MOLECULAR REPRODUCTION AND DEVELOPMENT 32:363368 (1992)
Prefertilization Treatment of Mice With Platelet Activating Factor Affects Pregnancy B.T. PODSIADLY,~L.M. ADAMSON,' J.D. STANGER: Y.C. SMART: AND T.K. ROBERTS 'Departments of Biological Science and 'Discipline of Surgical Science, University of Newcastle, N S W 2308, Australia; 3Lingard Fertility Centre, Lingard Hospital, Merewether, Australia
ABSTRACT Embryo-derived platelet activating factor (EPAF) is thought to be either biologically similar to platelet activating factor (PAF) or responsible for PAF liberation in vivo. We have previously shown that premating PAF treatment in the mouse renders the platelets nonresponsive to EPAF, leading to a reduced implantation rate in these animals. In this study, we have shown that females, injected with PAF before mating, show altered embryo development in vivo on day 4 postfertilization. This is manifested as an interruption of compaction, a reduced cell number per embryo, and reduced embryo number per mouse. Results suggest that EPAF represents an early pregnancy signal that supports embryo development. The most likely mechanism is via platelet activation, since only those mice that showed thrombocytopenia after fertilization were found to have normal embryos on day 4 postmating. 0 1992 Wiley-Liss, Inc.
ing (Adamson et al., 1987) and highlights the importance of EPAF-induced platelet activation during early pregnancy. In the present study, we have investigated the likely mechanism for the reduced pregnancy rate following PAF treatment prior to mating by examining the effects of this treatment on the embryo itself and on embryonic development in vivo and have also shown that thrombocytopenia is a necessary event for normal pregnancy.
MATERIALS AND METHODS Mice (C57BL/6 x CBA) F1 mice of 4-6 weeks were used for all in vitro embryo studies as embryos from this species did not express a two-cell developmental block in vitro. Quackenbush (QS) mice of 4-6 weeks were used for all in vivo work. Mice were housed at a constant temperature of 21-23°C with food and water freely available. Key Words: Thrombocytopenia, Platelets,Embryo, D e All mice were superovulated with 5-7.5 IU PMSG compaction, Implantation, Abortion (pregnant mare serum gonadotrophin, Intervet, Australia), followed 48 h later by 5-7.5 IU hCG (human chorionic gonadotrophin, Intervet, Australia) and then caged with fertile males overnight. Only animals showing vaginal plugs approximately 15 h after hCG injecINTRODUCTION tion were used. The day of vaginal plug appearance was Embryo-derived platelet activating factor (EPAF), designated day 1of pregnancy. Between 0900 and 1200 a n early pregnancy marker detected in the mouse and h on day 4 of pregnancy, the embryos were removed and human (O'Neill and Saunders, 1984; Roberts et al., scored according to their stage of development (cell 19871, is produced by the preimplantation embryo. number). EPAF is responsible for the transient thrombocytopenia that occurs in the pregnant mouse between days 1 Platelet Activating Factor and 5 postmating. Although currently controversial, Platelet activating factor (PAF) (L-a-phosphatidylEPAF is thought to be either biologically similar to choline, P-acetyl, y-alkyl) was obtained as a 2 mg/ml platelet activating factor (PAF) (O'Neill, 1985; Adam- solution in chloroform (Sigma, St. Louis, MO) and preson et al., 1987) or responsible for PAF liberation in pared a t a concentration able to mimic the thrombocyvivo (Angle et al., 1988; Amiel et al., 1989; Adamson topenia seen in pregnancy and capable of inducing a n et al., 1991). in vivo 30% reduction in mouse platelet levels within Injection of PAF, a phospholipid mediator of platelet 15 min, followed by a return to basal levels by 60 min activation, a t a concentration capable of decreasing in (i.e., reversible platelet activation) (Adamson et al., vivo platelet levels by 30% in the mouse, on 3 consecu- 1987). PAF was prepared at a concentration of 5 pg/ml tive days immediately prior to mating inhibits maternal platelet activation and the appearance of thrombocytopenia. This activity is due to desensitization of platelets by PAF at concentrations that induce reversReceived October 7,1991; accepted February 13,1992. ible platelet aggregation (Henson, 1977; Adamson Address reprint requests to Dr. T.K. Roberts, Department of Biologiet al., 1987). The net result of this desensitization is a cal Sciences, University of Newcastle, University Drive, Callaghan reduction in implantation as detected 10 days postmat- NSW 2308. Australia.
0 1992 WILEY-LISS, INC.
364
B.T. PODSIADLY ET AL.
(batch #127F-8405) and 3 pgiml (batch #66F-8395) in Dulbecco’s phosphate-buffered saline (PBS) (Ca2+ and Mg2+free, Sigma, St. Louis, MO) containing 0.25% bovine serum albumin (BSA) (fraction V, Sigma, St. Louis, MO). PAF was injected intraperitoneally in aliquots of 200-300 p1, depending on the weight of the mouse. Effect of PAF premating treatment on embryo development and EPAF release in vitro. (C57BL/6 x CBA) F1 superovulated mice (n = 8) were treated with PAF and PBS (n = 10) on three consecutive days. On the night of the third treatment day, the mice were caged with fertile males, and the one-cell embryos were collected the following day (day 1) from those mice showing vaginal plugs. Embryos were cultured in Tyrodes (T6) media containing 4 mg/ml BSA, a s previously described (Roberts et al., 1987; Adamson e t al., 1991). On the fifth day of culture, the embryo conditioned media (ECM), excluding embryo cells, were collected to test for EPAF activity in the splenectomized mouse bioassay (SMB), a s previously described (Roberts et al., 1987). Briefly, serial 10-fold dilutions of ECM in T6 medium were prepared and each sample was injected intraperitoneally into 4 splenectomized mice. Peripheral platelet counts of the mice were made immediately before injection (0 min) and again 15 min after treatment. A greater than 10 percentage reduction in platelet counts between 0 and 15 minutes was taken as a measure of EPAF activity.
development recorded. Fetal sac numbers were also recorded on day 10 postmating to indicate successful implantation.
Relationship of thrombocytopenia with embryo morphology on day 4 of pregnancy following PAF pretreatment. QS female mice (n = 26) were either desensitized with PAF (0.6 pg in 300 pJ PBS) or treated with 300 pl PBS on three consecutive days prior to mating. All mice were mated to intact males on the evening of the third desensitization day. Thrombocytopenia was assessed by measuring blood platelet levels on day 0 (day of mating) and day 2 pregnancy as previously described (Roberts et al., 1987). On day 4 of pregnancy, embryos were counted, and their morphology was scored as normal or decompacted. The relationship between thrombocytopenia and embryo morphology was analysed using regressional analysis and one-way analysis of variance (ANOVA). Histology Embryos collected on day 4 from PAF and PBS pretreated mice were assessed for blastomere number, using the staining method of Tarkowski (1966). Briefly, embryos were placed in 0.5% (w/v) sodium citrate solution to grease-free slides. The embryos were fixed with 2-5 microdrops of methanol: acetic acid (3:l) and stained with 10% Giemsa solution. The blastomere number, indicated by the nuclei number, was assessed using phase-contrast microscopy. Uterine tissue were collected from PAF and PBS pretreated mice on day 4 postmating and fixed in Bouin’s fixative. Routine histological processing was performed by a tissue processor (Elliot, Liverpool, Australia). The sections were stained with Harris haematoxylin and eosin and examined for evidence of decidualization and implantation.
Effect of PAF pretreatment on embryo number and development in vivo. QS females similarly treated with PAF and PBS (7 mice per group) and mated as described above, were killed on each of the first 4 postmating days, and the embryos were collected from different parts of the reproductive tract as follows: from the ampulla region on day 1,from the oviducts on day 2 and 3 and from the oviducts and uterine horns on days 4. The embryos were counted and assessed for RESULTS their development. Day 4 embryos (n = 100) from five Development of Embryos Collected From PAF experiments were cultured at 37°C and 5% CO, overand PBS Pretreated Mice In Vitro night and scored for blastocyst development the followOne-cell embryos were collected from both PAF and ing day. The results were pooled for the final analysis. PBS pretreated (C57BL/6 x CBA) F1 animals and culEffect of PAF treatment at different times on embryo development and implantation. PAF (0.6 tured in Tyrode’s media for 5 days to the blastocyst pg) was administered on 3 consecutive days to female stage of development. Results showed that the percentQS mice beginning on either -5, -4, -3, -2, -1 days age of blastocysts in each PAF (158/219; 72%) and PBS prior to mating, day 0 (day of mating) or day 1(vaginal (191f272; 70%) pretreated group was comparable. plug seen). Mice from each group were then sacrificed When injected into splenectomized mice, the embryo on either day 4 or day 10 postmating. Day 4 embryos conditioned media demonstrated similar PAF activity were counted and assessed for their stage of develop- (data not shown). These results indicated that PAF prement while the number of fetal sacs were enumerated treatment did not affect the embryos ability to release EPAF. for day 10 pregnant mice. Effect of PAF concentrations on embryo develPAF Pretreatment Affects Embryos opment and implantation. QS females were treated on Day 4 Pregnancy with different doses of PAF (0.2, 0.4, 0.6, 0.8, 1.0 pg; in Embryos collected from QS mice on days 1-3 after 300 pl of PBS) on three consecutive days prior to mating and placed with fertile males on the evening of the mating were similar in number and developmental stathird treatment day. Mice similarly treated with PBS tus. However on day 4 of pregnancy, significantly fewer alone served as controls. Embryos were flushed from embryos (P < 0.005) were recovered from the uteri of mice on day 4 postmating and the number and state of PAF pretreated mice compared to controls (Table 1).
PREFERTILIZATION PAF TREATMENT AFFECTS PREGNANCY
365
TABLE 1. Percentage of Embryos Recovered at Different Stages of Development From PAF- and PBS-Pretreated Mice During the First 4 Days of Pregnancy Stage of embryo development (cell number) (percentage of total number of embryos recovered) Fragmented or Day of No. 2-4 >8 >8 >32 degenerated of mice 1 (%) NCa(%) Cb(%) (%) (%) pregnancy Treatment 90.5% 9.5 9 1 PAF 9.3 87.2% 3.4 PBS 9 16.2 2 PAF 7 7.4% 76.3 11.8 10.5% 78.2 PBS 8 10 7.7% 4.5 38.6 35.5 13.6 3 PAF PBS 7 7.1% 1.5 35.7 39.7 15.9 4 PAF 36 2.9% 7.0 47.0* 15.0 9.2* 18.8 0.6% 1.7 8.7 28.9 51.0 8.9 PBS 29 aNoncompacted. bCompacted. *P 64 cells per embryo. Histological examination of the uteri from which day 4 embryos were collected showed no evidence of implanting embryos in either treatment group.
Effects of PAF Pretreatment at Different Times and Concentration on Embryo Development and Implantation QS mice pretreated with PAF starting on days 4-5 and 0-1 prior to pregnancy showed no difference in embryo morphology or implantation characteristics compared to controls (Fig. 2a,b). Profound effects were evident in the group of mice pretreated with PAF beginning 1-3 days prior to mating. Implantation rates were decreased (Fig. 2b), and the number of decompacted embryos was significantly higher than controls (Fig. 2a). Using a n optimal injection regime (days -2, - 1, 0) and varying the dose of PAF administered, the reduction in implantation numbers remained unchanged (data not shown). The degree of embryo decompaction was greatest using a PAF concentration capable of inducing a 30% decline in platelet levels or lower (Fig. 3). Absence of Thrombocytopenia Due to Platelet Desensitization to PAF, Directly Affects Embryo Development on Day 4 Postmating Thrombocytopenic activity was assessed for PAFand PBS-pretreated QS mice and correlated with embryo development on day 4 of pregnancy. Thrombocytopenia activity was indicative for mice whose platelet
levels dropped greater than 10% on day 2 of pregnancy a s compared with platelet levels on the day of mating (day 0). Significantly more embryos (P < 0.01) displaying normal morphology (see Fig. l a ) were recovered from mice exhibiting thrombocytopenia (Fig. 4)as compared with mice displaying platelet desensitization. Conversely, decompacted embryos (see Fig. l b ) were recovered from mice in which there was no platelet decrease after mating and platelet concentrations remaining at normal levels.
DISCUSSION PAF treatment on three consecutive days prior to mating has been shown to reduce implantation rates significantly, a s assessed by the decrease in fetal sac numbers on day 10ofpregnancy (Adamson et al., 1987). In this study, PAF pretreatment had no deleterious effect on embryo development in culture. Embryos collected from PAF pretreated animals displayed both “normal” in vitro development and EPAF release over a 5-day culture period. Analysis of embryo conditioned media resulting from embryos obtained from PAF- and PBS-pretreated mice, showed similar EPAF activity in the splenectomized mouse bioassay. These observations indicate the inhibition of embryo-induced thrombocytopenia by PAF pretreatment, is more likely due to platelet desensitization rather than the absence of embryonic EPAF release. Premating PAF treatment did not affect ovulation or fertilization rate, a s shown by the equal proportion of one- and two-cell embryos recovered from treated and control mice on days 1 and 2 postmating respectively (Table 1). Similarly, embryos collected on day 3 showed no adverse effect of PAF pretreatment, with most embryos at the eight-cell noncompacted or compacted morula stage in both groups. Examination of embryos collected on day 4, however, showed two striking effects of PAF pretreatment. Significantly fewer embryos (P < 0.005) were recovered from the uteri of PAF pretreated mice on this day and a significantly reduced proportion of these embryos ( P < 0.005) showed normal
366
B.T. PODSIADLY ET AL.
Fig. 1. a: Photomicrographs of embryos collected on day 4 of pregnancy from PBS-pretreated (days - 2, -1, 0) superovulated QS mice. ( ~ 2 0 0b: ) Photomicrographs of embryos collected on day 4 of pregnancy from PAF-pretreated (days -2, - 1 , O ) superovulated QS mice. (x200).
development (Table 1). While most control embryos were found a t the compacted morula or blastocyst stage on day 4 (see Fig. la), the majority of embryos from the PAF pretreated mice displayed decompacted or “grapelike” morphology (see Fig. l b ) . This embryo decompaction was observed in both (C57B116 x CBA) F1 and QS mouse strains. To determine whether these embryos were decompacted or fragmented, the cells were stained for nuclear material. Results, showing that each blastomere contained a nucleus, suggesting that decompaction occurred a t the 16- to 32-cell stage. The effect of PAF pretreatment on embryo development was based on the inability of platelets to respond to embryonic EPAF after fertilization. Since platelet turnover in the mouse is 4-5 days (Bannerman 1983),it was not surprising that PAF administration 5-6 days prior to pregnancy showed no adverse effect on embryo morphology and implantation (see Fig. 2a,b). Similarly, PAF pretreatment beginning day 0 had little effect, suggesting platelet desensitization was incomplete. The results have shown that PAF pretreatment 3 days
prior to pregnancy (days -2, -1, 0 ) induced the greatest proportion of decompacted embryos and subsequent low implantation rate (see Fig. 2a,b). Furthermore, PAF pretreatment 4 days prior to pregnancy generated a significantly large proportion of normal embryos (P < 0.005) that failed to implant. Increasing PAF concentration on the 3 days prior to pregnancy induced similar low implantation rates; however, the most pronounced PAF effect on embryo morphology was exhibited when low PAF doses were used (Fig. 3). These results suggest that PAF-induced platelet desensitization acts on the uterus, resulting in embryo decompaction. Even though the concept of a hostile uterus has been described elsewhere (Psychoyos and Casimiri, 1981), this is the first report proposing that the uterine environment may directly regulate embryo compaction. Thrombocytopenia during early pregnancy in the mouse appears to be important €or normal embryo development. Decompacted embryos were recovered from PAF-pretreated mice that displayed no thrombocytopenia indicating that platelets or platelet products were
PREFERTILIZATION PAF TREATMENT AFFECTS PREGNANCY 120
20
1
0
-
loo
1
367
T
1
. a
0
.
'
"
'
I
.
'
I
'
'
"
'
l
.
"
.
'
l
'
'
t
0
0.2
0.4
0.6
PAF CONCENTRATION fua
0.8
1
PER MOUSF)
Fig. 3. Effect of PAF concentration on embryo development on day
4 of pregnancy. The degree of of normal embryos (m) and of decompacted embryos ( 0 ) as recovered on day 4 of pregnancy is shown as percentages (-tSE) of total embryo number with increasing PAF concentration in kg.
-6 -5 -4 -3 -2 -1 0 1 2 DAY OF FIRST INJECTION RELATIVE TO DAY OF MATING
Fig. 2. a: Time window sensitivity of PAF pretreatment on embryo development. Groups of 310 mice were treated with PAF (0) and PBS ( 0 )on 3 consecutive days beginning on days -5, -4, -3, -2, - 1 before mating, day of mating (day 0 ) and day 1 of pregnancy. Each point is the mean (-tSE) of the number of normal embrybos recovered on day 4 of pregnancy. *Embryo numbers significantly different (P < 0.05) between PAF- and PBS-pretreated mice. b Time window sensitivity of PAF pretreatment on implantation. Groups of 210 mice were treated with PAF ( 0 )and PBS ( 0 )on 3 consecutive days beginning on days -5, -4, -3, -2, -1 before mating, day of mating (day 0) and day 1 of pregnancy. Each point is the mean ( i S E ) of the number of fetal sacs present on day 10 of pregnancy. "Fetal sac numbers significantly different (P < 0.05) between PAF- and PBS-pretreated mice.
important for normal blastocyst development. Significantly more compacted morulae and blastocysts were recovered from mice that showed a platelet decline on day 2 of pregnancy (Fig. 4). Although these studies provide no evidence for the mechanism by which embryos decompact in vivo, various in vitro models have been reported. The most likely cause for embryo decompaction is the removal of calcium since the initiation and maintenance of compaction is strongly dependent on calcium for the maximization of cell-cell contact (Wales, 1970; Ducibella and Anderson, 1975). It is possible that these decompacted embryos may have been the result of insufficient calcium levels in the uterus of
-30% to -10%
-10% to +lo%
THROMBOCYTOPENIA PERCENTAGE CHANGE IN PLATELET LEVELSON D A Y 2 OFPREGNANCY.
Fig. 4. Relationship between embryo-induced thrombocytopenia and embryo development on day 4 of pregnancy. Percentages ( r S E )of normal embryos (m) and of decompacted embryos ( 0 ) as recovered on day 4 of pregnancy from PAF- and PBS-pretreated mice are correlated with thrombocytopenia (>lo% decline in platelet level). "Normal embryo number significantly different (P < 0.01) to decompacted embryo number when thrombocytopenia is evident.
PAF pretreated mice. This is supported by the fact that if placed in culture, 42% of decompacted embryos underwent blastulation within 24 hs, which is similar to reports showing in vitro embryo recompaction in calcium media (Ducibella and Anderson, 1975). Alternatively, PAF pretreatment may have directly interfered with the function of cell adhesion molecules. The inhibition of uvomorulin (Hyafil et al., 1980) or p-1,4 galactosyltransferase activity (Bayna e t al., 1988) also
368
B.T. PODSIADLY ET AL.
causes embryo decompaction, the enzyme being the first cell adhesion molecule identified to participate during late morula compaction, subsequent to uvomorulin function (Bayna et al., 1988). Studies are currently under way to address these proposals. In summary, our results indicate t h a t EPAF and platelets have a role in the maintenance of early pregnancy. While controversy still surrounds the exact chemical nature of EPAF (Angle et al., 1988; Amiel et al., 1989; Adamson et al., 19911, our results have shown that desensitization of platelets to PAF, a highly specific phenomenon, inhibits the platelet activation associated with early pregnancy in the mouse (Adamson et al., 1987). PAF is therefore involved directly (ie. being homologous to EPAF) or indirectly (ie. being a secondary product of EPAF liberation) during early pregnancy. The pregnancy associated platelet activation could liberate such factors as calcium and growth factors, which may play a role in early embryonic development. Alternatively, EPAF might also be a n early embryonic signal for uterine receptivity as pre-exposure of platelets to PAF prior to mating does not interfere with the potential of the embryo to continue development when released from the uterine environment.
ACKNOWLEDGMENTS We thank the University of Newcastle and Lingard Fertility Centre, Merewether for financial support.
REFERENCES Adamson LM, Smart YC, Stanger JD, Murdoch RN, Roberts TK (1987):Mechanistic studies of early pregnancy associated thrombocytopenia (EPAT) in the mouse. Am J Reprod Immunol Microbiol 13:117-120.
Adamson LM, Podsiadly BT, Smart YC, Stanger JD, Roberts TK (1991):Studies on murine embryo-derived platelet activating factor (EPAF).Mol Reprod Dev 30:207-213. Amid ML, Duquenne C, Benveniste J , Testart J (1989):Platelet aggregating activity in human embryo culture media free of PAFacether. Hum Reprod 4:327-330. Angle MJ, Jones MA, McMannus LM, Pinckard RN, Harper MJK (1988): Platelet activating factor in the rabbit uterus during early pregnancy. J Reprod Fertil83:711-722. Bannermann RM (1983):Hematology. In HL Foster, J D Small, J G Fox (eds): “The Mouse in Biomedical Research,” Vol. 111. New York: Academic Press, pp 293-312. Bayna EM, Shaper JH, Shur BD (1988): Temporally specific involvement of cell surface p-1,4 galactosyltransferase during mouse embryo morula compaction. Cell 53:145-157. Ducibella T, Anderson E (1975):Cell shape and membrane changes in the eight cell mouse embryo: Prerequisite for morphogenisis of the blastocyst. Dev Biol47:45-58. Henson PM (1977): Activation of rabbit platelets by platelet activating factor derived from IgE - sensitized basophils. Characteristics of the aggregation and its dissociation from secretion. J Clin Invest 60:481-487. Hyafil F, Morello D, Babinet C, Jacob F (1980):A cell surface glycoprotein involved in the compaction of embryonal carcinoma cells and cleavage stage embryos. Cell 21:927-934. O’Neill C (1985):Examination of the causes of early pregnancy associated thrombocytopenia in mice. J Reprod Fertil73:567-577. ONeill C, Saunders DM (1984):Assessment of embryo quality. Lancet 1:1035. Psychoyos A and Casimiri V (1981): Uterine blastotoxic factors. In SR Glasser, DW Bullock (eds): “Cellular and Molecular Aspects of Implantation.” New York: Plenum Press, pp 327-334. Roberts TK, Adamson LM, Smart YC, Stanger JD, Murdoch RN (1987): An evaluation of peripheral blood platelet enumeration as a monitor of fertilization and early pregnancy. Fertil Steril 47:848858. Tarkowski AK (1966):An air-drying method for chromosome preparation from mouse eggs. Cytogenetics 5:39&400. Wales RG (1970):Effects of ions on the development of the preimplantation mouse embryo in vitro. Aust J Biol Sci 23:421429.
Prefertilization Treatment of Mice With Platelet Activating Factor Affects Pregnancy B.T. PODSIADLY,~L.M. ADAMSON,' J.D. STANGER: Y.C. SMART: AND T.K. ROBERTS 'Departments of Biological Science and 'Discipline of Surgical Science, University of Newcastle, N S W 2308, Australia; 3Lingard Fertility Centre, Lingard Hospital, Merewether, Australia
ABSTRACT Embryo-derived platelet activating factor (EPAF) is thought to be either biologically similar to platelet activating factor (PAF) or responsible for PAF liberation in vivo. We have previously shown that premating PAF treatment in the mouse renders the platelets nonresponsive to EPAF, leading to a reduced implantation rate in these animals. In this study, we have shown that females, injected with PAF before mating, show altered embryo development in vivo on day 4 postfertilization. This is manifested as an interruption of compaction, a reduced cell number per embryo, and reduced embryo number per mouse. Results suggest that EPAF represents an early pregnancy signal that supports embryo development. The most likely mechanism is via platelet activation, since only those mice that showed thrombocytopenia after fertilization were found to have normal embryos on day 4 postmating. 0 1992 Wiley-Liss, Inc.
ing (Adamson et al., 1987) and highlights the importance of EPAF-induced platelet activation during early pregnancy. In the present study, we have investigated the likely mechanism for the reduced pregnancy rate following PAF treatment prior to mating by examining the effects of this treatment on the embryo itself and on embryonic development in vivo and have also shown that thrombocytopenia is a necessary event for normal pregnancy.
MATERIALS AND METHODS Mice (C57BL/6 x CBA) F1 mice of 4-6 weeks were used for all in vitro embryo studies as embryos from this species did not express a two-cell developmental block in vitro. Quackenbush (QS) mice of 4-6 weeks were used for all in vivo work. Mice were housed at a constant temperature of 21-23°C with food and water freely available. Key Words: Thrombocytopenia, Platelets,Embryo, D e All mice were superovulated with 5-7.5 IU PMSG compaction, Implantation, Abortion (pregnant mare serum gonadotrophin, Intervet, Australia), followed 48 h later by 5-7.5 IU hCG (human chorionic gonadotrophin, Intervet, Australia) and then caged with fertile males overnight. Only animals showing vaginal plugs approximately 15 h after hCG injecINTRODUCTION tion were used. The day of vaginal plug appearance was Embryo-derived platelet activating factor (EPAF), designated day 1of pregnancy. Between 0900 and 1200 a n early pregnancy marker detected in the mouse and h on day 4 of pregnancy, the embryos were removed and human (O'Neill and Saunders, 1984; Roberts et al., scored according to their stage of development (cell 19871, is produced by the preimplantation embryo. number). EPAF is responsible for the transient thrombocytopenia that occurs in the pregnant mouse between days 1 Platelet Activating Factor and 5 postmating. Although currently controversial, Platelet activating factor (PAF) (L-a-phosphatidylEPAF is thought to be either biologically similar to choline, P-acetyl, y-alkyl) was obtained as a 2 mg/ml platelet activating factor (PAF) (O'Neill, 1985; Adam- solution in chloroform (Sigma, St. Louis, MO) and preson et al., 1987) or responsible for PAF liberation in pared a t a concentration able to mimic the thrombocyvivo (Angle et al., 1988; Amiel et al., 1989; Adamson topenia seen in pregnancy and capable of inducing a n et al., 1991). in vivo 30% reduction in mouse platelet levels within Injection of PAF, a phospholipid mediator of platelet 15 min, followed by a return to basal levels by 60 min activation, a t a concentration capable of decreasing in (i.e., reversible platelet activation) (Adamson et al., vivo platelet levels by 30% in the mouse, on 3 consecu- 1987). PAF was prepared at a concentration of 5 pg/ml tive days immediately prior to mating inhibits maternal platelet activation and the appearance of thrombocytopenia. This activity is due to desensitization of platelets by PAF at concentrations that induce reversReceived October 7,1991; accepted February 13,1992. ible platelet aggregation (Henson, 1977; Adamson Address reprint requests to Dr. T.K. Roberts, Department of Biologiet al., 1987). The net result of this desensitization is a cal Sciences, University of Newcastle, University Drive, Callaghan reduction in implantation as detected 10 days postmat- NSW 2308. Australia.
0 1992 WILEY-LISS, INC.
364
B.T. PODSIADLY ET AL.
(batch #127F-8405) and 3 pgiml (batch #66F-8395) in Dulbecco’s phosphate-buffered saline (PBS) (Ca2+ and Mg2+free, Sigma, St. Louis, MO) containing 0.25% bovine serum albumin (BSA) (fraction V, Sigma, St. Louis, MO). PAF was injected intraperitoneally in aliquots of 200-300 p1, depending on the weight of the mouse. Effect of PAF premating treatment on embryo development and EPAF release in vitro. (C57BL/6 x CBA) F1 superovulated mice (n = 8) were treated with PAF and PBS (n = 10) on three consecutive days. On the night of the third treatment day, the mice were caged with fertile males, and the one-cell embryos were collected the following day (day 1) from those mice showing vaginal plugs. Embryos were cultured in Tyrodes (T6) media containing 4 mg/ml BSA, a s previously described (Roberts et al., 1987; Adamson e t al., 1991). On the fifth day of culture, the embryo conditioned media (ECM), excluding embryo cells, were collected to test for EPAF activity in the splenectomized mouse bioassay (SMB), a s previously described (Roberts et al., 1987). Briefly, serial 10-fold dilutions of ECM in T6 medium were prepared and each sample was injected intraperitoneally into 4 splenectomized mice. Peripheral platelet counts of the mice were made immediately before injection (0 min) and again 15 min after treatment. A greater than 10 percentage reduction in platelet counts between 0 and 15 minutes was taken as a measure of EPAF activity.
development recorded. Fetal sac numbers were also recorded on day 10 postmating to indicate successful implantation.
Relationship of thrombocytopenia with embryo morphology on day 4 of pregnancy following PAF pretreatment. QS female mice (n = 26) were either desensitized with PAF (0.6 pg in 300 pJ PBS) or treated with 300 pl PBS on three consecutive days prior to mating. All mice were mated to intact males on the evening of the third desensitization day. Thrombocytopenia was assessed by measuring blood platelet levels on day 0 (day of mating) and day 2 pregnancy as previously described (Roberts et al., 1987). On day 4 of pregnancy, embryos were counted, and their morphology was scored as normal or decompacted. The relationship between thrombocytopenia and embryo morphology was analysed using regressional analysis and one-way analysis of variance (ANOVA). Histology Embryos collected on day 4 from PAF and PBS pretreated mice were assessed for blastomere number, using the staining method of Tarkowski (1966). Briefly, embryos were placed in 0.5% (w/v) sodium citrate solution to grease-free slides. The embryos were fixed with 2-5 microdrops of methanol: acetic acid (3:l) and stained with 10% Giemsa solution. The blastomere number, indicated by the nuclei number, was assessed using phase-contrast microscopy. Uterine tissue were collected from PAF and PBS pretreated mice on day 4 postmating and fixed in Bouin’s fixative. Routine histological processing was performed by a tissue processor (Elliot, Liverpool, Australia). The sections were stained with Harris haematoxylin and eosin and examined for evidence of decidualization and implantation.
Effect of PAF pretreatment on embryo number and development in vivo. QS females similarly treated with PAF and PBS (7 mice per group) and mated as described above, were killed on each of the first 4 postmating days, and the embryos were collected from different parts of the reproductive tract as follows: from the ampulla region on day 1,from the oviducts on day 2 and 3 and from the oviducts and uterine horns on days 4. The embryos were counted and assessed for RESULTS their development. Day 4 embryos (n = 100) from five Development of Embryos Collected From PAF experiments were cultured at 37°C and 5% CO, overand PBS Pretreated Mice In Vitro night and scored for blastocyst development the followOne-cell embryos were collected from both PAF and ing day. The results were pooled for the final analysis. PBS pretreated (C57BL/6 x CBA) F1 animals and culEffect of PAF treatment at different times on embryo development and implantation. PAF (0.6 tured in Tyrode’s media for 5 days to the blastocyst pg) was administered on 3 consecutive days to female stage of development. Results showed that the percentQS mice beginning on either -5, -4, -3, -2, -1 days age of blastocysts in each PAF (158/219; 72%) and PBS prior to mating, day 0 (day of mating) or day 1(vaginal (191f272; 70%) pretreated group was comparable. plug seen). Mice from each group were then sacrificed When injected into splenectomized mice, the embryo on either day 4 or day 10 postmating. Day 4 embryos conditioned media demonstrated similar PAF activity were counted and assessed for their stage of develop- (data not shown). These results indicated that PAF prement while the number of fetal sacs were enumerated treatment did not affect the embryos ability to release EPAF. for day 10 pregnant mice. Effect of PAF concentrations on embryo develPAF Pretreatment Affects Embryos opment and implantation. QS females were treated on Day 4 Pregnancy with different doses of PAF (0.2, 0.4, 0.6, 0.8, 1.0 pg; in Embryos collected from QS mice on days 1-3 after 300 pl of PBS) on three consecutive days prior to mating and placed with fertile males on the evening of the mating were similar in number and developmental stathird treatment day. Mice similarly treated with PBS tus. However on day 4 of pregnancy, significantly fewer alone served as controls. Embryos were flushed from embryos (P < 0.005) were recovered from the uteri of mice on day 4 postmating and the number and state of PAF pretreated mice compared to controls (Table 1).
PREFERTILIZATION PAF TREATMENT AFFECTS PREGNANCY
365
TABLE 1. Percentage of Embryos Recovered at Different Stages of Development From PAF- and PBS-Pretreated Mice During the First 4 Days of Pregnancy Stage of embryo development (cell number) (percentage of total number of embryos recovered) Fragmented or Day of No. 2-4 >8 >8 >32 degenerated of mice 1 (%) NCa(%) Cb(%) (%) (%) pregnancy Treatment 90.5% 9.5 9 1 PAF 9.3 87.2% 3.4 PBS 9 16.2 2 PAF 7 7.4% 76.3 11.8 10.5% 78.2 PBS 8 10 7.7% 4.5 38.6 35.5 13.6 3 PAF PBS 7 7.1% 1.5 35.7 39.7 15.9 4 PAF 36 2.9% 7.0 47.0* 15.0 9.2* 18.8 0.6% 1.7 8.7 28.9 51.0 8.9 PBS 29 aNoncompacted. bCompacted. *P 64 cells per embryo. Histological examination of the uteri from which day 4 embryos were collected showed no evidence of implanting embryos in either treatment group.
Effects of PAF Pretreatment at Different Times and Concentration on Embryo Development and Implantation QS mice pretreated with PAF starting on days 4-5 and 0-1 prior to pregnancy showed no difference in embryo morphology or implantation characteristics compared to controls (Fig. 2a,b). Profound effects were evident in the group of mice pretreated with PAF beginning 1-3 days prior to mating. Implantation rates were decreased (Fig. 2b), and the number of decompacted embryos was significantly higher than controls (Fig. 2a). Using a n optimal injection regime (days -2, - 1, 0) and varying the dose of PAF administered, the reduction in implantation numbers remained unchanged (data not shown). The degree of embryo decompaction was greatest using a PAF concentration capable of inducing a 30% decline in platelet levels or lower (Fig. 3). Absence of Thrombocytopenia Due to Platelet Desensitization to PAF, Directly Affects Embryo Development on Day 4 Postmating Thrombocytopenic activity was assessed for PAFand PBS-pretreated QS mice and correlated with embryo development on day 4 of pregnancy. Thrombocytopenia activity was indicative for mice whose platelet
levels dropped greater than 10% on day 2 of pregnancy a s compared with platelet levels on the day of mating (day 0). Significantly more embryos (P < 0.01) displaying normal morphology (see Fig. l a ) were recovered from mice exhibiting thrombocytopenia (Fig. 4)as compared with mice displaying platelet desensitization. Conversely, decompacted embryos (see Fig. l b ) were recovered from mice in which there was no platelet decrease after mating and platelet concentrations remaining at normal levels.
DISCUSSION PAF treatment on three consecutive days prior to mating has been shown to reduce implantation rates significantly, a s assessed by the decrease in fetal sac numbers on day 10ofpregnancy (Adamson et al., 1987). In this study, PAF pretreatment had no deleterious effect on embryo development in culture. Embryos collected from PAF pretreated animals displayed both “normal” in vitro development and EPAF release over a 5-day culture period. Analysis of embryo conditioned media resulting from embryos obtained from PAF- and PBS-pretreated mice, showed similar EPAF activity in the splenectomized mouse bioassay. These observations indicate the inhibition of embryo-induced thrombocytopenia by PAF pretreatment, is more likely due to platelet desensitization rather than the absence of embryonic EPAF release. Premating PAF treatment did not affect ovulation or fertilization rate, a s shown by the equal proportion of one- and two-cell embryos recovered from treated and control mice on days 1 and 2 postmating respectively (Table 1). Similarly, embryos collected on day 3 showed no adverse effect of PAF pretreatment, with most embryos at the eight-cell noncompacted or compacted morula stage in both groups. Examination of embryos collected on day 4, however, showed two striking effects of PAF pretreatment. Significantly fewer embryos (P < 0.005) were recovered from the uteri of PAF pretreated mice on this day and a significantly reduced proportion of these embryos ( P < 0.005) showed normal
366
B.T. PODSIADLY ET AL.
Fig. 1. a: Photomicrographs of embryos collected on day 4 of pregnancy from PBS-pretreated (days - 2, -1, 0) superovulated QS mice. ( ~ 2 0 0b: ) Photomicrographs of embryos collected on day 4 of pregnancy from PAF-pretreated (days -2, - 1 , O ) superovulated QS mice. (x200).
development (Table 1). While most control embryos were found a t the compacted morula or blastocyst stage on day 4 (see Fig. la), the majority of embryos from the PAF pretreated mice displayed decompacted or “grapelike” morphology (see Fig. l b ) . This embryo decompaction was observed in both (C57B116 x CBA) F1 and QS mouse strains. To determine whether these embryos were decompacted or fragmented, the cells were stained for nuclear material. Results, showing that each blastomere contained a nucleus, suggesting that decompaction occurred a t the 16- to 32-cell stage. The effect of PAF pretreatment on embryo development was based on the inability of platelets to respond to embryonic EPAF after fertilization. Since platelet turnover in the mouse is 4-5 days (Bannerman 1983),it was not surprising that PAF administration 5-6 days prior to pregnancy showed no adverse effect on embryo morphology and implantation (see Fig. 2a,b). Similarly, PAF pretreatment beginning day 0 had little effect, suggesting platelet desensitization was incomplete. The results have shown that PAF pretreatment 3 days
prior to pregnancy (days -2, -1, 0 ) induced the greatest proportion of decompacted embryos and subsequent low implantation rate (see Fig. 2a,b). Furthermore, PAF pretreatment 4 days prior to pregnancy generated a significantly large proportion of normal embryos (P < 0.005) that failed to implant. Increasing PAF concentration on the 3 days prior to pregnancy induced similar low implantation rates; however, the most pronounced PAF effect on embryo morphology was exhibited when low PAF doses were used (Fig. 3). These results suggest that PAF-induced platelet desensitization acts on the uterus, resulting in embryo decompaction. Even though the concept of a hostile uterus has been described elsewhere (Psychoyos and Casimiri, 1981), this is the first report proposing that the uterine environment may directly regulate embryo compaction. Thrombocytopenia during early pregnancy in the mouse appears to be important €or normal embryo development. Decompacted embryos were recovered from PAF-pretreated mice that displayed no thrombocytopenia indicating that platelets or platelet products were
PREFERTILIZATION PAF TREATMENT AFFECTS PREGNANCY 120
20
1
0
-
loo
1
367
T
1
. a
0
.
'
"
'
I
.
'
I
'
'
"
'
l
.
"
.
'
l
'
'
t
0
0.2
0.4
0.6
PAF CONCENTRATION fua
0.8
1
PER MOUSF)
Fig. 3. Effect of PAF concentration on embryo development on day
4 of pregnancy. The degree of of normal embryos (m) and of decompacted embryos ( 0 ) as recovered on day 4 of pregnancy is shown as percentages (-tSE) of total embryo number with increasing PAF concentration in kg.
-6 -5 -4 -3 -2 -1 0 1 2 DAY OF FIRST INJECTION RELATIVE TO DAY OF MATING
Fig. 2. a: Time window sensitivity of PAF pretreatment on embryo development. Groups of 310 mice were treated with PAF (0) and PBS ( 0 )on 3 consecutive days beginning on days -5, -4, -3, -2, - 1 before mating, day of mating (day 0 ) and day 1 of pregnancy. Each point is the mean (-tSE) of the number of normal embrybos recovered on day 4 of pregnancy. *Embryo numbers significantly different (P < 0.05) between PAF- and PBS-pretreated mice. b Time window sensitivity of PAF pretreatment on implantation. Groups of 210 mice were treated with PAF ( 0 )and PBS ( 0 )on 3 consecutive days beginning on days -5, -4, -3, -2, -1 before mating, day of mating (day 0) and day 1 of pregnancy. Each point is the mean ( i S E ) of the number of fetal sacs present on day 10 of pregnancy. "Fetal sac numbers significantly different (P < 0.05) between PAF- and PBS-pretreated mice.
important for normal blastocyst development. Significantly more compacted morulae and blastocysts were recovered from mice that showed a platelet decline on day 2 of pregnancy (Fig. 4). Although these studies provide no evidence for the mechanism by which embryos decompact in vivo, various in vitro models have been reported. The most likely cause for embryo decompaction is the removal of calcium since the initiation and maintenance of compaction is strongly dependent on calcium for the maximization of cell-cell contact (Wales, 1970; Ducibella and Anderson, 1975). It is possible that these decompacted embryos may have been the result of insufficient calcium levels in the uterus of
-30% to -10%
-10% to +lo%
THROMBOCYTOPENIA PERCENTAGE CHANGE IN PLATELET LEVELSON D A Y 2 OFPREGNANCY.
Fig. 4. Relationship between embryo-induced thrombocytopenia and embryo development on day 4 of pregnancy. Percentages ( r S E )of normal embryos (m) and of decompacted embryos ( 0 ) as recovered on day 4 of pregnancy from PAF- and PBS-pretreated mice are correlated with thrombocytopenia (>lo% decline in platelet level). "Normal embryo number significantly different (P < 0.01) to decompacted embryo number when thrombocytopenia is evident.
PAF pretreated mice. This is supported by the fact that if placed in culture, 42% of decompacted embryos underwent blastulation within 24 hs, which is similar to reports showing in vitro embryo recompaction in calcium media (Ducibella and Anderson, 1975). Alternatively, PAF pretreatment may have directly interfered with the function of cell adhesion molecules. The inhibition of uvomorulin (Hyafil et al., 1980) or p-1,4 galactosyltransferase activity (Bayna e t al., 1988) also
368
B.T. PODSIADLY ET AL.
causes embryo decompaction, the enzyme being the first cell adhesion molecule identified to participate during late morula compaction, subsequent to uvomorulin function (Bayna et al., 1988). Studies are currently under way to address these proposals. In summary, our results indicate t h a t EPAF and platelets have a role in the maintenance of early pregnancy. While controversy still surrounds the exact chemical nature of EPAF (Angle et al., 1988; Amiel et al., 1989; Adamson et al., 19911, our results have shown that desensitization of platelets to PAF, a highly specific phenomenon, inhibits the platelet activation associated with early pregnancy in the mouse (Adamson et al., 1987). PAF is therefore involved directly (ie. being homologous to EPAF) or indirectly (ie. being a secondary product of EPAF liberation) during early pregnancy. The pregnancy associated platelet activation could liberate such factors as calcium and growth factors, which may play a role in early embryonic development. Alternatively, EPAF might also be a n early embryonic signal for uterine receptivity as pre-exposure of platelets to PAF prior to mating does not interfere with the potential of the embryo to continue development when released from the uterine environment.
ACKNOWLEDGMENTS We thank the University of Newcastle and Lingard Fertility Centre, Merewether for financial support.
REFERENCES Adamson LM, Smart YC, Stanger JD, Murdoch RN, Roberts TK (1987):Mechanistic studies of early pregnancy associated thrombocytopenia (EPAT) in the mouse. Am J Reprod Immunol Microbiol 13:117-120.
Adamson LM, Podsiadly BT, Smart YC, Stanger JD, Roberts TK (1991):Studies on murine embryo-derived platelet activating factor (EPAF).Mol Reprod Dev 30:207-213. Amid ML, Duquenne C, Benveniste J , Testart J (1989):Platelet aggregating activity in human embryo culture media free of PAFacether. Hum Reprod 4:327-330. Angle MJ, Jones MA, McMannus LM, Pinckard RN, Harper MJK (1988): Platelet activating factor in the rabbit uterus during early pregnancy. J Reprod Fertil83:711-722. Bannermann RM (1983):Hematology. In HL Foster, J D Small, J G Fox (eds): “The Mouse in Biomedical Research,” Vol. 111. New York: Academic Press, pp 293-312. Bayna EM, Shaper JH, Shur BD (1988): Temporally specific involvement of cell surface p-1,4 galactosyltransferase during mouse embryo morula compaction. Cell 53:145-157. Ducibella T, Anderson E (1975):Cell shape and membrane changes in the eight cell mouse embryo: Prerequisite for morphogenisis of the blastocyst. Dev Biol47:45-58. Henson PM (1977): Activation of rabbit platelets by platelet activating factor derived from IgE - sensitized basophils. Characteristics of the aggregation and its dissociation from secretion. J Clin Invest 60:481-487. Hyafil F, Morello D, Babinet C, Jacob F (1980):A cell surface glycoprotein involved in the compaction of embryonal carcinoma cells and cleavage stage embryos. Cell 21:927-934. O’Neill C (1985):Examination of the causes of early pregnancy associated thrombocytopenia in mice. J Reprod Fertil73:567-577. ONeill C, Saunders DM (1984):Assessment of embryo quality. Lancet 1:1035. Psychoyos A and Casimiri V (1981): Uterine blastotoxic factors. In SR Glasser, DW Bullock (eds): “Cellular and Molecular Aspects of Implantation.” New York: Plenum Press, pp 327-334. Roberts TK, Adamson LM, Smart YC, Stanger JD, Murdoch RN (1987): An evaluation of peripheral blood platelet enumeration as a monitor of fertilization and early pregnancy. Fertil Steril 47:848858. Tarkowski AK (1966):An air-drying method for chromosome preparation from mouse eggs. Cytogenetics 5:39&400. Wales RG (1970):Effects of ions on the development of the preimplantation mouse embryo in vitro. Aust J Biol Sci 23:421429.
