Some cases of human male infertility are explained by abnormal in vitro human sperm activation

Transcription

1 Urology-andrology FERTILITY AND STERILITY Vol. 64, No. 3, September 1995 Copyright c 1995 American Society for Reproductive Medicine Printed on acid-free paper in U. S. A. Some cases of human male infertility are explained by abnormal in vitro human sperm activation David B. Brown, Ph.D.t:j: Erica J. Hayes, B.Sc.t Tatsuo Uchida, M.Sc. Manubai Nagamani, M.D.II University of Texas Medical Branch, Galveston, Texas Objective: To determine if a human sperm activation assay can be used to evaluate males exhibiting otherwise unexplained infertility. Design: Sperm from age-matched fertile and idiopathic infertile males were assayed in the human sperm activation assay and the results were compared. A portion of the sperm from the idiopathic infertile males also was used in assisted reproductive technology (ART) attempts at pregnancy. Patients: Idiopathic infertile couples who had extensive fertility testing with no identified problems that would explain their infertility. Fertile males that had fathered one or more children. Main Outcome Measures: Sperm nuclear decondensation-recondensation and DNA synthesis. Pregnancies resulting from ART using semen from a male whose sperm responded abnormally in the human sperm activation assay. Results: Thirteen (22%) of 59 idiopathic infertile males produced sperm that responded abnormally in the human sperm activation assay. Only 1 (1.7%) of 59 fertile males produced sperm that responded abnormally in the human sperm activation assay. The percentage of abnormal responders in the patient group exhibiting unexplained infertility was significantly higher than in the fertile male group. No sperm samples that responded abnormally in the human sperm activation assay resulted in pregnancies when used in ART. Conclusions: The human sperm activation assay is a new and independent indicator for some cases of infertility that otherwise would be unexplained. The human sperm activation assay appears to have utility in determining a sperm sample's efficacy for fertilization in ART attempts at pregnancy. Fertil Steril 1995;64: Key Words: Male infertility, assisted reproductive technologies, sperm activation, Xenopus laeuis frog egg extract, sperm penetration assay Although numerous tests are available for diagnosing infertility problems, 16% of all couples who Received October 13, 1994; revised and accepted March 21, * Supported in part by Advanced Research Program grant no (D.B.B.) from the Texas Higher Education Coordinating Board, Austin, Texas, and a supplement to this grant from Diagnostic Systems Laboratories, Inc., Webster, Texas. t Department of Human Biological Chemistry and Genetics. :/: Reprint requests: David B. Brown, Ph.D., Department ofhuman Biological Chemistry and Genetics, University of Texas Medical Branch, Galveston, Texas (FAX: ). 612 Brown et al. Human sperm activation seek medical treatment are diagnosed with unexplained infertility (1). Male members of inexplicably infertile couples produce sperm that appear normal in standard semen analyses, which include assessment of sperm density, progressive motility, sperm morphology, and semen volume. Because these methods have limited capacity for evaluating male fertility, other tests have been developed, including quantitative ultramorphology using transmission and scanning electron microscopy (2); in vitro nu- Office of Biostatistics. II Department of Obstetrics and Gynecology.

2 clear chromatin decondensation analysis (3); and the sperm chromatin structure assay (4). Tests of human sperm function also have been developed, including videomicrographic assessment of motility (straight line and curvilinear velocity and amplitude oflateral head displacement); analysis of membrane integrity; tests for sperm nuclear maturity; measurements of acrosomal status; acrosome reaction and acrosin activity; and the analysis of sperm penetration of hamster zona-free ova (sperm penetration assay; SPA); and zona binding (reviewed in Liu and Baker [5]). However, except for the SPA, none of these tests directly assay sperm activation events that occur after the sperm nucleus enters into the egg, a series of events critical for the entry of the zygote into the developmental program. Sperm activation includes the following sequence of events: chromatin decondensation, formation of a pronucleus, DNA synthesis, and then chromatin recondensation and breakdown of the pronucleus in preparation for the mitotic division that produces a two-cell embryo (6). The molecular mechanisms of these processes in mammalian systems have been hard to characterize because of the difficulty in obtaining sufficient quantities of female gametes for in vitro biochemical studies (7). It therefore was believed that an in vitro human sperm activation assay would be impossible to develop. In 1983, Lohka and Masui (8) devised an in vitro frog sperm activation assay while studying frog sperm activation. They demonstrated that incubating demembranated Xenopus laevis sperm in Rana pipiens egg extract resulted in complete sperm activation. These results were extended to the human system, where it was demonstrated that mixing lysolecithin-permeabilized human sperm with Xenopus laevis frog egg extract promoted human sperm activation (9,10). The events detected included chromatin decondensation, pronucleus formation, DNA synthesis, and chromatin recondensation (9, 10). The development of an in vitro human sperm activation assay now makes it possible to study the molecular mechanisms of human sperm activation and also provides the basis for a new diagnostic tool to study idiopathic male infertility. The human sperm activation assay uses Xenopus egg extract from unfertilized eggs arrested at the second meiotic metaphase. The frogs can be induced every 2 months to produce substantial numbers of relatively large eggs (1.2 mm diameter) at all times of the year, thus providing milliliter amounts of egg extract; 1 ml is sufficient for 10 assays. The extract contains all the factors necessary for sperm activation, including nucleoplasmin that is required for decondensation and protamine-histone exchange Vol. 64, No.3, September 1995 (11), cyclin~ and cyclin-dependent kinases that direct the cell cycle traverse (12), and the components needed for pronuclear assembly and DNA synthesis (13). In the human sperm activation assay, the human sperm activation events occur in a time frame determined by factors in the frog egg. Mter fertilization, a Xenopus frog sperm takes 1.5 hours for complete activation in vivo, a process that, as with human sperm, requires 3 hours upon incubating with frog egg extract in vitro (9, 10, 14). The first clinical use of the human sperm activation assay was in a pilot study analyzing sperm from 15 idiopathic infertile males and 15 fertile males (15). We chose to use the human sperm activation assay in evaluating samples from idiopathic infertile males because, by definition, the sperm of this patient population previously tested normal in the standard semen tests performed in the fertility clinic. Three of the idiopathic infertile males produced sperm that responded abnormally in the human sperm activation assay; none of the fertile males' sperm responded abnormally (15). These preliminary results suggested that the human sperm activation assay might be useful in evaluating idiopathic male infertility. However, that the sample size (n = 15) was not sufficient to establish the assay's utility with statistical significance provided the impetus for the present study in which sperm from 59 idiopathic infertile males and 59 fertile males were analyzed in the human sperm activation assay. Some of the idiopathic males who participated in this study also had sperm samples analyzed in the SPA (16-18). We compare these human sperm activation assay and SPA results in this report. Patient Selection MATERIALS AND METHODS In this study, the human sperm activation assay was used to analyze sperm from a total of 59 fertile males and 59 idiopathic infertile males. Fertile males are defined as having fathered one or more children. The idiopathic infertile males and their spouses were patients at the University of Texas Medical Branch Fertility Clinic. These couples were diagnosed as having unexplained infertility when the following series of tests were completed without finding a cause for the couple's infertility: the female partner had complete infertility evaluations, including postcoital tests (to rule out sperm motility loss in cervical mucus); a timed endometrial biopsy (to exclude luteal phase defect); a hysterosalpingogram Brown et al. Human sperm activation 613

3 (to establish tubal patency) and laparoscopy (to rule out pelvic pathology); analysis of the male partner's semen was normal on two occasions with counts >20 X 10 6 /ml, sperm motility >60%, and normal morphology in >60% ofthe sperm (19). The inexplicably infertile couples were involuntarily infertile for >2 years, with an average length of 5.7 years. We analyzed the portion of the semen sample from the idiopathic infertile males that was not used for other procedures performed at the fertility clinic. Semen samples were obtained from fertile males specifically for these studies. All donors provided signed consent forms before their sperm were analyzed in the human sperm activation assay. All donors abstained from ejaculation for 2 to 4 days before the collection of the semen sample obtained by masturbation. The average age of both the fertile and infertile males was 35 years, with a range of 22 to 52 years and 23 to 50 years, respectively. Sperm Preparation Semen samples were kept at 4 C for up to 7 days to perform parallel assays on samples collected throughout the week. Fresh samples are not needed for the human sperm activation assay; sperm isolated from semen that had been stored at 4 C for up to 1 month responded the same in the human sperm activation assay as did sperm isolated from the same fertile male's semen collected the day the assay was performed (15). On the day of the experiment, the semen samples from the control fertile male whose sperm had been shown previously to respond normally when analyzed in the human sperm activation assay, and the test subjects were incubated for 30 minutes at 37 C, and then each sample was suspended in 10 ml of nuclear isolation medium (200 mm sucrose, 2.4 mm MgCI 2, 10 mm Tris-HCI and 5 mm maleic acid, ph 7.4). A sperm count was taken and 20 X 10 6 sperm were aliquoted for use in the assay. The remainder of the sperm were pelleted by centrifugation (10 minutes, 400 X g), resuspended in 0.5 ml of phosphate-buffered saline (PBS) with 15% glycerol, and then frozen at -70 C before transfer to liquid nitrogen. Such sperm can be thawed and permeabilized, and the analysis can be repeated if desired. The aliquot was centrifuged for 10 minutes at 400 X g, and the pellet was resuspended in 10 ml of nuclear isolation medium containing 0.05% lysolecithin and 1 j.tg/ml soybean trypsin inhibitor. The mixture was incubated at room temperature for 5 minutes, the sperm were pelleted for 10 minutes at 400 X g and then washed, first in nuclear isolation medium with 3% bovine serum albumin (BSA) and then with nuclear isolation medium containing 0.4% 614 Brown et al. Human sperm activation BSA (15). The sperm finally were resuspended in Xenopus extract isolation medium (10 mm Tris HCL, ph 7.5; 1.5 mm MgCb; 100 mm KCI; and 50 mm dithiothreitol [DTT]; Calbiochem, La Jolla, CA) at a concentration of 25,000 spermlj.tl and kept on ice for 45 minutes before removing an aliquot to mix with the frog egg extract for the assay. Dithiothreitol pretreatment was found to enhance decondensation and DNA synthesis (15) and is probably required to allow removal of the protamines that in mammalian sperm are cross-linked by intermolecular disulfide bonds (20). We believe that the frog egg extract may lack reducing factor(s) found in mammalian eggs, making it necessary to reduce disulfide bonds experimentally. All work with human sperm was performed at Biosafety Level 2. Egg Extract Preparation To promote oocyte maturation, three adult Xenopus laevis frogs were injected with 500 IU hcg 1M (Sigma Chemical Co., St. Louis, MO) 18 hours before the experiment began. Mature eggs were collected by squeezing the frogs and collecting the eggs in a beaker. Eggs that were released during the night and spent an indeterminate amount of time in the holding tank water were not used. The quality of the freshly squeezed eggs was determined by noting the proportion of eggs that were in the process of being readsorbed by the frog. These are a solid grayishwhite color and are easy to differentiate from healthy eggs, which have a distinct pigmented animal hemisphere and white vegetal hemisphere. Only egg batches containing <5% of the readsorbing eggs were used for preparation of egg extract. All readsorbing eggs were separated from the healthy eggs with a Pasteur pipette. The eggs were then dejellied with 2% cysteine-hci, ph 7.6, washed, and incubated for 1 hour at room temperature in Barth's medium (88 mm NaCI, 1.0 mm KCI, 0.83 mm MgS0 4, 0.34 mm Ca [N03h, 0.41 mm CaCI 2, 7.5 mm Tris HCI, ph 7.6,10 j.tg/ml penicillin, 10 mglml streptomycin, and 2.4 mm Na[C0 3h). The eggs then were washed three times in Xenopus extract isolation medium. Excess buffer was removed and the eggs were lysed by centrifugation at 10,000 X g, 4 C, for 15 minutes in a swinging bucket rotor. The resulting middle layer between the lipid phase and the egg pellet was removed, recentrifuged, and the resulting middle layer was used as the stock extract solution. The extract was stored on ice until its use in the human sperm activation assay. Human Sperm Activation Assay The control assay mix was prepared by adding 8 j.tl of sperm suspension from the fertile male

4 (200,000 sperm) to 100 ILL of extract containing 8 ILCi (8 ILL) of tritiated thymidine triphosphate (lcn, Irvine, CA), giving a final volume of 116 ILL. We used phase-contrast microscopy to observe wet mounts prepared by placing 5 ILL aliquots of the control assay mix on glass slides, onto which coverslips were applied. We determined the percentage of nuclei that were decondensed fully at 5, 10, and 15 minutes after addition of sperm to the extract (during a 5- minute time interval, 2:50 nuclei were scored) and determined the time point at which >90% of the sperm had decondensed fully. Any extract preparation that did not decondense fully >90% of the control sperm during a 15-minute incubation was not used to analyze patient samples. The fertile male and patient sperm assay mixes were prepared as described previously. At the time point at which >90% of the sperm in the control assay mix were found to be decondensed fully, we removed 5 ILL of the sperm assay mix for use in preparing a wet mount. Again, we scored 2:50 nuclei for decondensation and determined the percent of fully decondensed nuclei. We then compared the sperm assay mix percent decondensation with the control assay mix percent decondensation results and normalized the data by determining percent of the control decondensation values (percent of control decondensation = percent sperm assay mix decondensation/percent control assay mix decondensation X 100). Examples of nuclei scored as fully decondensed sperm are shown in Figure 1. From here until the completion of the assay, the control assay mix and sperm assay mix were treated as follows: after a 15- to 20-minute incubation of the sperm in the extract, we removed 25 ILL of suspension and mixed it with 75 ILL of PBS. We made two cytoprep slides, putting 50 ILL of spec i men-pbs mix into each cytocentrifuge chamber. We centrifuged the specimen at 550 X g in a Shandon 3 Cytocentrifuge (Shandon Inc., Pittsburgh, PA) for 3 minutes (high acceleration setting). Any sperm that have decondensed smear onto the glass slide (see Figs. 2B and 3A). The cytoprep slides were fixed for 5 minutes in 1:3 glacial acetic acid-methanol (Baker, Phillipsburg, NJ) and then were stained with Giemsa (Baker). The 15- to 20-minute Giemsastained cytopreps were used to verify further abnormal decondensation, in that such sperm assay mix cytopreps will have less smearing than control cytopreps (see Fig. 3A and B). Mter 2 hours, we mixed the assay suspension and removed 25 ILL to mix with 75 ILL of PBS. We again made two cytopreps using 50 ILL of specimen-pbs mix per slide. We centrifuged at 250 X g for 5 minutes (medium acceleration setting). The cytoprep slides were fixed for 5 minutes as before, and then were dipped in Kodak nuclear Vol. 64, No.3, September 1995 track emulsion (NTB-2; Eastman Kodak, Rochester, NY), stored at 4 C for 2 weeks, and then stained with Giemsa through the emulsion after development. All microscopy used a Leitz Orthoplan microscope (Leitz, Wetzlar, Germany). For both the control assay mix and sperm assay mix slides, 200 nuclei were scored for labeling and the result for the sperm assay mix (percent of the patient's nuclei showing label above background) was compared with the result for the control assay mix sperm as to the percent of control that were labeled (percent of control DNA synthesis = sperm assay mix percent DNA synthesis/control assay mix percent DNA synthesis X 100). At 3 hours, we mixed the suspension and removed 25 ILL for mixing with 75 ILL of PBS. We made two cytopreps using 50 ILL of specimen-pbs mix. We then centrifuged at 250 X g for 3 minutes (medium acceleration). We also removed a 5-ILL aliquot to prepare a wet mount slide and, again using phase-contrast microscopy, determined the percent recondensed nuclei for both control assay mix and sperm assay mix, and then the percent of the control nuclei that had recondensed (percent of control recondensation = sperm assay mix percent recondensation/control assay mix percent recondensation X 100). The 3-hour Giemsa-stained cytopreps were used to verify that recondensation of the chromatin was the same in the control assay mix and sperm assay mix cytopreps. Sperm Penetration Assay Semen samples were collected by masturbation and then were allowed to liquefy for 2:30 minutes. Sperm were obtained from the samples using the swim-up procedure (18). The assay was performed as described by Rogers et al. (17) with minor modifications. Golden Syrian hamsters were injected with 30 IU pregnant mare serum. Two days later, the hamsters were injected with 30 IU hcg (Sigma Chemical Co., St. Louis, MO). Sixteen to 18 hours later, the hamsters were killed and the egg mass was harvested. Eggs were transferred into a drop of 0.1 % hyaluronidase and incubated 3 to 5 minutes until the cumulus mass was digested. The zona pellucidae were removed by digestion in 0.1% trypsin for approximately 2 minutes. Approximately 30 ova were inseminated with 1 X 10 6 sperm. Mter a 4- hour incubation at 37 C in a 5% CO 2 incubator, the ova were fixed in 3% glutaraldehyde, stained with acetolacmoid, and a phase-contrast microscope was used to examine the ova for penetration. The penetration index was determined (penetration index = number of eggs containing decondensed swollen sperm nuclei/total number eggs). A penetration index of 2: 10% was considered normal (18). Brown et al. Human sperm activation 615

5 Figure 1 Phase-contrast photographs of nuclei having normal and abnormal decondensation in the human sperm activation assay, taken using a 40x objective. Bar, 10 /-Lm. (A), Time zero. Permeabilized sperm from a fertile male with no extract treatment. (B to D), Permeabilized sperm from a male with unexplained infertility incubated 15 minutes in egg extract. In B is shown an example of sperm that did not decondense, near a partially decondensed sperm. Another partially decondensed sperm is shown in C. A fully decondensed sperm is shown in D. The idiopathic infertile male whose sperm are shown in B to D had a decondensation score that was 57% of the control. (E), A fully decondensed sperm from a fertile male incubated 15 minutes in egg extract. UI, unexplained infertility. RESULTS The human sperm activation assay was used to analyze sperm from 59 idiopathic infertile males and 59 fertile males. Typical examples of activated human sperm throughout the 3-hour human sperm activation assay are shown in Figure 2. Figure 4 shows the decondensation and DNA synthesis results expressed as the percent of control values. The distribution of the decondensation and DNA synthesis response of the 59 fertile males had the two responses clustered in the range of 85% to 110% (except for one male) and 90% to 110%, respectively. By contrast, the distribution of decondensation responses of the idiopathic infertile males showed two different populations: one clustered in the range of 84% to 107%, another with a range of9% to 73%. The distribution of DNA synthesis also showed two different populations: one clustered in the range of 90% to 110%, the other in the range of 15% to 72%. We have defined abnormal and normal populations as <80% 616 Brown et al. Human sperm activation and >80%, respectively, based on the wide gap observed near the 80% value, separating the patients (but not the fertile controls) into two groups (Fig. 4). This biphasic distribution was observed for both decondensation at 5 to 15 minutes and DNA synthesis at 2 hours (Fig. 4). Also, there was a low probability that sperm from a normal individual would respond with a value of <80% of the control when analyzed in the human sperm activation assay. Of the 59 fertile males whose sperm samples were analyzed in the human sperm activation assay in this study, the percent of the control values for decondensation and DNA synthesis were distributed around a mean ± SD of 100% ± 4.69% and 3.06%, respectively. Based on these data, the probability of sperm from a normal individual having a human sperm activation assay response of <80% of the control was < (Gaussian distribution approximation) for both decondensation and DNA synthesis. For these reasons, any sperm samples having human sperm activation assay response values <80% of the control

6 Figure 2 A pictorial representation of a normal response of permeabilized human sperm incubated in frog egg extract during a 3-hour time course. The sperm nuclei shown in A to H were photographed using a 40 x objective. I and J were photographed using a 25x objective in order to include more sperm nuclei in the field of view. Bar, 10 J.lm. Bright field microscopy was used to photograph Giemsa-stained cytopreps of nuclei shown in A to D, whereas E to H are nuclei photographed using phase-contrast microscopy, and C shows auto radiographs of Giemsa-stained nuclei. (A, E, and I), Time zero. Permeabilized sperm with no extract treatment. (B, F, and J), Fifteen minutes in egg extract. B shows sperm nuclei that did not remain intact during the centrifugation procedure used when making cytopreps; the nuclei became smeared during transfer onto the glass slide. Mter 5 to 15 minutes in the frog egg extract, >95% of the sperm have completely decondensed nuclei. A typical completely decondensed sperm nucleus is shown in F; all nuclei shown in J are decondensed fully. (C and G), Two hours in egg extract. The sperm nuclei have begun to recondense their chromatin. Notice the decreased size of the sperm nucleus in F compared with that in G. Recondensed sperm nuclei again withstand the cytocentrifugation procedure (note intact nuclei in C). The labeling of the nuclei in C (black grains over the nuclei) indicates that they have undergone DNA synthesis. Typically, >95% of control nuclei are labeled after 2 hours in frog egg extract containing tritiated thymidine. (D and H) Three hours in egg extract. Examples of recondensed nuclei are shown in D and H. In Dare shown nuclei that have recondensed their chromatin to the late G2 to early prophase chromosome level. Ten photographs of permeabilized sperm, decondensed sperm nuclei at the 15-minute time point, and recondensed sperm nuclei at the 3-hour time point were cut out and weighed separately using an analytic balance. The average weight of the nondecondensed cut-outs was :!: g (mean:!: SD); of the fully decondensed cut-outs, :!: g; and of the fully recondensed cut-outs, :!: g. These values translate to a sevenfold increase in size as sperm decondense and a threefold decrease in size as they recondense. The recondensed chromatin in the nucleus shown in H is no longer contained within a nuclear membrane, as the periphery of the nucleus is rough and appears to be delineated by recondensed chromatin. The outline of the nucleus is smooth when the nuclear membrane is intact. were considered abnormal. We did not observe any significant differences in the recondensation of sperm from the idiopathic infertile and fertile males Vol. 64, No.3, September 1995 (data not shown). Examples of abnormal sperm nuclear decondensation and DNA synthesis are shown in Figures 1 and 3.. Brown et a1. Human sperm activation 617

7 Figure 3 Photographs of Giemsa-stained cytopreps (A and B) and autoradiographs (C and D) of nuclei having normal and abnormal decondensation and DNA synthesis in the human sperm activation assay; a 63x oil emersion objective was used. Bar, 10 j-lm. For A and B, sperm were incubated 20 minutes in egg extract before cytopreps were made. (A), Examples of typical smeared nuclei seen in cytopreps of sperm from fertile males. (B), Examples of the smeared nuclei observed in cytopreps of sperm from males with unexplained infertility; these sperm had a decondensation score that was 51% of the control. The single arrow points to a nondecondensed sperm nucleus that did not smear, and the double arrows point to a partially decondensed and smeared sperm nucleus. Both of these sperm nuclei are next to a sperm nucleus smeared equivalent to that observed in A. For C and D, sperm were incubated 2 hours in extract containing tritiated thymidine before preparation of cytopreps and autoradiography. (C), Example of a labeled sperm nucleus from the control fertile male. (D), Example of unlabeled sperm nuclei from a male with unexplained infertility whose sperm had an abnormal DNA synthesis response in the human sperm activation assay (15% of the control). ur, unexplained infertility. Thirteen of the idiopathic infertile males produced sperm that responded abnormally in the human sperm activation assay, 10 having abnormal decondensation, 2 having abnormal DNA synthesis, and 1 having abnormal responses in both decondensation and DNA synthesis. Thus, the abnormal response rate is estimated at 22%, with a 95% confidence interval (CI) of 12.7% to 35.1%. Only 1 of 59 fertile males produced sperm that responded abnormally 618 Brown et al. Human sperm activation in the human sperm activation assay. The false negative rate is thus estimated as 1.7%, with a 95% CI of 0.1 % to 8.3% (but see Discussion). The percentage of abnormal responders in the unexplained infertility patient group was significantly higher than in the fertile male group (P = 0.001, X 2 test). A repeat ejaculate was obtained from the fertile male whose sperm responded abnormally in the human sperm activation assay and a repeat analysis

8 A B UNEXPLAINED INFERTILE MALES, q.,,, '"..., L... d... I FERTILE MALES,,,,,,,.,, I.11, DECONDENSATION (% of Control) UNEXPLAINED INFERTILE MALES,,,, ",,,I], ~~ I", FERTILE MALES DNA SYNTHESIS (% of control) Figure 4 Scatterplots of responses of sperm from 59 inexplicably infertile and 59 fertile males in the human sperm activation assay for (A) decondensation and (B) DNA synthesis. The open box denotes the patient who had both abnormal decondensation and abnormal DNA synthesis responses in the human sperm activation assay. Of 59 males having unexplained infertility whose sperm were analyzed in the human sperm activation assay, 13 had abnormal responses that were <80% of the control. was performed. Again, the sperm responded abnormally (75% of control). This individual's sperm concentration was normal (50 X 10 6 /ml); however, only 16% of his sperm had a normal morphology. To test the reproducibility of the assay, we obtained six repeat samples (collected 18 to 300 days after the initial sample collection date) from patients whose sperm initially responded abnormally (decondensation) in the human sperm activation assay, and the sperm in all but one of these samples again responded abnormally in the human sperm activation assay (data not shown); one patient's repeat sperm sample responded normally in the human sperm activation assay. A portion of this same sample was used in a successful GIFT attempt. We also obtained repeat samples (collected 19 to 455 days after the initial sample collection date) from 12 patients whose sperm initially responded normally in the human sperm activation assay; all these repeat samples again responded normally (data not shown). To determine if frozen sperm samples can be thawed and analyzed in the human sperm activation assay to give results comparable to those obtained Vol. 64, No.3, September 1995 initially, we have repeated the human sperm activation assay analysis of seven thawed patient sperm samples that initially responded abnormally, six showing abnormal decondensation and one showing abnormal DNA synthesis. All seven samples again responded abnormally in the human sperm activation assay (data not shown). We also have repeated the human sperm activation assay analysis of five patient sperm samples that responded normally in the initial human sperm activation assay. All five samples again responded normally in the human sperm activation assay (data not shown). Of 16 abnormal responders in the human sperm activation assay to date (15) (this study), 10 also had their sperm analyzed in the SPA. Five patients had penetration scores of <10% (considered abnormal, 18; all scored 0%) and five had scores >10%. Sperm penetration assay scores also were determined for 34 of 58 unexplained infertility patients whose sperm responded normally in the human sperm activation assay (15) (this study). Nine of 34 samples assayed had abnormal SPA scores of < 10% (three 0% scores). These data are summarized in Table l. In this study, we were assaying the portion of the semen sample that was not used in attempts at pregnancy via ARTs, including superovulation and timed lui with precapacitated sperm, GIFT, or IVF-ET. Table 1 Comparison of Human Sperm Activation Assay and SPA Results* Patient Abnormal response in the Sperm penetration no. human sperm activation assay assay score 1 Decondensation 60 2 Decondensation-DNA synthesis 0 3 Decondensation 16 4 Decondensation-DNA synthesis 69 5 Decondensation 0 6 DNA synthesis 64 7 Decondensation 28 8 Decondensation 0 9 Decondensation 0 10 Decondensation 0 Normal response Patient in the human sperm Sperm penetration nos. activation assay assay scores 11 to 44 Yes 0, 15, 15, 0, 30, 0, 20, 30, 24, 40, 4, 5, 18, 20, 27, 8, 59, 52, 23, 20, 70, 8, 70, 8, 53, 8, 33, 30, 14, 14, 58,67,34,17 * n = 12 (15) and n = 32 (this study). Boldface denotes scores < 10%. Brown et al. Human sperm activation 619 % %

9 We did not observe a single false positive in which the patient's sperm responded abnormally in the human sperm activation assay but resulted in a successful pregnancy when used in an ART attempt. To date, two of the idiopathic infertile male patients (one previously reported in Brown and N agamani [15] and one reported in this study) who produced sperm that initially responded abnormally in the human sperm activation assay have at a later time provided sperm that have responded normally in the human sperm activation assay and, when used in ART attempts, resulted in pregnancies. One of these patients provided a sample 4 months after the initial human sperm activation assay test date that responded normally in the human sperm activation assay (52% of the control decondensation score in initial assay, 100% of the control decondensation score in a later assay) (15). This later sample resulted in a pregnancy when used in the NF-ET procedure; however, the pregnancy ended with a spontaneous miscarriage at 6 weeks of pregnancy. The other patient (this study), provided a sample (just 18 days after the initial human sperm activation assay test) that responded normally in the human sperm activation assay (17% of the control decondensation score in initial assay, SPA score of 0%; 111% of the control in later assay, SPA score of 11%). Again the second sample resulted in a pregnancy when used in the GIFT procedure; the pregnancy went to term, with delivery of a normal infant. Seven inexplicably infertile couples were removed from this study when sperm from the male was used in ARTs and pregnancy resulted, i.e., they were no longer inexplicably infertile. All seven samples responded normally in the human sperm activation assay (data not shown). These preliminary data suggest that the human sperm activation assay can be used to determine a sperm sample's efficacy for fertilization. Confirming this will require a study to determine if human sperm activation assay results correlate with NF fertilization rates. DISCUSSION Of 74 fertile males whose sperm have so far been analyzed in the human sperm activation assay (n = 15; Brown and Nagamani [15]; n = 59, this study), only 1 produced sperm that responded abnormally in the human sperm activation assay. It is interesting to note that this male has only one child, who is 13 years old. During the past 12 years this man has worked on several chemical spill cleanups, during which he could have been exposed to toxic agents that may have had an adverse effect on his fertility. The human sperm activation 1issay may 620 Brown et ai. Human sperm activation have some utility in evaluating male sperm function after exposure to toxic agents known to affect male fertility. The average age of both the fertile and infertile males was 35 years, with a range of 22 to 52 years and 23 to 50 years, respectively. The average age of the abnormal responders was 34 years. Thus, there was no correlation between age and an abnormal response in the human sperm activation assay. Although other tests of human sperm function have been developed, only the human sperm activation assay and the SPA directly assay sperm activation events. The SPA tests the capacity of sperm to bind and penetrate hamster zona-free ova (16-18). The SPA score is the percentage of penetrated ova, determined by counting the ova that contain decondensed or activated sperm nuclei (pronuclei). Although both assays score activated sperm, the SPA cannot be used to study the efficiency of the decondensation process. During the SPA, the hamster ova routinely become bound with a large number of sperm that do not penetrate and thus do not decondense. Nondecondensed sperm bound to the ova cannot be differentiated from sperm that have penetrated ova and not decondensed as a result of not responding to activation signals in the ova. Interestingly, a 0% SPA score may in some cases reflect the sperm's inability to decondense instead of its penetration capabilities. In a study analyzing the sperm samples from 443 men having 0% SPA scores, and zero to three abnormal sperm parameters, 16.3% of these males had no readily detectable sperm abnormalities, i.e., unexplained infertility (18). It is not known whether the 0% SPA scores were a result of the sperm not penetrating ova, not decondensing in penetrated ova, or a combination of both. As for the possibility of using penetrated hamster ova to monitor DNA synthesis and recondensation, the labor and expense that would be required to obtain enough ova to analyze one patient would prohibit this approach as a routine clinical procedure. Conversely, the human sperm activation assay allows routine comparison of hundreds of a patient's activated sperm nuclei with those from a normal fertile male and easily detects any abnormal response, and thus is the clear assay of choice for assaying sperm activation. However, the human sperm activation assay cannot detect male infertility that is caused by the sperm's inability to penetrate the egg. The combined use of the human sperm activation assay and assays of the fertilization events occurring before sperm activation (e.g., the SPA, tests of the acrosome reaction, acrosin activity, and zona binding) will identify a greater percentage of infertile males than anyone assay alone.

10 Presently, there is no explanation for why some individuals' sperm do not activate normally in the human sperm activation assay. We speculate that the sperm have abnormal spermiogenesis, resulting in defective chromatin. During human spermiogenesis, the chromatin containing nucleosomal histones is converted to chromatin packaged mainly by three different protamine molecules (see Balhorn et al. [21]). A chromatin defect involving aberrant protamine ratios has been suggested as a possible explanation for some male infertility (21, 22). Further evidence linking defective chromatin and male infertility is provided in studies in which cytometric analysis of acridine orange-stained heated sperm from humans of low fertility and from fertile males revealed a significant decrease in resistance to in situ denaturation of the sperm from the subfertile males (4); acridine orange-stained human sperm from nonfertile donors were shown to have significantly fewer intact nuclei containing double-stranded DNA than did sperm from fertile donors (23); and in vitro nuclear chromatin decondensation and quantitative nuclear ultramorphology analysis of human sperm from males who had failed in vitro fertilization of the ova from their wives revealed a significant decrease in sperm chromatin stability and an increase in hypoelongated sperm heads from the infertile male group versus sperm from fertile males (24). In the future, we plan to analyze the chromatin of sperm that respond abnormally in the human sperm activation assay, to determine if the abnormal response is linked to defective chromatin. Future research directed at understanding the sperm activation events at the molecular level may provide information useful to the development of new therapies. The results from this study establish the human sperm activation assay as a useful tool for evaluating a subgroup of human male infertility patients who otherwise would have unexplained infertility. We believe that, before using expensive ARTs, an extensive evaluation of the couple's fertility should be performed, including analysis of the male's sperm in the human sperm activation assay. This will identify couples who cannot benefit from the available treatments of infertility and thus will save these patients the associated expense and discomfort. Although this study evaluated patients exhibiting unexplained infertility, the human sperm activation assay also may have utility in evaluating a sperm sample before use in the intracytoplasmic sperm injection procedure (25), which bypasses all sperm function defects except the sperm activation events measured by the human sperm activation assay. To extend the utility of the human sperm activation Vol. 64, No.3, September 1995 assay, we plan to determine if an abnormal response in the human sperm activation assay correlates with a lower intracytoplasmic sperm injection fertilization rate. Acknowledgments. We acknowledge helpful discussions and criticism of the manuscript by Phillip Lee, M.D., David Konkel, Ph.D., and Judith Yannariello-Brown, Ph.D. David B. Brown, Ph.D., is presently a part-time consultant with DSL, in the DSL-Human Sperm Activation Assay Laboratory. REFERENCES 1. Collins JA, Crosignani PG. Unexplained infertility: a review of diagnosis, prognosis, treatment efficacy and management. Int J Gynaecol Obstet 1992;39: Bartoov B, Eltes F, Weissenberg R, Lunenfeld B. Morphological characterization of abnormal human spermatozoa using transmission electron microcopy. Arch Androl 1980;5: Bedford JM, Bent MJ, Calvin HI. Variations in the structural character and stability of the nuclear chromatin in morphologically normal human spermatozoan. J Reprod Fertil 1973;33: Evenson DP, Darzynkiewiccz Z, Melamed MR. Relation of mammalian sperm chromatin heterogeneity to fertility. Science 1980;210: Liu DY, Baker HWG. Tests of human sperm function and fertilization in vitro. Fertil Steril 1992;58: Longo FJ, Kunkle M. Transformations of sperm nuclei upon insemination. CUIT Top Dev BioI 1978; 12: Wolgemuth DJ. Synthetic activities of the mammalian early embryo: molecular and genetic alterations following fertilization. In: Hartmann JF, editor. Mechanism and control of animal fertilization. New York: Academic Press, 1983: Lohka M, Masui Y. Formation in vitro of sperm pronuclei and mitotic chromosomes induced by amphibian ooplasmic components. Science 1983;220: Gordon K, Brown DB, Ruddle FH. In vitro activation of human sperm induced by amphibian egg extract. Exp Cell Res 1985; 157: Brown DB, Blake EJ, Wolgemuth DJ, Gordon K, Ruddle FH. Chromatin decondensation and DNA synthesis in human sperm activated in vitro by using Xenopus laevis egg extracts. J Exp Zool 1987;242: Philpott A, Leno GH, Laskey RA. Sperm decondensation in Xenopus egg cytoplasm is mediated by nucleoplasmin. Cell 1991;65: Lohka MJ, Hayes MK, Maller JL. Purification of maturationpromoting factor, an intracellular regulator of early mitotic events. Proc Natl Acad Sci USA 1988;85: Blow JJ, Laskey RA. Initiation of DNA replication in nuclei and purified DNA by a cell-free extract of Xenopus eggs. Cell 1986;47: KirschnerM, NewportJ, GerhartJ. The timing of early developmental events in Xenopus. Trends Genet 1985;1: Brown DB, Nagamani M. Use of Xenopus laevis frog egg extract in diagnosing human male unexplained infertility. Yale J BioI Med 1992;65: Yanagimachi R, Yanagimachi H, Rogers BJ. The use of zonafree animal ova as a test system for the assessment of the fertilizing capacity of human spermatozoa. BioI Reprod 1976;15: Brown et al. Human sperm activation 621

11 17. Rogers BJ, Van Campen H, Ueno M, Lambert H, Bronson R, Hale R. Analysis of human spermatozoal fertilizing ability using zona-free ova. Fertil Steril 1979;32: Rogers BJ. The sperm penetration assay: its usefulness reevaluated. Fertil Steril 1985;43: World Health Organization. WHO laboratory manual for the examination of human semen and semen-cervical mucus interaction. 2nd ed. Cambridge: The Press Syndicate of the University of Cambridge, 1987:3, Perreault SD, Barbee RR, Slott VL. The role of disulfide bond reduction during mammalian sperm nuclear decondensation in vivo. Dev BioI 1984; 101: Balhorn R, Reed S, Tanphaichitr N. Aberrant protamine liprotamine 2 ratios in sperm of infertile human males. Experientia 1988;44: Belokopytovak la, Kostyleva EI, Tomilin AN, Vorob'ev VI. Human male infertility may be due to a decrease ofthe protamine P2 content in sperm chromatin. Mol Reprod Devel 1993;34: Peluso JJ, Luciano AA, Nulsen JC. The relationship between alterations in spermatozoal deoxyribonucleic acid, heparin binding sites, and semen quality. Fertil Steril 1992;57: Lipitz S, Bartoov B, Rajuan C, Reichart M, Kedem P, Mashiach S, et al. Sperm head ultramorphology and chromatin stability of males with unexplained infertility who fail to fertilize normal human ova in vitro. Andrologia 1992;24: Palermo G, Joris H, Devroey P, Van Steirteghem AC. Pregnancies after intracytoplasmic injection of single spermatozoon into an oocyte. Lancet 1992;340: Brown et al. Human sperm activation