Reproductive.. animal research

Transcription

1 Reproductive.. animal research FERTILITY AND STERILITY Vol. 63, No.1, January 1995 Copyright 1995 American Society for Reproductive Medicine Printed on acid-free paper in U. S. A. Human sperm motility-enhancing agents have detrimental effects on mouse oocytes and embryos* Lynette Scott, M.Sc. t Samuel Smith, M.D. Division of Reproductive Endocrinology, Women's and Children's Services, Sinai Hospital of Baltimore, Baltimore, Maryland Objectives: To test the artificial activating properties of the human sperm motility-enhancing agents pentoxifylline, caffeine, 2-deoxyadenosine, and cyclic adenosine 3':5' monophosphate (camp) on mouse oocytes and determine if the agents exhibit an inhibitory effect on in vitro development of mouse embryos. Design: CD-1 mouse oocytes were exposed to 1, 2.5, 5, or 10 mm pentoxifylline, caffeine, 2-deoxyadenosine, or camp for 10, 30, or 60 minutes and their activation and development was scored over 96 hours of culture. A 10% ethanol solution and aging unstimulated oocytes served as controls. Pronuclear embryos from CD-1, CF-1, and B6C3 F1 hybrid mice were cultured in 0.16, 0.33, 0.66, 1.25, 2.5, 5.0, or 10 mm of pentoxifylline, caffeine, 2-deoxyadenosine, or camp and development was scored over 96 hours of culture. Results: Exposure to pentoxifylline, caffeine, and 2-deoxyadenosine, but not camp, artificially activated mouse oocytes in a concentration- and exposure time-dependent manner. The level of activation was significantly greater than that associated with oocyte aging but less than ethanol-induced activation. Agent-activated oocytes had limited developmental capacity compared with the ethanol-activated oocytes. Pentoxifylline and 2-deoxyadenosine were more toxic than caffeine, especially at the higher concentrations and after prolonged exposure. All of the agents affected embryo development in a dose-dependent manner with developmental inhibition and embryotoxicity that was often not evident until after one to three cell cycles. Conclusions: Pentoxifylline, caffeine, 2-deoxyadenosine, and camp have adverse effects on mouse oocytes or embryos at concentrations commonly used to activate sperm in human IVF. Therefore, care should be taken to minimize the exposure of human oocytes and embryos to these agents until their direct effects have been investigated more fully. Fertil Steril1995;63: Key Words: Sperm motility-enhancing agents, parthenogenesis, embryo development, 2-deoxyadenosine, pentoxifylline, caffeine, cyclic AMP In assisted reproductive technologies (ART) programs there is an increasing interest in the use of agents for the enhancement of sperm motility for assisted fertilization and where there has been a history offailed fertilization (1, 2). There have been Received March 4, 1994; revised and accepted July 22, * Presented in part at the conjoint meeting of The American Fertility Society and the Canadian Fertility and Andrology Society, Montreal, Quebec, Canada, October 11 to 14, t Reprint requests: Lynette Scott, M.Sc., Sinai Hospital of Baltimore, Assisted Reproductive Technology Program, 2411 West Belvedere Avenue, Weinberg Building, Suite 206, Baltimore, Maryland (FAX: ). several reports of the motility-enhancing capabilities of various agents on fresh and frozen human sperm (3, 4). Although some reports indicate an increase in fertilization and pregnancy rates (PR) when these agents are used clinically (2), others report a lack of correlation between the increased sperm motility parameters, fertilization, and PR in human ART (1, 5). The reasons for this lack of agreement are not clear. In most procedures, the spermatozoa are washed free of the stimulating agent before being placed with oocytes. Aitken et al. (3) have shown that the enhancement in motility induced by these agents generally is short lived. Thus, the transient nature of the enhancement may 166 Scott and Smith Detrimental effects of sperm enhancers Fertility and Sterility

2 account in part for the inconsistent reports in fertilization rates. Another factor may be carry over of an agent, resulting in exposure of oocytes. Imoedemhe et al. (6) observed a decrease in fertilization rate and a retardation or inhibition of human embryo development when caffeine was present in the fertilization medium. The oocytes of most mammalian species, including humans, are arrested at the second metaphase of meiosis after ovulation. When a spermatozoa enters an oocyte, it activates a cascade of events that allows resumption and completion of meiosis, resulting in zygote formation. These events include depolarization of the oolemma; release of the cortical granules; increases in the intracellular pool of Ca 2 +, which leads to the decondensation of the sperm nucleus; formation of the female and male pronuclei; extrusion of the second polar body; and, eventually, the first cleavage division (7). This cleavage event reestablishes the diploid nature of the embryo and denotes the successful completion of fertilization. In many species, including humans, some or all of the events can be stimulated by agents other than a spermatozoa, thus leading to the partial or complete development of parthenogenetic embryos (7-11). Both physical and chemical agents have been shown to induce parthenogenetic activation in many mammalian species. Fresh and aged human oocytes also have been reported to undergo varying degrees of parthenogenetic activation (9-11), although the resulting embryos have a limited developmental potential compared with the parthenotes induced in other species (9, 10). Furthermore, many of the agents that are used or are under investigation for human sperm motility enhancement are purine derivatives, which have been shown to have detrimental effects on various stages of mouse embryo development (12-15). The purpose of this study was to assess the effects of four human sperm motility-enhancing agents-caffeine, pentoxifylline, 2-deoxyadenosine, and cyclic adenosine 3':5' monophosphate (camp)-on mouse oocytes and embryos. Toward this end, we evaluated the parthenogenetic-activating properties of these agents on ovulated mouse oocytes and tested the effect these agents had on the development of mouse embryos in vitro. A mouse model was used because it provides a convenient starting point for studies on the potential toxic effects of additives to oocytes and embryos given the ethical considerations and limited num- hers of human oocytes available for experimental investigations. Animals MATERIALS AND METHODS Four- to five-week-old random-bred CD-1 or CF- 1 and B6C3 F1 hybrid virgin female mice (Charles River, Wilmington, NJ) were superovulated with 5 IU pregnant mare serum gonadotropin (G-4877; Sigma Chemical Co., St. Louis, MO) and hcg (CG5; Sigma Chemical Co.) given 44 to 48 hours apart. Females were maintained either for collection of oocytes or placed individually with stud males of proven fertility for mating. Mated females were checked the next morning for a vaginal plug as evidence of mating. Media The media used in these experiments were modifications of the Earle's balanced salt solution (EBSS) formulated as described by Scott et al. (16). These media have been shown to support >80% blastocyst formation from the one-cell stage of CD- 1, CF-1, and B6C3 F1 hybrid embryos (16). To the EBSS base was added 25 mm sodium bicarbonate, mm pyruvate, mm ethylenediaminetetraacetic acid, 0.01 mm glutamine, and 21 mm lactate to generate Earle's modified medium. Earle's modified medium was modified by supplementing with mm glucose and 1.32 mm sodium hydrogen phosphate to form Earle's modified medium with glucose and phosphate. For handling and washing of oocytes and embryos, the sodium bicarbonate concentration of Earle's modified medium with glucose and phosphate was reduced to 10 mm, HEPES was added to 15 mm, and the ph was adjusted to 7.3 (HEPES-Earle's modified medium with glucose and phosphate). All media were supplemented with 3 mg/ml bovine serum albumin (A9647; Sigma Chemical Co.). Oocyte and Embryo Collection N onmated female mice were killed by rapid cervical dislocation at 15.5 to 16 hours after administration of hcg. The oviducts were isolated and placed under equilibrated light liquid paraffin oil (0121-1; Fisher Scientific, Pittsburgh, PA). The oocyte-cumulus complexes used for the artificial activation studies were released from the oviduct into HEPES-buffered Earle's modified medium with Vol. 63, No.1, January 1995 Scott and Smith Detrimental effects of sperm enhancers 167

3 glucose and phosphate under warmed oil and the oocyte-cumulus complexes were pooled. All manipulations of oocyte-cumulus complexes and embryos were performed on a microscope stage maintained at 37 C. For the experiments using fertilized embryos, mated females were killed by rapid cervical dislocation 20 hours after hcg and the oocyte-cumulus complexes were released from the oviduct under warmed oil into HEPES-Earle's modified medium containing 2 mg/ml hyaluronidase (H3506; Sigma Chemical Co.) to remove the cumulus cells. Exposure of Oocytes and Embryos to the Spermactivating Agents Stock solutions (10 mm) of pentoxifylline (P1784; Sigma Chemical Co.), caffeine (C0750; Sigma Chemical Co.), 2-deoxyadenosine (D7400; Sigma Chemical Co.), and camp (A4237; Sigma Chemical Co.) were made in Earle's modified medium with glucose and phosphate. The stock solutions were diluted with fresh Earle's modified medium with glucose and phosphate to 1.0, 2.5, and 5 mm, filtered with a 0.22 J.Lm filter, and equilibrated overnight in a humidified atmosphere of 5% C0 2 in air. A fresh 10% ethanol solution was made in equilibrated Earle's modified medium with glucose and phosphate using 100% analar ethanol before each experiment. Oocyte cumulus complexes were collected using a flame-polished pasteur pipette in 5 to 10 J.LL medium. They were dispensed into 2 ml of activating medium under oil and placed in the incubator for 10, 30, or 60 minutes. Oocyte-cumulus complexes were washed through three 2 ml rinses of HEPES Earle's modified medium with glucose and phosphate and then placed in 100 J.LL of Earle's modified medium with glucose and phosphate under oil for 5 to 6 hours. At the end of this period, the cumulus cells were removed and the oocytes were washed through three 2 ml rinses ofhepes-earle's modified medium with glucose and phosphate, one 2 ml rinse of Earle's modified medium, and finally placed in 20 to 30 J.LL drops of Earle's modified medium under oil for assessment and culture. Nontreated oocyte-cumulus complexes were placed in Earle's modified medium with glucose and phosphate and subsequently processed as outlined above. Nontreated oocytes were used to control for handling and age-related parthenogenetic activation (7, 8). Exposure of fertilized embryos was carried out in 0.16, 0.33, 0.66, 1.25, 2.5, 5.0, or 10 mm pentoxifyl- line, caffeine, 2-deoxyadenosine, or camp prepared from 10 mm stock solutions. Groups of 10 embryos were placed in 20 to 30 J.LL drops of medium under oil and cultured for 48 to 52 hours. Any embryos continuing to develop were moved to Earle's modified medium with glucose and phosphate with the equivalent concentration of agent for continued culture. Sperm Motility-Enhancing Agent Activation of Oocytes and Exposure of Embryos Pooled oocyte-cumulus complexes from CD-1 mice were allocated randomly among treatment groups in any one experiment. All experiments were repeated three times with all time points and concentrations of the activating agent being used in each experiment. A minimum of two oocyte-cumulus complexes were used for each treatment group in each experiment. Three strains of mice, which have been shown to develop differently under standard culture conditions (16, 17), were used to study strain variability in the response of embryos to the sperm-activating agents. The strains used were an inbred strain of Swiss origin (CD-1) that suffers from a partial twocell block in conventional mouse embryo culture medium (18), an inbred strain of non-swiss origin (CF-1) that suffers from a complete two-cell block in conventional culture medium (17), and an F1 hybrid of B6C3 origin that does not suffer from any in vitro developmental blocks (16-18). Pronuclear embryos from the three strains of mice were isolated, washed, pooled, and then placed in culture as described above. Groups of 10 embryos were placed in 20 to 30 ILL drops of Earle's modified medium containing the agent and development scored was over 96 hours of culture. Control embryos were placed in Earle's modified medium after being washed and handled in a similar manner. After 48 to 52 hours of culture, any developing embryos were moved to fresh drops of Earle's modified medium with glucose and phosphate maintaining the same concentration of agent. Control embryos also were moved to Earle's modified medium with glucose and phosphate. Each experiment was repeated three times for each strain of embryos with all concentrations of each agent being used in each experiment. Evaluation of Activation and Development Oocytes used in the activation experiments were assessed using differential contrast optics on an in- 168 Scott and Smith Detrimental effects of sperm enhancers Fertility and Sterility!

4 verted microscope at magnifications of X320 and X510. Activated oocytes included those with one or two pronuclei, development of two polar bodies, or spontaneous cleavage to a two-cell embryo (7). The level of agent-induced activation of the oocytes was compared with the responses using ethanol, aging, and handling. The development of activated oocytes was compared with in vivo fertilized oocytes. Those oocytes activated during exposure to the agents were transferred to 20- to 30-~-tL drops of Earle's modified medium under oil in groups of up to 10 oocytes per drop. Any activated oocytes that showed signs of development were transferred to fresh drops of Earle's modified medium with glucose and phosphate after 48 to 52 hours of culture. After 24 hours of culture, oocytes were reassessed for activation, fragmentation, cleavage, lysis, and cytoplasmic disruptions. Activated and cleaving embryos were left in culture and assessed for continued development over a further 72 hours. The embryos were assessed for developmental stage, degree of fragmentation, cleavage arrest, lysis, cytoplasmic disruption, and delayed development after 24, 48, 72, and 96 hours of culture, 48, 72, 96, and 120 hours after hcg, respectively. The data were analyzed by Student's t-test, x 2 analysis, or analysis of variance where appropriate. Control Activations RESULTS In experiments to determine the response of CD I oocytes to a known parthenogenetic activator, ethanol (19), under the culture conditions used in this study, 86 of 117 (74%) oocytes showed some form of parthenogenetic activation 6 hours after exposure (Table 1). Seven additional oocytes were affected by becoming atretic or fragmented. Seventy-one of the 86 ethanol-activated zygotes cleaved to the two-cell stage (83% of those activated), 56 proceeded to the eight-cell-morula stage (65% of activated), and 41 formed fully expanded blastocysts (48% of activated). In contrast, none of the unfertilized unstimulated oocytes showed any sign of activation at 6 hours after being placed in culture and only 12 (14%) showed activation after 24 hours. None of these underwent cleavage. Control experiments also were carried out to assess the development of in vivo fertilized oocytes under the culture conditions in use. One hundred nineteen fertilized oocytes were cultured, of which 117 (98%) cleaved to the two-cell stage and 99 (83%) developed to the fully expanded blastocyst stage after 96 hours of culture. Sperm Motility-Enhancing Agent Activation of Oocytes Cyclic AMP at concentrations of 1, 2.5, 5.0, or 10 mm did not activate CD-I oocytes under the exposure conditions used in this study (data not shown). The results of exposing CD-1 oocytes to increasing concentrations of pentoxifylline for 10, 30, or 60 minutes are shown in Table 2. An exposure as low as 1 mm for 10 minutes resulted in 21% of oocytes being activated. This was significantly greater than that seen due to 24 hours of aging (P < 0.001) but less than the level achieved with ethanol. The level of activation increased with both the concentration and the exposure time to a maximum effect with 10 mm pentoxifylline for a 10-minute exposure (82% of oocytes activated). This was equivalent to the level of activation obtained with ethanol. The effects of pentoxifylline extended beyond simply activating the oocytes. The number of oocytes adversely affected (atretic and lysed) was significantly greater with increased exposure for each concentration tested with maximum toxic effects seen with 5 and 10 mm pentoxifylline at 60 minutes. The overall effect of the pentoxifylline rose to >90% affected. Regardless of the concentration or exposure time, the pentoxifylline-activated oocytes had little developmental capacity compared with oocytes activated with ethanol (Table 2). A maximum of 58% (7 /12) activated oocytes cleaved to the two-cell stage after a 30-minute exposure to 5 mm pentoxifylline but little further development was recorded. Almost all the two-cell embryos arrested and became atretic or fragmented and no blastocysts formed. Table 3 shows the results of exposing oocytes to increasing concentrations of caffeine. Significant levels of activation of CD-1 oocytes by 1, 2.5, and 5 mm caffeine were noted only at 30- and 60-minute exposure times. In contrast, even a 10-minute exposure to 10 mm caffeine resulted in 60% activation. Under the conditions used in these experiments, caffeine affected only a modest increase in the number of atretic and lysed oocytes when compared with aged oocytes (P < 0.07) or ethanol-activated oocytes (P < 0.05). However, the total proportion of affected oocytes was comparable to that Vol. 63, No. 1, January 1995 Scott and Smith Detrimental effects of sperm enhancers 169

5 Table 1 The Activation and In Vitro Development of Ethanol-activated Oocytes, Control Aged Oocytes, and In Vivo Fertilized In Vitro Cultured Embryos Activation response In vitro development of activated oocytes No. of Not Eight-cell- Fully expanded System oocytes* activatedt Activatedt Attretic-lysedt Affected t+ Two-cell morulae blastocysts 10% Ethanol (20) 86 (74) 7 (6) 93 (79) 71/86 (83) 56 (65) 41 (48) Aging 6 hours (93) 0 6 (7) 6 (7) hours (76) 12 (14) 8 (9) 20 (24) In vivo fertilized /119 (98) II 112 (94) 99 (83) * Total number of oocytes from three experiments. t Values in parentheses are of total oocytes. t Sum of activated oocytes plus attretic-lysed oocytes. Values in parentheses are percent of activated oocytes. II Values in parentheses are % of fertilized oocytes. observed with pentoxifylline. Also similar to pentoxifylline, the activated oocytes had limited developmental capacity, resulting in few two-cell embryos and no blastocysts (Table 3). Any embryos that did develop arrested at the two-cell stage or became atretic or fragmented. The few eight-cell and morulae that did develop ultimately fragmented. Although 2-deoxyadenosine had only a limited capacity to activate CD-1 oocytes during the shortest exposure time for the lower concentrations tested, longer exposures and an increase in the concentration of 2-deoxyadenosine resulted in significant levels of activation (Table 4). The extent of activation was comparable to that achieved with ethanol. In general, the numbers of atretic and lysed oocytes increased with both concentration and exposure time and was greater than that obtained with ethanol (P < 0.001), pentoxifylline (P < 0.05), caffeine (P < 0.01), or spontaneous activation due to aging (P < 0.001). The overall number of affected oocytes rose with both concentration and exposure to a maximum of 94%. The development of two-cell embryos from 2- deoxyadenosine-activated oocytes was significantly greater than that obtained with either pentoxifylline or caffeine (P < 0.01) but less than for ethanolactivated oocytes (P < 0.001). The continued development of these two-cell embryos was limited compared with ethanol-induced parthenotes, with very few proceeding to the blastocyst stage. The developing embryos that did not reach the blastocyst stage either arrested in development or fragmented. Table 2 The Parthenogenetic Activation and In Vitro Development of CD-1 Oocytes Exposed to Increasing Concentrations of Pentoxifylline for Increasing Lengths of Time Activation response In vitro development of activated oocytes No. of Not Eight-cell- Fully expanded Concentration Time oocytes* affectedt Activatedt Atretic-lysedt Affectedt Two-cell morulae blastocysts mm min (63) 9 (21) 7 (16) 16 (37) 1/9 (11) (45) 31 (49) 4 (6) 35 (55) 0/ (37) 22 (43) 10 (20) 32 (63) 0/ (50) 16 (30) 11 (20) 27 (53) 4/16 (25) 1 (6) (33) 32 (59) 4 (7) 36 (76) 3/32 (9) (41) 17 (36) 11 (23) 28 (59) 0/ (46) 12 (31) 9 (23) 21 (53) 7/12 (58) 2 (17) (8) 29 (61) 15 (31) 44 (91) 5/29 (17) 1 (3) (2) 24 (44) 29 (54) 53 (97) 0/ (5) 47 (83) 7 (12) 54 (94) 12/47 (26) (7) 37 (60) 20 (33) 57 (94) 4/37 (11) (18) 14 (37) 17 (45) 31 (82) 0/ * Sum of oocytes in three repeat experiments. t Sum of activated oocytes and lysed-attretic oocytes. t Values in parentheses are percent of total oocytes. Values in parentheses are percent of those activated. 170 Scott and Smith Detrimental effects of sperm enhancers Fertility and Sterility

6 Table 3 The Parthenogenetic Activation and In Vitro Development of CD-1 Oocytes Exposed to Increasing Concentrations of Caffeine for Increasing Lengths of Time Activation response In vitro development of activated oocytes No. of Not Eight-cell- Fully expanded Concentration Time oocytes* affectedt Activatedt Attretic-lysedt Affectedt:l: Two-cell morulae blastocysts mm min (96) (44) 22 (40) (46) 24 (38) (91) (16) 29 (59) (11) 36 (80) (66) 9 (19) (9) 46 (79) (4) 46 (90) (21) 36 (68) (26) 33 (61) (20) 32 (60) * Sum of oocytes in three repeat experiments. t Values in parentheses are percent of total oocytes. 3 (4) 3 (4) 0/ (14) 30 (54) 3/22 (14) (16) 34 (56) 1/24 (4) (9) 5 (9) 0/ (25) 41 (83) 5/29 (17) 2 (7) 0 4 (9) 40 (88) 3/36 (8) 2 (6) 0 7 (15) 16 (35) 1/9 (11) (12) 53 (92) 2/46 (4) (6) 49 (96) 1/46 (2) (11) 42 (79) 8/36 (22) 2 (6) 0 7 (13) 40 (82) 4/33 (12) 1 (3) 0 11 (20) 43 (79) 2/32 (6) 0 0 :j: Sum of activated oocytes and lysed-attretic oocytes. Values in parentheses are percent of those activated. Effect of Sperm Motility-Enhancing Agents on Embryo Development Nontreated embryos from all three strains resulted in >75% development to the fully expanded blastocyst stage in the culture system used in these experiments (CD-I, 50/60,83%; CF-1, 47/60, 78%; and B6C3 Fl hybrids, 53/60, 88%) after 96 hours of culture (data not shown). For treated embryos, all four agents had an adverse effect on the in vitro development of embryos from all three strains, re- suiting in cleavage arrest, developmental retardation; or lysis. Because the response was similar in all three mouse strains, indicating that the toxic effects were not strain-specific, only the results obtained from CD-I embryos are presented (Table 5). Over the concentration range of 0.16 to 10 mm, pentoxifylline had little effect on embryo progression through the first cleavage division. However, after the first cleavage division, there was a concentration-dependent increase in the incidence of developmental arrest and atresia. There was moder- Table 4 The Parthenogenetic Activation and In Vitro Development of CD-1 Oocytes Exposed to Increasing Concentrations of 2-Deoxyadenosine for Increasing Lengths of Time Activation response In vitro development of activated oocytes No. of Not Eight-cell- Fully expanded Concentration Time oocytes* affectedt Activatedt Attretic-lysedt Affectedt:l: Two-cell morulae blastocysts mm min (71) 5 (10) 9 (19) 14 (48) 1/5 (20) (40) 24 (45) 8 (15) 32 (60) 7/24 (29) 1 (4) 1 (4) (21) 19 (50) 11 (29) 30 (79) 4/19 (21) 2 (11) (51) 16 (28) 12 (21) 28 (49) 7/16(44) 3 (19) 1 (6) (4) 36 (69) 14 (27) 50 (96) 13/36 (36) 9 (25) 4 (11) (33) 18 (37) 15 (30) 33 (69) 7/18 (39) 4 (22) 2 (11) (34) 38 (56) 7 (10) 45 (66) 27/38 (71) 18 (47) 2 (5) (11) 46 (71) 12 (18) 58 (71) 26/46 (56) 14 (30) 2 (4) (27) 23 (40) 19 (33) 42 (73) 3/23 (13) (21) 31 (60) 10 (19) 41 (25) 18/31 (58) (8) 46 (75) 10 (17) 56 (91) 17/46(37) 5 (11) (3) 47 (70) 18 (27) 65 (97) 5/47 (11) 2 (4) 0 * Sum of oocytes in three repeat experiments. :j: Sum of activated oocytes and lysed-attretic oocytes. t Values in parentheses are percent of total oocytes. Values in parentheses are percent of those activated. Vol. 63, No. 1, January 1995 Scott and Smith Detrimental effects of sperm enhancers 171

7 Table 5 The In Vitro Development of CD-1 Embryos in Increasing Concentrations of the Four Agents From the Pronuclear Stage Over 96 Hours of Continuous Culture Per development at each concentration* Concentration (mm) Hours of Stage of culturet development:j: Pentoxifylline 24 Two-cell 97 ± ± ± ± ± ± ± Four-to-eight-cell 85 ± ± ± ± ± ± 13 58± Morulae 57± ± ± ± ± ± 2.9 2± Fully expanded blastocyst 5± 2.9 2± Caffeine 24 Two-cell 97 ± ± ± ± ± ± ± Four-to-eight-cell 95 ± ± ± ± ± ± Morulae 47 ± ± ± ± ± 10 7± Fully expanded blastocyst 5± 2.9 5± 5 0 3± 3 2± Deoxyadenosine 24 Two-cell 98 ± ± ± ± ± ± ± Four-to-eight-cell 73 ± ± ± 13 55± ± ± 7.3 2± Morulae 27 ± 4.4 7± 96 Fully expanded blastocyst 5± camp 24 Two-cell 93 ± ± 48 Four-to-eight-cell 95 ± ± 72 Morulae 63 ± ± 96 Fully expanded blastocyst 33 ± ± % ± ± ± ± ± ± ± ± ± ± ± ± 10 2± ± * Values are the mean ± SEM of three repeat experiments with 20 embryos per group per repeat for a total of 60 embryos for each agent at each concentration SEM. t Pronuclear stage embryos were placed in culture 24 hours after hcg. :j: The minimum stage required for a positive score at each scoring time point. ate inhibition of development after 48 hours of culture and severe inhibition after 72 and 96 hours of culture. The formation of fully expanded blastocysts was inhibited profoundly by even the lowest level of pentoxifylline compared with controls embryo cultures (P < ). Caffeine had a significant (P < 0.001) effect on the first cleavage division only at 10 mm. After 48 hours of culture there was a significant reduction in development that was concentration related. The development in all concentrations of caffeine was significantly reduced at 72 hours of culture (P < 0.001) with minimal blastocyst formation at 96 hours of culture (P < ). The embryos exposed to caffeine fragmented and became atretic in addition to arresting in development. Exposure ofcd-1 embryos to :2:1.25 mm 2-deoxyadenosine resulted in a significant (P < ) degree of cleavage inhibition and fragmentation after 24 hours of culture. After 48 hours of culture, there was a further reduction in the degree of development in all concentrations, with little further embryonic development in any concentration. The embryos become atretic, fragmented, and arrested in development. Embryos cultured in camp for the first cleavage division appeared to be unaffected. However, after an additional24 hours of culture, there was a significant reduction in development (P < 0.01) in concentrations > 1.25 mm. The development of embryos was reduced in all concentrations of camp by 72 hours of culture. Approximately half of the embryos cultured in levels < 1.25 mm compacted by 72 hours of culture. Of these, 20% to 33% formed fully expanded blastocysts by 96 hours of culture (120 hours after hcg) in :::;;0.66 mm camp. This was significantly lower than the compaction and blastocyst formation of controls (P < 0.001). No fully expanded blastocysts developed in concentrations of camp > 1.25 mm. DISCUSSION Three of four agents that are known to enhance the motility of human sperm in vitro were demonstrated to have the ability to artificially activate unfertilized mouse oocytes. The activation was observed at concentrations currently used in assisted reproduction programs. The activating agents also were shown to inhibit the development of fertilized 172 Scott and Smith Detrimental effects of sperm enhancers Fertility and Sterility

8 mouse embryos at concentrations as low as mm. It has been shown previously that both pentoxifylline (20, 21) and 2-deoxyadenosine (20) have detrimental effects on mouse embryo development and/or mouse oocyte IVF (20). The data presented here confirm the inhibitory effects of these agents on mouse embryo development and demonstrate that two other sperm motility-enhancing agents, caffeine and camp, also are detrimental to in vitro development of mouse embryos. Both the activating properties of these agents on mouse oocytes and the detrimental effects on embryo development have serious implications for the use of sperm motility-activating agents in human IVF. Because parthenogenetic embryos can resemble fertilized ones in the early cleavage stages, it may be difficult to recognize them at the time an ET would normally be performed. The only unambiguous way to identify these cleaving parthenogenetic embryos would be by chromatin content, which is not feasible when one is transferring the embryos. Also, depending on the agent and the concentration used, the effects on mouse embryo development often were not manifested until at least one or two cleavage divisions had occurred. If these agents have similar effects on human embryos, the effects may go unrecognized because ETs generally are performed after 48 to 72 hours of culture, when the effects of the agents may not be evident yet. Furthermore, if one of the human sperm motilityenhancing agents can initiate a partial or full activation response in human oocytes, its presence in the fertilizing medium could theoretically render those oocytes refractory to fertilization by the spermatozoa because it is likely that the agent will affect the oocyte before the sperm can bind to the oocyte and initiate a productive interaction. This could explain the decrease in fertilization rate observed when caffeine was left in the fertilizing sperm suspension (6). Ethanol, traditionally used to activate mouse oocytes with a high degree of success (17), has been shown to give a poor to moderate response in human oocytes depending on whether they were fresh (9) or aged and previously exposed to sperm (11). In this series of experiments, ethanol was used as a control activating agent. Ethanol acts on the oolemma and causes the oocytes to release internal calcium pools (7, 17). This increase in the concentration of free cystolic calcium leads to the resumption of meiosis. The resulting zygotes have been shown to have very good in vitro developmental capacity (8). Ethanol appears to stimulate the whole cascade of events leading to the formation of the mouse embryo, thus allowing continued development. Other activating agents are not as successful as this and the resulting activated oocytes are able to undergo only a partial completion of events that lead to a release from the meiotic block. Regardless of the pathway, once the activation events have begun, oocytes become refractory to fertilization by spermatozoa. Aging unfertilized oocytes also have been shown to undergo spontaneous activation (8). The low levels of parthenogenetic activation obtained after 24 hours of culture indicate that the methods used in the experimental design were not contributing to the results obtained and that parthenogenetic activation was being stimulated by the agent used. Pentoxifylline, which is a methylxanthine derivative and a phosphodiesterase inhibitor, stimulates sperm motility by inhibiting 3'5' nucleotide phosphodiesterase, which catalyzes the conversion of camp to 5'AMP (3). This results in a buildup of camp in the sperm, which in turn stimulates glycolysis, thus leading to increased sperm activity. Pentoxifylline may be activating oocytes by its effect on the Ca 2 + pump (22), which will cause a modification of the internal Ca 2 + concentration, leading to a partial or full activation response. The whole cascade of events may not be initiated but pentoxifylline may release the oocyte from the meiotic block and result in the various forms of parthenogenetic activation seen and the subsequent atresia and fragmentation of the oocytes. The effects ofpentoxifylline on embryo development could stem from a combination of its phosphodiesterase-inhibiting properties and its Ca 2 + pump inhibition. Increasing levels of camp in the embryo will lead to developmental arrest and a decrease in protein phosphorylation (23). Because Ca 2 + and camp are two of the main internal signaling systems in most cells, disrupting their balance could lead to a disruption of the entire functioning of the cell. The purine analogue 2-deoxyadenosine could be enhancing sperm motility in much the same way as pentoxifylline (3). Other purine analogues also have been shown to inhibit mouse embryo development (12-15) and induce mitotic cellular arrest (24). 2-Deoxyadenosine could be acting directly on the oocyte through its properties as a Ca 2 + antagonist and by its ability to increase the concentration of camp. The oocytes activated with 2-deoxyadenosine were the most viable of the agent-activated oocytes, with development proceeding through at least two to three cleavage divisions. The toxicity of Vol. 63, No. 1, January 1995 Scott and Smith Detrimental effects of sperm enhancers 173

9 2-deoxyadenosine on the development of embryos in vitro is presumably due to its mitotic arresting properties (25). Caffeine, which is also a phosphodiesterase inhibitor, enhances sperm motility by the same mechanisms as pentoxifylline (3, 4). The parthenogenetic-activating mechanism of caffeine could be through interference with the Ca 2 + channels and Ca 2 + concentration in the oocyte. These caffeineinduced parthenogenetic oocytes had little developmental capacity. Caffeine also had a marked effect on embryo development, which could be due to its ability to interfere with actin-myosin interactions in the microfilaments in the cell, thus disrupting the internal cellular transport mechanisms and cell division. It has been reported that the use of medium from which caffeine was not removed before insemination was associated with reduced fertilization, reduced cleavage, and increased fragmentation (6). Cyclic AMP had no effect on mouse oocytes in terms of an artificial activation response. However, embryo development was affected by camp. As the length of culture or concentration increased, the development of embryos decreased. Cyclic AMP causes a G 1 cell cycle arrest in eucaryotic cells, which is mediated through a camp-dependent protein kinase (24, 25) and is reversible if exposure is not protracted. The kinase phosphorylates serine and threonine residues on certain proteins that are important in maintaining the cell cycle. As either the concentration or the length of exposure increased, an increasing degree of developmental arrest was noted, which is consistent with this theory. The results presented here demonstrate that both mouse oocytes and embryos are affected adversely by certain human sperm motility-enhancing agents when the oocytes and embryos were exposed to concentrations currently used in ART practice. There are distinct species differences in the extent of parthenogenetic-activating abilities of oocytes (7-9) and the developmental potential of any resulting embryos also is species specific. Aged and fresh human oocytes have been shown to be refractory either partially or fully to parthenogenetic activation by agents most commonly used to activate mouse oocytes (9-11). However, human oocytes can be activated artificially under certain conditions, although the resulting parthenogenetic embryos have limited developmental capacity (9). The present studies and those reported by Lacham Kaplan and Trounson (20) and Tournaye et al. (21) suggest that the direct effects of sperm motility-enhancing agents on human oocytes and embryos be investigated more thoroughly before extensive use is made of them in clinical ART settings. Acknowledgments. The authors acknowledge Alan Scott, Ph.D., Johns Hopkins University, School of Public Health and Hygiene, Baltimore, Maryland, for his encouragement and comments and editing of the manuscript. REFERENCES 1. Palermo G, Joris H, Devroey P, Van Steirteghem A. Induction of acrosome reaction in human spermatozoa used for subzonal insemination. Hum Reprod 1992;7: Yovich JM, Edirisinghe WR, Cummins JM, Yovich JL. Influence of pentoxifylline in severe male factor infertility. Fertil Steril 1990;53: Aitken RJ, Mattei A, Irvine S. Paradoxical stimulation of human sperm motility by 2-deoxyadenosine. J Reprod Fertil 1986;78: Hammitt DG, Bedia E, Rogers PR, Syrop CH, Donovan JF, Williamson RA. Comparison of motility stimulants for cryo-preserved human sperm. Fertil Steril1989;52: Tournaye H, Janssens R, Camus M, Staessen C, Devroey P, Van Steirteghem A. Pentoxifylline is not useful in enhancing sperm function in cases with previous in vitro fertilization failure. Fertil Steril 1993;59: Imoedemhe DAG, Sigue AB, Pacpaco ELA, Olazo AB. The effect of caffeine on the ability of spermatozoa to fertilize mature human oocytes. J Assisted Reprod Genet 1992; 9: Whittingham DG. Parthenogenesis in mammals. In: Finn CA, editor. The Oxford reviews of reproductive biology. Oxford: Clarendon Press, 1980: Kaufman MH. The experimental production of mammalian parthenogenetic embryos. In: Daniel JC, editor. Methods in mammalian reproduction. New York: Academic Press, 1978: Winston N, Johnson M, Pickering S, Braude P. Parthenogenetic activation and development of fresh and aged human oocytes. Fertil Steril1991;56: Pickering S, Johnson MH, Braude PR, Houliston E. Cytoskeletal organization in fresh, aged and spontaneously activated human oocytes. Hum Reprod 1988;3: Balakier H, Casper RF. Experimentally induced activation of human oocytes. Hum Reprod 1993;8: Nureddin A, Epsaros E, Kiessling A. Purines inhibit the development of mouse embryos in vitro. J Reprod Fertil 1990;90: Downs SM, Dow MPD. Hypoxanthine-maintained two-cell block in mouse embryos: dependence on glucose and effect of hypoxanthine phosphoribosyltransferase inhibitors. Biol Reprod 1991;44: Loutradis D, John D, Kiessling AA. Hypoxanthine causes a 2-cell block in random-bred mouse embryos. Biol Reprod 1987;37: Downs SM, Coleman DL, Eppigg JJ. Maintenance of murine oocyte meiotic arrest: uptake and metabolism of hypoxanthine and adenosine by cumulus cell-enclosed and denuded mouse oocytes. Dev Biol1986;117: Scott and Smith Detrimental effects of sperm enhancers Fertility and Sterility

10 16. Scott LF, Sundaram SG, SmithS. The relevance and use of mouse embryo bioassays for quality control in an assisted reproductive technology program. Fertil Steril1993;60: Chatot CL, Ziomek CA, Bavister BD, Lewis JL, Torres I. An improved culture medium supports development of random-bred 1-cell mouse embryos in vitro. J Reprod Fertil 1989:86: Goddard MJ, Pratt HPM. Control of events during early cleavage of the mouse embryo: an analysis of the "2-cell block". J Embryo! Exp Morphol1983:73: Cuthbertson KSR. Parthenogenetic activation of mouse oocytes in vitro with ethanol and benzyl alcohol. J Exp Zoo! 1983;226: Lacham-Kaplan 0, Trounson A. The effects of the sperm motility activators 2-deoxyadenosine and pentoxifylline used for micro-injection on mouse and human embryo development. Hum Reprod 1993;8: Tournaye H, Van der Linden M, Van den Abbeel E, Devroey P, Van Stertingham A. Effects ofpentoxifylline on in-vitro development of preimplantation mouse embryos. Hum Reprod 1993;8: Blancato J, Seyler D. Effect of calcium-modifying drugs on mouse in vitro fertilization and preimplantation development. Int J Fertil 1990;35: Abramczuk JW, Lopata A. Resistance of human follicular oocytes to parthenogenetic activation: DNA distribution and content in oocytes maintained in vitro. Hum Reprod 1990;5: How let SK, Bolton VN. Sequence and regulation of morphological and molecular events during the first cell cycle of mouse embryogenesis. J Embryo! Exp Morpho! 1985;87: Albert DA, Nodzenski E, Heredia C, Kuchibholtla J, Kowalski J. Deoxyadenosine and cyclic AMP-induced cell cycle arrest and toxicity. Exp Cell Res 1991;197: Vol. 63, No. 1, January 1995 Scott and Smith Detrimental effects of sperm enhancers 175