1 Department of Cell Biology and Human Anatomy, 2 Department of Anatomy, Physiology and Cell Biology and 3 California National Primate Research Center, University of California, Davis, California, USA
4 To whom correspondence should be addressed at: Department of Cell Biology and Human Anatomy, University of California, One Shields Avenue, Davis, CA 95616 USA. E-mail: acenders{at}ucdavis.edu
![]() |
Abstract |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Key words: animal model/implantation/MDCK/trophoblast
![]() |
Introduction |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Lindenberg et al. (1986) and Bentin-Ley et al. (2000)
pioneered the in vitro study of human blastocysts cultured on cells derived from the endometrium. In an important ultrastructural study (Bentin-Ley et al.,2000) it was demonstrated that trophoblast cells could intrude between uterine epithelial cells under co-culture conditions. These authors also demonstrated that trophoblast cells could share desmosomes with the epithelial cells. There was, however, a peculiar development of stromal cells within the blastocyst cavity in these conditions. Lopata and Borg (1998)
demonstrated that marmoset blastocysts cultured on uterine epithelial cells produced areas of polar cellular and syncytial trophoblast adjacent to the uterine cells and that trophoblast protrusions from this area displaced uterine epithelial cells from a Matrigel substrate.
Meseguer et al. (2001) showed that human blastocysts apparently modify the glycocalyx of endometrial epithelial cells when co-cultured. The blastocysts adhered to these uterine epithelial cells but showed no further development. It also appeared that a single blastocyst could induce apoptosis in underlying human endometrial epithelial cells when adhering (Galan et al.,2000a
,b
). Carver et al. (2003)
demonstrated that hatched human blastocysts co-cultured on human endometrial stromal cell monolayers showed trophoblast outgrowth and invasion into the stromal layer. Co-culture of in vitro-fertilized ova to the blastocyst stage in the presence of endometrial cells has also been used to enhance blastocyst development prior to transfer in the human (Mercader et al.,2003
).
While in vitro studies with blastocysts cannot completely mimic the intrauterine environment, they can illustrate some of the potential interactions that may occur, as well as possibly providing a situation in which some of the parameters of interaction may be manipulated.
In an unpublished study, in vitro-fertilized oocytes were cultured with buffalo rat liver cells since these cells had been shown to improve blastocyst development during in vitro culture (Zhang et al.,1994; Weston et al.,1996
). Although the blastocysts varied somewhat in morphology, they tended to hatch at
1011 days after IVF. One of the specimens was left an additional 48 h; when this specimen was examined it was found that the trophoblast had penetrated the otherwise intact epithelial layer of hepatic cells. This suggested that a variety of epithelial cells from different sources and species might be used to assess the ability of blastocysts to adhere to and penetrate epithelial cells in vitro. Subsequently we examined the interaction of blastocysts with MDCK cells, UtMVEC cells, ME-180 cells, CaSki cells, and Matrigel plated on micropore filters.
![]() |
Materials and methods |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Substrates and co-culture conditions
The following cell lines were obtained from the American Type Culture Collection, Manassas, VA, USA; Madin Darby Canine Kidney cells (MDCK, ATCC# CCL34), ME-180 (human cervical carcinoma, ATCC# HTB33), CaSki (human cervical carcinoma, ATCC# CRL1550), buffalo rat liver cells (BRL3A). Human uterine microvascular endothelial cells (UtMVEC) were purchased from Cambrex. These cells were cultured on collagen-coated filters (5 µm pore size) in Dulbeccos modified Eagles medium with 5% (vol/vol) bovine calf serum (Hyclone) and penicillin/streptomycin. Hatched blastocysts were then added. An additional group of blastocysts was placed on Matrigel matrix (9) or just the uncoated filter (1). The cultures were examined periodically to assess development of the blastocyst and whether or not the blastocyst was adherent to the substrate. When blastocysts adhered to the substrate sufficiently to withstand gentle movement of the culture dishes, the culture fluid was withdrawn and Karnovskys fixative was added to the well. Subsequently the blastocysts together with the substrate were cut from the wells, postfixed in 1% (wt/vol) osmium tetroxide, dehydrated in ethanol, and embedded in Araldite epoxy resin. Thick sections were stained with Azure B and thin sections stained for electron microscopic examination. Oocytes from more than one female were used to provide blastocysts for the Matrigel, MDCK cell, and UtMVEC cell experiments.
A single blastocyst was obtained from an aspirated oocyte, injected with a spermatozoon, and cultured continuously in the presence of buffalo rat liver cells. The blastocyst was adherent to the cell layer at 14 days after fertilization, and was fixed and prepared as described above.
Statistical analysis
Blastocysts cultured on different cellular substrates and on Matrigel were examined for the presence or absence of selected morphological features. Results for MDCK cells or UtMVEC were separately compared to the Matrigel data using 2x2 contingency tables. Statistical significance (P < 0.05) was tested using Fishers exact test (Prizm; GraphPad, San Diego, CA, USA).
![]() |
Results |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Blastocysts on MDCK cells
In general the conditions of the blastocysts on MDCK cells (Figures 13) were superior to those on other substrates. Of nine blastocysts plated on these cells, eight were attached and one was adherent. All of the blastocysts had intact mural trophoblast. Two of the blastocysts had rather poor inner cell mass (ICM) development. The blastocysts cultured for 38 h on MDCK cells had good ICM but no evidence of the polarization of ICM cells that precedes amnion formation in this species (Figure 2A). Two of the blastocysts that were incubated for 48 h had the beginnings of amnion formation with polarized epiblast cells radiating around a small amniotic cavity (Figures 2B, C and 3B).
|
|
|
All of the attached blastocysts had penetrated though the MDCK cell layer. All but one of these had extensive areas of trophoblast resting on the underlying Millipore filter. The MDCK layer was otherwise intact and unilaminar in all specimens, and in some regions MDCK cells had migrated though the filter pores, forming stretches of unilaminar epithelium on the bottom of the filter. Most of the blastocysts had regions of syncytial trophoblast, commonly situated at the margins, but mostly cytotrophoblast rested on the Millipore filter. At the periphery of the area of attachment, the trophoblast abutted MDCK cells, and shared desmosomes with these cells (see Figure 3A). One of the blastocysts that did not have an ICM penetrated the MDCK cells in only two places, and was overlying some of the MDCK cells (Figure 2D). This blastocyst also did not show any regions of syncytial trophoblast.
Blastocysts on UtMVEC cells
Of the six blastocysts placed on UtMVEC cells, two were adherent and four were attached. All but one of these blastocysts had a discernible ICM. However, none of them showed amnion formation, although all were incubated a minimum of 48 h and one was incubated for 70 h after hatching.
Of those blastocysts that were attached, there was always contact with UtMVEC cells (Figure 4A). However, the UtMVEC cells formed a discontinuous layer, and many of the UtMVEC cells migrated though the pores in the Millipore filter, some of them adhering to the far side. It could therefore not be determined whether the blastocysts penetrated the UtMVEC cells as well as adhering to them.
|
ME-180 and CaSki
Two other epithelial cell lines, ME-180 and CaSki, were also used. Both of these cell lines formed incomplete unilaminar stretches of cells under our culture conditions. The two blastocysts on CaSki cells that were adherent after 72 h post-hatching became displaced during processing. Of these two blastocysts the larger appeared healthy, with a good ICM and a mass of syncytial trophoblast adjacent to the ICM, but no indication of incipient amnion formation (Figure 4B).
Both of the blastocysts on ME-180 cells (Figure 5 and 6) were attached 72 h after hatching. One of these blastocysts had a compact ICM but rather small blastocyst cavity, and was underlain by endothelial cells. It had a mass of syncytial trophoblast at the margin of the attachment and was underlain by epithelial cells (Figure 5A). The other attached blastocyst, although adhering to underlying ME-180 cells, had no well-formed ICM and extensive debris within the blastocyst cavity (Figure 5B, C).
|
|
Blastocysts on Matrigel
Nine blastocysts were placed on Matrigel. Two of the nine blastocysts were not adherent but were partially collapsed although they had been incubated only a little over 1 day. Two blastocysts that were incubated for 2 days were adherent but came free during processing. One of these appeared reasonably well developed, the other looked more like a morula. Of the four attached blastocysts, none had appropriate development; they had little or no blastocyst cavity (Figure 7A). Only one of the four showed outgrowth along the Matrigel (Figure 7B) but all four were slightly indented into the Matrigel.
|
Blastocyst on a hepatocyte cell layer
The blastocyst placed on buffalo rat hepatic cells was large, and had penetrated through the hepatocyte layer which was otherwise continuous (Figure 8). In places the trophoblast of the blastocyst was in contact with connective tissue cells; in other places it was in contact with the support membrane. However, since the distribution of connective tissue cells under the hepatocyte layer was irregular, it could not be determined whether the trophoblast penetrated the connective tissue. At the periphery of the area of attachment the trophoblast cells shared desmosomal-type junctions with the hepatocytes. Although there was an appreciable endodermal layer underlying the trophoblast, there was no discernible ICM in this specimen.
|
Summary of results
Table I summarizes the incidence of major morphological features of the blastocysts when cultured on different substrates. Statistical analyses of these data using contingency tables showed that blastocysts cultured on MDCK cells or on UtMVEC had significantly greater incidence of blastocyst cavity and inner cell mass formation than blastocysts cultured on Matrigel. Blastocysts cultured on ME-180 cells or CaSki cells were not included in the analyses due to the small sample numbers.
|
![]() |
Discussion |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
The blastocysts that showed the most advanced development had an amniotic cavity as well as cytotrophoblast and syncytial trophoblast situated along the supporting membrane. The distribution with the syncytial trophoblast at the margin of the site and cytotrophoblast centrally is similar to that seen normally at the trophoblastic plate stage in this species (Enders, 1995). However, there was less syncytial trophoblast than is normally found at this stage, and the total time from IVF to the trophoblastic plate stage was delayed by
2 days. Although syncytial trophoblast is the initial type of trophoblast invading the endometrium by penetrating between uterine epithelial cells in vivo, these studies did not demonstrate whether cytotrophoblast without syncytium could or could not penetrate the epithelial cell layer.
Of particular interest is the observation that trophoblast can attach to and even penetrate diverse heterologous epithelial sheets from different organs and species. These results suggest, as does the presence of ectopic implantation in the human, that the limited opportunities for implantation in vivo are not a result of the lack of invasiveness by the trophoblast but rather modification of the endometrial environment, including the glycocalyx of the luminal epithelial cells, to decrease the facility of trophoblast to attach appropriately. Extensive expression of MUC1 may be deleterious to attachment of trophoblast to epithelial cells in a number of species (Carson et al., 2000), and a recent study suggests that human blastocysts may induce paracrine cleavage of endometrial cell MUC1 at the implantation site (Meseguer et al., 2001
). In this respect it is noteworthy that normal hepatocytes do not express MUC1 histochemically (Cao et al., 1999
), and that MDCK cells can be transfected to produce abundant expression of MUC1 (Lavelle et al.,1997
).
The use of a cell line such as MDCK cells which has been thoroughly studied and can be manipulated has a number of advantages as a means of identifying factors that may increase or decrease adhesion and epithelial invasion by intact blastocysts. Use of such cells does not completely substitute for the use of uterine epithelial cells, which have been used to model implantation in the human (Dominguez et al., 2001). However, it avoids the need to provide macaque endometrial luminal epithelial cells that respond in culture to hormones in a similar fashion to endometrium at implantation. Since even short-term supravital culture of endometrium results in changes including increased cell death, which in itself produces adhesion, this method also has severe limitations.
Further studies of co-culture of blastocysts on variously transfected MDCK cells and on macaque uterine cells or human luminal epithelial cell lines could provide further information on possible limits to trophoblast penetration of epithelial cell layers.
![]() |
Acknowledgements |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
![]() |
References |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Bentin-Ley U, Horn T, Sjogren A, Sorensen S, Falck Larsen J and Hamberger L (2000) Ultrastructure of human blastocystendometrial interactions in vitro. J Reprod Fertil 120,337350.[CrossRef][ISI][Medline]
Boatman DE (1987) In Vitro Growth of Non-human Primate Pre- and Peri-implantation Embroys. Plenum Press, New York, pp 273308.
Cao Y, Karsten U, Otto G and Bannasch P (1999) Expression of MUC1, ThomsenFriedenreich antigen, Tn, sialosyl-Tn, and alpha2,6-linked sialic acid in hepatocellular carcinomas and preneoplastic hepatocellular lesions. Virchows Arch 434,503509.[CrossRef][Medline]
Carson DD, Bagchi I, Dey SK, Enders AC, Fazleabas AT, Lessey BA and Yoshinaga K (2000) Embryo implantation. Dev Biol 223,217237.[CrossRef][ISI][Medline]
Carver J, Martin K, Spyropoulou I, Barlow D, Sargent I and Mardon H (2003) An in-vitro model for stromal invasion during implantation of the human blastocyst. Hum Reprod 18,283290.
Dominguez F, Martin J, Navarro J, Pellicer A and Simon C (2001) Embryoendometrial interactions. In Lessey BA (ed) Infertility and Reproductive Medicine Clinics of North America. WB Saunders Company, Philadelphia, pp 373388.
Enders AC (1995) Transition from lacunar to villous stage of implantation in the macaque, including establishment of the trophoblastic shell. Acta Anat (Basel) 152,151169.[Medline]
Enders AC, Boatman D, Morgan P and Bavister BD (1989) Differentiation of blastocysts derived from in vitro-fertilized rhesus monkey ova. Biol Reprod 41,715727.
Fishel SB, Edwards RG and Evans CJ (1984) Human chorionic gonadotropin secreted by preimplantation embryos cultured in vitro. Science 223,816818.[ISI][Medline]
Galan A, Herrer R, Remohi J, Pellicer A and Simon C (2000a) Embryonic regulation of endometrial epithelial apoptosis during human implantation. Hum Reprod 15(Suppl 6),7480.
Galan A, OConnor JE, Valbuena D, Herrer R, Remohi J, Pampfer S, Pellicer A and Simon C (2000b) The human blastocyst regulates endometrial epithelial apoptosis in embryonic adhesion. Biol Reprod 63,430439.
Lavelle JP, Negrete HO, Poland PA, Kinlough CL, Meyers SD, Hughey RP and Zeidel ML (1997) Low permeabilities of MDCK cell monolayers: a model barrier epithelium. Am J Physiol 273,F6775.[ISI][Medline]
Lindenberg S, Hyttel P, Lenz S and Holmes PV (1986) Ultrastructure of the early human implantation in vitro. Hum Reprod 1,533538.[Abstract]
Lopata A and Borg J (1998) Studies on marmoset monkey blastocyst implantation in vitro. Proc Fertil Soc Aust, p.122.
Lopata A, Berka J, Simula A, Norman R and Otani T (1995a) Differential distribution of mRNA for the alpha- and beta-subunits of chorionic gonadotrophin in the implantation stage blastocyst of the marmoset monkey. Placenta 16,335346.[CrossRef][ISI][Medline]
Lopata A, Kohlman DJ, Bowes LG and Watkins WB (1995b) Culture of marmoset blastocysts on matrigel: a model of differentiation during the implantation period. Anat Rec 241,469486.[CrossRef][ISI][Medline]
Lopata A, Oliva K, Stanton PG and Robertson DM (1997) Analysis of chorionic gonadotrophin secreted by cultured human blastocysts. Mol Hum Reprod 3,517521.[Abstract]
McKiernan SH and Bavister BD (2000) Culture of one-cell hamster embryos with water soluble vitamins: pantothenate stimulates blastocyst production. Hum Reprod 15,157164.
Mercader A, Garcia-Velasco JA, Escudero E, Remohi J, Pellicer A and Simon C (2003) Clinical experience and perinatal outcome of blastocyst transfer after coculture of human embryos with human endometrial epithelial cells: a 5-year follow-up study. Fertil Steril 80,11621168.[CrossRef][ISI][Medline]
Meseguer M, Aplin JD, Caballero-Campo P, OConnor JE, Martin JC, Remohi J, Pellicer A and Simon C (2001) Human endometrial mucin MUC1 is up-regulated by progesterone and down- regulated in vitro by the human blastocyst. Biol Reprod 64,590601.
Pope CE, Pope VZ and Beck LR (1982) Development of baboon preimplantation embryos to post-implantation stages in vitro. Biol Reprod 27,915923.
Sarason RL, VandeVoort CA, Mader DR and Overstreet JW (1991) The use of nonmetal electrodes in electroejaculation of restrained but unanesthetized macaques. J Med Primatol 20,122125.[ISI][Medline]
Sjoblom C, Wikland M and Robertson SA (2002) Granulocyte-macrophage colony-stimulating factor (GM-CSF) acts independently of the beta common subunit of the GM-CSF receptor to prevent inner cell mass apoptosis in human embryos. Biol Reprod 67,18171823.
VandeVoort CA and Tarantal AF (1991) The macaque model for in vitro fertilization: superovulation techniques and ultrasound-guided follicular aspiration. J Med Primatol 20,110116.[ISI][Medline]
VandeVoort CA and Tarantal AF (2001) Recombinant human gonadotropins for macaque superovulation: repeated stimulations and post-treatment pregnancies. J Med Primatol 30,304307.[CrossRef][ISI][Medline]
Weston AM, Zelinski-Wooten MB, Hutchison JS, Stouffer RL and Wolf DP (1996) Developmental potential of embryos produced by in-vitro fertilization from gonadotrophin-releasing hormone antagonist-treated macaques stimulated with recombinant human follicle stimulating hormone alone or in combination with luteinizing hormone. Hum Reprod 11,608613.[Abstract]
Zhang L, Weston AM, Denniston RS, Goodeaux LL, Godke RA and Wolf DP (1994) Developmental potential of rhesus monkey embryos produced by in vitro fertilization. Biol Reprod 51,433440.[Abstract]
Submitted on March 22, 2005; resubmitted on May 6, 2005; accepted on May 31, 2005.
|