The use of whole rat embryo culture as a technique for investigating potential serum toxicity in recurrent miscarriage patients

Matthew J. Hewitt1,4, Margaret K. Pratten2, Lesley Regan3, Siobhan M. Quenby1 and Philip N. Baker1

1 School of Human Development and 2 School of Biomedical Sciences, University of Nottingham, UK and 3 Imperial College of Science and Technology, UK


    Abstract
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
Previously, the rat embryo model has been used as an experimental technique in investigations of the aetiology of idiopathic recurrent miscarriage. The aim of the present study was to validate it as a tool in the investigation of the aetiology of this condition. Subjects (n = 36) with a history of recurrent miscarriage were recruited from two dedicated recurrent miscarriage clinics and compared with control women with at least one previous pregnancy resulting in a live birth (n = 23). Serum from each woman was used as culture medium in the rat embryo model. Cultured embryos were scored for growth and differentiation. No statistical difference was found in any parameter between the two groups. Furthermore, patients from the recurrent miscarriage group whose serum demonstrated a trend towards lower scores, subsequently conceived and underwent uncomplicated pregnancies.

Key words: rat embryo model/recurrent miscarriage


    Introduction
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
Recurrent miscarriage (RM) is defined as three consecutive pregnancy losses, and is a cause of considerable psychological morbidity (Regan, 1997Go). Evidence that the presence of chronic pathology contributes to the pathogenesis of RM is provided by the fact that 1% of the female population suffer from RM (Albermann, 1988Go). This figure exceeds that expected from repeated sporadic losses. Recognized causes of RM include abnormal maternal or paternal karyotype (De Braekeleer and Dao, 1990Go), uterine malformation, cervical incompetence and autoimmune factors (including circulating maternal antiphospholipid antibodies) (Regan, 1991Go). However, in 50% of women with RM the aetiology is unknown (Drakeley et al., 1998Go).

It has been proposed that there is a circulating factor(s) that causes idiopathic RM. This has been demonstrated using a mouse embryo model (Cameo et al., 1999Go) and a rat embryo model (Ferrari et al., 1991Go; Abir et al., 1994Go). The rat embryo model, which was developed previously (New et al., 1976Go), involves explantation of rodent conceptuses and their subsequent culture in vitro in experimental serum, and is a recognized tool in teratology (Chatot et al., 1984Go). Following culture, the conceptuses can be scored for growth and differentiation. Previous studies have suggested that sera from women with idiopathic RM are more embryopathic than sera from normal women, when examined with this model (Ferrari et al., 1991Go; Abir et al., 1994Go). The effect of sera from women with a past history of RM but who were not pregnant at the time of sampling has been compared with the effect of sera from control, currently pregnant women (Ferrari et al., 1991Go). The validity of this comparison is questionable. Sera from pregnant women are inherently different from sera from non-pregnant women, regardless of the women's reproductive history. Similarly, others (Abir et al., 1994Go) used a second-trimester pregnant control group and women following normal vaginal delivery, and made no attempt to follow the clinical course of the RM cohort. The majority of miscarriages occur in the first trimester, and thus the comparison of a first-trimester miscarriage cohort with women in the second and third trimester may be unfair, as there is evidence that the embryopathicity of the serum changes during pregnancy (Ferrari et al., 1991Go). The model has also been used to investigate the aetiology of women with poor reproductive histories, including those with recurrent miscarriage, but did not indicate any statistical difference between subjects and controls, though only 15 women with recurrent miscarriage were recruited (Scialli et al., 1993Go).

The apparent dichotomy between the results of the studies emphasizes the need to define recurrent miscarriage precisely, to exclude women with identifiable causes of miscarriage, and to select appropriate controls.

The study reported here has the advantage of using a large sample of patients from two dedicated miscarriage clinics, and has used only women who are recurrent miscarriers by the now accepted criteria of three consecutive miscarriages (Regan, 1997Go). These women were carefully screened for confounding factors such as diet, lifestyle and occupation, and were not pregnant at the time of sampling. They were compared with non-pregnant pre-menopausal women of comparable backgrounds, who had a history of viable pregnancies and who were not using hormonal contraception. All experiments were performed using a blind protocol. The aim of the present study therefore was to investigate the use of the rat embryo model as a potential tool in elucidating the cause(s) of idiopathic RM.

The women in the RM group were followed up to determine the outcome of any subsequent pregnancy and attempts were made to correlate this outcome with the results from embryo culture.


    Materials and methods
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
Subjects and controls
Ethical approval was obtained from the local Hospital Ethics Committee and written, informed consent was obtained from all patients recruited into the study, as well as a detailed personal history at the time of phlebotomy.

Women were recruited from two dedicated recurrent miscarriage clinics at the Nottingham City Hospital and St Mary's Hospital, London. Screening included ultrasound examination to diagnose uterine abnormality and/or polycystic ovaries in conjunction with serum concentrations of LH, maternal and paternal karyotyping and autoimmune screening (including anticardiolipin antibodies and lupus anticoagulant). The diagnosis of the antiphospholipid syndrome (RM with demonstrable circulating antiphospholipid antibodies) was only made by two repeated positive assays for lupus anticoagulant and anticardiolipin antibodies at least 6 weeks apart (Rai et al., 1995Go). Women with a demonstrable cause of RM were excluded from the study; this included women who had a history suggestive of cervical incompetence, i.e. second-trimester miscarriage associated with spontaneous rupture of membranes and a short pain-free miscarriage of a live fetus. Any woman from whom aborted products of conception were analysed and revealed an abnormal karyotype was excluded from the study. Any subsequent achievement of a spontaneous viable pregnancy was documented.

Controls were women who had a history of viable pregnancy and no more than one miscarriage followed by a subsequent live birth. Furthermore, women taking hormonal contraception were excluded. The demographics of subjects and controls are detailed in Table IGo.


View this table:
[in this window]
[in a new window]
 
Table I. Patient demographic details
 
Serum collection and preparation
Serum was obtained and processed under sterile conditions from non-pregnant subjects and controls. Following phlebotomy, the serum was immediately centrifuged at 3000 r.p.m. for 10 min. The fibrin clot in the supernatant was squeezed to expel the serum, after which the sample was re-centrifuged for a further 5 min. The supernatant was then extracted and stored at –10°C. Before culturing, the serum was gently defrosted and heat inactivated for 30 min at 56°C.

Rat embryo model
Pregnant Wistar rats were anaesthetized at 9.5 days gestation, and at laparotomy the uterine wall was removed in segments. Using a dissection microscope, the decidual mass was split to expose the conceptus, which was gently teased free and immediately placed in Hanks' balanced salt solution. Reicherts's membrane was then removed and the conceptus inspected closely. Any embryos damaged during this process were discarded. The stage of the embryos was then assessed by determining the number of chambers in the conceptus. Those of a three stage were discarded. It is not always possible to determine the exact stage of the conceptuses, and it is feasible that some conceptuses were of an earlier gestation. To avoid any one serum sample receiving embryos of an earlier gestation, embryos from the three rats culled on each experimental day were mixed and assigned randomly to the three serum samples under investigation on that day.

Culture medium consisted of 1 ml human serum per embryo, 30 µl heat-inactivated rat serum, 4 µl penicillin/streptomycin, and 1 mg of glucose. Four or five embryos were placed in each culture bottle, and the bottles were placed on a roller incubator at 37°C. The culture bottles were gassed with 5% oxygen at commencement, 20% oxygen at 24 h, and 40% oxygen at 44 h of culture (New et al., 1976Go).

At 48 h the embryos were removed from culture and examined by an individual who was blinded to the origin of the culture medium. The yolk sac diameter and crown–rump length were measured. Using the a previously devised system (Van Maele-Fabry et al., 1990Go), the embryos were scored for growth and differentiation. This is a morphological scoring system that is adapted from an earlier method (Brown and Fabro, 1981Go) and which subjectively scores the yolk sac, allantois, embryo turning, neural tube closure, branchial bar number, heart, optic, otic and limb development and somite number. The original scoring system (Van Maele-Fabry et al., 1990Go) was devised for the scoring of mouse embryos; however, there are clear indications within the scheme of how it can be modified to be used with rat embryos. The advantage of this scheme is that it extends certain of the scoring parameters which are crucial in the evaluation of embryonic growth and therefore gives a more accurate picture.

In order to avoid the inclusion of data derived from blighted or damaged concepti unrecognized at the start of culture, the two lowest scoring embryos cultured in each individual's serum were eliminated from the analysis.

The embryos and yolk sacs were immediately frozen and their protein content determined using a standard assay (Lowry et al., 1951Go).

Statistical analysis
Yolk sac diameter, crown–rump length and somite number were not normally distributed, and so were analysed using non-parametric statistical tests. Morphological score was also analysed using non-parametric statistical tests as this parameter was based upon the subjective ranking of features by the investigator and is therefore classed as ordinal data and cannot be assessed using parametric tests. A median embryo score from each patient was determined and a Mann–Whitney U-test performed between the two groups using a `Statview' package for PC.

Validation of the scoring process and reproducibility of the results
To assess inter-observer error, 10 random embryos were chosen and counted by two observers. The correlation between the observers was 0.9 (P = 0.02; Spearman rank correlation). Similarly, to assess the intra-observer error of the scoring process, 10 randomly chosen embryos were set aside and rescored later on the afternoon of the original count. The correlation was found to be 0.88 (P < 0.02). The intra-assay variation of the measured parameters for each sample was <10%. To assess the reproducibility of the model, 10 serum samples were used as culture medium on two different dates and the scores compared; the correlation was 0.8 (P < 0.03).


    Results
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
There was no statistical difference between the recurrent miscarriage and control group in terms of age, body mass index, smoking habit and folate intake. The RM group had a mean of 3.5 miscarriages per subject (range 3–5). The mean parity of the control group was 1.8 (range 1–4) (Table IGo). The parity of the six women with secondary recurrent miscarriage was 1, 1, 5, 2, 1, and 1.

No statistical difference was detected in morphological score, crown–rump length, yolk sac diameter and somite number in rat embryos grown in serum from recurrent miscarriage group and serum from the control group (Table IIGo). In both cases there were ~70% normal embryos in total, without neural tube or other major anomalies (RM 69%, control 71%). There was no statistical difference between the subjects recruited from the two centres, when any of the measured parameters of the cultured embryos were compared.


View this table:
[in this window]
[in a new window]
 
Table II. Growth and differentiation of embryos cultured in serum from recurrent miscarriage (RM) and control subjects
 
A small percentage of rat serum was included in the culture medium to improve its growth-promoting properties (Gulamhusein et al., 1990Go). The inclusion of 3% rat serum in the culture medium eliminated the anaemia frequently observed with human serum culture, but failed to improve the abnormality rate to that of the control embryos cultured in rat serum alone (Table IIGo).

Dividing the recurrent miscarriage cohort data into primary (those patients with no viable pregnancies) and secondary (those patients with one or more viable pregnancies) did not demonstrate a statistical difference between any of the measured parameters (Table IIIGo).


View this table:
[in this window]
[in a new window]
 
Table III. Growth and differentiation of embryos cultured in serum from patients with primary recurrent miscarriage (RM) (no viable pregnancies) and secondary recurrent miscarriage (one or more viable pregnancies)
 
Subdivision of the recurrent miscarriage cohort into groups depending on the number of consecutive miscarriages revealed no statistical difference between any of the measured parameters of the cultured embryos and the number of consecutive miscarriages (Table IVGo).


View this table:
[in this window]
[in a new window]
 
Table IV. Growth and differentiation of embryos cultured in serum from patients with recurrent miscarriage(s)
 
Women in the recurrent miscarriage cohort who subsequently became pregnant spontaneously and delivered viable fetuses, demonstrated no statistical differences in any of the measured parameters of the cultured embryos compared with those women who had not achieved a viable pregnancy in the study period (Table VGo).


View this table:
[in this window]
[in a new window]
 
Table V. Growth and differentiation of embryos cultured in serum from patients with a history of recurrent miscarriage; patients who subsequently became pregnant spontaneously and delivered viable fetuses within the study period
 
There was no statistical difference in the cultured embryo parameters of the control cohort who were supplementing their diet with folate at the time of sampling compared with those women who were not (Table VIGo).


View this table:
[in this window]
[in a new window]
 
Table VI. Growth and differentiation of embryos cultured in serum from patients with a history of recurrent miscarriage; comparison of patients with or without dietary folate supplement
 

    Discussion
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
No difference was detected in the morphological score, yolk sac diameter, crown–rump length and somite number of embryos cultured in serum from women with idiopathic RM as compared to those cultured in serum from normal controls (Table IIGo). There may be several reasons for these findings.

Firstly, 50% of miscarriages are sporadic and occur due to an isolated chromosomal abnormality (Daniely et al., 1998Go) in the aborted fetus or embryo. These are chance occurrences, but they can occur on multiple occasions. Any women with a history of RM may have experienced such a series of sporadic losses, and this is a limiting factor in any study of a cohort of women with RM. The inclusion of such patients in the RM group may account for why we were unable to demonstrate an embryopathic factor in the sera of the RM cohort. Women who passed products of conception with abnormal karyotypes were excluded from the study. However, on occasion the karyotype cannot be determined and karyotypes from the majority of first- and second-trimester miscarriages were not available. This study specifically incorporated a large number of subjects to reduce the chance of selecting women with serial sporadic miscarriages.

The importance of subdividing RM patients into primary (those without a previous viable pregnancy) and secondary (those with a previous viable pregnancy) has been highlighted by some authors (McIntyre et al., 1989Go). Such a sub-classification did not reveal a difference in the embryopathic effect of the serum between these two groups (Table IIIGo).

It has been suggested previously that the activity of the putative circulating factor(s) implicated in RM may be directly proportional to the number of miscarriages the subject has experienced, as it is statistically more likely that an increasing number of miscarriages is associated with a chronic maternal pathology (Ferrari et al., 1991Go). In contrast, we have been unable to demonstrate a correlation between low morphological scores and increasing numbers of miscarriage (Table IVGo), though a limiting factor in our study may be the small numbers of subjects recruited with a history of more than three miscarriages.

We have demonstrated a range of morphological scores among the control group with a 30% neural tube defect rate. This has also been highlighted by previous studies, which have reported rates of embryopathicity of between 27% and 40% (Ferrari et al., 1991Go; Scialli et al., 1993Go; Abir et al., 1994Go). However, in these studies the culture medium did not contain rat serum, and the absence of this factor has previously been shown to result in poorer embryo growth compared with those embryos cultured with additional rat serum (Anwar et al., 1989Go). It was felt to be inappropriate to increase the percentage of rat serum because of the danger of masking any embryopathic effect of the human serum. The growth and differentiation of embryos cultured in human serum was similar to that of embryos grown in rat serum only. The embryos grown in 100% rat serum acted as internal quality and audit controls of the explantation and culture process.

If lower morphological scores (representing embryopathicity) are associated with miscarriage, then it may be expected that those women who culture poor embryos would be less likely to carry a viable fetus in a subsequent pregnancy, compared with women who culture well-formed embryos. This was found not to be the case (Table VGo). The results are consistent with the finding that women with idiopathic RM have subsequent high rates of viable pregnancies (Clifford et al., 1994Go), and this reinforces the need to incorporate a control group in any study involving RM.

There was a relatively small number of controls supplementing their diet with folate at the time of sampling compared with those in the RM cohort. Folate deficiency is associated with neural tube defects (MRC Vitamin Study Research Group, 1991Go). It has previously been demonstrated, using the rat embryo model, that women who have experienced a pregnancy complicated by this abnormality have serum that is more embryopathic than that of controls (Anwar et al., 1989Go). Although RM subjects are not thought to have a reduced serum concentration of folate, differences in folate concentrations may not be demonstrable in assays (Sutterlin et al., 1997Go). It is therefore possible that the lower number of dietary folate supplementers in the control group masked an embryopathic effect in the recurrent miscarriage cohort. However, subdivision of the control cohort into supplementers and non-supplementers revealed no significant difference in measured parameters of the cultured embryos (Table VIGo).

Failure to demonstrate an embryopathic factor may be due the processing of the serum. Heat inactivation is a recognized and necessary procedure for serum to be used in rat embryo culture. Inevitably there is some risk that any factor present at the time of collection may be modified or lost during storage and/or heat inactivation. These processes are used routinely in the vast majority of laboratories undertaking rat embryo culture. Since one of the purposes of this study was to evaluate the validity of the method and relate this to previous studies, it seems appropriate to retain the standard protocol.

In summary, we have failed to demonstrate a role for the rat embryo model in the investigation of the cause of idiopathic RM. Although RM was precisely defined and an appropriate non-pregnant control group was included, no embryopathic effect was demonstrated. It is possible that the putative circulating factor(s) implicated in the aetiology of this condition may only be present in the serum at the time of pregnancy and miscarriage. To this end, future work will be directed towards investigating a cohort of women with a history of RM at the time of a miscarriage and matched with a control group of normal pregnant women.


    Acknowledgments
 
We are very grateful to Nicky Lindley and Tracey McGrath for their help in patient recruitment and sample collection. We would like to thank all the patients and controls who volunteered and donated blood to the project.


    Notes
 
4 To whom correspondence should be addressed at: School of Human Development, Division of Obstetrics and Gynaecology,City Hospital, Hucknall Road, Nottingham NG5 1PB, UK. E-mail: matthew.hewitt{at}nottingham.ac.uk Back


    References
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
Abir, R., Ornoy, A., Ben Hur, H. et al. (1994) The effects of sera from women with spontaneous abortions on the in vitro development of early somite stage rat embryos. Am. J. Reprod. Immunol., 32, 73–81.[ISI][Medline]

Albermann, E. (1988) The epidemiology of repeated abortion. In Beard, R.W. and Sharp, F. (eds), Early Pregnancy Loss: Mechanisms and Treatment, R.C.O.G., London, pp. 9–17.

Anwar, M., MacVicar, J. and Beck, F. (1989) Serum from pregnant women carrying a fetus with neural tube defect is teratogenic for rat embryos in culture. Br. J. Obstet. Gynaecol., 96, 33–37.[ISI][Medline]

Brown, N.A. and Fabro, S. (1981) Quantification of rat embryonic development in-vitro: a morphological scoring system. Teratology, 24, 65–78.[ISI][Medline]

Cameo, M., Fontan, V., Cameo, P. et al. (1999) Similar embryotoxic effects of sera from infertile patients and exogenous interferon-{gamma} on long-term in-vitro development of mouse embryos. Hum. Reprod., 14, 959–963.[Abstract/Free Full Text]

Chatot, C.L., Klein, N.W., Clapper, M.L. et al. (1984) Human serum teratogenicity studied by rat embryo culture: epilepsy, anticonvulsant drugs, and nutrition. Epilepsia, 25, 205–216.[ISI][Medline]

Clifford, K., Rai, R., Watson, H. and Regan, L. (1994) An informative protocol for the investigation of recurrent miscarriage: preliminary experience of 500 consecutive cases (see comments). Hum. Reprod., 9, 1328–1332.[Abstract]

Daniely, M., Aviram-Goldring, A., Barkai, G. and Goldman, B. (1998) Detection of chromosomal aberration in fetuses arising from recurrent spontaneous abortion by comparative genomic hybridization. Hum. Reprod., 13, 805–809.[Abstract]

De Braekeleer, M. and Dao, T.N. (1990) Cytogenetic studies in couples experiencing repeated pregnancy losses. Hum. Reprod., 5, 519–528.[Abstract]

Drakeley, A.J., Quenby, S. and Farquharson, R.G. (1998) Mid-trimester loss – appraisal of a screening protocol. Hum. Reprod., 13, 1975–1980.[Abstract]

Ferrari, T.J., Gilles, P.A. and Klein, N.W. (1991) Serum teratogenicity to cultured rat embryos in the serum of women with histories of spontaneous abortion. Teratology, 43, 460–468.

Gulamhusein, A.P., Pratten, M.K., Williams, C.A. and Beck, F. (1990) The effect of macromolecular rat serum fractions on conceptuses cultured in human serum: role of transferrin. J. Anat., 168, 113–121.[ISI][Medline]

Lowry, D.H., Rosebrough, N.J., Farr, A.L. and Randal, R.J. (1951) Protein measurements with Folin phenol reagents. J. Biol. Chem., 193, 265–275.[Free Full Text]

McIntyre, J.A., Coulam, C.B. and Faulk, W.P. (1989) Recurrent spontaneous abortion. Am. J. Reprod. Immunol., 21, 100–104.[ISI][Medline]

MRC Vitamin Study Research Group (1991) Prevention of neural tube defects: results of the Medical Research Council vitamin study. Lancet, 338, 131–137.[ISI][Medline]

New, D.A., Coppola, P.T. and Cockroft, D.L. (1976) Comparison of growth in vitro and in vivo of post-implantation rat embryos. J. Embryol. Exp. Morphol., 36, 133–144.[ISI][Medline]

Rai, R.S., Clifford, K., Cohen, H. and Regan, L. (1995) High prospective fetal loss rate in untreated pregnancies of women with recurrent miscarriage and antiphospholipid antibodies. Hum. Reprod., 10, 3301–3304.[Abstract]

Regan, L. (1991) Recurrent miscarriage (editorial) (see comments). Br. Med. J., 302, 543–544.[ISI][Medline]

Regan, L. (1997) Sporadic and recurrent miscarriage. In Grudzinskas, J.G. and O'Brien, P.M.S. (eds), Problems in Early Pregnancy, Advances in Diagnosis and Management. R.C.O.G., London, pp. 31–52.

Scialli, A.R., Flynn, T.J. and Gibson, R.R. (1993) Rat embryo culture to detect nutritional deficiency in women with poor reproductive histories. Reprod. Toxicol., 7, 581–587.[ISI][Medline]

Sutterlin, M., Bussen, S., Ruppert, D. and Steck, T. (1997) Serum levels of folate and cobalamin in women with recurrent spontaneous abortion. Hum. Reprod., 12, 2292–2296.[Abstract]

Van Maele-Fabry, G., Delhaise, F. and Picard, J.J. (1990) Morphogenesis and quantification of the development of postimplantation mouse embryos. Toxicol. In Vitro, 4, 149–156.[ISI]

Submitted on January 6, 2000; accepted on June 15, 2000.