- Email: [email protected]
Inflammatory mediators and endometrial function—focus on the perivascular cell
Journal of Reproductive Immunology 57 (2002) 81–93 www.elsevier.com/locate/jreprimm
Inflammatory mediators and endometrial function—focus on the perivascular cell Rodney W. Kelly a,*, Anne E. King a, Hilary O.D. Critchley b a
Human Reproducti6e Sciences Unit, Medical Research Council, Centre for Reproducti6e Biology, Uni6ersity of Edinburgh, 37 Chalmers Street, Edinburgh EH3 9ET, UK b Department of Reproducti6e and De6elopmental Science, Centre for Reproducti6e Biology, Uni6ersity of Edinburgh, 37 Chalmers Street, Edinburgh EH3 9ET, UK
Abstract Human endometrium has a unique vascular architecture that allows menstruation, the shedding of a well-vascularised tissue layer, with limited bleeding. Blood loss is controlled at least in part by constriction of the perivascular cells, myofibroblasts that surround the spiral arterioles and have contractile activity. These perivascular cells, which are coupled to endothelial cells by processes, are responsive to changes in progesterone levels and express chemokines, cytokines and prostaglandins (PG) crucial to the control of leukocyte entry into endometrium. In this location the chemokine interleukin-8 (IL-8) and prostaglandin E (PGE) will have synergistic effects on leukocyte entry. CD40 is also expressed on the perivascular cells. Activation of CD40 by CD40 ligand is known to increase COX-2 and IL-8 expression in endometrial fibroblasts. The likely source of CD40 ligand in the uterus is platelets. Thus ingress of platelets will up-regulate NFkB by activating CD40 and increase agents such as PGE which will stimulate further the ingress of platelets. There is thus the possibility of a spiralling inflammatory response. This response however, is normally modulated by progesterone raising the threshold of the NFkB pathway and in the presence of high progesterone levels activation of CD40 will be ineffective. When progesterone falls at the end of the ovarian cycle and the restrictions on activation are lost, the perivascular cells will respond, initiating leukocyte entry, vasoconstriction–vasodilatation cycles with associated hypoxia and consequent sloughing off of the endometrium. The perivascular cell in endometrium is
* Corresponding author. E-mail address: [email protected] (R.W. Kelly). 0165-0378/02/$ - see front matter © 2002 Elsevier Science Ireland Ltd. All rights reserved. PII: S 0 1 6 5 - 0 3 7 8 ( 0 2 ) 0 0 0 0 8 - 6
82
R.W. Kelly et al. / Journal of Reproducti6e Immunology 57 (2002) 81–93
pivotal in both menstruation and early pregnancy and we need to understand this cell better to devise more effective medical treatment for menstrual dysfunction. © 2002 Elsevier Science Ireland Ltd. All rights reserved. Keywords: Inflammatory mediators; Endometrial function; Perivascular cell
1. Introduction One major function of the vasculature of endometrium is to restrict blood loss at the time of menstruation. The means by which this is achieved and the hormonal control of the process are still poorly understood but the cells with smooth muscle character that surround the blood vessels are likely to be relevant to this mechanism. In addition, cells in this location are a source of cytokines such as interleukin-15 (IL-15) which could stimulate growth of sub-populations of stromal cells. This review identifies cytokines and prostaglandins which are released by these cells and relates these events to hormonal changes and the initiation of menstruation.
2. The perivascular cells of human endometrium Although there is heterogeneity of the perivascular cells of the human endometrium (Roberts et al., 1992) this may reflect the heterogeneity found among pericytes in general (Hirschi and D’Amore, 1996). These cells in endometrium maintain close contact with endothelial cells and with each other by means of specialised projections which are maintained throughout the cycle (Roberts et al., 1992). The distribution of the progesterone receptors which are prominent in the perivascular cells but absent in the endothelium (Kohnen et al., 2000) suggests that progesterone induced changes are primarily mediated via the perivascular rather than the endothelial cell. An area of increasing interest is the interaction between the perivascular cells— a source of IL-15, the CD56++ cells —a source of IFN-g and the endothelial cells. A proper interaction is necessary for the IFN-g-dependent decidual blood vessel changes at the time of implantation (Ashkar et al. 2000). Cells associated with small blood vessels vary in character depending on location and those found in human endometrium are myofibroblasts in that they express smooth muscle a-actin and the fibroblast marker vimentin (Kohnen et al., 2000). These endometrial cells express a wide range of cytokines and vasoactive factors and are restricted to a discrete area surrounding the blood vessels and thus the term perivascular cell is used
R.W. Kelly et al. / Journal of Reproducti6e Immunology 57 (2002) 81–93
83
here. The group of cells, with a clearly perivascular location which stain for cytokines and prostaglandin (Fig. 1) may well include pericytes or be differentiated from these pluripotent cells (Hirschi and D’Amore, 1996). Such cells have a sentinel role in control of oedema and cellular ingress into the tissue but they are also thought to be associated with early signs of decidualisation. Whether these cells are the origin of the decidual cell is controversial but the decidual cell of early pregnancy has myofibroblast characteristics (Oliver et al., 1999) and the process of decidualisation commences around the blood vessels. In addition, in other sites, cell types such as adipocytes are known to originate from perivascular cells (Hirschi and D’Amore, 1996). The cells surrounding the spiral arterioles in human endometrium are a source of inflammatory mediators and there is good evidence that the expression of some of these mediators is controlled by progesterone. The signal for menstruation is the fall in ovarian progesterone and the perivascular cells will respond to this change with cytokine release and increased permeability. In contrast, high progesterone levels in early pregnancy decidua will regulate cell migration into the tissue and maintain pregnancy.
3. Control of leukocyte influx Prostaglandins (PGs) are very relevant to the function of the perivascular cell since they are important mediators of inflammation and in human endometrium PGs are under hormonal control. Non-steroidal anti-inflammatory drugs, which inhibit prostaglandin production, have been widely used to alleviate both dysmenorrhoea and menorrhagia (Cameron et al., 1990) and the likely site of action of such agents is at or around the blood vessels. The impact of prostaglandins on inflammatory mediators is synergistic: PGE is known to accentuate the effects of the inflammatory peptide bradykinin in oedema induction and both oedema and influx of polymorphonuclear leukocytes is synergistically increased by the combination of interleukin-8 (IL-8) and PGE (Colditz, 1990; Foster et al., 1989; Rampart et al., 1989). Since the expression of IL-8 in human endometrium is very marked in perivascular cells (Critchley et al., 1994; Milne et al., 1999), the concerted induction of both PGE and IL-8 would be expected to be a powerful drive for neutrophil influx into the tissue. In a hormonally responsive tissue such as endometrium, the ingress of leukocytes is likely to be modulated by steroids. This influx is indirectly mediated by steroids since the leukocytes of endometrium do not express the progesterone receptor (King et al., 1996; Stewart et al., 1998) whereas the stromal cells around the vessels maintain these receptors (Fig. 1) even in the secretory phase when
R.W. Kelly et al. / Journal of Reproducti6e Immunology 57 (2002) 81–93
Fig. 1.
84
R.W. Kelly et al. / Journal of Reproducti6e Immunology 57 (2002) 81–93
85
progesterone receptors are lost from the glands. The relevance of neutrophil entry into tissue is that the neutrophil is a potent source of enzymes, particularly collagenase and elastase. Collagenase from an activated neutrophil will disrupt inter-cellular anchoring and thus facilitate the process of tissue shedding. A further role for the neutrophil is to exert a secondary effect on blood vessels and increase their permeability (Wedmore and Williams, 1981) and thus augment oedema.
4. Progesterone control in perivascular cells The majority of perivascular cells in endometrium have abundant progesterone receptors (Fig. 1). Progesterone receptors have been identified in this location throughout the cycle and in early pregnancy (Perrot-Aplanat et al., 1994; Perrot-Applanat et al., 1988; Wang et al., 1998) and the intense staining in this location led to the conclusion that progesterone acting on these cells was essential for modulating blood-flow during pregnancy (Perrot-Aplanat et al., 1994), a point that is underscored by the absence of progesterone receptors from the placenta. Progesterone is also implicated in the stimulation of proliferation of smooth-muscle a-actin + ve cells that surround the spiral arterioles (Rogers et al., 2000) and fewer of these cells have been observed in women suffering from menorrhagia (Abberton et al., 1999). The suggestion has been made that the growth of these cells must accord with the development of the spiral arterioles and may be related to specific, progesterone induced, growth factors such as keratinocyte growth factor (FGF-7; Koji et al., 1994) and any defect could lead to impaired control of blood loss and consequent menstrual aberration. The effects of progesterone on the perivascular cell are revealed when anti-progestins such as RU486 (Mifepistone) are administered in vivo. The appearance of PGE in characteristic whorls around the small blood vessels
Fig. 1. Perivascular distribution of prostaglandins and cytokines. The perivascular cells of human endometrium possess progesterone receptors (g) endothelial cells (E) and perivascular cells (P) are marked. Although CD40 appears to be present at all stages of the menstrual cycle (f) cyclooxygenase (COX)-2 can be induced in vitro treatment with phorbol ester (e). In decidua, progesterone withdrawal with RU486 leads to an increased expression of PGE (d) and a decreased expression of 13,14-dihydro-15-keto PGE (c), the main PGE metabolite (PGEM). IL-8 is also prominent in the small blood vessels in decidua (b). Panel (a) shows CD45+ve (leukocyte common antigen) leukocytes clustered around small blood vessels in an endometrial sample from a patient exposed to an anti-progestin 24 h earlier. Scale bar=100 mM.
86
R.W. Kelly et al. / Journal of Reproducti6e Immunology 57 (2002) 81–93
of decidua follows 24 h after the progesterone antagonist is administered (Cheng et al., 1993a,b; Fig. 1). The mechanism by which PGE appearance is mediated is not fully understood but one element is the removal of the protecting catabolic enzyme 15-hydroxy prostaglandin dehydrogenase. Immunohistochemistry studies to detect 15-hydroxy prostaglandin dehydrogenase showed a marked absence of this enzyme around the blood vessels (Cheng et al., 1993a,b). Indirect evidence that this enzyme was induced by progesterone has been available for some time (Casey et al., 1980; Kelly and Bukman, 1990) but with the elucidation of the promoter region for this enzyme, the progesterone dependence has been made explicit (Greenland et al., 2000). Cytokine control in these perivascular cells is also progesterone dependent. Both IL-8 and monocyte chemotactic protein (MCP)-1 levels in these cells increase in the late secretory phase of the menstrual cycle at the time that leukocyte accumulation is evident (Jones et al., 1997). In both endometrium (Critchley et al., 1999) and decidua (Critchley et al., 1996), IL-8 rises perivascularly after progesterone antagonism with RU486. Thus progesterone appears to be maintaining low levels of chemokines and leukocyte concentration in endometrium and decidua. Such findings are consistent with elegant earlier animal data that show leukocyte entry into endometrium to be dependent on progesterone withdrawal (Staples et al., 1983). The mechanism by which progesterone modulates cytokine production is likely to be similar to the mechanisms employed by glucocorticoids, via NFkB (Caldenhoven et al., 1995). Progesterone can modulate the NFkB pathway in two ways. Progesterone bound to its receptor and NFkB both act as transcription factors competing for sites on the gene (Kalkhoven et al., 1996) or alternatively progesterone can stimulate synthesis of the inhibitory protein IkBa which maintains NFkB in the cytosol of the cell in an inactive state (Wissink et al., 1998; Fig. 2). Recent reports have shown that IL-15, a growth factor that shares the b and g receptor with IL-2, is released from the perivascular cells of human endometrium (Kitaya et al., 2000; Okada et al., 2000a,b) and that the release of this cytokine is progesterone dependent. The relevance of IL-15 is that it stimulates proliferation of the CD56+ ve large granular lymphocytes (Verma et al., 2000) that are a characteristic stromal component in late secretory phase endometrium and early decidua (Loke and King, 1995). The CD56+ve cells are associated with the spiral arterioles when first evident and this would be consistent with a perivascular location of their growth factor. IL-15 expression is thought to be under control of NFkB (Washizu et al., 1998) but this is unlikely to be the
R.W. Kelly et al. / Journal of Reproducti6e Immunology 57 (2002) 81–93
87
Fig. 2. Progesterone control mechanisms. IL-8 and PGE act synergistically to bring in leukocytes into tissue and both are controlled, at least in part, by the NFkB system. Progesterone also controls prostaglandins through induction of prostaglandin catabolism. The CD40–CD40L system in the perivascular cells would be progesterone responsive through the action of progesterone on the NFkB cascade.
mechanism of progesterone control because most effects of progesterone on the NFkB pathway are negative. One activator of the NFkB control pathway residing in the perivascular cells of human endometrium and myometrium is the CD40–CD40 ligand system. CD40 is a member of the TNF receptor family and, on ligation, is known to mediate prostaglandin synthesis by the inducible enzyme responsible for prostaglandin synthesis (COX-2; Zhang et al., 1998). This pathway can also be used to stimulate cytokine release from fibroblasts derived from human uterine tissue (King et al., 2001). However, whereas CD40 is expressed on perivascular cells and expression in cultured endometrial stromal cells (ESC) can be stimulated with interferon-g, the source of the CD40 ligand (CD40L) is not clear. One major source of CD40L is platelets (Henn et al., 1998) and thus any CD40 on the perivascular cell surface could be activated by extravasation of platelets. If the result of CD40 activation is stimulation of both prostaglandin and IL-8 then neutrophil ingress and oedema would follow. Moreover, the vasoactive effects of prostaglandin release will increase the exposure of CD40 to platelets and thus lead to a positive feedback (Figs. 3 and 4) but this will only operate in the absence of progesterone which will negate the feedforward loop at the level of NFkB and by inducing 15-hydroxy prostaglandin dehydrogenase. Thus, from the mid-secretory phase of the cycle onwards, progesterone will play an important role in maintaining the selective exclusion of leukocytes, keeping neutrophil numbers low but allowing entry of CD56+ ve cells.
88
R.W. Kelly et al. / Journal of Reproducti6e Immunology 57 (2002) 81–93
5. Models to study perivascular cell responses Cells grown from stroma of human endometrium resemble fibroblasts in morphology and are heterogeneous (Koumas et al., 2001). A feature of the ESC is that a sub-population can be grown in a progesterone and growth factor dependent manner. This parallels the in vivo situation where perivascular smooth muscle cells proliferate at the time of high progesterone concentrations (Abberton et al., 1999; Koji et al., 1994). In vitro, in the presence of growth factors such as epidermal growth factor (EGF) and basic-fibroblast growth factor (bFGF) growth is progesterone dependent (Irwin et al., 1991). Cells grown in the presence of bFGF do not decidualise even when high progesterone levels are maintained (Irwin et al., 1991). The classical trade-off between proliferation and differentiation appears to be present in this situation since progesterone will cause decidualisation, but in the absence of growth factors—usually engineered in vitro by low fetal-calf serum concentrations. In human endometrium pre-decidualisation occurs in the secretory phase (Noyes et al., 1950) and in early pregnancy this process develops with the earliest signs of differentiation seen in the cells surrounding the spiral arterioles (Buckley and Fox, 1989). Although the in vivo initiator of decidualisation is likely to be progesterone, in vitro experiments underline the role of cAMP in inducing this change in endometrium stromal cells
Fig. 3. Perivascularrole of the CD40 – CD40 ligand system. One activator of the CD40 on perivascular cells could be the CD40 ligand derived from activated platelets. This would lead to prostaglandin and cytokine production which would cooperatively induce increased vascular permeability thus leading to a feed forward loop. But, in the presence of progesterone this pathway would be blocked. Progesterone withdrawal would be permissive for an activation of the system which would lead to leukocyte entry and menstruation.
R.W. Kelly et al. / Journal of Reproducti6e Immunology 57 (2002) 81–93
89
Fig. 4. Differential control of cytokine release from ESC — the effect of decidualisation. IL-8 and MCP-1 release after 24 h in culture from five separate ESC preparations grown 3 – 5 weeks with progestins (MPA) and growth factor (bFGF). Cells were decidualised (or not) with 8-bromo cAMP (0.25 mM) and MPA for 5 days prior to treatment with IFN-g 25 ng/ml and/or CD40L (10 ng/ml) (Alexis). CD40 is expressed only when cells are stimulated with IFN-g and progesterone receptor is unaffected by decidualisation (data not shown). Both IL-8 and MCP-1 were measured in the culture medium after 24 h by ELISA. Results are shown as mean 9S.E.M., N=5. **, Represents difference from respective control PB0.005; *, difference B0.05; ‘a’ denotes difference in IL-8 release resulting from CD40 ligand treatment in decidualised and non-decidualised cells PB 0.001 and ‘b’ represents difference in MCP-1 concentrations between CD40L + IFN-g in decidualised and non-decidualised cells PB0.05.
(Brosens et al., 1999) which is characterised by prolactin and insulin-like growth factor binding protein-1 (IGFBP-1) release. In vivo, the mechanism by which progesterone induces pre-decidualisation and decidualisation in human endometrium and decidua is poorly understood. However the effects of progesterone and PGE are likely to be involved and the modulation of cAMP will be enhanced by the action of progesterone stimulating the expression of the EP2 (cyclic AMP coupled) receptor (Brar et al., 1997). ESC grown in the presence of bFGF and progestin are one model for examining cytokine expression in response to mediators such as CD40 L. Such cells express CD40 in response to interferon-g and the action of CD40 L on these cells is to stimulate IL-8 release (King et al., 2001). The monocyte-chemoattractant, MCP-1 is not significantly stimulated by CD40L but is released in response to interferon-g. Interestingly, decidualisation of these cells by treatment with cAMP analogues reduces IL-8 release in response to stimulation but has no effect on MCP-1 levels.
90
R.W. Kelly et al. / Journal of Reproducti6e Immunology 57 (2002) 81–93
6. Summary Given the assumed importance of the perivascular cells in controlling blood loss both at the time of menstruation and also at the time of parturition, we need to examine further the role of these cells in the initiation of menstruation (Kelly et al., 2001) and further studies must examine the effect of cell–cell contact between endothelial cells and perivascular myofibroblasts. In addition, we need to elucidate the factors altering the ratio of smooth-muscle to fibroblast character in endometrium stromal cells. Clinical treatments for excessive blood-loss (menorrhagia) include progestins-loaded intrauterine systems which induce a regressed endometrium, anti-fibrinolytic agents to stabilise blood-clotting and nonsteroidal anti-inflammatory agents which reduce prostaglandin synthesis. This last approach will act directly on the perivascular cells, those cells that surround the spiral arterioles of the human endometrium. New medical approaches to menorrhagia would be invaluable and agents that inhibit the CD40 – CD40-ligand system are potential treatments that would reduce the sensitivity of the vasoactive system. Finally, it is essential to understand the effect of progesterone on these cells since the retention of the progesterone receptor in the presence of high progesterone concentrations means that differentiation of these cells (e.g. by cAMP) could not only induce decidualisation by progesterone but also effect other changes in cell phenotype that will allow growth factors such as IL-15 to be released.
References Abberton, K.M., Taylor, N.H., Healy, D.L., Rogers, P.A., 1999. Vascular smooth muscle cell proliferation in arterioles of the human endometrium. Hum. Reprod. 14, 1072 –1079. Ashkar, A.A., Di Santo, J.P., Croy, B.A., 2000. Interferon-g contributes to initiation of uterine vascular modification, decidual integrity, and uterine natural killer cell maturation during normal murine pregnancy. J. Exp. Med. 192, 259 –270. Brar, A.K., Frank, G.R., Kessler, C.A., Cedars, M.I., Handwerger, S., 1997. Progesteronedependent decidualization of the human endometrium is mediated by cAMP. Endocrine 6, 301 –307. Brosens, J.J., Hayashi, N., White, J.O., 1999. Progesterone receptor regulates decidual prolactin expression in differentiating human endometrial stromal cells. Endocrinology 140, 4809 –4820. Buckley, C.H., Fox, H., 1989. Biopsy Pathology of the Endometrium, vol. p45. Chapman & Hall, London. Caldenhoven, E., Liden, J., Wissink, S., Vandestolpe, A., Raaijmakers, J., Koenderman, L., Okret, S., Gustafsson, J.A., Vandersaag, P.T., 1995. Negative cross-talk between RelA
R.W. Kelly et al. / Journal of Reproducti6e Immunology 57 (2002) 81–93
91
and the glucocorticoid receptor —a possible mechanism for the antiinflammatory action of glucocorticoids. Mol. Endocrinol. 9, 401 –412. Cameron, I.T., Haining, R., Lumsden, M.A., Thomas, V.R., Smith, S.K., 1990. The effects of mefenamic acid and norethisterone on measured menstrual blood loss. Obstet. Gynecol. 76, 85 –88. Casey, M.L., Hemsell, D.L., Johnston, J.M., MacDonald, P.C., 1980. NAD dependent 15-hydroxyprostaglandin dehydrogenase activity in human endometrium. Prostaglandins 19, 115 –122. Cheng, L., Kelly, R.W., Thong, K.J., Hume, R., Baird, D.T., 1993a. The effect of mifepristone (RU486) on the immunohistochemical distribution of prostaglandin E and its metabolite in decidual and chorionic tissue in early pregnancy. J. Clin. Endocrinol. Metab. 77, 873 –877. Cheng, L., Kelly, R.W., Thong, K.J., Hume, R., Baird, D.T., 1993b. The effects of mifepristone (RU486) on prostaglandin dehydrogenase in decidual and chorionic tissue in early pregnancy. Hum. Reprod. 8, 705 –709. Colditz, I.G., 1990. Effect of exogenous prostaglandin E2 and actinomycin D on plasma leakage induced by neutrophil activating peptide-1/interleukin-8. Immunol. Cell Biol. 68, 397–403. Critchley, H.O.D., Kelly, R.W., Kooy, J., 1994. Perivascular expression of chemokine interleukin-8 in human endometrium. Hum. Reprod. 9, 1406 –1409. Critchley, H.O., Kelly, R.W., Lea, R.G., Drudy, T.A., Jones, R.L., Baird, D.T., 1996. Sex steroid regulation of leukocyte traffic in human decidua. Hum. Reprod. 11, 2257 –2262. Critchley, H.O., Jones, R.L., Lea, R.G., Drudy, T.A., Kelly, R.W., Williams, A.R., Baird, D.T., 1999. Role of inflammatory mediators in human endometrium during progesterone withdrawal and early pregnancy. J. Clin. Endocrinol. Metab. 84, 240 – 248. Foster, S.J., Aked, D.M., Schroder, J.M., Christophers, E., 1989. Acute inflammatory effects of a monocyte-derived neutrophil activating peptide in rabbit skin. Immunology 67, 181–183. Greenland, K.J., Jantke, I., Jenatschke, S., Bracken, K.E., Vinson, C., Gellersen, B., 2000. The human NAD+-dependent 15-hydroxyprostaglandin dehydrogenase gene promoter is controlled by Ets and activating protein-1 transcription factors and progesterone. Endocrinology 141, 581 –597. Henn, V., Slupsky, J.R., Grafe, M., Anagnostopoulos, I., Forster, R., Muller-Berghaus, G., Kroczek, R.A., 1998. CD40 ligand on activated platelets triggers an inflammatory reaction of endothelial cells. Nature 391, 591 –594. Hirschi, K.K., D’Amore, P.A., 1996. Pericytes in the microvasculature. Cardiovasc. Res. 32, 687–698. Irwin, J.C., Utian, W.H., Eckert, R.L., 1991. Sex steroids and growth factors differentially regulate the growth and differentiation of cultured human endometrial stromal cells. Endocrinology 129, 2385 –2392. Jones, R.L., Kelly, R.W., Critchley, H.O., 1997. Chemokine and cyclooxygenase-2 expression in human endometrium coincides with leukocyte accumulation. Hum. Reprod. 12, 1300 –1306. Kalkhoven, E., Wissink, S., Vandersaag, P.T., Vanderburg, B., 1996. Negative interaction between the relA(P65) subunit of NF-kB and the progesterone-receptor. J. Biol. Chem. 271, 6217 –6224. Kelly, R.W., Bukman, A., 1990. Antiprogestagenic inhibition of uterine prostaglandin inactivation: a permissive mechanism for uterine stimulation. J. Steroid Biochem. 37, 97–101.
92
R.W. Kelly et al. / Journal of Reproducti6e Immunology 57 (2002) 81–93
Kelly, R.W., King, A.E., Critchley, H.O., 2001. Cytokine control in human endometrium. Reproduction 121, 3 –19. King, A., Gardner, L., Loke, Y.W., 1996. Evaluation of oestrogen and progesterone receptor expression in uterine mucosal lymphocytes. Hum. Reprod. 11, 1079 –1082. King, A.E., Kelly, R.W., Critchley, H.O., Malmstrom, A., Sennstrom, M., Phipps, R.P., 2001. CD40 expression in uterine tissues: a key regulator of cytokine expression by fibroblasts. J. Clin. Endocrinol. Metab. 86, 405 – 412. Kitaya, K., Yasuda, J., Yagi, I., Tada, Y., Fushiki, S., Honjo, H., 2000. IL-15 expression at human endometrium and decidua. Biol. Reprod. 63, 683 –687. Kohnen, G., Campbell, S., Jeffers, M.D., Cameron, I.T., 2000. Spatially regulated differentiation of endometrial vascular smooth muscle cells. Hum. Reprod. 15, 284 –292. Koji, T., Chedid, M., Rubin, J.S., Slayden, O.D., Csaky, K.G., Aaronson, S.A., Brenner, R.M., 1994. Progesterone-dependent expression of keratinocyte growth factor mRNA in stromal cells of the primate endometrium: keratinocyte growth factor as a progestomedin. J. Cell Biol. 125, 393 – 401. Koumas, L., King, A.E., Critchley, H.O., Kelly, R.W., Phipps, R.P., 2001. Fibroblast heterogeneity: existence of functionally distinct Thy 1( +) and Thy 1(−) human female reproductive tract fibroblasts. Am. J. Pathol. 159, 925 – 935. Loke, Y.W., King, A., 1995. Human Implantation. Cambridge University Press, Cambridge. Milne, S.A., Critchley, H.O., Drudy, T.A., Kelly, R.W., Baird, D.T., 1999. Perivascular interleukin-8 messenger ribonucleic acid expression in human endometrium varies across the menstrual cycle and in early pregnancy decidua. J. Clin. Endocrinol. Metab. 84, 2563–2567. Noyes, R.W., Hertig, A.T., Rock, J., 1950. Dating the endometrial biopsy. Fertil. Steril. 1, 3–25. Okada, H., Nakajima, T., Sanezumi, M., Ikuta, A., Yasuda, K., Kanzaki, H., 2000a. Progesterone enhances interleukin-15 production in human endometrial stromal cells in vitro. J. Clin. Endocrinol. Metab. 85, 4765 –4770. Okada, S., Okada, H., Sanezumi, M., Nakajima, T., Yasuda, K., Kanzaki, H., 2000b. Expression of interleukin-15 in human endometrium and decidua. Mol. Hum. Reprod. 6, 75–80. Oliver, C., Montes, M.J., Galindo, J.A., Ruiz, C., Olivares, E.G., 1999. Human decidual stromal cells express a-smooth muscle actin and show ultrastructural similarities with myofibroblasts. Hum. Reprod. 14, 1599 – 1605. Perrot-Applanat, M., Groyer-Picard, M.T., Garcia, E., Lorenzo, F., Milgrom, E., 1988. Immunocytochemical demonstration of estrogen and progesterone receptors in muscle cells of uterine arteries in rabbits and humans. Endocrinology 123, 1511 –1519. Perrot-Aplanat, M., Deng, M., Fernandez, H., Lelaidier, C., Meduri, G., Bouchard, P., 1994. Immunohistochemical localization of estradiol and progesterone receptors in human uterus throughout pregnancy: expression in endometrial blood vessels. J. Clin. Endocrinol. Metab. 78, 216 –224. Rampart, M., Van Damme, J., Zonnekyn, L., Herman, A.G., 1989. Granulocyte chemotactic protein/interleukin-8 induces plasma leakage and neutrophil accumulation in rabbit skin. Am. J. Pathol. 135, 21 –25. Roberts, D.K., Parmley, T.H., Walker, N.J., Horbelt, D.V., 1992. Ultrastructure of the microvasculature in the human endometrium throughout the normal menstrual cycle. Am. J. Obstet. Gynecol. 166, 1393 –1406. Rogers, P.A., Plunkett, D., Affandi, B., 2000. Perivascular smooth muscle a-actin is reduced in the endometrium of women with progestin — only contraceptive breakthrough bleeding. Hum. Reprod. 15 (Suppl. 3), 78 – 84.
R.W. Kelly et al. / Journal of Reproducti6e Immunology 57 (2002) 81–93
93
Staples, L.D., Heap, R.B., Wooding, F.B.P., King, G.J., 1983. Migration of leukocytes into the uterus after acute removal of ovarian progesterone during early pregnancy in the sheep. Placenta 4, 339 –350. Stewart, J.A., Bulmer, J.N., Murdoch, A.P., 1998. Endometrial leucocytes: expression of steroid hormone receptors. J. Clin. Pathol. 51, 121 – 126. Verma, S., Hiby, S.E., Loke, Y.W., King, A., 2000. Human decidual natural killer cells express the receptor for and respond to the cytokine interleukin 15. Biol. Reprod. 62, 959–968. Wang, H., Critchley, H.O.D., Kelly, R.W., Shen, D., Baird, D.T., 1998. Progesterone receptor subtype B is differentially regulated in human endometrial stroma. Mol. Hum. Reprod. 4, 407 –412. Washizu, J., Nishimura, H., Nakamura, N., Nimura, Y., Yoshikai, Y., 1998. The NF-kB binding site is essential for transcriptional activation of the IL-15 gene. Immunogenetics 48, 1 –7. Wedmore, C.V., Williams, T.J., 1981. Control of vascular permeability by polymorphonuclear leukocytes in inflammation. Nature 289, 646 – 650. Wissink, S., van Heerde, E.C., van der Burg, B., van der Saag, P.T., 1998. A dual mechanism mediates repression of NF-kB activity by glucocorticoids. Mol. Endocrinol. 12, 355 – 363. Zhang, Y., Cao, H.J., Graf, B., Meekins, H., Smith, T.J., Phipps, R.P., 1998. CD40 engagement up-regulates cyclooxygenase-2 expression and prostaglandin E2 production in human lung fibroblasts. J. Immunol. 160, 1053 –1057.
Inflammatory mediators and endometrial function—focus on the perivascular cell Rodney W. Kelly a,*, Anne E. King a, Hilary O.D. Critchley b a
Human Reproducti6e Sciences Unit, Medical Research Council, Centre for Reproducti6e Biology, Uni6ersity of Edinburgh, 37 Chalmers Street, Edinburgh EH3 9ET, UK b Department of Reproducti6e and De6elopmental Science, Centre for Reproducti6e Biology, Uni6ersity of Edinburgh, 37 Chalmers Street, Edinburgh EH3 9ET, UK
Abstract Human endometrium has a unique vascular architecture that allows menstruation, the shedding of a well-vascularised tissue layer, with limited bleeding. Blood loss is controlled at least in part by constriction of the perivascular cells, myofibroblasts that surround the spiral arterioles and have contractile activity. These perivascular cells, which are coupled to endothelial cells by processes, are responsive to changes in progesterone levels and express chemokines, cytokines and prostaglandins (PG) crucial to the control of leukocyte entry into endometrium. In this location the chemokine interleukin-8 (IL-8) and prostaglandin E (PGE) will have synergistic effects on leukocyte entry. CD40 is also expressed on the perivascular cells. Activation of CD40 by CD40 ligand is known to increase COX-2 and IL-8 expression in endometrial fibroblasts. The likely source of CD40 ligand in the uterus is platelets. Thus ingress of platelets will up-regulate NFkB by activating CD40 and increase agents such as PGE which will stimulate further the ingress of platelets. There is thus the possibility of a spiralling inflammatory response. This response however, is normally modulated by progesterone raising the threshold of the NFkB pathway and in the presence of high progesterone levels activation of CD40 will be ineffective. When progesterone falls at the end of the ovarian cycle and the restrictions on activation are lost, the perivascular cells will respond, initiating leukocyte entry, vasoconstriction–vasodilatation cycles with associated hypoxia and consequent sloughing off of the endometrium. The perivascular cell in endometrium is
* Corresponding author. E-mail address: [email protected] (R.W. Kelly). 0165-0378/02/$ - see front matter © 2002 Elsevier Science Ireland Ltd. All rights reserved. PII: S 0 1 6 5 - 0 3 7 8 ( 0 2 ) 0 0 0 0 8 - 6
82
R.W. Kelly et al. / Journal of Reproducti6e Immunology 57 (2002) 81–93
pivotal in both menstruation and early pregnancy and we need to understand this cell better to devise more effective medical treatment for menstrual dysfunction. © 2002 Elsevier Science Ireland Ltd. All rights reserved. Keywords: Inflammatory mediators; Endometrial function; Perivascular cell
1. Introduction One major function of the vasculature of endometrium is to restrict blood loss at the time of menstruation. The means by which this is achieved and the hormonal control of the process are still poorly understood but the cells with smooth muscle character that surround the blood vessels are likely to be relevant to this mechanism. In addition, cells in this location are a source of cytokines such as interleukin-15 (IL-15) which could stimulate growth of sub-populations of stromal cells. This review identifies cytokines and prostaglandins which are released by these cells and relates these events to hormonal changes and the initiation of menstruation.
2. The perivascular cells of human endometrium Although there is heterogeneity of the perivascular cells of the human endometrium (Roberts et al., 1992) this may reflect the heterogeneity found among pericytes in general (Hirschi and D’Amore, 1996). These cells in endometrium maintain close contact with endothelial cells and with each other by means of specialised projections which are maintained throughout the cycle (Roberts et al., 1992). The distribution of the progesterone receptors which are prominent in the perivascular cells but absent in the endothelium (Kohnen et al., 2000) suggests that progesterone induced changes are primarily mediated via the perivascular rather than the endothelial cell. An area of increasing interest is the interaction between the perivascular cells— a source of IL-15, the CD56++ cells —a source of IFN-g and the endothelial cells. A proper interaction is necessary for the IFN-g-dependent decidual blood vessel changes at the time of implantation (Ashkar et al. 2000). Cells associated with small blood vessels vary in character depending on location and those found in human endometrium are myofibroblasts in that they express smooth muscle a-actin and the fibroblast marker vimentin (Kohnen et al., 2000). These endometrial cells express a wide range of cytokines and vasoactive factors and are restricted to a discrete area surrounding the blood vessels and thus the term perivascular cell is used
R.W. Kelly et al. / Journal of Reproducti6e Immunology 57 (2002) 81–93
83
here. The group of cells, with a clearly perivascular location which stain for cytokines and prostaglandin (Fig. 1) may well include pericytes or be differentiated from these pluripotent cells (Hirschi and D’Amore, 1996). Such cells have a sentinel role in control of oedema and cellular ingress into the tissue but they are also thought to be associated with early signs of decidualisation. Whether these cells are the origin of the decidual cell is controversial but the decidual cell of early pregnancy has myofibroblast characteristics (Oliver et al., 1999) and the process of decidualisation commences around the blood vessels. In addition, in other sites, cell types such as adipocytes are known to originate from perivascular cells (Hirschi and D’Amore, 1996). The cells surrounding the spiral arterioles in human endometrium are a source of inflammatory mediators and there is good evidence that the expression of some of these mediators is controlled by progesterone. The signal for menstruation is the fall in ovarian progesterone and the perivascular cells will respond to this change with cytokine release and increased permeability. In contrast, high progesterone levels in early pregnancy decidua will regulate cell migration into the tissue and maintain pregnancy.
3. Control of leukocyte influx Prostaglandins (PGs) are very relevant to the function of the perivascular cell since they are important mediators of inflammation and in human endometrium PGs are under hormonal control. Non-steroidal anti-inflammatory drugs, which inhibit prostaglandin production, have been widely used to alleviate both dysmenorrhoea and menorrhagia (Cameron et al., 1990) and the likely site of action of such agents is at or around the blood vessels. The impact of prostaglandins on inflammatory mediators is synergistic: PGE is known to accentuate the effects of the inflammatory peptide bradykinin in oedema induction and both oedema and influx of polymorphonuclear leukocytes is synergistically increased by the combination of interleukin-8 (IL-8) and PGE (Colditz, 1990; Foster et al., 1989; Rampart et al., 1989). Since the expression of IL-8 in human endometrium is very marked in perivascular cells (Critchley et al., 1994; Milne et al., 1999), the concerted induction of both PGE and IL-8 would be expected to be a powerful drive for neutrophil influx into the tissue. In a hormonally responsive tissue such as endometrium, the ingress of leukocytes is likely to be modulated by steroids. This influx is indirectly mediated by steroids since the leukocytes of endometrium do not express the progesterone receptor (King et al., 1996; Stewart et al., 1998) whereas the stromal cells around the vessels maintain these receptors (Fig. 1) even in the secretory phase when
R.W. Kelly et al. / Journal of Reproducti6e Immunology 57 (2002) 81–93
Fig. 1.
84
R.W. Kelly et al. / Journal of Reproducti6e Immunology 57 (2002) 81–93
85
progesterone receptors are lost from the glands. The relevance of neutrophil entry into tissue is that the neutrophil is a potent source of enzymes, particularly collagenase and elastase. Collagenase from an activated neutrophil will disrupt inter-cellular anchoring and thus facilitate the process of tissue shedding. A further role for the neutrophil is to exert a secondary effect on blood vessels and increase their permeability (Wedmore and Williams, 1981) and thus augment oedema.
4. Progesterone control in perivascular cells The majority of perivascular cells in endometrium have abundant progesterone receptors (Fig. 1). Progesterone receptors have been identified in this location throughout the cycle and in early pregnancy (Perrot-Aplanat et al., 1994; Perrot-Applanat et al., 1988; Wang et al., 1998) and the intense staining in this location led to the conclusion that progesterone acting on these cells was essential for modulating blood-flow during pregnancy (Perrot-Aplanat et al., 1994), a point that is underscored by the absence of progesterone receptors from the placenta. Progesterone is also implicated in the stimulation of proliferation of smooth-muscle a-actin + ve cells that surround the spiral arterioles (Rogers et al., 2000) and fewer of these cells have been observed in women suffering from menorrhagia (Abberton et al., 1999). The suggestion has been made that the growth of these cells must accord with the development of the spiral arterioles and may be related to specific, progesterone induced, growth factors such as keratinocyte growth factor (FGF-7; Koji et al., 1994) and any defect could lead to impaired control of blood loss and consequent menstrual aberration. The effects of progesterone on the perivascular cell are revealed when anti-progestins such as RU486 (Mifepistone) are administered in vivo. The appearance of PGE in characteristic whorls around the small blood vessels
Fig. 1. Perivascular distribution of prostaglandins and cytokines. The perivascular cells of human endometrium possess progesterone receptors (g) endothelial cells (E) and perivascular cells (P) are marked. Although CD40 appears to be present at all stages of the menstrual cycle (f) cyclooxygenase (COX)-2 can be induced in vitro treatment with phorbol ester (e). In decidua, progesterone withdrawal with RU486 leads to an increased expression of PGE (d) and a decreased expression of 13,14-dihydro-15-keto PGE (c), the main PGE metabolite (PGEM). IL-8 is also prominent in the small blood vessels in decidua (b). Panel (a) shows CD45+ve (leukocyte common antigen) leukocytes clustered around small blood vessels in an endometrial sample from a patient exposed to an anti-progestin 24 h earlier. Scale bar=100 mM.
86
R.W. Kelly et al. / Journal of Reproducti6e Immunology 57 (2002) 81–93
of decidua follows 24 h after the progesterone antagonist is administered (Cheng et al., 1993a,b; Fig. 1). The mechanism by which PGE appearance is mediated is not fully understood but one element is the removal of the protecting catabolic enzyme 15-hydroxy prostaglandin dehydrogenase. Immunohistochemistry studies to detect 15-hydroxy prostaglandin dehydrogenase showed a marked absence of this enzyme around the blood vessels (Cheng et al., 1993a,b). Indirect evidence that this enzyme was induced by progesterone has been available for some time (Casey et al., 1980; Kelly and Bukman, 1990) but with the elucidation of the promoter region for this enzyme, the progesterone dependence has been made explicit (Greenland et al., 2000). Cytokine control in these perivascular cells is also progesterone dependent. Both IL-8 and monocyte chemotactic protein (MCP)-1 levels in these cells increase in the late secretory phase of the menstrual cycle at the time that leukocyte accumulation is evident (Jones et al., 1997). In both endometrium (Critchley et al., 1999) and decidua (Critchley et al., 1996), IL-8 rises perivascularly after progesterone antagonism with RU486. Thus progesterone appears to be maintaining low levels of chemokines and leukocyte concentration in endometrium and decidua. Such findings are consistent with elegant earlier animal data that show leukocyte entry into endometrium to be dependent on progesterone withdrawal (Staples et al., 1983). The mechanism by which progesterone modulates cytokine production is likely to be similar to the mechanisms employed by glucocorticoids, via NFkB (Caldenhoven et al., 1995). Progesterone can modulate the NFkB pathway in two ways. Progesterone bound to its receptor and NFkB both act as transcription factors competing for sites on the gene (Kalkhoven et al., 1996) or alternatively progesterone can stimulate synthesis of the inhibitory protein IkBa which maintains NFkB in the cytosol of the cell in an inactive state (Wissink et al., 1998; Fig. 2). Recent reports have shown that IL-15, a growth factor that shares the b and g receptor with IL-2, is released from the perivascular cells of human endometrium (Kitaya et al., 2000; Okada et al., 2000a,b) and that the release of this cytokine is progesterone dependent. The relevance of IL-15 is that it stimulates proliferation of the CD56+ ve large granular lymphocytes (Verma et al., 2000) that are a characteristic stromal component in late secretory phase endometrium and early decidua (Loke and King, 1995). The CD56+ve cells are associated with the spiral arterioles when first evident and this would be consistent with a perivascular location of their growth factor. IL-15 expression is thought to be under control of NFkB (Washizu et al., 1998) but this is unlikely to be the
R.W. Kelly et al. / Journal of Reproducti6e Immunology 57 (2002) 81–93
87
Fig. 2. Progesterone control mechanisms. IL-8 and PGE act synergistically to bring in leukocytes into tissue and both are controlled, at least in part, by the NFkB system. Progesterone also controls prostaglandins through induction of prostaglandin catabolism. The CD40–CD40L system in the perivascular cells would be progesterone responsive through the action of progesterone on the NFkB cascade.
mechanism of progesterone control because most effects of progesterone on the NFkB pathway are negative. One activator of the NFkB control pathway residing in the perivascular cells of human endometrium and myometrium is the CD40–CD40 ligand system. CD40 is a member of the TNF receptor family and, on ligation, is known to mediate prostaglandin synthesis by the inducible enzyme responsible for prostaglandin synthesis (COX-2; Zhang et al., 1998). This pathway can also be used to stimulate cytokine release from fibroblasts derived from human uterine tissue (King et al., 2001). However, whereas CD40 is expressed on perivascular cells and expression in cultured endometrial stromal cells (ESC) can be stimulated with interferon-g, the source of the CD40 ligand (CD40L) is not clear. One major source of CD40L is platelets (Henn et al., 1998) and thus any CD40 on the perivascular cell surface could be activated by extravasation of platelets. If the result of CD40 activation is stimulation of both prostaglandin and IL-8 then neutrophil ingress and oedema would follow. Moreover, the vasoactive effects of prostaglandin release will increase the exposure of CD40 to platelets and thus lead to a positive feedback (Figs. 3 and 4) but this will only operate in the absence of progesterone which will negate the feedforward loop at the level of NFkB and by inducing 15-hydroxy prostaglandin dehydrogenase. Thus, from the mid-secretory phase of the cycle onwards, progesterone will play an important role in maintaining the selective exclusion of leukocytes, keeping neutrophil numbers low but allowing entry of CD56+ ve cells.
88
R.W. Kelly et al. / Journal of Reproducti6e Immunology 57 (2002) 81–93
5. Models to study perivascular cell responses Cells grown from stroma of human endometrium resemble fibroblasts in morphology and are heterogeneous (Koumas et al., 2001). A feature of the ESC is that a sub-population can be grown in a progesterone and growth factor dependent manner. This parallels the in vivo situation where perivascular smooth muscle cells proliferate at the time of high progesterone concentrations (Abberton et al., 1999; Koji et al., 1994). In vitro, in the presence of growth factors such as epidermal growth factor (EGF) and basic-fibroblast growth factor (bFGF) growth is progesterone dependent (Irwin et al., 1991). Cells grown in the presence of bFGF do not decidualise even when high progesterone levels are maintained (Irwin et al., 1991). The classical trade-off between proliferation and differentiation appears to be present in this situation since progesterone will cause decidualisation, but in the absence of growth factors—usually engineered in vitro by low fetal-calf serum concentrations. In human endometrium pre-decidualisation occurs in the secretory phase (Noyes et al., 1950) and in early pregnancy this process develops with the earliest signs of differentiation seen in the cells surrounding the spiral arterioles (Buckley and Fox, 1989). Although the in vivo initiator of decidualisation is likely to be progesterone, in vitro experiments underline the role of cAMP in inducing this change in endometrium stromal cells
Fig. 3. Perivascularrole of the CD40 – CD40 ligand system. One activator of the CD40 on perivascular cells could be the CD40 ligand derived from activated platelets. This would lead to prostaglandin and cytokine production which would cooperatively induce increased vascular permeability thus leading to a feed forward loop. But, in the presence of progesterone this pathway would be blocked. Progesterone withdrawal would be permissive for an activation of the system which would lead to leukocyte entry and menstruation.
R.W. Kelly et al. / Journal of Reproducti6e Immunology 57 (2002) 81–93
89
Fig. 4. Differential control of cytokine release from ESC — the effect of decidualisation. IL-8 and MCP-1 release after 24 h in culture from five separate ESC preparations grown 3 – 5 weeks with progestins (MPA) and growth factor (bFGF). Cells were decidualised (or not) with 8-bromo cAMP (0.25 mM) and MPA for 5 days prior to treatment with IFN-g 25 ng/ml and/or CD40L (10 ng/ml) (Alexis). CD40 is expressed only when cells are stimulated with IFN-g and progesterone receptor is unaffected by decidualisation (data not shown). Both IL-8 and MCP-1 were measured in the culture medium after 24 h by ELISA. Results are shown as mean 9S.E.M., N=5. **, Represents difference from respective control PB0.005; *, difference B0.05; ‘a’ denotes difference in IL-8 release resulting from CD40 ligand treatment in decidualised and non-decidualised cells PB 0.001 and ‘b’ represents difference in MCP-1 concentrations between CD40L + IFN-g in decidualised and non-decidualised cells PB0.05.
(Brosens et al., 1999) which is characterised by prolactin and insulin-like growth factor binding protein-1 (IGFBP-1) release. In vivo, the mechanism by which progesterone induces pre-decidualisation and decidualisation in human endometrium and decidua is poorly understood. However the effects of progesterone and PGE are likely to be involved and the modulation of cAMP will be enhanced by the action of progesterone stimulating the expression of the EP2 (cyclic AMP coupled) receptor (Brar et al., 1997). ESC grown in the presence of bFGF and progestin are one model for examining cytokine expression in response to mediators such as CD40 L. Such cells express CD40 in response to interferon-g and the action of CD40 L on these cells is to stimulate IL-8 release (King et al., 2001). The monocyte-chemoattractant, MCP-1 is not significantly stimulated by CD40L but is released in response to interferon-g. Interestingly, decidualisation of these cells by treatment with cAMP analogues reduces IL-8 release in response to stimulation but has no effect on MCP-1 levels.
90
R.W. Kelly et al. / Journal of Reproducti6e Immunology 57 (2002) 81–93
6. Summary Given the assumed importance of the perivascular cells in controlling blood loss both at the time of menstruation and also at the time of parturition, we need to examine further the role of these cells in the initiation of menstruation (Kelly et al., 2001) and further studies must examine the effect of cell–cell contact between endothelial cells and perivascular myofibroblasts. In addition, we need to elucidate the factors altering the ratio of smooth-muscle to fibroblast character in endometrium stromal cells. Clinical treatments for excessive blood-loss (menorrhagia) include progestins-loaded intrauterine systems which induce a regressed endometrium, anti-fibrinolytic agents to stabilise blood-clotting and nonsteroidal anti-inflammatory agents which reduce prostaglandin synthesis. This last approach will act directly on the perivascular cells, those cells that surround the spiral arterioles of the human endometrium. New medical approaches to menorrhagia would be invaluable and agents that inhibit the CD40 – CD40-ligand system are potential treatments that would reduce the sensitivity of the vasoactive system. Finally, it is essential to understand the effect of progesterone on these cells since the retention of the progesterone receptor in the presence of high progesterone concentrations means that differentiation of these cells (e.g. by cAMP) could not only induce decidualisation by progesterone but also effect other changes in cell phenotype that will allow growth factors such as IL-15 to be released.
References Abberton, K.M., Taylor, N.H., Healy, D.L., Rogers, P.A., 1999. Vascular smooth muscle cell proliferation in arterioles of the human endometrium. Hum. Reprod. 14, 1072 –1079. Ashkar, A.A., Di Santo, J.P., Croy, B.A., 2000. Interferon-g contributes to initiation of uterine vascular modification, decidual integrity, and uterine natural killer cell maturation during normal murine pregnancy. J. Exp. Med. 192, 259 –270. Brar, A.K., Frank, G.R., Kessler, C.A., Cedars, M.I., Handwerger, S., 1997. Progesteronedependent decidualization of the human endometrium is mediated by cAMP. Endocrine 6, 301 –307. Brosens, J.J., Hayashi, N., White, J.O., 1999. Progesterone receptor regulates decidual prolactin expression in differentiating human endometrial stromal cells. Endocrinology 140, 4809 –4820. Buckley, C.H., Fox, H., 1989. Biopsy Pathology of the Endometrium, vol. p45. Chapman & Hall, London. Caldenhoven, E., Liden, J., Wissink, S., Vandestolpe, A., Raaijmakers, J., Koenderman, L., Okret, S., Gustafsson, J.A., Vandersaag, P.T., 1995. Negative cross-talk between RelA
R.W. Kelly et al. / Journal of Reproducti6e Immunology 57 (2002) 81–93
91
and the glucocorticoid receptor —a possible mechanism for the antiinflammatory action of glucocorticoids. Mol. Endocrinol. 9, 401 –412. Cameron, I.T., Haining, R., Lumsden, M.A., Thomas, V.R., Smith, S.K., 1990. The effects of mefenamic acid and norethisterone on measured menstrual blood loss. Obstet. Gynecol. 76, 85 –88. Casey, M.L., Hemsell, D.L., Johnston, J.M., MacDonald, P.C., 1980. NAD dependent 15-hydroxyprostaglandin dehydrogenase activity in human endometrium. Prostaglandins 19, 115 –122. Cheng, L., Kelly, R.W., Thong, K.J., Hume, R., Baird, D.T., 1993a. The effect of mifepristone (RU486) on the immunohistochemical distribution of prostaglandin E and its metabolite in decidual and chorionic tissue in early pregnancy. J. Clin. Endocrinol. Metab. 77, 873 –877. Cheng, L., Kelly, R.W., Thong, K.J., Hume, R., Baird, D.T., 1993b. The effects of mifepristone (RU486) on prostaglandin dehydrogenase in decidual and chorionic tissue in early pregnancy. Hum. Reprod. 8, 705 –709. Colditz, I.G., 1990. Effect of exogenous prostaglandin E2 and actinomycin D on plasma leakage induced by neutrophil activating peptide-1/interleukin-8. Immunol. Cell Biol. 68, 397–403. Critchley, H.O.D., Kelly, R.W., Kooy, J., 1994. Perivascular expression of chemokine interleukin-8 in human endometrium. Hum. Reprod. 9, 1406 –1409. Critchley, H.O., Kelly, R.W., Lea, R.G., Drudy, T.A., Jones, R.L., Baird, D.T., 1996. Sex steroid regulation of leukocyte traffic in human decidua. Hum. Reprod. 11, 2257 –2262. Critchley, H.O., Jones, R.L., Lea, R.G., Drudy, T.A., Kelly, R.W., Williams, A.R., Baird, D.T., 1999. Role of inflammatory mediators in human endometrium during progesterone withdrawal and early pregnancy. J. Clin. Endocrinol. Metab. 84, 240 – 248. Foster, S.J., Aked, D.M., Schroder, J.M., Christophers, E., 1989. Acute inflammatory effects of a monocyte-derived neutrophil activating peptide in rabbit skin. Immunology 67, 181–183. Greenland, K.J., Jantke, I., Jenatschke, S., Bracken, K.E., Vinson, C., Gellersen, B., 2000. The human NAD+-dependent 15-hydroxyprostaglandin dehydrogenase gene promoter is controlled by Ets and activating protein-1 transcription factors and progesterone. Endocrinology 141, 581 –597. Henn, V., Slupsky, J.R., Grafe, M., Anagnostopoulos, I., Forster, R., Muller-Berghaus, G., Kroczek, R.A., 1998. CD40 ligand on activated platelets triggers an inflammatory reaction of endothelial cells. Nature 391, 591 –594. Hirschi, K.K., D’Amore, P.A., 1996. Pericytes in the microvasculature. Cardiovasc. Res. 32, 687–698. Irwin, J.C., Utian, W.H., Eckert, R.L., 1991. Sex steroids and growth factors differentially regulate the growth and differentiation of cultured human endometrial stromal cells. Endocrinology 129, 2385 –2392. Jones, R.L., Kelly, R.W., Critchley, H.O., 1997. Chemokine and cyclooxygenase-2 expression in human endometrium coincides with leukocyte accumulation. Hum. Reprod. 12, 1300 –1306. Kalkhoven, E., Wissink, S., Vandersaag, P.T., Vanderburg, B., 1996. Negative interaction between the relA(P65) subunit of NF-kB and the progesterone-receptor. J. Biol. Chem. 271, 6217 –6224. Kelly, R.W., Bukman, A., 1990. Antiprogestagenic inhibition of uterine prostaglandin inactivation: a permissive mechanism for uterine stimulation. J. Steroid Biochem. 37, 97–101.
92
R.W. Kelly et al. / Journal of Reproducti6e Immunology 57 (2002) 81–93
Kelly, R.W., King, A.E., Critchley, H.O., 2001. Cytokine control in human endometrium. Reproduction 121, 3 –19. King, A., Gardner, L., Loke, Y.W., 1996. Evaluation of oestrogen and progesterone receptor expression in uterine mucosal lymphocytes. Hum. Reprod. 11, 1079 –1082. King, A.E., Kelly, R.W., Critchley, H.O., Malmstrom, A., Sennstrom, M., Phipps, R.P., 2001. CD40 expression in uterine tissues: a key regulator of cytokine expression by fibroblasts. J. Clin. Endocrinol. Metab. 86, 405 – 412. Kitaya, K., Yasuda, J., Yagi, I., Tada, Y., Fushiki, S., Honjo, H., 2000. IL-15 expression at human endometrium and decidua. Biol. Reprod. 63, 683 –687. Kohnen, G., Campbell, S., Jeffers, M.D., Cameron, I.T., 2000. Spatially regulated differentiation of endometrial vascular smooth muscle cells. Hum. Reprod. 15, 284 –292. Koji, T., Chedid, M., Rubin, J.S., Slayden, O.D., Csaky, K.G., Aaronson, S.A., Brenner, R.M., 1994. Progesterone-dependent expression of keratinocyte growth factor mRNA in stromal cells of the primate endometrium: keratinocyte growth factor as a progestomedin. J. Cell Biol. 125, 393 – 401. Koumas, L., King, A.E., Critchley, H.O., Kelly, R.W., Phipps, R.P., 2001. Fibroblast heterogeneity: existence of functionally distinct Thy 1( +) and Thy 1(−) human female reproductive tract fibroblasts. Am. J. Pathol. 159, 925 – 935. Loke, Y.W., King, A., 1995. Human Implantation. Cambridge University Press, Cambridge. Milne, S.A., Critchley, H.O., Drudy, T.A., Kelly, R.W., Baird, D.T., 1999. Perivascular interleukin-8 messenger ribonucleic acid expression in human endometrium varies across the menstrual cycle and in early pregnancy decidua. J. Clin. Endocrinol. Metab. 84, 2563–2567. Noyes, R.W., Hertig, A.T., Rock, J., 1950. Dating the endometrial biopsy. Fertil. Steril. 1, 3–25. Okada, H., Nakajima, T., Sanezumi, M., Ikuta, A., Yasuda, K., Kanzaki, H., 2000a. Progesterone enhances interleukin-15 production in human endometrial stromal cells in vitro. J. Clin. Endocrinol. Metab. 85, 4765 –4770. Okada, S., Okada, H., Sanezumi, M., Nakajima, T., Yasuda, K., Kanzaki, H., 2000b. Expression of interleukin-15 in human endometrium and decidua. Mol. Hum. Reprod. 6, 75–80. Oliver, C., Montes, M.J., Galindo, J.A., Ruiz, C., Olivares, E.G., 1999. Human decidual stromal cells express a-smooth muscle actin and show ultrastructural similarities with myofibroblasts. Hum. Reprod. 14, 1599 – 1605. Perrot-Applanat, M., Groyer-Picard, M.T., Garcia, E., Lorenzo, F., Milgrom, E., 1988. Immunocytochemical demonstration of estrogen and progesterone receptors in muscle cells of uterine arteries in rabbits and humans. Endocrinology 123, 1511 –1519. Perrot-Aplanat, M., Deng, M., Fernandez, H., Lelaidier, C., Meduri, G., Bouchard, P., 1994. Immunohistochemical localization of estradiol and progesterone receptors in human uterus throughout pregnancy: expression in endometrial blood vessels. J. Clin. Endocrinol. Metab. 78, 216 –224. Rampart, M., Van Damme, J., Zonnekyn, L., Herman, A.G., 1989. Granulocyte chemotactic protein/interleukin-8 induces plasma leakage and neutrophil accumulation in rabbit skin. Am. J. Pathol. 135, 21 –25. Roberts, D.K., Parmley, T.H., Walker, N.J., Horbelt, D.V., 1992. Ultrastructure of the microvasculature in the human endometrium throughout the normal menstrual cycle. Am. J. Obstet. Gynecol. 166, 1393 –1406. Rogers, P.A., Plunkett, D., Affandi, B., 2000. Perivascular smooth muscle a-actin is reduced in the endometrium of women with progestin — only contraceptive breakthrough bleeding. Hum. Reprod. 15 (Suppl. 3), 78 – 84.
R.W. Kelly et al. / Journal of Reproducti6e Immunology 57 (2002) 81–93
93
Staples, L.D., Heap, R.B., Wooding, F.B.P., King, G.J., 1983. Migration of leukocytes into the uterus after acute removal of ovarian progesterone during early pregnancy in the sheep. Placenta 4, 339 –350. Stewart, J.A., Bulmer, J.N., Murdoch, A.P., 1998. Endometrial leucocytes: expression of steroid hormone receptors. J. Clin. Pathol. 51, 121 – 126. Verma, S., Hiby, S.E., Loke, Y.W., King, A., 2000. Human decidual natural killer cells express the receptor for and respond to the cytokine interleukin 15. Biol. Reprod. 62, 959–968. Wang, H., Critchley, H.O.D., Kelly, R.W., Shen, D., Baird, D.T., 1998. Progesterone receptor subtype B is differentially regulated in human endometrial stroma. Mol. Hum. Reprod. 4, 407 –412. Washizu, J., Nishimura, H., Nakamura, N., Nimura, Y., Yoshikai, Y., 1998. The NF-kB binding site is essential for transcriptional activation of the IL-15 gene. Immunogenetics 48, 1 –7. Wedmore, C.V., Williams, T.J., 1981. Control of vascular permeability by polymorphonuclear leukocytes in inflammation. Nature 289, 646 – 650. Wissink, S., van Heerde, E.C., van der Burg, B., van der Saag, P.T., 1998. A dual mechanism mediates repression of NF-kB activity by glucocorticoids. Mol. Endocrinol. 12, 355 – 363. Zhang, Y., Cao, H.J., Graf, B., Meekins, H., Smith, T.J., Phipps, R.P., 1998. CD40 engagement up-regulates cyclooxygenase-2 expression and prostaglandin E2 production in human lung fibroblasts. J. Immunol. 160, 1053 –1057.