted into Interphase-Extract, where pronucleus formation and DNA replication take place. Figure 1 demonstrates that efficient replication in pretreated human sperm nuclei in the Interphase-Extract was critically dependent on a limited period of preincubation in MII-Extract. In 5 separate experiments, sperm samples not preincubated in MII-Extracts failed to replicate DNA upon dilution into Interphase-Extract (data not shown). Cytological examination revealed that human sperm nuclei were still very compact following lysolecithin permeabilization and disulfide reduction, but they swelled very rapidly when they were incubated in MII-Extract (Fig. 2). The point of maximum swelling was reached after 10–18 min (average: 12 min, from 25 separate experiments) depending on the particular batch of MII-Extract. Thereafter, nuclear size decreased slightly and the contour of each nucleus appeared more ragged (Fig. 2). Nuclei eventually formed dispersed chromosomal fibers (data not shown).
|
|
Nuclei preincubated in MII-Extract to the point when they first achieved maximum swelling went on to synthesize DNA rapidly once diluted into Interphase-Extract (compare Figs. 1 and 3). In contrast, sperm nuclei not exposed to MII-Extract did not synthesize DNA in Interphase-Extract. Nuclei preincubated in MII-Extract for more than the optimal length of time exhibited lower overall replication and stopped DNA synthesis sooner (compare 15-min and 20-min samples in Fig. 1). Thus, measurement of nuclear area in MII-Extract provided a convenient means of optimizing subsequent replication in Interphase-Extract.
Because activated Xenopus eggs progress through the cell cycle very quickly, small differences in the time or temperature at which activated eggs are incubated prior to preparation of Interphase-Extract have major effects on the intrinsic capacity of the extract to support DNA synthesis. For instance, extracts prepared from eggs incubated for 28 min at 20°C after activation with calcium ionophore A23187 exhibit the highest DNA synthetic capacity in Xenopus erythrocyte nuclei [13]. In order to establish which Interphase-Extract exhibited the highest capacity for DNA synthesis in pretreated human sperm nuclei, a large batch of unfertilized Xenopus eggs was synchronously activated at 20°C; groups of eggs were removed after 12, 17, 22, and 26 min and were then used to prepare separate extracts. The results demonstrate that all 4 extracts supported DNA synthesis in human sperm pronuclei, but the extract prepared from eggs incubated for 22 min after activation had the highest DNA synthetic capacity (Fig. 4). These findings were reproduced on 4 separate occasions, and results from 2 of those experiments are shown in Figure 4. The reasons for the difference between the optima for frog erythrocyte and human sperm nuclei remain to be explored but may relate to the fact that sperm chromatin is composed of protamines while erythrocyte chromatin contains histones.
|
The Kinetics of DNA Synthesis in the Optimized System
Chemically pretreated sperm nuclei incubated in MII-Extract and then diluted into a 22-min Interphase-Extract decondensed, formed pronuclei, and replicated DNA (Fig. 5). The overall rate of DNA synthesis in such reactions was readily measured by [-32P]dCTP incorporation. The results in Figure 6 (and below) demonstrate that the lag period and rate of DNA synthesis were rather similar in many separate experiments. DNA synthesis ended abruptly and was followed by a plateau period, during which [-32P]-labeled DNA was recovered as high molecular weight molecules (not shown). Addition of aphidicholin, an inhibitor of DNA polymerase , blocked subsequent incorporation of [-32P]dCTP completely (results not shown), indicating that [-32P]dCTP incorporation was not due to DNA repair.
|
|
Cytological Analysis of the Synchrony of Nuclear Replication
Labeling with [-32P]dCTP is the most convenient technique for establishing relative rates of DNA synthesis, but it does not reveal the percentage of pretreated nuclei that initiate and proceed through DNA replication. In order to measure this aspect of replication, we employed continuous incorporation of biotinylated dUTP followed by staining with Texas Red/streptavidin (see Materials and Methods). Prior to the start of S-phase, all nuclei are small and compact and do not stain red (i.e., do not incorporate biotinylated dUTP). When S-phase begins, nuclei swell and stain light red, indicating that some DNA synthesis has occurred. As S-phase continues, nuclei become larger and redder, indicating that more and more newly synthesized DNA has accumulated in each nucleus (see Fig. 5). Using this approach we scored the percentages of not-red, light-red, and bright-red nuclei in samples collected before and after S-phase. Table 1 displays those data obtained in 5 separate experiments. These results demonstrate that 78–97% of nuclei (average 87%) replicated enough DNA by the end of S-phase to be scored as bright-red. The remaining nuclei, 3–22% (average 13%), were either still unreplicated (not-red) or partially replicated (light-red), even after 165 min.
|
Deconvolution of Fluorescent Images Also Revealed the Homogeneity of Replicating Nuclei
Conventional fluorescent imaging is useful for assessing relative levels of Texas Red/streptavidin staining among biotin-labeled nuclei, but it does not generate particularly clear images of nuclear structure. This limitation is typical of virtually all fluorescent objects and is attributable to the fact that light emitted from all out-of-focus planes is superimposed on the light emanating from the in-focus plane being examined. Mathematical deconvolution of fluorescent images allows the out-of-focus light to be reassigned back to its plane of origin [19, 20]. Figure 7 illustrates deconvolved images of 12 different human sperm pronuclei during the course of S-phase. Two characteristics of these images stand out: 1) the images demonstrate that the chromatin within decondensing nuclei changed in structure from very thick, ropelike fibers to very thin, threadlike fibers; 2) they show that at each point in time, different nuclei had very similar structures.
|
Quantitative Analysis of the Extent of Genome Replication via CsCl Labeling
The biotinylated dUTP-labeling experiments described above demonstrate that 78–97% of human sperm pronuclei replicated DNA within 2 h in the optimized system, but they do not rigorously demonstrate the extent of genome replication. Quantitative analysis of total DNA synthesis was carried by labeling replicating DNA with Br-dUTP and then analyzing it via CsCl density gradient centrifugation. Figure 8 reveals that at the start of S-phase (60 min), all of the DNA was unlabeled and was therefore recovered from the light-light (LL) position on the gradient. In contrast, at the end of S-phase (180 min), 21.4% of the total DNA remained unreplicated [LL], while 76.7% had replicated once and was recovered as heavy-light (HL) DNA. An additional 1.9% of the DNA was of intermediate density. The percentage of LL DNA present after 180 min (21.4%) was consistent with the maximum percentage of not-red and light-red nuclei present at the end of S-phase in biotinylated dUTP-labeled samples (22%, see Table 1). This situation may reflect the fact that Br-dUTP partially inhibits DNA synthesis when added at the concentrations needed for extensive substitution of thymidine [10]. Taken together, these results lead us to conclude that biotin-labeled nuclei scored as bright-red by the end of S-phase replicated all or nearly all of each of their genomes.
|
In order to account for complete genome replication in such a very short S-phase in our in vitro system, virtually all replicons in each nucleus must initiate DNA synthesis at approximately the same time. This conclusion is consistent with Newport's direct measurements of replicon elongation in Xenopus sperm nuclei [21] and with classical observations of replicating DNA in early embryonic nuclei [22]. Random shearing of partially replicated DNA molecules labeled with Br-dUTP would be expected to generate fragments of HL DNA from fully replicated replicons, fragments of LL DNA from unreplicated DNA, and DNA fragments of intermediate density corresponding to transient replication intermediates consisting of unreplicated and replicated DNA (LLH DNA). CsCl density gradient analysis of samples prepared partway through S-phase (120 min in Fig. 8C) demonstrate the presence of both an HL DNA peak and a peak that has a density slightly greater than that of LL DNA. We call this material LLH DNA, and we consistently observed that the LLH peak decreased as the HL DNA increased toward the end of S-phase.
Use of the Xenopus Egg Extract System to Compare the Responses of Sperm from Fertile and Subfertile Men
To assess the diagnostic potential of our Xenopus egg extract system, we have started to compare the responses of sperm samples from known fertile and subfertile patients. Thus far samples have been prepared from 10 subfertile men undergoing fertility treatment at the Boston Fertility Laboratories, Inc. All samples scored in the 0–3% range for normal morphology. Control samples were obtained from 5 proven fertile donors displaying 6–14% normal sperm morphology.
Regardless of the fertility status of each donor, all sperm samples underwent nuclear envelope breakdown and chromatin reorganization in the MII-Extract and went on to acquire envelopes, form pronuclei, and undergo DNA replication upon dilution into the Interphase-Extract. Analysis of the kinetics of sperm DNA replication from 15 separate experiments using 5 proven fertile donors established the range of responses that could be observed in normal sperm samples (Fig. 9A). Two of the samples from the 8 subfertile patients displayed rates of [-32P]dCTP incorporation that were faster than any of those observed for the proven fertile donors. These differences in replication rates were most evident after 90 min, just after DNA synthesis had begun. Fluorescent microscopy did not reveal why the 2 subfertile samples synthesized DNA more rapidly (data not shown).
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
To our knowledge this is the first fully optimized system for reactivation and replication of human sperm nuclei in Xenopus egg extracts. Xu et al. [8] recently reported complete human sperm nuclear replication in a system employing only a single extract prepared from activated Xenopus eggs. However, DNA synthesis was only 60% complete after 4 h and required 9 h to achieve genome doubling. Fertilized Xenopus eggs divide approximately 12 times and synthesize at least 1000 genomes worth of DNA in that amount of time. Although we have not directly compared our system to that of Xu et al. [8], our results suggest that the rate of DNA synthesis is greatly increased by preincubation of sperm nuclei in MII-Extract prior to dilution into Interphase-Extract. We further propose that the reason Xu et al. [8] observed DNA replication at all is that they prepared their extract from metaphase-II-arrested eggs that were activated at the time of extract preparation. Extracts prepared under these conditions contain transient levels of H1 kinase activity.
We foresee that in vitro systems for human sperm pronuclear formation and replication will prove useful for a variety of future studies. For instance, we are currently using fluorescent in situ hybridization to analyze the location and unfolding of centromeres and telomeres in decondensing sperm nuclei. Our results confirm that most centromeres co-localize in a chromocenter in the condensed sperm genome [25], and they demonstrate that the chromocenter persists during the initial decondensation process in MII-Extract (personal communication with Serra et al.).
We are also utilizing our optimized system to study the properties of sperm from normal and subfertile men. Preliminary comparisons of the responses of known fertile and subfertile sperm in our optimized system show no correlation between low percentage normal morphology and the ability of sperm from subfertile men to form pronuclei and replicate in vitro. However, relative to proven fertile sperm, 2 of 8 subfertile sperm samples exhibited slightly faster-than-normal kinetics of DNA replication. Further analysis will be required to determine whether differences in the kinetics of pronuclear DNA synthesis can be used to distinguish fertile from subfertile sperm samples and whether a heterologous system (Xenopus egg extract) can produce data that are representative of a homologous system (human eggs).
Brown and coworkers [26] have reported that a significant percentage of sperm samples prepared from infertile men undergo abnormal nuclear swelling and DNA synthesis in an egg extract assay, while virtually no control samples from fertile men fail the same assays. These investigators use a single extract prepared from activated eggs, as compared to our two-extract system. It is possible that we have not observed major differences in sperm nuclear decondensation and replication among samples from infertile men because of the high efficiency of our two-extract system. Unlike our experimental system, the one-extract system described by Brown and coworkers [26–28] provides no evidence for complete genome replication, even among control sperm samples. As we have tried to emphasize and demonstrate in this paper, mere detection of nuclear swelling, or incorporation of a radioactive precursor, does not demonstrate whether DNA synthesis is efficient, extensive, or synchronous.
Finally, we suggest that our in vitro system may provide insights into the critical events that must take place within the human egg after ICSI to achieve normal development. For instance, we have observed that the length of time that a sperm nucleus is incubated in the MII-Extract greatly affects the rate and extent of DNA replication in the subsequent S-phase. Maximum replication occurs when maximum nuclear swelling takes place prior to S-phase; but fully swelled nuclei gradually become more compact if they remain in the MII-phase cytoplasm too long, and they no longer replicate efficiently. These phenomena warrant detailed biochemical investigation in light of reports that eggs fertilized by ICSI show slower-than-normal rates of cell cycle activation [29] and delayed patterns of nuclear decondensation in monkeys [30].
|
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
2 Correspondence: L.J. Wangh, Department of Biology, Brandeis University, 415 South Street, Waltham, MA 02454-9110. FAX: 781 736 3107; wangh{at}brandeis.edu
3 These authors contributed equally to this work.
Accepted: May 10, 1999.
Received: September 30, 1998.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
