Download PDF - Prairie Swine Centre

Placenta 30, Supplement A, Trophoblast Research, Vol. 23 (2009) S26–S31
Contents lists available at ScienceDirect
Placenta
journal homepage: www.elsevier.com/locate/placenta
Comparison of Immune Cell Recruitment and Function in Endometrium During
Development of Epitheliochorial (Pig) and Hemochorial (Mouse and Human)
Placentas
B.A. Croy a, *, J. Wessels b, N. Linton b, C. Tayade b
a
b
Department of Anatomy and Cell Biology, Queen’s University, Room 924 Botterell Hall, Kingston, ON K7L 3N6, Canada
Department of Biomedical Sciences, Ontario Veterinary College, University of Guelph, Guelph, ON, Canada N1G 2W1
a r t i c l e i n f o
a b s t r a c t
Article history:
Accepted 30 September 2008
The role of maternal immune cells in early implantation sites has received special attention from
reproductive biologists because immune cells participate in tissue transplant rejection. During normal
pregnancy, endometrial immune cells differ from those in blood by subset distribution and appear to be
activated but non-destructive of conceptuses. The immune system evolved well before placental
mammals. By comparing the regulation and functions of endometrial immune cells between species in
two phylogenetic clades that model differently evolved placental types (pig (Sus scrofa) versus mouse
(Mus musculus) and human (Homo sapiens)), we seek to understand how ‘‘non-self’’ trophoblast cells
thrive in most pregnancies. Our studies suggest recruitment of specific immune cells to conceptusassociated endometrium and immune cell-promoted endometrial angiogenesis are of key importance for
mammalian conceptus well-being.
Ó 2009 Published by IFPA and Elsevier Ltd.
Keywords:
Angiogenesis
Chemokine
Chemokine decoy
Dendritic cell
Lymphocyte
Trophoblast
Uterine Natural Killer cell
1. INTRODUCTION
The immune system evolved millions of years before placental
mammals appeared (Fig. 1A) [1,2]. Components of innate immunity,
including cells that resembled Natural Killer (NK) lymphocytes,
molecules related to NK cell receptors and the major histocompatibility complex are estimated to have evolved >600 million years ago
(Mya), prior to the first whole genome duplication. About 60 Mya
afterwards, the recombinase activating genes (RAG) appeared that
control somatic cell recombination in T and B lymphocytes, and thus
adaptive immunity. Prior to this, antigen specific receptors, clonal
selection and immunological memory would not have existed [3].
Evolution of placental mammals followed that of oviparous
mammals and began w100 Mya. It is estimated that mouse (Mus
musculus) placentation appeared w32 Mya, that of humans (Homo
sapiens) w35 Mya and that of pigs (Sus scrofa) w55 Mya [4]. These
relative relationships are shown in Fig. 1. Mammalian evolution
would then result as reproductive strategies not discarded as
incompatible with immune mechanisms and different immunological strategies may have evolved between species.
* Corresponding author. Tel.: þ1 613 533 2859; fax: þ1 613 533 2566.
E-mail address: [email protected] (B.A. Croy).
0143-4004/$ – see front matter Ó 2009 Published by IFPA and Elsevier Ltd.
doi:10.1016/j.placenta.2008.09.019
Many investigators have addressed the functions of the
maternal immune system during pregnancy in humans and mice
and found considerable similarities. Both species belong to the
molecular phylogenetic clade Euarchontoglires and are classified as
having hemochorial placentation (Fig. 1B). That is, uterine epithelial
cells are lost during conceptus implantation and an endometrial
decidual response is induced. Many fewer immunological investigations have been reported on other species. We study the pig as an
earlier placental mammal from a distinct molecular phylogenetic
clade, the Laurasiatheria [4]. Pigs use epitheliochorial placentation.
This means that epithelial cells are retained at the uterine lumen
surface, there is no induction of endometrial stromal cell decidualization and that trophoblast does not invade maternal tissue but
rather expands to cover a large maternal surface [5]. Here, our
major immunological findings to date from studies of a commercial
meat pig, the Yorkshire, are reviewed and contrasted with current
understanding of murine and human implantation site immunology. Our major approaches have been real-time polymerase
chain reaction (PCR) and immunohistochemistry (IHC) for
comparisons of gene expression between endometrium and
trophoblast from the same conceptus attachment site. In many
cases, laser capture microdissection (LCM) was used to isolate
immune cells or endometrial endothelial cells from histological
sections prepared from conceptus attachment sites. Relative gene
B.A. Croy et al. / Placenta 30, Supplement A, Trophoblast Research, Vol. 23 (2009) S26–S31
S27
2.2. Endometrial immune cells in early pig pregnancy
Fig. 1. Illustration of the estimated evolutionary time scale in millions of years (Myr)
for development of the immune system and placental mammals (left) and for the
phylogenetic clades of mammals (right). Self–non-self discrimination was the first
immune trait to evolve. It used innate immune processes such as evolution of transplantation antigens (MHC) and receptor (R)-bearing Natural Killer (NK) cells.
Appearance of the duplicated Recombinase activating genes (RAG1, RAG2) permitted
somatic recombination and evolution of adaptive immunity. RAG1 and RAG2 are used
by T and B lymphocytes to produce their wide ranges of exquisitely specific antigen
surface receptors.
expression in these pure populations of cells was compared with
that in the endometrial and trophoblast biopsies.
2. FUNCTIONAL MICRODOMAINS ARE PRESENT WITHIN THE
PORCINE GESTATIONAL UTERUS
2.1. Gene expression is not uniform throughout the porcine
conceptus attachment site
Pregnancy is 114 days in commercial pigs. Blastocysts begin to
attach at gestation day (gd)12 on the mesometrial side of the uterus
but several days elapse before firm attachment is achieved [6]. At
this time, porcine blastocysts produce oestradiol (E2), interferon
(IFN)-gamma and -delta [7]. These are not major known products of
mouse or human blastocysts [8,9]. At gd19, we examined the levels
of transcription of endometrial IFNG, tumour necrosis factor-alpha
(TNF) and vascular endothelial growth factor (VEGFA), products of
murine and human uterine Natural Killer (uNK) cells. In comparison to mesometrial samples from virgin gilts (female pigs at
puberty), gd19 mesometrial tissue had significantly elevated transcription of these three genes [10]. In anti-mesometrial endometrium, only transcription of TNF was increased. A greater increase in
TNF occurred mesometrially. Thus, positionally-defined microdomains of immune gene expression occur in porcine endometrium during conceptus attachment.
The mouse blastocyst attaches on the opposite side of the uterus
(anti-mesometrial), inducing there, decidualization and changes in
gene transcription. Lateral regions of the uterus then decidualize
and finally the mesometrial region [11] where, in mice and pigs, the
suspensory mesentery serves as a conduit for the uterine vessels
that will supply placentae. No decidualization occurs between
mouse implantation sites.
We examined gene expression in mesometrial pig endometrium
between gd19 conceptus attachment sites. Neither IFNG nor VEGFA
transcription was elevated above that in non-pregnant tissue but
TNF transcripts were equivalent in abundance between and at
attachment sites [10]. This suggests gestational TNF transcription in
porcine endometrium may be induced endocrinologically rather
than by conceptus attachment.
Oestrous cycles and pregnancy modify immune cell subsets in
the porcine endometrium and uterine epithelium [12,13]. Again,
this is similar to mice, humans and many other mammals. We
observed that pregnancy induced mild skewing in these subsets
with enrichment of uNK cells and their transient expression of
cytotoxicity from gd12 that was abruptly terminated at gd28
[14,15]. Porcine uNK cells show association with uterine glands as
seen in humans [16] but not mice, and with blood vessels, as seen in
both humans and mice [16,17]. Porcine uNK cells are found below
the luminal epithelium, absent in humans and mice as well as in the
endometrial stroma, equivalent to the uNK cell-rich decidua basalis
of humans and mice. Uterine NK cells become abundant rapidly
during decidualization in mice (post-implantation) and humans
(late secretory phase of each menstrual cycle and early pregnancy)
but cytotoxic activity is not easily demonstrated ex vivo [16,17].
Recent work using multiparous sows of a different meat breed and
earlier study time points suggests uNK cells are lost from pig
attachment regions and move into deeper endometrial regions
[13]. This would position the porcine uNK cell population closer to
the uNK cell positions found in mice.
We reported that presence of a blastocyst was essential to shift
lymphocyte subsets in pigs [18]. This is not true in mice or humans.
In mice, artificial deciduomata induce differentiation of numerous
uNK cells, likely via initiation of synthesis of interleukin-15 (IL15),
an essential viability factor for the progenitors of NK cells [19–21].
In women, ectopic pregnancy sites are devoid of decidua and uNK
cells but both are found within the uterus as long as the ectopic
gestation continues [22]. Thus, progesterone (P4) has more obvious
effects on endometrial immune cells in mice and humans than in
pigs. This effect appears to be indirect as lymphocytes in murine
implantation sites do not express progesterone receptor (PGR) [23].
Studies in humans have suggested that immature dendritic cells
(DC) recognized by anti- CD209/DC-SIGN have a contact relationship with uNK cells in early decidua [24,25]. We found that antihuman CD209 marked a relatively rare cell of stellate morphology
in virgin and in gd20 and gd50 pig endometrium. These DC-like
cells frequently localized with blood vessels but appeared to have
random positional relationships with lymphocytes [26].
2.3. Chemokines and chemokine decoy receptors at the
maternal–fetal interface
Chemokines are small cytokine molecules important in the
positioning of immune cells, endothelial progenitors cells and other
cell types within developing tissues. There are >40 known chemokines that are subclassified structurally. An alternative classification is whether a chemokine contributes to recruitment of cells
that promote inflammation or to cells that promote homeostasis.
The former are more common. Most chemokines bind to more than
one signalling receptor but also to non-signalling, decoy receptors
[27]. Three of these are known and reported to be expressed by
mouse and human trophoblast. Decoy receptors D6 and DARC bind
and target the degradation of inflammatory chemokines [28] while
Chemocentryx decoy receptor (CCX CKR) targets degradation of
homeostatic chemokines (Fig. 2) [29,30]. Only two chemokines are
known that are not ligands of the characterized decoy receptors.
These are CCL1 and CXCL12. Decoy function is required in vivo to
resolve inflammation [31,32].
We addressed the expression of chemokine decoy receptors
in porcine attachment site endometrial biopsies and in sitematched trophoblast [33]. We successfully amplified porcine D6
and CCX CKR (GenBank accession numbers: DQ437505.1 and
EU168334.1) but were unsuccessful using multiple primer pairs,
in amplifying DARC, a gene not yet sequenced in pigs. Both D6
S28
B.A. Croy et al. / Placenta 30, Supplement A, Trophoblast Research, Vol. 23 (2009) S26–S31
Fig. 2. A list chemokines that bind to the decoy receptor molecules D6, DARC and CCX CKR. Some inflammatory chemokines bind to both D6 and DARC.
and CCX CKR were expressed by porcine trophoblast. Both genes
were also expressed at more abundant levels by co-localized
endometrium during early pregnancy [33]. We also established
that at least one ligand for all three known chemokine decoy
receptors is found in porcine endometrium and trophoblast [33].
D6 and CCX CKR were demonstrated at gd20 and 50 by IHC
using cross-reacting, anti-human reagents. Fig. 3 shows
a photomicrograph of gd50 porcine trophoblast dual labelled for
D6 and its ligand CCL4 (previously called MIP-1-beta, Act2 or
Scya4). Expression of both D6 and CCL4 by trophoblasts and by
endometrium suggests that intraepithelial as well as stromal
positioning of cells chemoresponsive to CCL4 (i.e., those
expressing CCR5), is co-regulated by D6.
Pregnancies have been studied in mice genetically ablated for
D6 [34]. Although implantation site morphology is slightly altered
in this strain, large litters are carried to term. If pregnant D6 null
mice are challenged by immunological protocols that provoke
abortion (lipopolysaccharide or human anti-phospholipid antibodies), they are exquisitely sensitive to pregnancy failure. Thus, D6
is considered a non-redundant, conceptus-sparing molecule in
pregnant mice [27,34]. Commercial meat pigs show two waves of
spontaneous fetal loss. Rescue experiments indicate these are
genetically normal conceptuses with full term potential (unpublished data). We did not find transcriptional changes in maternal or
fetal D6 during porcine peri-attachment or mid pregnancy losses. In
contrast, mid pregnancy loss was associated with both maternal
Fig. 3. Photomicrographs of D6 decoy receptor expression (A), expression of CCL4, a D6 ligand (B) and an overlay to display receptor/ligand co-expression (C) in healthy gd50
paraffin-embedded porcine trophoblast. Sections were stained with rat anti-human D6 (R&D Systems) and anti-mouse and rat CCL4 (eBioscience Inc.) antibodies. D6: green; CCL4:
red. Scale bar is 400 mm. Inset (D) is negative control. Background reactivity is non-specific binding to RBCs. Scale bar is 200 mm.
B.A. Croy et al. / Placenta 30, Supplement A, Trophoblast Research, Vol. 23 (2009) S26–S31
S29
and fetal elevation in CCX CKR [33]. This suggests that removal of
both maternal and fetal signals for homeostasis is a pathway
contributing to fetal death in pigs. CCX CKR has not been studied in
the endometrium of other species, leaving undefined the relevance
of this strategy to pregnancy failure in other mammals.
lymphocytes always made significant contribution to HIF1A abundance in endometrium. However, at sites of conceptus arrest,
lymphocytes did not transcribe HIF1A, suggesting perhaps their
early participation in blocking the maternal neoangiogenesis
required to support a conceptus.
3. IMMUNE CELL FUNCTIONS WITHIN PORCINE CONCEPTUS
ATTACHMENT SITES
3.2. Endometrial lymphocytes and DCs are angiogenic
3.1. Endometrial lymphocytes are sensors of their
microenvironment
Pattern recognition is a more primitive method of immune cell
recognition than is use of antigen specific receptors. One of the
major families of pathogen-associated recognition pattern molecules (PAMPs) is the Toll-like receptor (TLR) family, originally
described in invertebrates [35]. TLRs signal in a manner analogous
to IL1B and several TLRs play key roles in innate immune responses.
Immature DC and monocytes/macrophages have the most abundant expression of TLRs. TLRs and other pattern recognition
receptors have been described at the human and mouse maternal–
fetal interface where they are postulated to prevent microbial
infections of implantation sites [36]. First trimester human villous
and extravillous trophoblasts express TLR2 and TLR4. TLR1 and
TLR6 can form heterodimers with TLR2 and be co-localized with it
[37]. Human endometrial cells also express TLRs, particularly
lumen epithelial cells and uNK cells. The latter do not appear to
respond directly to TLR ligands but become activated to produce
IFNG by TLR-responding accessory cells [38].
TLR expression is less well studied at the mouse maternal–fetal
interface. Sobill and her colleagues examined TLR transcription and
function in cultured mouse uterine epithelial cells and found
consistent expression of TLR1–6 and sporadic expression of TLR7–9
[39]. In random-bred mice, uterine expression of TLR2, 3, 4 and 9
was reported by Gonzalez et al., between midgestation and term in
comparison to non-cycle-timed, virgin uterus. Transcripts for these
four TLRs increased gradually as pregnancy progressed. Placenta
also expressed these molecules dynamically with TLR4 decreasing
significantly in late gestation. Western blotting and IHC verified
transcription of the receptors [40]. Using LPS challenge of late
gestation, inbred mice, Salminen et al, found induction of uterine
and placental TLR2 transcription but no change in that of TLR4 [41]
while Zhang and colleagues report outcomes in DBA-mated CBA
females indicative of functional endometrial TLR3 [42].
Based upon these reports, we cloned TLRs1–10 for Sus scrofa [43]
and asked whether the TLRs expressed by human trophoblast
(TLR1, 2, 4, 6) are expressed at the porcine maternal–fetal interface.
TLR2 expression was not found but porcine trophoblast and endometrium expressed TLR1, 4 and 6. Levels of TLR1 and TLR6 were
similar in both trophoblasts and endometrium at both gd20 and 50.
In contrast, peri-attachment TLR4 was greater in endometrium than
in trophoblast while at midgestation, TLR4 transcripts had become
more abundant in trophoblast than in endometrium. CD209/DCSIGNþ cells were among the porcine endometrial cell types
expressing TLR4 and TLR6 but they did not express TLR1 [44].
Activation of TLR4 and other TLRs in mouse macrophages induces
the expression of VEGF [45].
Another molecular sensor that appears to have a key role in
human and mouse implantation sites is hypoxia inducible factor
(HIF1) [46–48]. We examined porcine HIF1A expression in virgin
endometrium and peri-implantation stage maternal and fetal
tissues. Levels of transcription rose gradually within attachment
sites (gd15, 19, 21, 23) from basal levels found in virgin endometrium and there were always relatively more endometrial than
trophoblastic transcripts [10]. Transcripts from endometrial
lymphocytes dissected at healthy attachment sites suggested that
The process of enlargement of maternal vascular supply to
placentas is different between pigs and mice. In pigs, expansion of
the sub-epithelial capillary network is the major event while in
mice, as in humans, structural and functional changes to spiral
arteries are key [17,49]. VEGFA is the key molecule maintaining
endothelial cell viability and driving their proliferation to initiate
angiogenesis. To study porcine attachment sites, primers for VEGFA,
Placenta growth factor (PGF) and VEGFRI and RII were used on
samples from virgin, gd20 and gd50 gilts. Lymphocytes had a much
greater abundance of VEGFA transcripts at all three time points than
endometrial endothelial cells and than trophoblast from matched
pregnancy sites [50]. PGF transcription was at similar abundance in
lymphocytes and in endothelial cells in virgin endometrium but,
during pregnancy, endothelium had more transcripts than
lymphocytes and both had more transcripts than trophoblast.
Lymphocytes preferentially expressed transcripts for VEGFRI rather
than VEGFRII, the reverse of trophoblast. IHC confirmed expression
of these four molecules [49,50]. Conceptus arrest was associated
with a combined transcriptional loss of VEGFA and gain of PGF in
lymphocytes and relatively constant receptor transcription. PGF has
a decoy function for VEGFA that may enhance the reduction of
endothelial cell viability achieved through loss of VEGFA
transcription [51].
In human and mouse implantation sites, lymphocytes are
reported to produce VEGFA, PGF and other angiogenic molecules
[52–54]. Disruption of this pathway has been associated with
human gestational complications, particularly pre-eclampsia ([55].
In mice, genetic loss of VEGFA is an embryonic lethal while loss of
PGF has more minor, non-lethal effects [56].
3.3. Endometrial lymphocytes transcribe FasL, Fas
and pro-inflammatory type 1 cytokines
Maturation and activation of lymphocytes leads to transcription
of death pathway effector molecules and inflammatory cytokines.
To address whether lymphocytes in virgin porcine endometrium
and at conceptus attachment sites between gd15 and 50 were fully
differentiated, activated cells, analyses of transcripts for FASLG, FAS,
IFNG and TNF were undertaken. Basal transcription of both FASLG
and FAS in lymphocytes from virgin uteri was not altered at gd15
but levels had begun to rise by gd19 and were significantly higher
by gd 21 [50]. From gd6, mouse uNK cells express FASLG in their
cytoplasmic granules and surface FAS [57]. Both molecules are also
reported as functional in human uNK cells [58–60].
Pregnancy induced a gain in porcine lymphocyte transcription
of IFNG but not of TNF unless the conceptus was experiencing
midgestation arrest [50]. Interestingly, trophoblast showed much
greater transcription of TNF than endometrium or endometrial
endothelium or lymphocytes in healthy attachment sites. From day
6, mouse uNK cells release IFNG [61] and, by gd8, TNF [62]. The
latter cytokine is granule localized [62]. Human uNK cells transcribe both molecules early during normal pregnancy [58].
4. SUMMARY
The biology of uterine and placenta attachment site specific
lymphocytes in pigs appears to be very similar to that in mice and
humans. To date, only minor differences in usage of TLRs have been
S30
B.A. Croy et al. / Placenta 30, Supplement A, Trophoblast Research, Vol. 23 (2009) S26–S31
identified as distinct. Co-ordinated maternal and fetal elevation of
CCX CKR, the decoy receptor responsible for degradation of
homeostatic chemokines, during midgestation conceptus arrest,
has been uniquely described but remains to be investigated in other
species. Given the more ancient phylogenetic development of
porcine placentation versus human and mouse, a more primitive
porcine usage of immune cells might have been expected. The clade
separation of these species also appears to be without effect,
although blastocyst attachment site positions and placental structure differ greatly. Endometrial immune cells for the species we
compared as samples from two of the four molecularly-defined
clades of placental evolution appear to be environmental sensing
cells able to participate in the process of ‘‘non-hard-wired’’ angiogenesis at individual placenta attachment sites. This convergence is
consistent with the concept that placental evolution was accompanied by modification of the functional programs of an established
primitive immune system in ways that increase the success of
placental reproduction.
5. CONFLICT OF INTEREST
The authors do not have any potential or actual personal,
political, or financial interest in the material, information, or
techniques described in this paper.
Acknowledgements
These studies have been supported by NSERC, OMAFRA, Ontario
Pork, Agriculture and Agri-Food Canada, Bioniche Life Science, Inc.
and the Canada Research Chairs’ Program. We thank Mr Kota Hatta
for assistance in preparation of the illustrations.
References
[1] Kasahara M, Suzuki T, Pasquier LD. On the origins of the adaptive immune
system: novel insights from invertebrates and cold-blooded vertebrates.
Trends Immunol 2004;25:105–11.
[2] Kasahara M. The 2R hypothesis: an update. Curr Opin Immunol 2007;19:
547–52.
[3] Du Pasquier L, Zucchetti I, De Santis R. Immunoglobulin superfamily receptors
in protochordates: before RAG time. Immunol Rev 2004;198:233–48.
[4] Murphy WJ, Eizirik E, Johnson WE, Zhang YP, Ryder OA, O’Brien SJ.
Molecular phylogenetics and the origins of placental mammals. Nature
2001;409:614–8.
[5] King GJ. Comparative placentation in ungulates. J Exp Zool 1993;266:588–602.
[6] Keys JL, King GJ, Kennedy TG. Increased uterine vascular permeability at the
time of embryonic attachment in the pig. Biol Reprod 1986;34:405–11.
[7] Cencic A, La Bonnardiere C. Trophoblastic interferon-gamma: current knowledge and possible role(s) in early pig pregnancy. Vet Res 2002;33:139–57.
[8] Sasaki N, Nagaoka S, Itoh M, Izawa M, Konno H, Carninci P, et al. Characterization of gene expression in mouse blastocyst using single-pass sequencing of
3995 clones. Genomics 1998;49:167–79.
[9] Stanton JL, Green DP. A set of 1542 mouse blastocyst and pre-blastocyst genes
with well-matched human homologues. Mol Hum Reprod 2002;8:149–66.
[10] Tayade C, Black GP, Fang Y, Croy BA. Differential gene expression in endometrium, endometrial lymphocytes, and trophoblasts during successful and
abortive embryo implantation. J Immunol 2006;176:148–56.
[11] Bany BM, Cross JC. Post-implantation mouse conceptuses produce paracrine
signals that regulate the uterine endometrium undergoing decidualization.
Dev Biol 2006;294:445–56.
[12] Bischof RJ, Brandon MR, Lee CS. Cellular immune responses in the pig uterus
during pregnancy. J Reprod Immunol 1995;29:161–78.
[13] Dimova T, Mihaylova ASP, Georgieva R. Superficial implantation in pigs is
associated with decreased numbers and redistribution of endometrial NK-cell
populations. Am J Reprod Immunol 2008;59:359–69.
[14] Engelhardt H, Croy BA, King GJ. Evaluation of natural killer cell recruitment to
embryonic attachment sites during early porcine pregnancy. Biol Reprod
2002;66:1185–92.
[15] Croy BA, Waterfield A, Wood W, King GJ. Normal murine and porcine embryos
recruit NK cells to the uterus. Cell Immunol 1988;115:471–80.
[16] Bulmer JN, Lash GE. Human uterine natural killer cells: a reappraisal. Mol
Immunol 2005;42:511–21.
[17] Croy BA, van den Heuvel MJ, Borzychowski AM, Tayade C. Uterine Natural
Killer cellsda specialized differentiation regulated by ovarian hormones.
Immunol Rev 2006;214:161–85.
[18] Engelhardt H, Croy BA, King GJ. Conceptus influences the distribution of
uterine leukocytes during early porcine pregnancy. Biol Reprod
2002;66:1875–80.
[19] Herington JL, Bany BM. Effect of the conceptus on uterine natural killer cell
numbers and function in the mouse uterus during decidualization. Biol Reprod
2007;76:579–88.
[20] Peel S. Granulated metrial gland cells. Adv Anat Embryol Cell Biol 1989;115:
1–112.
[21] Ye W, Zheng LM, Young JD, Liu CC. The involvement of interleukin (IL)-15 in
regulating the differentiation of granulated metrial gland cells in mouse
pregnant uterus. J Exp Med 1996;184:2405–10.
[22] Ho HN, Chao KH, Chen CK, Yang YS, Huang SC. Activation status of T and NK
cells in the endometrium throughout menstrual cycle and normal and
abnormal early pregnancy. Hum Immunol 1996;49:130–6.
[23] Oh MJ, Croy BA. A map of relationships between uterine natural killer cells and
progesterone receptor expressing cells during mouse pregnancy. Placenta
2008;29:317–23.
[24] Dietl J, Honig A, Kammerer U, Rieger L. Natural killer cells and dendritic cells at
the human feto-maternal interface: an effective cooperation? Placenta
2006;27:341–7.
[25] Kammerer U, Kruse A, Barrientos G, Arck PC, Blois SM. Role of dendritic cells in
the regulation of maternal immune responses to the fetus during mammalian
gestation. Immunol Invest 2008;37:499–533.
[26] Wessels JM, Linton NF, Croy BA, Tayade C. A review of molecular contrasts
between arresting and viable porcine attachment sites. Am J Reprod Immunol
2007;58:470–80.
[27] Borroni EM, Bonecchi R, Buracchi C, Savino B, Mantovani A, Locati M. Chemokine decoy receptors: new players in reproductive immunology. Immunol
Invest 2008;37:483–97.
[28] Borroni EM, Buracchi C, de la Torre YM, Galliera E, Vecchi A, Bonecchi R, et al.
The chemoattractant decoy receptor D6 as a negative regulator of inflammatory responses. Biochem Soc Trans 2006;34:1014–7.
[29] Heinzel K, Benz C, Bleul CC. A silent chemokine receptor regulates steady-state
leukocyte homing in vivo. Proc Natl Acad Sci USA 2007;104:8421–6.
[30] Haraldsen G, Rot A. Coy decoy with a new ploy: interceptor controls the levels
of homeostatic chemokines. Eur J Immunol 2006;36:1659–61.
[31] McKimmie CS, Graham GJ. Leucocyte expression of the chemokine scavenger
D6. Biochem Soc Trans 2006;34:1002–4.
[32] McKimmie CS, Fraser AR, Hansell C, Gutierrez L, Philipsen S, Connell L, et al.
Hemopoietic cell expression of the chemokine decoy receptor D6 is dynamic
and regulated by GATA1. J Immunol 2008;181:3353–63.
[33] Wessels JM. Chemokine and chemokine decoy receptor contributions to
fetal success and failure during porcine pregnancy. Thesis. University of
Guelph; 2008.
[34] Martinez dlT Buracchi C, Borroni EM, Dupor J, Bonecchi R, Nebuloni M, et al.
Protection against inflammation- and autoantibody-caused fetal loss by the
chemokine decoy receptor D6. Proc Natl Acad Sci USA 2007;104:2319–24.
[35] Muzio M, Polentarutti N, Bosisio D, Manoj Kumar PP, Mantovani A. Toll-like
receptor family and signalling pathway. Biochem Soc Trans 2000;28:563–6.
[36] Abrahams VM. Pattern recognition at the maternal–fetal interface. Immunol
Invest 2008;37:427–47.
[37] Abrahams VM, Aldo PB, Murphy SP, Visintin I, Koga K, Wilson G, et al. TLR6
modulates first trimester trophoblast responses to peptidoglycan. J Immunol
2008;180:6035–43.
[38] Eriksson M, Meadows SK, Basu S, Mselle TF, Wira CR, Sentman CL. TLRs
mediate IFN-gamma production by human uterine NK cells in endometrium.
J Immunol 2006;176:6219–24.
[39] Soboll G, Shen L, Wira CR. Expression of Toll-like receptors (TLR) and
responsiveness to TLR agonists by polarized mouse uterine epithelial cells in
culture. Biol Reprod 2006;75:131–9.
[40] Gonzalez JM, Xu H, Ofori E, Elovitz MA. Toll-like receptors in the uterus, cervix,
and placenta: is pregnancy an immunosuppressed state? Am J Obstet Gynecol
2007;197:296.
[41] Salminen A, Paananen R, Vuolteenaho R, Metsola J, Ojaniemi M, AutioHarmainen H, et al. Maternal endotoxin-induced preterm birth in mice: fetal
responses in toll-like receptors, collectins, and cytokines. Pediatr Res
2008;63:280–6.
[42] Zhang J, Wei H, Wu D, Tian Z. Toll-like receptor 3 agonist induces impairment
of uterine vascular remodeling and fetal losses in CBA x DBA/2 mice. J Reprod
Immunol 2007;74:61–7.
[43] Linton NF, Wessels JM, Cnossen SA, Croy BA, Tayade C. Immunological
mechanisms affecting angiogenesis and their relation to porcine pregnancy
success. Immunol Invest 2008;37:611–29.
[44] Linton NF. Cytokines, dendritic dells and toll-like receptors in the success and
failure of porcine pregnancy. Thesis. University of Guelph; 2008.
[45] Pinhal-Enfield G, Ramanathan M, Hasko G, Vogel SN, Salzman AL, Boons GJ,
et al. An angiogenic switch in macrophages involving synergy between Tolllike receptors 2, 4, 7, and 9 and adenosine A(2A) receptors. Am J Pathol
2003;163:711–21.
[46] Fryer BH, Simon Hypoxia MC. HIF and the placenta. Cell Cycle 2006;5:495–8.
[47] Kanasaki K, Palmsten K, Sugimoto H, Ahmad S, Hamano Y, Xie L, et al. Deficiency in catechol-O-methyltransferase and 2-methoxyoestradiol is associated
with pre-eclampsia. Nature 2008;453:1117–21.
[48] Zamudio S, Wu Y, Ietta F, Rolfo A, Cross A, Wheeler T, et al. Human placental
hypoxia-inducible factor-1alpha expression correlates with clinical outcomes
in chronic hypoxia in vivo. Am J Pathol 2007;170:2171–9.
B.A. Croy et al. / Placenta 30, Supplement A, Trophoblast Research, Vol. 23 (2009) S26–S31
[49] Winther H, Ahmed A, Dantzer V. Immunohistochemical localization of vascular
endothelial growth factor (VEGF) and its two specific receptors, Flt-1 and KDR,
in the porcine placenta and non-pregnant uterus. Placenta 1999;20:35–43.
[50] Tayade C, Fang Y, Hilchie D, Croy BA. Lymphocyte contributions to altered
endometrial angiogenesis during early and midgestation fetal loss. J Leukoc
Biol 2007;82:877–86.
[51] Carmeliet P, Moons L, Luttun A, Vincenti V, Compernolle V, De Mol M, et al.
Synergism between vascular endothelial growth factor and placental growth
factor contributes to angiogenesis and plasma extravasation in pathological
conditions. Nat Med 2001;7:575–83.
[52] Li XF, Charnock-Jones DS, Zhang E, Hiby S, Malik S, Day K, et al. Angiogenic
growth factor messenger ribonucleic acids in uterine natural killer cells. J Clin
Endocrinol Metab 2001;86:1823–34.
[53] Wang C, Tanaka T, Nakamura H, Umesaki N, Hirai K, Ishiko O, et al. Granulated
metrial gland cells in the murine uterus: localization, kinetics, and the functional
role in angiogenesis during pregnancy. Microsc Res Tech 2003;60:420–9.
[54] Hanna J, Goldman-Wohl D, Hamani Y, Avraham I, Greenfield C, NatansonYaron S, et al. Decidual NK cells regulate key developmental processes at the
human fetal-maternal interface. Nat Med 2006;12:1065–74.
[55] Maynard S, Epstein FH, Karumanchi SA. Preeclampsia and angiogenic imbalance. Annu Rev Med 2008;59:61–78.
S31
[56] Tayade C, Hilchie D, He H, Fang Y, Moons L, Carmeliet P, et al. Genetic deletion
of placenta growth factor in mice alters uterine NK cells. J Immunol
2007;178:4267–75.
[57] Kusakabe K, Otsuki Y, Kiso Y. Involvement of the fas ligand and fas system in
apoptosis induction of mouse uterine natural killer cells. J Reprod Dev
2005;51:333–40.
[58] Saito S, Nishikawa K, Morii T, Enomoto M, Narita N, Motoyoshi K, et al.
Cytokine production by CD16-CD56bright natural killer cells in the human
early pregnancy decidua. Int Immunol 1993;5:559–63.
[59] Darmochwal-Kolarz D, Rolinski J, Leszczynska-Gorzelak B, Oleszczuk J. Fas
antigen expression on the decidual lymphocytes of pre-eclamptic patients.
Am J Reprod Immunol 2000;43:197–201.
[60] Crncic TB, Laskarin G, Frankovic KJ, Tokmadzic VS, Strbo N, Bedenicki I,
et al. Early pregnancy decidual lymphocytes beside perforin use
Fas ligand (FasL) mediated cytotoxicity. J Reprod Immunol 2007;73:
108–17.
[61] Ashkar AA, Croy BA. Interferon-gamma contributes to the normalcy of murine
pregnancy. Biol Reprod 1999;61:493–502.
[62] Hunt JS, Miller L, Roby KF, Huang J, Platt JS, DeBrot BL. Female steroid
hormones regulate production of pro-inflammatory molecules in uterine
leukocytes. J Reprod Immunol 1997;35:87–99.