Embryo Implantation - The Molecular Mechanism

RBMOnline - Vol 13 No 6. 2006 833-839 Reproductive BioMedicine Online; www.rbmonline.com/Article/ www.rbmonline.com/Article/2454 on web 5 October 2006

Outlook Embryo implantation: the molecular mechanism remains elusive John Aplin received his initial training in chemistry at Queen’s College Oxford, and in biophysics at the University of British Columbia where he first became interested in glycobiology and the cell surface. As a ‘post-doc’ at the MRC laboratories in Mill Hill, London his interests turned to the biology of cell adhesion. Chance events led him into reproductive biology, most notably implantation and placental development. He is currently Professor of Reproductive Biomedicine at the University of Manchester where he is a member of the Maternal and Fetal Health Research Centre in the Medical School, with a cross-appointment in the Faculty of Life Sciences.

Dr John Aplin JD Aplin Division of Human Development, Medical School, University of Manchester, UK Correspondence: Research Floor, St Mary’s Hospital, Manchester M13 0JH, UK; e-mail: john.aplin@manchester. ac.uk

Abstract Low rates of implantation are an impediment to more efficient assisted reproduction techniques. Improved endometrial receptivity and embryo preparation should lead to higher pregnancy rates, lower rates of early pregnancy failure and fewer multiple pregnancies. As the first site of contact between embryo and endometrium, the luminal epithelium (LE) is responsible for the non-receptive status of proliferative and early secretory tissue, and transformation to receptivity in the mid-secretory phase presumably requires alterations in expression, organization or activation of adhesion systems. Luminal cells are less abundant than their glandular counterparts, and are under-represented in global tissue datasets. Furthermore, alterations in cell surface composition can be readily accomplished by mechanisms that do not rely on altered transcription or translation. Current data from in-vitro models are consistent with initial attachment to mucin in the apical glycocalyx, perhaps via a carbohydrate-mediated interaction, after which the epithelial phenotype is modified by a medium- or short-range embryonic signal. A cascade of interactions follows, mediating embryo migration across the epithelium. Strikingly, numerous potential mediators of adhesion at implantation are located in the lateral rather than the apical surface of LE cells. Attached embryos appear to gain rapid access to this highly adhesive lateral membrane domain. Keywords: adhesion molecule, endometrium, epithelial polarity, glycocalyx, implantation, mucin

The implantation window is regulated by endometrial luminal epithelium Observations made following embryo replacement indicate that the human endometrium is receptive to an implanting embryo from about 6–10 days after the LH peak (Bergh and Navot, 1992; Aplin, 2000). Hatched blastocysts attach non-selectively to endometrial stromal cells in vitro (Carver et al., 2003) as well as to tissue culture dishes, indicating that it is the luminal epithelium that confers upon the uterus its unique property of resistance to implantation in phases other than the mid-secretory. Though both originate from progenitor cells in the basal glands during postmenstrual regeneration, luminal epithelial (LE) cells are distinct from their glandular (GE) counterparts (Dockery, 2002): they

are cuboidal rather than columnar; they show less pronounced secretory and organellar changes in the secretory phase; they develop mid-secretory uterodomes or pinopodes, apical cell surface swellings that have been suggested as an attribute of receptive tissue (Murphy, 2000; Lopata et al., 2002); and there are numerous molecular markers that indicate compositional distinctions between the two (Seif and Aplin, 1990; Aplin et al., 1998; ). In the mouse, comparative microarray analysis of the GE and LE transcriptomes at the time of implantation (Niklaus and Pollard, 2006) indicates that though most products are common, about 280 genes significantly differ in expression. These fall into numerous functional categories, but one important observation is the presence of immune defence gene products in the LE, a reminder that it has a primary barrier function in repelling microorganisms in the upper reproductive tract. Discussion of the effect of ovarian stimulation regimes and hormone replacement therapy on receptivity has been clouded by the lack of a precise

833

Outlook - Molecular mechanism of embryo implantation - JD Aplin understanding of the key attributes (molecular or structural) that confer receptive status on endometrium. In fact, there is evidence that a range of phenotypes will support implantation; it may well be that healthy embryos can to some extent overcome endometrial defects.

a primary polarity and an apical barrier. In carcinoma cells, this tightly defined polarity of MUC1 expression is lost and novel glycoforms appear. There is evidence that the glycocalyx is altered at the surface of uterodomes in the mid-secretory LE, with loss of MUC1 epitopes (Horne et al., 2005).

Identifying the molecular mediators of implantation

First phase attachment

Common assumptions that underpin experimental approaches to the identification of components involved in embryo attachment and subsequent phases of implantation deserve closer scrutiny: (i) it appears imperative that protein expression should occur in the implantation phase LE and in the trophectoderm of the hatched blastocyst; (ii) since initial attachment occurs at the apical side of maternal cells, display at this surface appears important; (iii) candidate components on the maternal and embryonic surfaces should be capable of receptor−ligand interaction; (iv) in order to explain varying receptivity, it is often assumed that expression will be regulated with highest levels in the mid-secretory phase; and (v) if mice null for the respective gene product show impaired implantation, the hypothesis that there is involvement in humans is strengthened, though by no means proven. Microarray studies based on assumption (iv) have sought to identify a molecular signature characteristic of receptive endometrium (see, e.g. Ponnampalam et al., 2004; Talbi et al., 2006). Though informative in other respects, these studies have not thus far been particularly helpful in identifying candidate gene products or pathways that might play a role in embryo attachment. One problem is that LE contributes a minority of the cells present in a typical endometrial biopsy; another is that endometrial tissue composition has inherent variability, as seen very obviously in glandular and stromal histology (Buckley and Fox, 1989). This leads to uncertainty in histological dating (Coutifaris et al., 2004; Myers et al., 2004). Furthermore, the proportion of epithelial and stromal cell-derived mRNA in similarly timed biopsies can vary widely (Jasper et al., 2006).

Epithelial polarity and implantation Work carried out in earlier decades led to recognition that epithelial polarity plays a key role in uterine function (Denker, 1993; Glasser and Mulholland, 1993). Thus, for example, key aspects of the steroid response in endometrial epithelial cells are controlled indirectly by paracrine signals reaching the basal epithelial surface from receptor-positive, steroid-responsive stromal cells (Cooke et al., 2002). The apical epithelial surface, as well as providing an immune barrier (innate and acquired via immunoglobulin A), is clearly the first site of interaction with the embryo. To prevent epithelial cavities from sealing up, apical surfaces must normally be non-self-adhesive. Thus, the hypothesis arose that implantation might occur in the context of altered maternal epithelial polarity.

834

The epithelial glycocalyx, which contains MUC1 (Aplin et al.,, 1998; Meseguer et al., 2001) and other higher molecular mass mucin glycoproteins MUC4 (Idris and Carraway, 2000), MUC6 (Gipson et al., 1997) and MUC16 (CA125; Mylonas et al., 2003), is a key attribute of the apical epithelial surface. Its absence from basolateral cell surfaces in normal luminal (and glandular) endometrial epithelium defines these cells as showing

MUC1 can play a dual role in relation to cellular interactions. By virtue of high expression at the cell surface, density of glycosylation and extended conformation, it is capable of sterically hindering interaction with other cell surfaces mediated by families of conformationally smaller adhesion molecules such as integrins and cadherins (Hilkens et al., 1992). On the other hand, some isoforms of MUC-1 are able to bind intracellular adhesion molecule (ICAM)-1 (via the exposed core protein) (Regimbald et al., 1996), and MUC-1 glycoforms carrying selectin ligands can be observed in the endometrium (Hey and Aplin, 1996). This has led to the suggestion that L-selectin on the trophectodermal surface of the blastocyst might mediate initial interaction with ligand on the LE (Genbacev et al., 2003), but other studies have failed to detect L-selectin on the blastocyst (Campbell et al., 1995b; Bloor et al., 2002) and more work is required to clarify this important point. MUC-1 is polymorphic, with expression at the cell surface of two alleles differing in the number of tandem repeats in the mucin domain. The suggestion has been made that variation at this locus may be associated with infertility, but this requires further investigation (Goulart et al., 2004).

Clearance of apical surface mucin by attaching embryos In mouse and rat, MUC1 is strongly expressed at the apical LE surface, but it is removed under maternal hormonal control in advance of embryo attachment (reviewed in Aplin and Kimber, 2004). This occurs even in leukaemia inhibitory factor (LIF)-null female mice, which have a block to implantation (Chen et al., 2000). Here, embryos appear to attach normally to the epithelial surface, but implantation arrests thereafter (Fouladi-Nashta et al., 2005). MUC1 is also lost from luminal epithelium in macaques under maternal control (Julian et al., 2005). In humans, MUC-1 is continuously present at the luminal surface of the endometrial cavity throughout the cycle, with an increase in the secretory phase (Hey et al., 1994). However, the human embryo attaches to primary cultured MUC1+ endometrial epithelial monolayers, and MUC1 immunostaining is lost from cells beneath and immediately adjacent to the attachment site (Meseguer et al., 2001). Surprisingly, mouse embryos also appear to be capable of removing MUC1 from the surface of epithelial cells beneath and adjacent to their site of attachment, suggesting that a rescue pathway may be available in this species for circumstances in which MUC1 is not fully cleared under maternal control. The mechanism of clearance is still unknown, though it has been shown that both ADAM 17 (TACE) and matrix metalloproteinase (MMP)-14 (MT-MMP1) are capable of releasing MUC1 from the cell surface (Thathiah et al., 2003; Thathiah and Carson, 2004). It is clear that soluble signals pass locally between the embryo and receptors on maternal cells: chemokines (Dominguez et al., 2003), leptin (Cervero et al., 2005) and chorionic gonadotrophin (Zhou et al., 1999; Cameo et al., 2004) may all play a role in mediating this early dialogue.

Outlook - Molecular mechanism of embryo implantation - JD Aplin Based on these findings, one can postulate that initial attachment to glycocalyx is followed by lateral clearance of mucin, leading to interaction of trophoblast with higher affinity epithelial adhesion systems (Figure 1). Thus, the glycocalyx could act both

as an adhesion substrate and a barrier. Uterodomes represent a specialized area of LE plasma membrane that could be used for attachment, but this has not been observed (Bentin-Ley, 2000).

a

b Figure 1. (a) Initial attachment is to the glycocalyx and may be mediated by a lectin−carbohydrate interaction. The lateral luminal epithelium (LE) membranes are rich in adhesion molecules. (b) Second phase attachment follows the clearance of apical glycocalyx in response to an embryonic signal. The lateral LE borders open and trophectodermal processes insert between the LE cells. MUC 1 = mucin 1; OPN = osteopontin.

835

Outlook - Molecular mechanism of embryo implantation - JD Aplin This initial interaction locates the embryo within a specific area of the uterine surface. Short range signalling to maternal epithelial cells can then occur, leading to glycocalyx clearance. This is a test of viability for the embryo, and poor quality (e.g. monosomic) blastocysts may fail at this stage. At the same time, it places the state of receptivity in a new light: the ability of the epithelium to respond to embryo-derived signals (Quenby et al., 2002).

Second phase attachment: candidate adhesion systems First phase attachment is presumed to be followed closely by adhesion of trophectoderm to non-mucin receptors at the epithelial surface. There are numerous hurdles to the identification of molecular mediators, of which the most serious is that it is not possible to obtain sufficient human embryos to carry out well-controlled studies of molecular pathways. Mouse embryos may provide useful clues, and studies of fertility in mouse gene knockouts are in some cases instructive. Candidate adhesion systems have been reviewed in detail elsewhere (Aplin, 1997; Kimber and Spanswick, 2000; Aplin and Kimber, 2004). A few are mentioned here in brief. In assessing possible mediators, the Human Protein Atlas, part of the Human Proteome Organisation (HUPO) human antibody initiative (http://www.proteinatlas.org), is a very useful resource that aims eventually to show the distribution of all proteins in a wide range of human tissues. Several of the proteins discussed below are included in current releases of the atlas with high quality immunolocalization in endometrium. Basigin, also known as CD147 and EMMPRIN, is a membranespanning immunoglobulin superfamily member with multiple binding partners at the cell surface. It may be activated by direct homotypic interaction between adjacent cells, but is best known as an activator of MMP activity. In both mice and rats it is expressed in LE on day 1 of pregnancy under the influence of oestrogen, is down-regulated but then reappears locally in response to an embryonic stimulus on day 4 (Xiao et al., 2002a,b). Implantation rates are severely compromised in animals lacking basigin (Igakura et al., 1998; Kuno et al., 1998). Basigin is expressed most prominently at the lateral epithelial surface of endometrial epithelial cells in human and rodent. It is also expressed on pre-implantation embryos. CD44 is another single pass transmembrane glycoprotein. It shows a complex pattern of alternative splicing with several forms observed in endometrium (Behzad et al., 1994). It is expressed in pre-implantation embryos (Campbell et al., 1995a). It is associated with cell migration and has numerous binding partners including hyaluronate and osteopontin. Null mice are fertile. In endometrium, CD44 shows a predominantly lateral distribution in both GE and LE (Behzad et al., 1994).

836

Osteopontin (OPN; also known as SPP1) is a secreted glycoprotein that is regulated by progesterone (Apparao et al., 2001) and reaches a maximum in the secretory phase of the cycle when it immunolocalizes predominantly to the apical cytoplasm of LE and GE cells. It has multiple binding partners including integrins α4β1, αvβ3, αvβ5, αvβ6 and specific splice variants of CD44 (v3, v6). Null mice are fertile. In ruminants exhibiting

epitheliochorial implantation, OPN localizes precisely to the interface between trophectoderm and LE (Johnson et al., 2003). The possibility that OPN could bridge between cell surface ligands such as integrins and CD44 on the trophectoderm and LE has not been tested directly. Integrins of the β1 and αv subfamilies have been widely discussed as potential mediators of implantation (see Aplin and Kimber, 2004). Mice lacking the integrin β1 subunit fail to implant, but this intervention deletes at least 10 different integrins. In contrast, mice null for αv are fertile. Integrin αvβ3 exhibits a mid-secretory phase increase in expression by virtue of an increase in β3 abundance (Lessey, 2002). Experimental evidence has been reported (blocking antibody action in vivo) that suggests a role for integrin α4β1 in mouse implantation (Basak et al., 2002). Integrin expression on epithelial cells is highly variable between different areas of both GE and LE. In general, expression is substantially higher in the former. Integrins of both types are again more abundant on lateral (and basal) epithelial surfaces than at the apical surface, though some apical expression is detectable. Integrins are also widely expressed on stromal cells. Trophinin is a membrane component discovered in a screen for components mediating the attachment of teratocarcinoma cells to a uterine epithelial cell line (Fukuda et al., 1995). There is evidence that human chorionic gonadotrophin can induce local expression of trophinin by maternal epithelial cells (Nakayama et al., 2003). Functional trophinin associates with two cytoplasmic proteins, tastin and bystin, and the complex appears to be apically displayed (Fukuda and Nozawa, 1999). It is not absolutely required for early implantation in mice (Nadano et al., 2002), but remains a possible candidate in primates. CD9 is a member of the tetraspanin family of accessory proteins. It can act as a receptor for the family of pregnancyspecific glycoproteins (PSG) produced by trophoblast (Wynne et al., 2006), as well as associating laterally with β1 integrins in the plasma membrane (Park et al., 2000). Its distribution in endometrium is again largely in the lateral membranes of epithelial cells, including the LE. It is also expressed in blastocysts. Blocking antibodies to CD9 added in vitro to mouse embryos attaching to epithelial monolayers stimulated trophoblast outgrowth, though there was no effect on attachment rate. MMP-2 production was stimulated via the phosphatidylinositol 3-kinase pathway. Intrauterine injection of the antibodies on day 4 of mouse pregnancy increased the number of implantation sites (Liu et al., 2006). These experiments suggest CD9 as an inhibitor or regulator of implantation, and in agreement with this hypothesis, CD9-replete embryos implant normally in CD9-null mothers (Wynne et al., 2006).

Regulation of adhesion systems that mediate implantation It is striking that several molecular systems that might potentially play a role in the epithelial phases of embryo implantation are predominantly localized to the lateral cell membranes of LE cells when studies are carried out in non-conception cycles. Furthermore, few show substantial evidence of hormonal regulation, either when studied directly at protein level or when large scale transcriptomic screens have been carried out. There

Outlook - Molecular mechanism of embryo implantation - JD Aplin Table 1. Non-transcriptional, non-translational mechanisms for altering receptivity at the endometrial luminal epithelium. Gross membrane morphology, e.g. uterodomes Proteolysis and shedding Redistribution of surface components between membrane domains Association with accessory molecules Trafficking from endosomal compartments to and from the cell surface Activation, e.g. via phosphorylation Other enzymatic modifications, e.g. cross-linking, glycosylation Cytoskeletal association Lipid association

is compelling evidence that local signalling occurs between the embryo and LE at implantation and it is entirely plausible that such signalling triggers a reorganization of the LE surface with alteration of polarity and redistribution of components between membrane domains. Table 1 lists several mechanisms by which alteration in plasma membrane composition or activity could occur independent of transcription or translation.

receptor systems be displayed at the apical LE surface in the mid-secretory phase of non-conception cycles. Mouse mutants with impaired implantation are useful in hypothesis generation, but the extent to which embryo attachment in human uses the same ligand–receptor systems remains to be demonstrated.

The lateral borders of LE cells undergo striking changes at the time of implantation in several species. For example, desmosomes are reduced in the mouse uterine luminal epithelium during the preimplantation period of pregnancy (Illingworth et al., 2000), and gap junctional connections between LE cells are exquisitely modulated by embryonic signals in the implantation chamber (Grummer et al., 2004). Tight junctions move to a deeper level in LE in mid-secretory phase human endometrium (Murphy et al., 1982), and the tight junction proteins claudins 1, 4 and 5 are concentrated in the lower parts of lateral LE cell borders (http://www.proteinatlas.org). It is noteworthy that when human embryos attach to primary human epithelial cells, transmission electron microscopic examination reveals direct membraneto-membrane interactions between the trophectoderm and the lateral (not apical) surfaces of endometrial cells (Bentin-Ley, 2000). This appears to occur after disruption of apical junctional complexes, with opening up of spaces between the lateral borders of epithelial cells, and intrusion of trophectodermal processes. There is sharing of apical junctional complexes and desmosomes between trophectoderm and LE cells (Lopata et al., 2002; Figure 1).

Aplin JD 2000 The cell biological basis of human implantation. Baillière’s Best Practice and Research. Clinical Obstetrics and Gynaecology 14, 757–764. Aplin JD 1997 Adhesion molecules in implantation. Reviews of Reproduction 2, 84–93. Aplin JD, Kimber SJ 2004 Trophoblast-uterine interactions at implantation. Reproductive Biology and Endocrinology 2, 48. Aplin JD, Hey NA, Graham RA 1998 Human endometrial MUC1 carries keratan sulfate: characteristic glycoforms in the luminal epithelium at receptivity. Glycobiology 8, 269–276. Apparao KB, Murray MJ, Fritz MA et al. 2001 Osteopontin and its receptor αvβ3 integrin are coexpressed in the human endometrium during the menstrual cycle but regulated differentially. Journal of Clinical Endocrinology and Metabolism 86, 4991–5000. Basak S, Dhar R, Das C 2002 Steroids modulate the expression of alpha4 integrin in mouse blastocysts and uterus during implantation. Biology of Reproduction 66, 1784–1789. Behzad F, Seif MW, Campbell S, Aplin JD 1994 Expression of two isoforms of CD44 in human endometrium. Biology of Reproduction 51, 739–747. Bentin-Ley U 2000 Relevance of endometrial pinopodes for human blastocyst implantation. Human Reproduction Supplement 6, 67–73. Bergh PA, Navot D 1992 The impact of embryonic development and endometrial maturity on the timing of implantation. Fertility and Sterility 583, 537–542. Bloor DJ, Metcalfe AD, Rutherford A et al. 2002 Expression of cell adhesion molecules during human preimplantation embryo development. Molecular Human Reproduction 8, 237–245. Buckley CH, Fox H 1989 Biopsy Pathology of the Endometrium. Chapman and Hall, London. Cameo P, Srisuparp S, Strakova Z, Fazleabas AT 2004 Chorionic gonadotropin and uterine dialogue in the primate. Reproductive Biology and Endocrinology 2, 50. Campbell S, Swann HR, Aplin JD et al. 1995a CD44 is expressed throughout pre-implantation human embryo development. Human Reproduction 10, 425–430. Campbell S, Swann HR, Seif MW et al. 1995b Cell adhesion molecules on the oocyte and preimplantation human embryo. Human Reproduction 10, 1571–1578. Carver J, Martin K, Spyropoulou I et al. 2003 An in-vitro model for stromal invasion during implantation of the human blastocyst.

Thus, the interaction between the embryo and the apical LE surface is transient. Interactions occurring at the highly hydrated glycocalyx are difficult to visualize in electron microscopy because of dehydration artefacts arising at processing. The store of cell adhesion molecules in lateral LE membranes may be called into action early in the epithelial phase of implantation, either following attachment to the glycocalyx or immediately after its clearance. Returning to the criteria listed at the start of this article, transcriptional regulation of components mediating attachment during the menstrual cycle is a questionable assumption. Given local signalling at the implantation site, and the many mechanisms available to cells for rapid alteration of cell surface composition, neither is it essential that the key adhesion ligand–

References

837

Outlook - Molecular mechanism of embryo implantation - JD Aplin

838

Human Reproduction 18, 283–290. Cervero A, Horcajadas JA, Dominguez F et al. 2005 Leptin system in embryo development and implantation: a protein in search of a function. Reproductive BioMedicine Online 10, 217–223. Chen JR, Cheng JG, Shatzer T et al. 2000 Leukemia inhibitory factor can substitute for nidatory estrogen and is essential to inducing a receptive uterus for implantation but is not essential for subsequent embryogenesis. Endocrinology 141, 4365–4372. Cooke PS, Buchanan DL, Kurita T et al. 2002 In: Glasser SR, Aplin JD, Giudice LN Tabibzadeh S (eds) The Endometrium. Taylor and Francis, London, pp. 151–166. Coutifaris C, Myers ER, Guzick DS et al. 2004 NICHD National Cooperative Reproductive Medicine Network. Histological dating of timed endometrial biopsy tissue is not related to fertility status. Fertility and Sterility 82, 1264–1272. Denker HW 1993 Implantation: a cell biological paradox. Journal of Experimental Zoology 266, 541–558. Dockery P 2002 The fine structure of the mature human endometrium. In: Glasser SR, Aplin JD, Giudice LN, Tabibzadeh S (eds) The Endometrium. Taylor and Francis, London, pp. 21–38. Dominguez F, Galan A, Martin JJ et al. 2003 Hormonal and embryonic regulation of chemokine receptors CXCR1, CXCR4, CCR5 and CCR2B in the human endometrium and the human blastocyst. Molecular Human Reproduction 9, 189–198. Fouladi-Nashta AA, Jones CJ, Nijjar N et al. 2005 Characterization of the uterine phenotype during the peri-implantation period for LIFnull, MF1 strain mice. Developmental Biology 281, 1–21. Fukuda MN, Nozawa S 1999 Trophinin, tastin, and bystin: a complex mediating unique attachment between trophoblastic and endometrial epithelial cells at their respective apical cell membranes. Seminars in Reproductive Endocrinology 17, 229–234 Fukuda MN, Sato T, Nakayama J et al. 1995 Trophinin and tastin, a novel cell adhesion molecule complex with potential involvement in embryo implantation. Genes Development 9, 1199–1210. Genbacev OD, Prakobphol A, Foulk RA et al. 2003 Trophoblast Lselectin-mediated adhesion at the maternal−fetal interface. Science 299, 405–408. Gipson IK, Ho SB, Spurr-Michaud SJ et al. 1997 Mucin genes expressed by human female reproductive tract epithelia. Biology of Reproduction 56, 999–1011. Glasser SR, Mulholland J 1993 Receptivity is a polarity dependent special function of hormonally regulated uterine epithelial cells. Microscopy Research and Technique 25, 106–120. Goulart LR, Vieira GS, Martelli L et al. 2004 Is MUC 1 polymorphism associated with female infertility? Reproductive BioMedicine Online 8, 477–482. Grummer R, Hewitt SW, Traub O et al. 2004 Different regulatory pathways of endometrial connexin expression: preimplantation hormonal-mediated pathway versus embryo implantation-initiated pathway. Biology of Reproduction 71, 273–281. Hey NA, Aplin JD 1996 Sialyl-Lewis X and sialyl-Lewis A are associated with MUC1 in human endometrium. Glycoconjugate Journal 13, 769–779. Hey NA, Graham RA, Seif MW, Aplin JD 1994 The polymorphic epithelial mucin MUC1 in human endometrium is regulated with maximal expression in the implantation phase. Journal of Clinical Endocrinology and Metabolism 78, 337–342. Hilkens J, Ligtenberg MJ, Vos HL, Litvinov SV 1992 Cell membraneassociated mucins and their adhesion-modulating property. Trends in Biochemical Sciences 17, 359–363. Horne AW, Lalani EN, Margara RA et al. 2005 The expression pattern of MUC1 glycoforms and other biomarkers of endometrial receptivity in fertile and infertile women. Molecular Reproduction and Development 72, 216–229. Idris N, Carraway KL 2000 Regulation of sialomucin complex/Muc4 expression in rat uterine luminal epithelial cells by transforming growth factor-beta: implications for blastocyst implantation. Journal of Cellular Physiology 185, 310–316. Igakura T, Kadomatsu K, Kaname T et al. 1998 A null mutation in basigin, an immunoglobulin superfamily member, indicates its important roles in peri-implantation development and

spermatogenesis. Developmental Biology 194, 152–165. Jasper MJ, Tremellen KP, Robertson SA 2006 Reduced expression of IL-6 and IL-1α mRNA in secretory phase endometrium of women with recurrent miscarriage. Journal of Reproductive Biology, in press. Johnson GA, Burghardt RC, Bazer FW, Spencer TE 2003 Osteopontin: roles in implantation and placentation. Biology of Reproduction 69, 1458–1471. Julian J, Enders AC, Fazleabas AT, Carson DD 2005 Compartmental distinctions in uterine Muc-1 expression during early pregnancy in cynomolgous macaque and baboon. Human Reproduction 20, 1493–1503. Kimber SJ, Spanswick C 2000 Blastocyst implantation: the adhesion cascade. Seminars in Cell and Developmental Biology 11, 77–92. Kuno N, Kadomatsu K, Fan QW et al. 1998 Female sterility in mice lacking the basigin gene, which encodes a transmembrane glycoprotein belonging to the immunoglobulin superfamily. FEBS Letters 425 191–194. Lessey BA 2002 Uterine factors in implantation. In: Glasser SR, Aplin JD, Giudice L, Tabibzadeh S (eds) The Endometrium. Taylor and Francis, London, pp. 208–228. Liu WM, Cao YJ, Yang YJ et al. 2006 Tetraspanin CD9 regulates invasion during mouse embryo implantation. Journal of Molecular Endocrinology 36, 121–130. Lopata A, Bentin-Ley U, Enders A 2002 ‘Pinopodes’ and implantation. Reviews in Endocrine and Metabolic Disorders 3, 77–86. Meseguer M, Aplin JD, Caballero-Campo P et al. 2001 Human endometrial mucin MUC1 is up-regulated by progesterone and down-regulated in vitro by the human blastocyst. Biology of Reproduction 64, 590–601. Murphy CR 2000 Understanding the apical surface markers of uterine receptivity: pinopods–or uterodomes? Human Reproduction 15, 2451–2454. Murphy CR, Swift JG, Need JA et al. 1982 A freeze-fracture electron microscopic study of tight junctions of epithelial cells in the human uterus. Anatomy and Embryology 163, 367–370. Myers ER, Silva S, Barnhart K et al. 2004 NICHD National Cooperative Reproductive Medicine Network. Interobserver and intraobserver variability in the histological dating of the endometrium in fertile and infertile women. Fertility and Sterility 82, 1278–1282. Mylonas I, Makovitzky J, Richter DU et al. 2003 Immunohistochemical expression of the tumour marker CA-125 in normal, hyperplastic and malignant endometrial tissue. Anticancer Research 23, 1075–1080. Nadano D, Sugihara K, Paria BC et al. 2002 Significant differences between mouse and human trophinins are revealed by their expression patterns and targeted disruption of mouse trophinin gene. Biology of Reproduction 66, 313–321. Nakayama J, Aoki D, Suga T et al. 2003 Implantation-dependent expression of trophinin by maternal fallopian tube epithelia during tubal pregnancies: possible role of human chorionic gonadotrophin on ectopic pregnancy. American Journal of Pathology 163, 2211–2219. Niklaus AL, Pollard JW 2006 Mining the mouse transcriptome of receptive endometrium reveals distinct molecular signatures for the luminal and glandular epithelium. Endocrinology April 20; [Epub ahead of print]. Park KR, Inoue T, Ueda M et al. 2000 CD9 is expressed on human endometrial epithelial cells in association with integrins alpha(6), alpha(3) and beta(1). Molecular Human Reproduction 6, 252–257. Ponnampalam AP, Weston GC, Trajstman AC et al. 2004 Molecular classification of human endometrial cycle stages by transcriptional profiling. Molecular Human Reproduction 10, 879–893. Quenby S, Vince G, Farquharson R, Aplin J 2002 Recurrent miscarriage: a defect in nature’s quality control? Human Reproduction 17 1959–1963. Regimbald LH, Pilarski LM, Longenecker BM et al. 1996 The breast mucin MUCI as a novel adhesion ligand for endothelial intercellular adhesion molecule 1 in breast cancer. Cancer Research 56, 4244–4249.

Outlook - Molecular mechanism of embryo implantation - JD Aplin Seif MW, Aplin JD 1990 Cell surface components of human endometrial epithelium: monoclonal antibody studies. Trophoblast Research 4, 339–356. Talbi S, Hamilton AE, Vo KC et al. 2006 Molecular phenotyping of human endometrium distinguishes menstrual cycle phases and underlying biological processes in normo-ovulatory women. Endocrinology 147, 1097–1121. Thathiah A, Carson DD 2004 MT1-MMP mediates MUC1 shedding independent of TACE/ADAM17. Biochemical Journal 382, 363–373. Thathiah A, Blobel CP, Carson DD 2003 Tumor necrosis factor-alpha converting enzyme/ADAM 17 mediates MUC1 shedding. Journal of Biological Chemistry 278, 3386–3394. Wynne F, Ball M, McLellan AS et al. 2006 Mouse pregnancyspecific glycoproteins: tissue-specific expression and evidence of association with maternal vasculature. Reproduction 131, 721–732. Xiao LJ, Chang H, Ding NZ et al. 2002a Basigin expression and hormonal regulation in mouse uterus during the peri-implantation period. Molecular Reproductive Development 63, 47–54.

Xiao LJ, Diao HL, Ma XH et al. 2002b Basigin expression and hormonal regulation in the rat uterus during the peri-implantation period. Reproduction 124, 219–225. Zhou XL, Lei ZM, Rao CV 1999 Treatment of human endometrial gland epithelial cells with chorionic gonadotropin/luteinizing hormone increases the expression of the cyclooxygenase-2 gene. Journal of Clinical Endocrinology and Metabolism 84, 3364– 3377.

Paper based on contribution presented at the International Serono Symposium ‘Human implantation: the new frontiers of human assisted reproductive technologies’ in Erice, Sicily, Italy, May 5–6, 2006. Received 7 June, 2006; refereed 14 July, 2006; accepted 25 August, 2006.

839