Successful uterine transplantation in the mouse: pregnancy and post-natal development of offspring

Randa Racho El-Akouri1,3, Göran Kurlberg2 and Mats Brännström1

1 Department of Obstetrics and Gynecology and 2 Department of Surgery, The Sahlgrenska Academy at Göteborg University, Göteborg, Sweden

3 To whom correspondence should be addressed at: Department of Obstetrics and Gynecology, Sahlgrenska University Hospital, S-413 45 Göteborg, Sweden. e-mail: randa.racho{at}obgyn.gu.se


    Abstract
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
BACKGROUND: Uterine transplantation could serve as a tool in studies of the physiology of implantation/pregnancy, and is also a possible future treatment for patients with absolute uterine infertility. Here, the first live-born offspring in any uterine transplantation model is reported. METHODS: A syngeneic mouse model with a uterus transplanted, by end-to-side aorta/vena cava vascular anastomoses, alongside the native uterus was used. The cervix was attached to a cutaneous stoma. Pregnancy rate and offspring (birth weight, growth and fertility) was evaluated after blastocyst transfer to the native and the grafted uterus of transplanted mice and to controls. RESULTS: Pregnancy rates were comparable in the grafted uterus (8/12 animals became pregnant) and the native uterus (9/12 pregnant) of transplanted animals and controls (8/13 pregnant). In a separate set of animals, the native uterus was removed at transplantation to exclude influences from the native uterus on the pregnancy potential of the graft; two of four animals became pregnant after blastocyst transfer. The weights/lengths of fetuses (gestational day 18) and gestational lengths were similar in all groups. Offspring were delivered and the growth trajectories (up to 8 weeks) of offspring delivered from grafted or native uteri of transplanted mice were similar as compared with controls, and all were fertile. The second-generation offspring from transplanted animals were all fertile with normal birth weights. CONCLUSIONS: These observations document the capacity of a transplanted uterus to harbour pregnancies to term, and reveal that offspring from a transplanted uterus develop to normal fertile adults.

Key words: embryo transfer/infertility/mouse/transplantation/uterus


    Introduction
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
Tremendous advances in the field of assisted reproductive techniques have been made during the past two decades. It is now possible, with gonadotrophin stimulation and/or IVF/ICSI procedures to treat successfully the majority of the approximately 15% (Gray, 1990Go) of reproductive-aged couples in the western world that are infertile. Uterine infertility, due to congenital hypoplasia/agenesis or hysterectomy is untreatable. However, minor uterine malformations such as septate uterus is treatable by hysteroscopic surgery in about 75% of patients (Grimbizis et al., 2001Go). The only available option to become a genetic mother for the women with absolute uterine infertility is through use of a surrogate gestational carrier, a procedure that is not approved in most countries due to legal or ethical reasons. Transplantation of a uterus, followed by an IVF procedure, to a woman with absolute uterine infertility and preserved ovaries, could provide a treatment. Moreover, a uterine transplantation model would allow studies on the effects of certain genes on implantation/pregnancy in experiments involving transplantation of a uterus from a gene-deleted mouse to a wild-type or a uterus from a wild-type mouse to a gene-deleted mouse.

Previously, efforts have been made to develop methods for uterine transplantation in animals. These experiments mostly involved methods where the uterus was explanted and then returned (autografted) to the same animal. Using this approach to test for revascularization, the survival of an autografted uterus was demonstrated in ewes (Baird et al., 1976Go), dogs (Mattingly et al., 1970Go) and primates (Scott et al., 1971Go). The first—and so far the only—attempt to perform a human uterine transplantation from one women to another woman was recently reported (Fageeh et al., 2002Go). The transplantation was partly successful since, under immunosuppression, the uterus survived for several weeks, with endometrial response to hormone replacement therapy before the uterus had to be surgically removed due to massive vascular thrombosis. It should be emphasized that further trials in humans should not be performed until results from basic studies in animal species are available. Thus, it is important to demonstrate that implantation and pregnancy can occur in a transplanted uterus, and also to study the offspring from a transplanted uterus. The methodology for heterotopic uterine transplantation in the mouse with proven capacity to implant embryos was recently reported (Racho El-Akouri et al., 2002Go). The latter model demonstrated a poor implantation rate due to insufficient drainage of fluids from the transplanted uterus. In the present study, the technique was modified to include a cervical-cutaneous stoma instead of the cervix ending intra-abdominally. Herein, the live-born offspring from a transplanted uterus and their post-natal development is reported for the first time.


    Materials and methods
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
Animals
F1-hybrids of inbred female C57BL/6xCBA/ca (B6CBAF1) mice aged 6–8 weeks were used as organ donors and organ recipients. In the experiments to obtain pseudopregnancy and blastocysts, 8- to 10-week-old female and 10- to 14-month-old normal and vasectomized male B6CBAF1 mice were used. Female B6CBAF1 mice aged 8 weeks were used as lactational mothers. All animals were obtained from M&B A/S (Ry, Denmark). The experiments were approved by the local animal ethics committee, and carried out according to the principles and procedures outlined in the NIH Guide for Use of Laboratory Animals.

Surgery on the donor
The graft procurement procedure was to isolate the right horn of the uterus, including the cervix, with its connecting vasculature. The animal was anaesthetized with 2% isoflurane (Baxter, Kista, Sweden), placed in a supine position, and a mid-line laparotomy performed. During most of the surgical procedure, the surgical field was observed through a stereo microscope (Leica M651; Leica Microsystems, Sollentuna, Sweden) using magnifications from x6 to x40. The surgical method has been described previously in detail (Racho El-Akouri et al., 2002Go) and is outlined briefly below.

The small intestine was retracted to the left side of the animal and the colon transected above the rectum after cauterization and division of the inferior mesenteric artery and ureters. The inferior epigastric vessels were then cauterized and cut just above the ilio-inguinal ligament to fully expose the external iliac vessels, which were ligated and severed. The pudendal vessels and vessels supplying the caudal portion of the cervix and bladder were cauterized; this enabled removal of the bladder and visualization of the uterine artery.

The right horn of the uterus was dissected free from the underlying tissue, and the left uterine horn was ligated and excised at a site just cranial to its branching from the common uterine cavity. The cervix was transected just caudal to its attachments to the bladder, so that the cervical part of the uterus could be lifted to enable identification of the dorsally positioned uterine vessels. A titanium clip was attached onto the superior gluteal vessels and a suture applied around the caudal artery. A ligature was then placed en bloc around the oviduct and the blood vessels branching from the ovarian artery/vein at the tip of the right uterine horn, which was dissected free from the dorsal peritoneum. The left common iliac vessels were then doubly ligated and severed, while the lumbar vessels, branching from the aorta and vena cava, were cauterized and cut. The aorta and the vena cava were tied off caudal to the branching of ovarian vessels. The aorta was then cannulated and the specimen gently flushed with ice-cold NaCl (0.154 mol/l) supplemented with heparin sulphate (100 IU/ml; Leo Pharma AB, Malmö, Sweden) and xylocaine (0.2 mg/ml; Astra Zeneca, Göteborg, Sweden) until transparent fluid flowed through an incision in the vena cava. The uterus was then removed and the mouse killed by exsanguination and cervical dislocation. This donor uterus was then kept in ice-cold NaCl (0.154 mol/l) for ~20 min, when surgical preparation of the recipient was performed and when also the distal parts of the aorta and vena cava of the specimen were cleared from extravascular tissue to prepare for vascular anastomosis.

Surgery on the recipient
The surgical procedure of the recipient was performed to enable the uterus to be placed in a heterotopic position ventral to and somewhat cranial to the native uterus. Essentially the same surgical procedure has been described in detail (Racho El-Akouri et al., 2002Go) and is outlined below. However, one major modification was included as the cervix of the transplant was connected to a cutaneous stoma in the present study (Figure 1), in contrast to placement intra-abdominally in the previous report (Racho El-Akouri et al., 2002Go). Under anaesthesia, heparin sulphate (15 IU) was given s.c. and a midline laparotomy was performed. To enable mobilization of the aorta and the vena cava, the lumbar vessels were cauterized and cut. Haemostatic clamps were placed around the aorta and the vena cava at sites cranial to the branching of the inferior mesenteric vessels and caudal to the branching of ovarian vessels. The graft aorta and vena cava were attached to the aorta and vena cava of the recipient by end-to-side-anastomosis using continuous 11/0 gauge nylon sutures. The cervix of the graft was exteriorized as a stoma on the right side of the abdominal wall (Figure 1), and the midline incision of abdominal wall was closed in two layers.



View larger version (114K):
[in this window]
[in a new window]
 
Figure 1. Graft uterus at 2 weeks after transplantation.

 
In four animals, the native uterus was surgically removed. This was accomplished by placing double ligatures around the oviduct and branching ovarian vessels just cranial to the tip of each uterine horn and division in between. The uterus was then removed after a ligature had been placed en bloc around the cervix and the uterine vessels, with great caution being taken not to interfere with the ureters and inferior vesical vessels. The vascular anastomoses and fixation of cervix was then carried out as described above.

Finally, 1 ml NaCl (0.154 mol/l) and 1 ml glucose (50 mg/ml) were given s.c. to all transplanted animals. Animals were placed under a heat lamp until fully awake and given heparin sulphate (5 IU) s.c. at 4 h and 24 h after surgery.

Embryo transfer
Sexually mature (10-week-old) vasectomized B6CBAF1 males were used. To verify that they were sterile, the males were initially placed in a cage with 4-week-old B6CBAF1 female mice that had been superovulated by an i.p. injection of 5 IU pregnant mare’s serum gonadotrophin (Sigma-Aldrich Co., St Louis, MO, USA) followed 46 h later by an i.p. injection of 5 IU hCG (Sigma-Aldrich Co.). The females were examined for vaginal copulation plugs the next morning, and 46 h after hCG treatment the oviducts were flushed with Gamete-100 medium (Vitrolife AB, Göteborg, Sweden) using a 30-gauge needle. In all cases the unfertilized stage of the oocytes was confirmed at examination under a dissection microscope.

The female mice with transplanted uteri received embryos at 1–3 weeks after transplantation. A control group of normal, non-transplanted 8- to 10-week-old B6CBAF1 females were also used as embryo recipients. All these mice were caged with vasectomized B6CBAF1 males, and mating was confirmed by the presence of a vaginal plug (day of plug = day 1).

As blastocyst donors, 8- to 10-week-old B6CBAF1 females were used 4 days after mating with fertile B6CBAF1 males. The mated females were killed by cervical dislocation at 12:00 on day 4 and the uteri were flushed with Gamete-100 medium to obtain blastocysts. A laparotomy (under isoflurane anaesthesia) was then performed on the transplanted females at 12:00 on day 3–4 of pseudopregnancy. The blastocysts were held in Gamete-100 medium for 20–30 min at 37°C and under 5% CO2. Viability of the graft (normal size, colour and clear arterial pulsations in the uterine artery) was judged by visual inspection. Blastocysts were placed inside the uterine cavities by transmyometrial approach as previously described (Hogan et al., 1994Go). Three to six blastocysts were transferred to the lumen of each of the grafted and native uteri using a 30-gauge needle initially to penetrate the uterine wall; this was followed by the transfer of blastocysts through a glass transfer pipette. The abdominal wall was closed in two separate layers.

As control animals, non-transplanted females also received blastocysts on day 3–4 of pseudopregnancy. Two dorsolateral incisions (10 mm) were made on both sides of the midline. The ovary, oviduct and proximal end of the uterus were externalized on each side, and the blastocysts were transferred to each uterus horn using the technique described above. The incisions were closed using single-layer sutures. All mice were given 40 mg/kg cefuroxime (Glaxo Wellcome, Mölndal, Sweden) i.m. after blastocyst transfer and placed under heat until fully awake.

Assessment of implantation rate and pregnancy
In the first set of experiments, pregnancies were followed up to day 18 of pregnancy. The mice were then anaesthetized and laparotomized (see above) and the number of viable fetuses (visible heartbeat and normal size; length 19–23 mm) was recorded. Fetuses and placentae were carefully dissected out and immediately weighed and measured. The number of resorbing implantation sites was also recorded.

Birth and post-natal follow-up
In the second set of experiments, transplanted mice were taken to term to allow assessment of post-natal development of the offspring. Post-natal growth trajectory (body weight) of the offspring for three groups was followed; offspring from no-transplanted females (control group); offspring from native uterus of transplanted animals; and offspring from grafted uterus of transplanted animals. The transplanted mice were placed in individual cages from day 18 of pregnancy and then examined every 6–12 h until spontaneous delivery occurred. The pups of the control group were weighed within 24 h of spontaneous parturition. Immediately after spontaneous delivery from the native uterus of a transplanted mouse, the animal was anaesthetized and underwent laparotomy with removal of the entire graft uterus containing the fetuses. This uterus was then cut open and the pups retrieved. All pups of a transplanted mother were weighed, marked by ear incision and then placed together with a foster mother. The laparotomy incision was then closed, and the mouse placed under a heat lamp. When the mouse was fully awake it was placed in the same cage as its pups and foster mother. The foster mother was kept in the cage for the first 3 days to allow the transplanted mice to recover fully after surgery. The body weights of progeny from the three groups (normal, native and graft uterus) were determined at pregnancy day 19 (day of delivery) and then once every week until 8 weeks of age. The litters were weaned on day 21 and then kept in separate cages. The fertility of these mice was examined by placing them separately with fertile mice at 8 weeks of age, and following-up for up to 4 weeks to examine pregnancy and delivery of offspring.

Statistical analysis
For comparison of pregnancy rates in control animals and in the native uterus of the transplanted animals a logit-model, where animals were nested within group, was used. Mantel-Haenzel’s test was used to compare the pregnancy rate of the native and transplanted uterus within the transplanted animal. Data relating to weight of fetuses, offspring and placentae were presented as individual data points.


    Results
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
Implantation and pregnancy
Out of 19 transplanted uterus recipients, 12 had viable grafts and were used as embryo recipients. After blastocyst transfer, pregnancies occurred in eight of 13 control animals and in nine of 12 native uteri of transplanted mice (P = 0.79; Figure 2). Eight of the nine females with fetuses in the native uterus, also had fetuses within the transplanted uterus (P = 0.54; Figure 2). Resorbing fetuses were present in two of 12 graft uteri, in one of 12 native uteri, and in four of 13 control uteri (Figure 2); there were no significant differences between the groups.



View larger version (34K):
[in this window]
[in a new window]
 
Figure 2. Number of blastocysts transferred and pregnancies in uterus of control animals (A–M) and uteri of transplanted animals (1–12). Open bars: total bar length = transferred blastocysts; hatched bars = absorbed fetuses; solid bars = fetuses.

 
Weights of fetuses and placentas
The lengths and weights (Figure 3) of the fetuses (day 18), as well as weights of the placentae (Figure 4), were similar in pregnancies in native uterus, grafted uterus and uterus of control animals.



View larger version (16K):
[in this window]
[in a new window]
 
Figure 3. Lengths and weights of day 18 fetuses from control and transplanted animals. Four of the females with grafts had fetuses collected from both the native and grafted uterus. Individual values are shown.

 


View larger version (15K):
[in this window]
[in a new window]
 
Figure 4. Weights of each placenta from control and grafted animals. Two of the females with grafts had placentae collected from both the native and grafted uterus. Placentae from control animals 1–2 and transplanted animals (native uterus) 1–4 were not collected and weighed. Individual values are shown.

 
Development and fertility of offspring from transplanted uteri
The progeny from transplanted mice and controls were compared in order to investigate post-natal growth trajectory. Pups delivered spontaneously (controls, native uterus of transplanted mice) or by Caesarean section (graft uterus) on day 19–20 of pregnancy were followed from the time of birth up to 8 weeks of age. The three groups of animals followed a similar growth curve (Figure 5). The fertility of offspring from native and transplanted uteri was tested at 8 weeks of age. All tested female and male offspring proved to be fertile and their pups had birth weights within the normal range (Figure 6).



View larger version (19K):
[in this window]
[in a new window]
 
Figure 5. Growth trajectory of offspring from graft (filled triangle), native (filled square) and normal controls (filled circle). Medians are indicated by large symbols; ranges are indicated by small symbols. Inserted box-and-whisker plots [medians, 25–75% (boxes) and ranges (whiskers) show weights at birth and 8 weeks].

 


View larger version (20K):
[in this window]
[in a new window]
 
Figure 6. Birth weights of second-generation offspring. The progeny were derived from males and females from grafted (G) and native (N) uteri of transplanted mice and from control (C) (number of pups in each group 23–94). Female-C = female is normal control; Male-C = male is normal control; Female-N uterus = female born from native uterus of transplanted animal; Male-N uterus = male born from native uterus of transplanted animal; Female-G uterus = female born from grafted uterus of transplanted animal; Male-G uterus = male born from grafted uterus of transplanted animal.

 
Spontaneous delivery through stoma
In one heterotopically transplanted mouse, spontaneous delivery occurred from the native uterus and through the stoma from the grafted uterus at a time between the 6- to 12-hourly observations and before planned Caesarean section. Five normal pups were delivered spontaneously (three blastocysts were transferred to native uterus, four blastocysts to transplanted uterus), so at least two of the pups were delivered through the stoma. Two of the four females from which the native uterus had been removed became pregnant. Five and three blastocysts respectively were transferred to each of the grafted uteri in these two females and 20 days later, seven pups (4/5 in one female and 3/3 in the other female) were delivered spontaneously through the stoma (Figure 7).



View larger version (171K):
[in this window]
[in a new window]
 
Figure 7. A spontaneous delivery from the transplanted uterus through the stoma. The tail of the fetus is visible.

 

    Discussion
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
Transplantation of the uterus in humans may become a feasible treatment for female infertility, for women with a uterus that does not have the capacity to implant an embryo and/or to carry a pregnancy, as well as for women with an absence of the uterus due to congenital malformation or following hysterectomy. The present study reports the first series of live-born pups in a transplanted uterus in any animal species, where a uterus has been transplanted from one animal to another. Previous reports in this area have been on live offspring in a uterus which had been removed from an animal and then repositioned into the same animal (Eraslan et al., 1966Go; Scott et al., 1971Go; Barzilai et al., 1973Go).

The main results of the present study were that the pregnancy and implantation rates were not different in the transplanted uterus as compared with the controls (native uterus of transplanted animal and control animal). Furthermore, the weights of the fetuses and placentae, as well as the post-natal development of animals from transplanted (native and grafted uterus) and control animals, were comparable.

This is the first study in which the ability of a transplanted uterus to implant inserted blastocysts and to carry a pregnancy has been evaluated. In a previous report (Racho El-Akouri et al., 2002Go), the blood flow and histology of uteri that were heterotopically transplanted in mice but with the cervix placed intra-abdominally were evaluated. In follow-up experiments involving blastocyst transfer to these mice, extremely low implantation and pregnancy rates were seen in the transplanted uterus (present authors, unpublished data). The reason for this low pregnancy and implantation rate is not clear, but it may well have been related to inadequate drainage of mucus and fluids that were secreted from the endometrium and endocervix of the grafted uterus. Thus, these uteri were swollen with great accumulation of fluids in the lumen. It is likely that the endometrium would be affected by this increased intraluminal pressure and that the implantation capacity would then be reduced. In order to overcome these problems, the surgical method of transplantation was modified in the present study to create a cervical-cutaneous stoma to allow drainage extra-abdominally. The findings of the present study show that the pregnancy rate in these uteri is comparable with that in the native uterus. It could be anticipated that the rate of ascending infections would be high in these transplanted uteri, where the cervix is in direct contact with the skin and not separated from the skin by the vagina. However, the normal implantation and pregnancy rates in the transplanted uteri with cervical-cutaneous stoma shows that the cervix by itself is an important barrier with antimicrobial activity, as previously pointed out (Hein et al., 2002Go).

In the present study, the choice was made to transfer blastocysts at day 4, based on findings in a previous report which showed that embryos transferred at day 3–4 result in the highest implantation rate (Wakuda et al., 1999Go). The pregnancy rate in the latter study was within the same range as that in the uteri of transplanted mice in the present study, further indicating that these transplanted uteri are functionally normally.

The placental and fetal weights at late gestation were also evaluated in order to examine the intrauterine development of implanted embryos in a transplanted uterus. In this regard, the mouse would be a suitable experimental model for extrapolation of results to the human situation, as major similarities exist between the mouse and primates with regard to placentation (Pijnenborg et al., 1981Go). In both species, the terminal blood space is lined with trophoblast cells rather than endothelial cells. The results of the present study, where a similar placenta weight in transplanted and control uteri was demonstrated, suggest that the development of the placenta is not impaired in the transplanted uterus.

The transplanted uterus was anastomosed to the abdominal aorta, and it may be possible that blood flow through the uterine arteries of both the transplanted and native uterus would be suboptimal during gestation, when uterine blood flow is markedly increased (Bernstein et al., 2002Go). In a previous study, it was shown that in the non-pregnant transplanted mouse the blood flow is similar in the transplanted and native uterus (Racho El-Akouri et al., 2002Go). Earlier studies have shown that reduced blood flow to the uterus during gestation results in lower placental weights and intrauterine growth retardation (IUGR) (North et al., 1994Go; Boyle et al., 1996Go). There were no signs of IUGR of fetuses in either native or grafted uteri of transplanted mice in the present study. It is of major importance to rule out that the fetuses of a transplanted mouse have a lower birth weight, since it is known from epidemiological data in humans that this is associated with an increased risk of adult diseases such as hypertension and non-insulin-dependent diabetes (Godfrey and Barker, 2001Go).

It is proposed that the present mouse model of uterine transplantation is a useful tool with which to acquire essential knowledge with regard to uterine transplantation before further attempts are made in humans. The potential difficulties in translating data obtained from this mouse model to the human are differences in gestational lengths, immunological mechanisms for rejection and anatomical differences. Furthermore, this mouse model may be beneficial for basic studies on the physiology of implantation and pregnancy.


    Acknowledgements
 
These studies were supported by grants from the Swedish Research Council (11607), the Medical Faculty of Sahlgrenska Academy and Hjalmar Svensson’s Research Foundation.


    References
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
Baird, D.T., Land, R.B., Scaramuzzi, R.J. and Wheeler, A.G. (1976) Functional assessment of the autotransplanted uterus and ovary in the ewe. Proc. R. Soc. Lond. B Biol. Sci., 1109, 463–474.

Barzilai, A., Paldi, E., Gal, D. and Hampel, N. (1973) Autotransplantation of the uterus and ovaries in dogs. Isr. J. Med. Sci., 1, 49–52.

Bernstein, I.M., Ziegler, W.F., Leavitt, T. and Badger, G.J. (2002) Uterine artery hemodynamic adaptations through the menstrual cycle into early pregnancy. Obstet. Gynecol., 4, 620–624.[CrossRef]

Boyle, D.W., Lecklitner, S. and Liechty, E.A. (1996) Effect of prolonged uterine blood flow reduction on fetal growth in sheep. Am. J. Physiol., 1 (Pt 2), R246–R253.

Eraslan, S., Hamernik, R.J. and Hardy, J.D. (1966) Replantation of uterus and ovaries in dogs, with successful pregnancy. Arch. Surg., 1, 9–12.

Fageeh, W., Raffa, H., Jabbad, H. and Marzouki, A. (2002) Transplantation of the human uterus. Int. J. Gynaecol. Obstet., 3, 245–251.

Godfrey, K.M. and Barker, D.J. (2001) Fetal programming and adult health. Public Health Nutr., 2B, 611–624.

Gray, R.H. (1990) Epidemiology of infertility. Curr. Opin. Obstet. Gynecol., 2, 154–158.[Medline]

Grimbizis, G.F., Camus, M., Tarlatzis, B.C., Bontis, J.N. and Devroey, P. (2001) Clinical implications of uterine malformations and hysteroscopic treatment results. Hum. Reprod. Update., 2, 161–174.

Hein, M., Valore, E.V., Helmig, R.B., Uldbjerg, N. and Ganz, T. (2002) Antimicrobial factors in the cervical mucus plug. Am. J. Obstet. Gynecol., 1, 137–144.[CrossRef]

Hogan, B., Beddington, R. and Constantini, F. (1994) Manipulating the mouse embryo. Cold Spring Harbor Laboratory Press, USA.

Mattingly, R.F., Clark, D.O., Lutsky, I.I., Huang, W.Y., Stafl, A. and Maddison, F.E. (1970) Ovarian function in-utero ovarian homotransplantation. Am. J. Obstet. Gynecol., 5, 773–794.

North, R.A., Ferrier, C., Long, D., Townend, K. and Kincaid-Smith, P. (1994) Uterine artery Doppler flow velocity waveforms in the second trimester for the prediction of preeclampsia and fetal growth retardation. Obstet. Gynecol., 3, 378–386.

Pijnenborg, R., Robertson, W.B., Brosens, I. and Dixon, G. (1981) Review article: trophoblast invasion and the establishment of haemochorial placentation in man and laboratory animals. Placenta, 1, 71–91.

RachoEl-Akouri, R., Kurlberg, G., Dindelegan, G., Molne, J., Wallin, A. and Brannstrom, M. (2002) Heterotopic uterine transplantation by vascular anastomosis in the mouse. J. Endocrinol., 2, 157–166.

Scott, J.R., Pitkin, R.M. and Yannone, M.E. (1971) Transplantation of the primate uterus. Surg. Gynecol. Obstet., 3, 414–418.

Wakuda, K., Takakura, K., Nakanishi, K., Kita, N., Shi, H., Hirose, M. and Noda, Y. (1999) Embryo-dependent induction of embryo receptivity in the mouse endometrium. J. Reprod. Fertil., 2, 315–324.

Submitted on May 6, 2003; accepted on June 12, 2003.