Similar embryotoxic effects of sera from infertile patients and exogenous interferon-{gamma} on long-term in-vitro development of mouse embryos

Monica Cameo1,5, Vanina Fontana1, Paula Cameo1, Liliana G. Vauthay2, Juana Kaplanand3 and Marta Tesone4

1 Laboratorio Biología de la Reproducción, Ecuador 1465 2°B (1425), 2 Dpto. Bioterio y Cancer Experimental, Instituto de Oncología, A.H- Roffo (UBA), 3 Clínica y MaternidadSuizo-Argentina, 4 Instituto de Biología y Medicina Experimental (CONICET-UBA), Buenos Aires, Argentina


    Abstract
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
Circulating embryotoxic factors could be responsible for reproductive failures observed in patients suffering from recurrent spontaneous abortions (RSA) and endometriosis. The mouse bioassay has been widely used to detect such factors, since sera from these patients inhibit early embryonic development. This bioassay consists in the in-vitro culture of two-cell mouse embryos in the presence of different sera up to the blastocyst stage (72 h of culture). In the present study experiments were performed over long culture times (3–7 days), from two-cell to spreading stages, to determine the in-vitro effect of sera obtained from RSA or endometriosis patients, as well as the effect of interferon (INF)-{gamma} on embryo development. An embryotoxicity cut-off value of 45% blastocyst formation was established using control sera. When development to the blastocyst stage was considered only 25% of RSA and 20% of endometriosis sera were embryotoxic. However, all RSA sera significantly inhibited hatching (P < 0.05) and spreading stages (P < 0.01). IFN-{gamma} (10 µg/ml) (P < 0.001) did not impair early embryo development, but significantly inhibited blastocyst spreading. These observations suggest that culture to advanced embryonic stages increases the sensitivity of the bioassay and that IFN-{gamma} alters in-vitro peri-implantation mouse embryo development.

Key words: embryotoxic factors/endometriosis/IFN-{gamma}/recurrent spontaneous abortions


    Introduction
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
A large number of women affected by infertility problems suffer recurrent spontaneous abortions (RSA), classically defined as three or more abortions within the first trimester of pregnancy (Edmonds et al., 1982Go). In many cases, the causes of the abortion can be identified as chromosomal, endocrine, anatomical, infectious or immunological anomalies, but in others the cause of the disease is unknown. In addition, it has been established that 30–40% of patients suffering endometriosis are infertile (Abu-Musa et al., 1992Go). The most commonly proposed causes for endometriosis-related infertility are: anovulation (Roennberg et al., 1984Go), affected tubal transport (Croxatto et al., 1978Go), increased follicular fluid prostaglandin levels (Rock et al., 1982Go) and/or activated macrophages (Haney et al., 1981Go), and early loss of pregnancy (Wheeler et al., 1983Go).

During the last few years, evidence has accumulated suggesting that circulating embryotoxic factors could be responsible for reproductive failure in RSA and endometriosis patients. The in-vitro culture of mouse embryos is the most common experimental model for the detection of embryotoxins in sera. This technique is based on the great sensitivity of murine embryo development to substances present in the medium, and consists of the in-vitro culture of two-cell embryos up to the blastocyst stage under highly controlled conditions, in medium supplemented with the serum to be tested. Studies carried out up to now postulate the presence of embryotoxic factors in those cases in which an inhibition of development to blastocyst is observed within 72 h of culture.

Oksenberg and Brautbar (1986) found that 70% of sera from RSA patients inhibited murine embryo development in vitro. In addition, Dokras et al. (1993) demonstrated that 54.5% of sera from RSA patients affected development to blastocyst when compared to controls. This inhibition coincides with a decrease in cell proliferation, measured as incorporation of 3H-thymidine. In this same group of patients, a 0% pregnancy rate was detected, compared to an 80% rate for RSA patients who did not present this inhibitory effect. After validation of the assay (Roussev et al., 1995Go), Thomason et al. (1995) found embryotoxicity in 24% of 160 RSA patients. In addition, Damewood et al. (1990) compared the development of mouse embryos in the presence of sera from tubal factor, male factor and endometriosis patients. They observed an inhibition of development by sera from endometriosis patients compared with the other two groups. Simon et al. (1994) studied the effects of sera from endometriosis patients, and demonstrated that there exists a significant inhibition of embryo development in the presence of these sera. Hill et al. (1992) studied the effect of activated mononuclear cell supernatants from RSA patients on in-vitro embryo development, and detected an inhibition of development to the blastocyst stage. They concluded that a Th1-type cytokine, IFN-{gamma}, is responsible for the toxic effect on embryonic and trophoblast cells in vitro. The majority of these studies determined embryonic development after only 3 days of culture.

The present study uses the mouse model to determine the effects on embryo development of sera obtained from RSA or endometriosis patients in the culture medium. Experiments were performed over long culture periods (3–7 days), from the two-cell to the hatching, attachment and spreading blastocyst stages. The effect of IFN-{gamma} on mouse embryo development was also tested.


    Materials and methods
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
Human sera
Sera from 59 RSA and 30 endometriosis patients were studied. Patients included women with a history of RSA, who had been previously evaluated to discard infection, chromosomal, anatomical and/or hormonal anomalies (RSA group); and women with minimal to mild endometriosis diagnosed by laparoscopy and classified according to the criteria of the American Fertility Society (1985) (endometriosis group). The control group consisted of sera from five women and 10 men, all young and in good health, and five sera from an in-vitro fertilization programme in which fertilization and cleavage had been observed during human embryo culture. These sera were kindly donated by the Gynecological and Reproduction Team of the Hospital Italiano, Buenos Aires. Synthetic serum (SS) (Irvine Scientific, Santa Ana, CA, USA) was used as an additional control sera capable of maintaining in-vitro embryo development up to the blastocyst stage. All sera were obtained under sterile conditions, centrifuged 15 min at 1200 g, heated to 56°C for complement inactivation, and stored at –18°C until use. Sera were not charcoal stripped.

Superovulation and embryo recovery
B6CF1 hybrid mice (2–3 months of age), were stimulated for ovulation induction by subcutaneous injection of 5 IU of follicle-stimulating hormone (FSH, Instituto Massone, Buenos Aires, Argentina) followed 45 h later by 5 IU of human chorionic gonadotrophin (HCG, Instituto Massone, Buenos Aires, Argentina). Two hours after the last injection, each female was mated with a male BALB/c mouse. Copulation was confirmed the following day by the presence of a vaginal plug. Allowing 36–40 h after the HCG injection, those females presenting a vaginal plug were killed by cervical dislocation, their oviducts dissected, and the embryos recovered by flushing with phosphate-buffered saline (PBS). All the two-cell embryos recovered were pooled in pre-equilibrated human tubal fluid medium (HTF, Irvine Scientific), and the embryos kept at 37°C in an atmosphere of 5% CO2 until seeding.

Embryo culture
The embryos were randomly allocated to the control or patient groups. Up to five embryos were grown in each well, containing 1 ml of medium supplemented with 50% control or patient inactivated serum. The serum concentration was selected in preliminary experiments in which several sera showed embryotoxic activity at 50%, but were negative at 10%. Each assay was carried out in duplicate, and cultures were maintained at 37°C in 5% CO2 in air. The embryos were inspected at 3, 5 and 7 days of culture for progression.

After 3 days (72 h) of culture we evaluated: (i) the proportion of total blastocysts, i.e. blastocysts that underwent expansion, regardless of whether hatching had occurred, and (ii) the proportion of hatched blastocysts, i.e. blastocysts that had both expanded, and partially or totally lost their zona pellucidae.

After 5 days (120 h) of culture we determined the proportion of attachment, i.e. blastocysts that had attached to the culture dish and could not be separated by gentle agitation.

After 7 days (168 h) of culture we evaluated the percentage of normal spreading, i.e. characterized by the differentiation and migration of the trophectoderm which allows one to distinguish a rounded inner cell mass (ICM) and a surrounding flattened area of trophoblast outgrowth. The morphology of the trophoblast and ICM was examined and photographed by inverted microscopy. Abnormal spreading was identified according to Chavez and McIntyre's (1984) criteria, i.e. aggregation of the inner cell mass surrounded by only a few trophectodermal cells.

The mean percentages and the respective standard errors (SE) were calculated for each of the groups.

Effect of IFN-{gamma}
Two hundred and thirty-three embryos were grown in HTF medium containing 10% heat-inactivated non-embryotoxic serum, with or without human IFN-{gamma} (Boehringer Ingelheim, Vienna, Austria; 10 µg/ml). This serum concentration was the lower one that allowed a normal embryo development at long culture periods (data not shown). IFN-{gamma} was added to the culture medium at three different times: day 0, day 3 or day 5 of culture. The cultures were maintained at 37°C in 5% CO2 in air, and embryos inspected as described above. The results obtained were expressed as the percentage of the total number of embryos cultured.

Statistical analysis
Results were statistically evaluated by one-way analysis of variance (ANOVA), Student's or {chi}2 test as appropriate. Values of P < 0.05 were considered statistically significant. To determine the lower limit of tolerance for normality, or cut-off value, the mean minus two standard deviations was calculated using the control values, as proposed by Hill et al. (1992).


    Results
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
72 h of culture
No significant differences were found between the values obtained for the different types of control sera used (sera from young, healthy subjects of both sexes, and sera from subjects with successful fertilization and cleavage). Thus, an average value of 76.6 ± 3.5% blastocysts was established as the control value (Table IGo). Using this control value, a cut-off value of 45% was calculated (76.6% – [2x15.7]). This allowed the definition of any serum in whose presence the development to blastocyst was below 45%, as embryotoxic. Values obtained with SS were similar (78.0 ± 2.0 %, n = 27).


View this table:
[in this window]
[in a new window]
 
Table I. Embryotoxicity assay in the presence of control sera
 
The embryotoxicity assay was carried out on a total of 59 sera from RSA patients, and 30 sera from endometriosis patients. In 25% RSA and 20% endometriosis cases, the mean percentage of total blastocysts after 3 days of culture was significantly lower than in the control group (P < 0.001, Table IIGo). In 36 of the assays with sera from RSA patients, we also determined whether blastocysts had reached the hatching stage or had merely expanded after 3 days of culture. The average percentage of hatched blastocysts for sera from RSA women was significantly lower than for the control group (P < 0.05). The same inhibitory behaviour was observed for the SS group (P < 0.05). On the other hand, when the average percentage of hatched blastocysts for sera from endometriosis patients was compared to control values, no inhibitory effect could be observed (Table IIIGo).


View this table:
[in this window]
[in a new window]
 
Table II. Embryotoxicity of sera from recurrent spontaneous abortion and endometriosis patients
 

View this table:
[in this window]
[in a new window]
 
Table III. Percentages of total and hatched blastocysts in the presence of different seraa
 
5 days (120 h) of culture
Embryonic development after longer periods of culture was evaluated in the presence of control (C) (n = 6), RSA (n = 27), and endometriosis (n = 14) sera. After 5 days of culture, in 29% (8/27) of RSA and 21% (3/14) of endometriosis sera, the mean percentage of attachment was significantly lower than in the control group (13.5 ± 8.2% RSA, 0% endometriosis versus 66 ± 7.5% C, P < 0.05).

7 days (168 h) of culture
After 7 days of culture, the spreading stage was significantly inhibited for embryos cultured in medium supplemented with sera from RSA patients (n = 13), compared to embryos cultured in control serum (n = 10). A total of 81.6 ± 3.8% of embryos in the control group exhibited the beginnings of ICM and trophoblast formation; with an ICM surrounded by flat, elongated trophectodermal cells in the form of a monolayer (Figure 1AGo). On the other hand only 49.0 ± 9.3% of embryos in the RSA group (P < 0.01 versus C) showed ICM and trophoblast monolayer formation, while the remaining 51.1 ± 9.2% of embryos showed either a narrow line of cells instead of a surrounding monolayer (Figure 1BGo), or no detectable monolayer was formed (Figure 1CGo). Endometriosis sera were not assayed at this culture time.



View larger version (160K):
[in this window]
[in a new window]
 
Figure 1. Culture of two-cell embryos in different conditions to the spreading stage. Embryos originating in presence of control sera (A) exhibited the inner cell mass (arrow) surrounded by flat elongated trophectodermal cells; aggregates composed of closely applied single cells (arrowheads) can be observed. The presence of sera from recurrent spontaneous abortion patients affected trophectoderm growth which appeared in the form of a narrow line of cells (B) or totally inhibited (C). The addition of IFN-{gamma} (10 µg/ml) at the beginning of culture inhibited trophectoderm growth (D). Scale bars = 90 µm (A), 50 µm (B), 45 µm (C), 50 µm (D).

 
Effect of IFN-{gamma}
Table IVGo shows the effect of IFN-{gamma} (10 µg/ml), added at different culture times, on embryo development. The addition of IFN-{gamma} at the beginning of culture (day 0), did not impair early embryo development but significantly (P < 0.001) inhibited the spreading stage. After 7 days of culture (Figure 1dGo) none of the 43 embryos in the study group exhibited an ICM surrounded by typical flat, elongated trophectoderm cells, while 33/59 (56%) of the embryos in the control group showed a normal appearance. The addition of IFN-{gamma} on day 3 of culture produced a similar effect, a typical spreading was observed in only 12/61 (20%) of embryos in the study group, versus 56% in the control. The addition of IFN-{gamma} on day 5 also inhibited spreading of embryos (3/37, 8 versus 56% for IFN-{gamma} versus control). Lower concentrations of IFN-{gamma} did not produce significant effects (data not shown).


View this table:
[in this window]
[in a new window]
 
Table IV. The effect of interferon (IFN)-{gamma} (10 µg/ml) added at different cultures times (days 0, 3 and 5) on the development of 2-cell embryos to the blastocyst, attachment and spreading stagesa
 

    Discussion
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
The present study demonstrates that the addition of heat-inactivated sera from women with case histories of recurrent spontaneous abortions (RSA) or endometriosis, significantly inhibits the in-vitro development of mouse embryos from the two-cell to the spreading stages. This conclusion is based on the direct observation of embryo development by inverted microscopy, and the comparison of the effects produced by the addition of control sera and sera from RSA or endometriosis patients.

The present results confirm the subdivision of the RSA and endometriosis patients into two groups, according to the in-vitro embryo development to the blastocyst stage: embryotoxic and non embryotoxic. Sera were considered embryotoxic or inhibitory when the proportion of two-cell embryos that developed to the blastocyst stage at 3 days of culture was below 45%, which is very similar to the 50% cut-off value reported by Hill et al. (1992). Within the RSA and endometriosis groups, 25 and 20% of patients respectively showed embryotoxic activity. It is noteworthy that the incidence of embryotoxicity in the RSA group was similar to that described by Thomason et al. (1995) for a group of 160 patients.

In the present study, only those sera that inhibited the expansion of mouse blastocysts have been classified as embryotoxic. However, if the hatching from the zona pellucida or spreading was considered, the average percentage corresponding to all RSA tested was significantly lower than the corresponding value in the control group. The fact that attachment was not similarly altered suggests that hatching and spreading stages would be the most sensitive stages of the system. These results demonstrate the importance of culturing for periods of time longer than the 72 h usually employed. On the other hand in the endometriosis group (data not shown), the proportion of inhibiting sera remained the same even when the percentage of hatched blastocysts was considered. In addition, it is important to note that the presence of SS in the culture medium also resulted in a significant decrease in blastocyst hatching, compared to controls. The negative effect of SS could be due to the absence of proteins and/or growth factors that are present in control sera, which might be necessary for blastocyst hatching and spreading.

Despite the numerous studies that suggest the presence of embryotoxic factors in human sera, these have not yet been specifically characterized. Oksenberg and Brautbar (1986) postulate that the factor responsible for embryotoxicity could be an immunoglobulin G (IgG), though Hill et al. (1992) mention IFN-{gamma} as the factor responsible for the inhibition of development to the blastocyst stage. The long-term culture system (7 days) used in our experimental design made it possible to reveal deleterious effects of IFN-{gamma} on in-vitro embryo development which were similar to those observed for RSA sera. Further experiments are necessary to demonstrate whether IFN-{gamma}, probably present in RSA sera, is responsible for this effect. Taking into account that human and mouse IFN-{gamma} receptors are structurally similar and exhibit an important homology (Cofano et al., 1990Go), human IFN-{gamma} was used in this study. However, the fact that the inhibitory effect was only observed at high IFN-{gamma} concentrations (10 µg/ml) could be attributable to different affinities for both cytokines.

On a different note we should also consider that the inhibitory effect on embryo development observed in the sera of some infertile patients might be due to the low levels or absence of a factor(s) indispensable for the in-vitro development of embryos.

Embryotoxic effect has been also studied in hydrosalpingeal fluid (HF). Beyler et al. (1997) reported that HF has a significant deleterious effect upon mouse embryo cleavage and development to the expanded and hatched blastocyst stage. However, Granot et al. (1998) have presented contradictory results demonstrating that the development of human fertilized eggs exhibiting three pronuclei (3PN) up to 16-cells was not affected by the presence of HF. They suggest that constant passage of HF into the uterine cavity could possibly introduce a mechanical interference that might result in implantation failure.

We propose two-cell mouse embryo culture for 7 days, as a non-invasive embryotoxicity assay that adds extra information that may be helpful when considering possible treatments for individual patients. For example, using a similar system for detection of embryotoxins, Simon et al. (1994) observed significantly decreased embryotoxicity in endometriosis patients after 12 days of glucocorticoid treatment, while Zigril et al. (1991) observed that leukocyte immunization clearly reversed the embryotoxic effect of RSA sera, and Abu-Musa et al. (1992) observed that embryotoxicity in endometriosis patients was lost after treatment of the disease with danazol or buserelin acetate.

Preliminary results obtained in our laboratory suggest the existence of more than one embryotoxic factor. Sera from some patients maintain their embryotoxicity level despite numerous cycles of freeze-thawing, while in others embryotoxicity decreases with a resulting increase in embryo development. These results could be explained by the selective cryoprecipitation of serum substances; Hossain et al. (1997) observed spectrophotometric changes in frozen sera due to a result of freezing-induced changes in specific components.

Our laboratory is currently working on the identification of the possible embryotoxic factor(s) as well as on the development of possible therapeutic strategies to be adopted during the treatment of patients with circulating embryotoxins.

In conclusion, we have shown that: (i) culture to advanced embryonic stages using a long-term culture system increases the sensitivity of the bioassay; (ii) in these conditions, addition of IFN-{gamma} or RSA sera to the culture medium produced similar effects on in-vitro embryo development; (iii) IFN-{gamma} could be one of the embryotoxic factors present in RSA sera.


    Acknowledgments
 
The authors are grateful to Dra Lucrecia Calvo for critical reading of the manuscript, to Clinica y Maternidad Suizo Argentina for their support in the realization of this study. FSH and HCG were generously supplied by Instituto Massone of Buenos Aires. The present work was funded by grants from the CONICET and UBA, Buenos Aires, Argentina.


    Notes
 
5 To whom correspondence should be addressed Back


    References
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
Abu-Musa, A., Takanashi, K. and Kitao, M. (1992) The effect of serum obtained before and after treatment for endometriosis on in vitro development of two-cell mouse embryos. Fertil. Steril., 57, 1098–1102.[ISI][Medline]

American Fertility Society (1985) Revised American Fertility Society classification of endometriosis. Fertil. Steril., 43, 351–352.[Medline]

Beyler, S.A., James, K.P., Fritz, M.A. et al. (1997) Hydrosalpingeal fluid inhibits in-vitro embryonic development in a murine model. Hum. Reprod., 12, 2724–2728.[Abstract]

Chavez, D.J. and McIntyre, J.A. (1984) Sera from women with histories of repeated pregnancy losses cause abnormalities in mouse peri-implantation blastocysts. J. Reprod. Immunol., 6, 273–281.[ISI][Medline]

Cofano, F., Moore, S.K., Tanaka, S. et al. (1990) Affinity purification, peptide analysis, and cDNA sequence of the mouse interferon g receptor. J.Biol. Chem., 265, 4064–4071.[Abstract/Free Full Text]

Croxatto, H.B., Ortiz, M.E., Guiloff, E. et al. (1978) Effect of 15(S)- 15-methyl prostaglandin F2 {alpha} on human oviductal motilty and ovum transport. Fertil. Steril., 30, 408–414.[ISI][Medline]

Damewood, M.D., Hesla, J.S., Schalff, W.D. et al. (1990) Effect of serum from patients with minimal to mild endometriosis on mouse embryo development in vitro. Fertil. Steril., 54, 917–920.[ISI][Medline]

Dokras, A., Sargent, I.L., Redman, C.W.G. et al. (1993) Sera from women with unexplained infertility inhibit both mouse and human embryo growth in vitro. Fertil. Steril., 60, 285–292.[ISI][Medline]

Edmonds, D.K., Lindsay, K.I., Miller, J.F. (1982) Early embryonic mortality in women. Fertil. Steril., 38, 447–453.[ISI][Medline]

Granot, I., Dekel, N., Segal, I. et al. (1998) Is hydrosalpinx fluid cytotoxic? Hum. Reprod., 13, 1620–1624.[Abstract]

Haney, A.F., Mucato, J.J. and Weinberg, J.B. (1981) Peritoneal fluid cell populations in infertility patients. Fertil. Steril., 35, 696–698.[ISI][Medline]

Hill, J.A., Polgar, K., Harlow, B.L. et al. (1992) Evidence of embryo- and trophoblast-toxic cellular immune response(s) in women with recurrent spontaneous abortion. Am. J. Obstet. Gynecol., 166, 1044–1052.[ISI][Medline]

Hossain,A.M., Bari,S., Rizk,B. et al. (1997) Assessment of the quality of frozen serum by spectrophotometric analysis and sperm bioassay. Arch. Andro., 39, 119–125.[ISI]

Oksenberg, J.R. and Brautbar, C. (1986) In vitro suppression of murine blastocyst growth by sera from women with reproductive disorders. Am. J. Reprod. Immunol. Microbiol., 11, 118–124.[Medline]

Rock, J.A., Dubin, N.H., Ghodgaonkar, R.B. et al. (1982) Cul-de-sac fluid in women with endometriosis: fluid volume prostanoid concentration during the proliferative phase of the cycle – days 8 to 12. Fertil. Steril., 37, 747–750.[ISI][Medline]

Roennberg, L., Kauppila, A. and Rajaniemi, H. (1984) Luteinizing hormone receptor disorder in endometriosis. Fertil. Steril., 42, 64–68.[ISI][Medline]

Roussev, R.G., Stern J.J., Thorsell, L.P., et al. (1995) Validation of an embryotoxicity assay. Am. J. Reprod. Immunol., 33, 171- 5.[ISI][Medline]

Simon, C., Frances, A., Piquette, G.N. et al. (1994) Embryonic implantation in mice is blocked by interleukin-1 receptor antagonist (IL-1ra). Endocrinology, 134, 521–528.[Abstract]

Thomason, E.J., Roussev, R.G., Jaroslav, Stern, J. et al. (1995) Prevalence of embryotoxic factor in sera from women with unexplained recurrent abortion. Am. J. Reprod. Immunol., 34, 338–341.[ISI][Medline]

Wheeler, J.M., Johnston, B.M. and Malina, L.R. (1983) The relationship of endometriosis to spontaneous abortion. Fertil. Steril., 39, 656–660.[ISI][Medline]

Zigril, M., Fein, A., Carp, H. et al. (1991) Immunopotentiation reverses the embryotoxic effect of serum from women with pregnancy loss. Fertil. Steril., 56, 653–659.[ISI][Medline]

Submitted on July 28, 1998; accepted on December 17, 1998.