human reproductive failure ii: immunogenetic and interacting factors

Human reproductive failure II: Immunogenetic andinteracting factors

Shormila Roy Choudhury and Leslie A.Knapp1

Department of Biological Anthropology, University of Cambridge, Downing Street, Cambridge, CB2 3DZ, UK

1To whom correspondence should be addressed. E-mail: [email protected]

Studies in humans suggest that reproductive failure may be in¯uenced by immunological factors or by genesencoding immunological factors and regulatory mechanisms controlling immunological expression. Using molecularmethods, immunological factors can be clearly studied in an immunogenetic context. One example, the majorhistocompatibility complex (MHC), known as the human leukocyte antigens (HLA) in humans and MHC in othermammals, affects many different stages of reproduction. Studies in some outbred, and in closely related, humanpopulations indicate that HLA, or HLA-linked, genes and HLA regulatory factors affect gamete development,embryo cleavage, blastocyst and trophoblast formation, implantation, fetal development and survival. Studies in non-human mammals indicate that MHC, or MHC-linked, genes such as the grc complex, Ped/Qa-2, t haplotypes andMHC regulatory factors, have similar reproductive effects. Human reproductive failure may also be a consequenceof disruption of interacting factors, including interactions between HLA antigens, cytokines and natural killer (NK)cells. In this review, we highlight the importance of immunogenetic and interacting factors in human reproductivefailure. We argue that studies in closely related human populations and animal models may contribute to a betterunderstanding of the ways in which immunogenetic and interacting factors are involved in human reproduction.

Key words: early pregnancy loss/HLA/pregnancy/recurrent spontaneous abortion/reproductive failure

TABLE OF CONTENTS

Introduction

Immunogenetic factors

HLA expression and function

HLA expression and function at the maternal±fetal interface

HLA sharing in outbred individuals

HLA linkage disequilibrium

HLA sharing in closely related individuals

HLA regulation

HLA and gene interactions

MHC expression and sharing in animal models

Interactions: cytokines and HLA expression

Interactions: natural killer cells and HLA expression

Interactions: cytokines and natural killer cells

Other interacting factors

Summary and conclusion

Acknowledgements

Introduction

As discussed in Part I (Roy Chaudhury and Knapp, 2001), many

factors involved in humoral and cellular immune responses may

in¯uence human reproductive failure. Traditionally, most of these

factors have been studied from an immunological prespective and

are identi®ed by immunological assaying for antibodies and/or

antigens. Unfortunately, most immunological assays for anti-

bodies and/or antigens are unstandardized and dif®cult to

replicate. For this reason, controversy surrounds the relative

importance of many immunological factors in human reproduc-

tive failure. In contrast, molecular methods may provide greater

reproducibility, reliability, accuracy and may be used to

standardize immunological assays. Molecular methods may also

enable researchers to investigate how genes encoding immuno-

logical factors or regulatory mechanisms controlling gene

expression affects reproduction. Until recently, few studies have

investigated the in¯uence of immunogenetic factors during

reproductive failure and their importance in reproduction remains

unclear and often controversial. Animal models have also

improved our understanding of immunogenetic factors.

Increasingly, it is becoming clear that human reproduction is a

dynamic interplay between immunological and immunogenetic

factors (interacting factors) during pregnancy. The disruption of

Human Reproduction Update, Vol.7, No.2 pp. 135±160, 2001

Ó European Society of Human Reproduction and Embryology 135

these immunological and immunogenetic networks may often

lead to reproductive failure. Although, this also remains a

relatively unexplored area of research, our review demonstrates

that interactions between immunological and immunogenetic

factors must also be studied.

Due to the complicated nomenclature for human reproduction

(see Part I: Roy Chaudhury and Knapp, 2001), different categories

of reproductive failure will be de®ned as follows. Infertility is the

failure to conceive after frequent unprotected intercourse (Reiss,

1998). Unexplained infertility occurs when no cause of infertility

can be identi®ed after full clinical investigation of both partners

(Reiss, 1998). Miscarriage will be de®ned as the spontaneous loss

of a pregnancy with a gestational age of <24 weeks (Regan,

1997; Reiss, 1998). Recurrent miscarriage is de®ned as the loss of

three or more consecutive pregnancies prior to week 24 of

gestation (Reiss, 1998). Depending on the study, recurrent

spontaneous abortion (RSA) is de®ned as the loss of two or

more (or even three or more), clinically detectable pregnancies

without reference to the week of gestation. Reproductive failure is

de®ned as the inability to conceive or to carry a pregnancy to

term. Reproductive success is de®ned as the ability to conceive

and carry a pregnancy to term and reproductive outcome refers to

both reproductive failure and success.

Immunogenetic factors

While an extensive spectrum of immunological factors can affect

reproductive outcome, numerous genetic associations with human

reproductive failure suggest that immunogenetic factors may also

in¯uence reproduction (Figure 1). These factors include proteins

encoded by genes and the transcriptional and translational

mechanisms, which regulate gene expression during different

stages of reproduction. Some of these factors have been

traditionally studied from an immunological prespective.

However, it is important to determine how the genes and

regulatory mechanisms of these factors in¯uence pregnancy.

Genetic methods such as cloning, sequencing and expression

analysis are now utilized to identify and study immunogenetic

factors. However, to de®nitely demonstrate the direct in¯uence of

genes in pregnancy, the production of animals with inactivated

(null or knockout) genes are invaluable models when human

studies are not possible. Knockout mice have now provided

convincing evidence of the importance of factors, e.g. cytokines,

leukaemia inhibitory factor (LIF), mucin-1 (MUC1) and natural

killer (NK) cells during pregnancy, supporting the need for further

immunogenetic research.

Historically, many immunogenetic studies have focused upon

the in¯uence of human leukocyte antigen (HLA) genes during

reproduction. Major histocompatability complex (MHC) antigens

have been well-studied in humans and many other mammals, and

it has been suggested that the MHC may play a dual role in

reproduction. First, it may in¯uence mate choice in some species

(for review, see Grob et al., 1998) and second, the MHC may

affect many different stages of reproduction (Fernandez et al.,

1999). Thus, studies of the immunogenetic effects of HLA, or

HLA-linked, genes in human reproduction provide an example, or

framework, to study the function of other immunogenetic factors

during pregnancy.

HLA expression and function

The human MHC, or HLA, may play critical roles throughout

pregnancy by in¯uencing gamete development, embryo cleavage,

blastocyst and trophoblast formation, implantation, fetal devel-

opment and survival (Figure 1). The HLA complex contains over

150 loci spanning ~4 Mb of DNA on chromosome 6p21

(Browning and McMichael, 1996). These genes encode proteins

controlling cell±cell interactions and regulating immune re-

sponses (Ober and van der Ven, 1997) by presenting peptides to

T-cells to form immunogenic complexes capable of T-cell

stimulation (Browning and McMichael, 1996). The MHC is

divided into three regions, class I, II and III (Ober and van der

Ven, 1997). In class I and II loci allelic polymorphism is the

highest amongst all functional mammalian genes, with some loci

exceeding 100 alleles (Apple and Erlich, 1996). This diversity is

signi®cant for MHC regulation of cellular interactions and organ

transplantation (Browning and McMichael, 1996). The class I

region includes the classical, or class Ia genes (HLA-A, HLA-B

and HLA-C), the non-classical, or class Ib genes (HLA-E, HLA-F

and HLA-G) and the pseudogenes, HLA-H, HLA-J, HLA-K and

HLA-L (Geraghty et al., 1992a,b; Messer et al., 1992). Class I

molecules are heterodimers of HLA-encoded a chains with the

non-MHC-encoded b2m. Generally, the function of class I

molecules is to present peptides derived from endogenously

synthesized, cytosolic proteins to CD8 cytotoxic cells. Similarly,

the class II HLA-DR, -DP and -DQ molecules consist of HLA-

encoded a and b chain heterodimers (Apple and Erlich, 1996)

whose function is to present peptides derived from exogenous

proteins or from endogenously synthesized cell-surface glycopro-

teins, to CD4 helper cells (for review, see Browning and

McMichael, 1996). Non-HLA class II genes encode proteins

involved in antigen-processing and transport, e.g. TAP1, TAP2,

LMP7 and LMP2. The class III region also contains genes

involved in immune response, although these molecules do not

directly contribute to antigen processing or T-cell presentation

(Browning and McMichael, 1996; for review, see Aguado et al.,

1996). MHC genes are inherited as haplotypes and are co-

dominantly expressed. Historically, HLA typing was performed

by microcytotoxicity (Terasaki et al., 1964; Amos et al., 1969).

However, recent advances in molecular genetics have enabled

more precise and extensive HLA typing.

HLA distribution varies throughout gametic and somatic

tissues. HLA expression on male gametes remains controversial.

Early studies detected HLA antigens on ejaculated spermatozoa,

but some groups reported their absence (for review, see Fernandez

et al., 1999). Sperm cell precursors, or germ cells, have

consistently shown class Ia expression prior to sperm maturation

(Kurpisz et al., 1986, 1987; Janitz et al., 1993, 1994; Guillaudeux

et al., 1996). Class Ib, HLA-E and -F, expression has been

detected on sperm cell precursors (Guillaudeux et al., 1996;

Fiszer et al., 1997). Chiang et al. (1994) reported HLA-G

expression on mature sperm cells. However, a recent study (Hiby

et al., 1999) did not verify this. Class II male gametic expression

has been less thoroughly investigated than class I expression and

sample contamination in early studies may have confounded

detection, since class II gametic expression has not been recently

reported. Generally, there is some HLA expression on male

gametes, particularly class I expression on immature spermato-

S.Roy Choudhury and L.A.Knapp

136

zoa. There have been fewer, even less conclusive studies on

female gametic HLA expression (Fernandez et al., 1999). Another

group (Jurisicova et al., 1996) reported HLA-G mRNA in

unfertilized oocytes using reverse transcription±polymerase chain

reaction (RT±PCR). Most other researchers have not detected

oocyte class I or class II products (Dohr et al., 1987; Roberts et

al., 1992; Fenichel et al., 1995; Patankar et al., 1997). Reports of

class I and II expression in rodent gametes is also contradictory

(for review, see Kurpisz et al., 1995; Fernandez et al., 1999). The

lack of, or low, gametic HLA expression suggests that fertiliza-

tion does not depend on HLA expression. However, HLA gametic

antigens may function as signal molecules, or they may be

involved in gamete differentiation and maturation (for review, see

Hutter and Dohr, 1998; Fernandez et al., 1999). A schematic

diagram of HLA expression during early development is

presented in Figure 2.

In somatic tissues, HLA distribution varies with each

antigen. Class II expression tends to be restricted to B

lymphocytes, macrophages endothelial cells and activated T-

cells. Class Ia genes, HLA-A, -B and -C, are ubiquitously

distributed and expressed on nearly all nucleated cells.

Recently, studies have reported more extensive tissue distribu-

tion, and greater potential functions, for class Ib HLA-E, -F

and -G than previously considered (Hammer et al., 1997a; Le

Bouteiller and Blaschitz, 1999). For example, HLA-E is

expressed in most tissues, including adult and fetal thymus

and liver, lymph nodes, spleen, resting T and B cells,

activated T-cells, skin, mucosa colon and eosinophils. HLA-F

transcripts are found in resting T and B cells, activated T-

cells, peripheral lymphocytes, fetal thymus, liver, skin, adult

tonsil and lymphoblastoid cell lines (Ober and van der Ven,

1997; Wainwright et al., 2000). Low HLA-G mRNA expres-

sion levels have been detected in adult peripheral B and T

lymphocytes (Kirszenbaum et al., 1994), mononuclear phago-

cytes (Yang et al., 1996) and progenitor haematopoietic stem

cells (CD34+) from umbilical cord blood (Kirszenbaum et al.,

1995). In mature tissues, HLA-G mRNA has been found in

keratinocytes (Le Bouteiller and Lenfant, 1996), lymphoid and

melanoma malignant cells (Amiot et al., 1996; Paul et al.,

1998), thymus (Crisa et al., 1997), the anterior portion of the

eye, skin, lung, kidney, ovary, colon and intestine (Carosella

et al., 1996). Importantly, HLA-G expression is unlike other

class I genes, since it can be transcribed as different isoforms

and is predominantly expressed at the maternal±fetal interface.

HLA expression and function at the maternal±fetalinterface

Fetal and maternal cells are in close contact with one another at

the maternal±fetal interface, especially at the implantation site, in

Figure 1. A summary of the immunological and immunogenetic factors, and their interactions, during primiparous pregnancy. Many of these factors may also bepresent during multiparous pregnancies, if they ®rst developed during primiparous pregnancy. Immunological factors are discussed in Part I (Roy Choudhury andKnapp, 2001). See text for further details of immunogenetic factors during reproduction. * = HLA, or HLA-linked, genes; red line indicates interactions betweenfactors, as currently demonstrated in studies from either rodents or humans.

Part II: Immunogenetic and interacting factors

137

the maternal arteries and in the intervillous space (Benirschke and

Kaufmann, 1995). Generally, ®rst trimester trophoblast cells

transcribe more class I mRNA and protein than term trophoblast

cells (Kovats et al., 1990; Wei and Orr, 1990). During various

stages of gestation, distinct trophoblast subpopulations contain

different levels of class I mRNA and translated proteins (Hammer

et al., 1997b). Generally, HLA-A and -B membrane-bound

molecules have not been detected in any trophoblast subpopula-

tions during pregnancy (Galbraith et al., 1981; Sunderland et al.,

1981a; Faulk et al., 1982; Johnson and Stern 1986; Kovats et al.,

1990; Shorter et al., 1990; Hunt, 1992; Le Bouteiller, 1994).

However, HLA-A and -B transcripts and translated intracellular

products have been detected in several trophoblast cell types,

including villous cytotrophoblast cells (Guillaudeux et al., 1995).

In contrast, both HLA-C mRNA and protein expression has been

identi®ed in invasive extravillous cytotrophoblast cells in the cell

islands, placental septa, and the basal plate, and interstitial,

intramural and intraluminal cytotrophoblast cells, as well as

trophoblastic giant cells and endovascular cells (Chumbley et al.,

1994; King et al., 1996a; Hutter et al., 1996; for review, see Le

Bouteiller and Lenfant, 1996; Hammer et al., 1997a; Proll et al.,

1999; King et al., 2000a). Finally, Hammer et al. (1997b) reported

that all class Ia molecules are heterogeneously expressed in the

amnion epithelium, an embryoblast-derived cell layer covering

the amnion cavity which is not in contact with maternal tissues.

The class Ib molecules have different distributions at the

maternal±fetal interface. HLA-E transcripts have been detected in

several trophoblast cell subpopulations isolated from ®rst

trimester or term human placenta, in term chorionic membrane

and in amnion epithelial cells (Wei and Orr, 1990; Guillaudeux, et

al., 1995; Houlihan et al., 1995). In vivo, HLA-E membrane-

bound proteins have been detected in amnion epithelial cells and

trophoblast cells (Houlihan et al., 1995; King et al., 2000b).

Recently, HLA-E protein has been found to associate with the

class III molecules, TAP and calreticulin, and dissociates upon

binding of class I leader sequences. Without these leader

sequences, HLA-E cannot be expressed upon cell surfaces

(Braud et al., 1998a,b; Lee et al., 1998). HLA-F also binds TAP

and calreticulin, although it is not expressed as a membrane-

bound protein (Wainwright et al., 2000).

Various HLA-G expression patterns at the maternal±fetal

interface have been described using different methodologies (Le

Bouteiller and Blaschitz, 1999). The most precise distributions

have been determined by speci®c anti-HLA-G monoclonal

antibodies (Bensussan et al., 1995; Lee et al., 1995; McMaster

et al., 1995, 1998; Loke et al., 1997; Chu et al., 1998). In vivo,

HLA-G protein is expressed in ®rst trimester placentae, in

populations of extravillous cytotrophoblast cells including invad-

ing extravillous cytotrophoblast, endovascular and interstitial

trophoblast, placental bed giant cells and cytotrophoblast of the

chorion laeve (Bulmer et al., 1991; Chumbley et al., 1994; Hutter

et al., 1996; King et al., 1996b; Le Bouteiller and Blaschitz,

1999). HLA-G mRNA expression in extravillous cytotrophoblasts

has been reported in second trimester placenta and term

membranes (Sunderland et al., 1981b; Ellis et al., 1990; Kovats

et al., 1990; Wei and Orr, 1990; Yelavarthi et al., 1991;

Chumbley et al., 1994) and HLA-G protein has been detected

in term extravillous cytotrophoblast (Blaschitz et al., 1997). Only

one group has reported in-vivo expression of HLA-G in

syncytiotrophoblast (Chu et al., 1998). HLA-G expression has

also been reported in the endothelial cells of fetal vessels present

in the mesenchymal core of chorionic villi (Blaschitz et al., 1997)

and in amniochorion and amniotic ¯uid (McMaster et al., 1998).

In fetal tissues, HLA-G has been reported in the eye, thymus

(Shukla et al., 1990), ®rst trimester liver (Houlihan et al., 1995),

lung, heart and kidney (Carosella et al., 1996).

Although placental HLA-G expression occurs throughout

gestation (Le Bouteiller et al., 1999), the onset of HLA-G

expression during early development remains uncertain (Hiby et

al., 1999). HLA-G mRNA and protein was detected in 40% of

human pre-implantation embryos at each cleavage stage up to the

blastocyst and high expression was correlated with a high

embryonic cleavage rate. Blastocysts with detectable HLA-G

mRNA had signi®cantly greater mean numbers of blastomeres per

embryo at 24 and 48 h after fertilization than HLA-G-negative

blastocysts, suggesting that HLA-G expression re¯ects increased

embryo cleavage rates and is correlated with more successful

implantation (Juriscova et al., 1996). In contrast, other authors

(Hiby et al., 1999) did not detect HLA-G mRNA in human 2±8-

cell embryos, nor in blastocysts. Moreover, some research groups

reported that HLA-G expression decreases as gestation proceeds

(Kovats et al., 1990; Wei and Orr, 1990); others found high levels

of HLA-G mRNA in term placenta (Hiby et al., 1999). Thus, there

is currently no consensus regarding the onset and progression of

HLA-G expression during human development.

Using RT±PCR and different sets of HLA-speci®c primers, six

different HLA-G transcriptional isoforms (HLA-G1 to -G6) have

been detected in unpuri®ed trophoblast cells from human ®rst

trimester and term placentae (Ishitani and Geraghty, 1992; Fujii et

al., 1994; Kirszenbaum et al., 1994; Onno et al., 1994; Moreau et

al., 1995), from villous cytotrophoblast cells of puri®ed term

placenta and in-vitro differentiated syncytiotrophoblast

(Guillaudeux et al., 1995). Recently, Hiby et al. (1999) detected

all six isoforms in ®rst trimester and term placentae, highly

puri®ed villous and extravillous trophoblast cells and JEG-3 and

221-G cell lines. Unfortunately, there remain technical dif®culties

in isoform detection due to variations in methodological

preparation (for review, see Le Bouteiller et al., 1999).

The importance of HLA-G expression during pregnancy

remains unknown. Functionally, HLA-G differs from other class

Ib genes (for review, see Ober et al., 1996; Ober and Aldrich,

1997) since it preferentially binds nonamer peptides (Lee et al.,

1995; Diehl et al., 1996) and may present viral antigens to T-cells,

thereby playing a role in immunosurveillance during pregnancy.

Few viruses infect fetal cells, so limited HLA-G diversity may

suf®ce for T-cell viral presentation. Although fetal extravillous

trophoblast cells express HLA-G, pathogen infection through

maternal blood occurs primarily in the intervillous space, where

exposed fetal villous syncytiotrophoblast generally lack HLA-G

expression (Loke et al., 1999). Low polymorphism may eliminate

strong maternal immune responses or induce maternal tolerance.

Maejima et al. (1997) suggested that HLA-G expression may

modulate maternal tolerance of pregnancy by shifting Th1/Th2

cytokine pro®les in human decidua. Hiby et al. (1999) suggested

that HLA-G may be relatively unimportant in gametogenesis or

early embryogenesis and may, instead, play a role in later

developmental processes such as implantation. This proposal is

supported by a study by McMaster et al. (1995), in which HLA-G


138

production by early gestation cytotrophoblast stem cells was

critical for in-vivo and in-vitro cytotrophoblast differentiation

during invasion of maternal tissues. Moreover, mouse studies

indicate that non-classical class I expression increases during and

after implantation (Fernandez et al., 1999). HLA-G may also

sustain trophoblast growth and prevent reproductive loss.

Research by Hamai et al. (1999), who studied interleukin (IL)-2

supplementation on the growth of HLA-G positive (BeWo, JEG-

3) and negative (JAR) trophoblastic cell lines for IL-2 receptors,

found HLA-G positive cell lines were not in¯uenced by IL-2

addition. However, negative lines exhibited signi®cantly de-

creased proliferation when cultured with IL-2. HLA-G transfec-

tion of negative lines eliminated the IL-2 growth-inhibitory effect.

The authors suggested that since IL-2 stimulates T-cell prolifera-

tion and induces lymphocyte-activated-killer (LAK) cells, HLA-G

expression may enable trophoblast cells to evade cell damage.

Inadequate HLA-G expression could lead to IL-2 mediated growth

suppression and IL-2 expression may be up-regulated, or HLA-G

expression may be down-regulated, in women experiencing

reproductive failure (Hamai et al., 1997; Lim et al., 1998).

Interestingly, HLA-G expression was either absent, or reduced, in

extravillous trophoblast cells in a sample of 10 pre-eclamptic

placentae, suggesting that low fetal HLA-G expression may result

in maternal rejection (Goldman-Wohl et al., 2000).

Recently, Le Bouteiller and Blaschitz (1999) suggested that

HLA-G could regulate placental angiogenesis and soluble HLA-

G isoforms may act as immunosuppressors during pregnancy.

Regulated and quantitative shifts of HLA-G isoforms during

development have been observed. Thus, isoforms may perform

diverse functions in the placental environment (Ishitani and

Geraghty, 1992). Unfortunately, the speci®c roles of HLA-G

isoforms during pregnancy remain unknown. Limited in-vitro

and in-vivo studies present con¯icting results on the

importance of the HLA-G1 isoform. Mallet et al. (2000)

reported that HLA-G1 was expressed as a membrane-bound

protein at the cell surface of murine and human (JAR)

trophoblast cell lines, but HLA-G2, -G3 and -G4 transcripts

were retained in the endoplasmic reticulum. This suggests that

HLA-G1 expression may be important during pregnancy.

However, Ober et al. (1998a) identi®ed one adult and one

®rst term placenta, which were homozygous for a single base

pair deletion in exon 3 of HLA-G1. These ®ndings suggest

that HLA-G1 expression is not essential for survival. While

there is no consensus for humans, the observation that other

primates possess HLA-G homologues with isoforms (Boyson et

al., 1997) supports the argument that HLA-G isoforms could

be important during pregnancy.

HLA-G polymorphism may also be a target for maternal

anti-fetal immune responses (Klein, 1990). `Abnormal' HLA-G

alleles may fail to be recognized by the maternal immune

system during pregnancy, resulting in reproductive loss

(Yamashita et al., 1999). While initial studies reported that

HLA-G was non-polymorphic (Geraghty et al., 1987; Kovats et

al., 1990), the majority of recent studies have reported limited

HLA-G polymorphism at both the nucleotide and amino acid

levels (Alizadeh et al., 1993; Morales et al., 1993; Ober et al.,

1996; Yamashita et al., 1996; Hviid et al., 1997; Hiby et al.,

1999). One study of 45 African Americans detected extensive

non-synonymous HLA-G variation (van der Ven and Ober,

1994). This may re¯ect the general ®nding that many loci in

African populations are more polymorphic than the same loci

in other populations (Tishkoff et al., 1996). Notably, most

studies focus on HLA-G polymorphisms in exons 2 and 3

(encoding the a1 and a2 domains) because of class Ia

homology. However, if HLA-G performs novel functions,

polymorphisms in other regions of the molecule may also be

important (for review, see Ober and Aldrich, 1997).

To summarize, HLA may play critical roles during many

different stages of pregnancy. Although the importance of HLA

antigens in human gamete development and fertilization remains

controversial, these antigens could in¯uence early embryonic

cleavage, blastocyst/trophoblast development and implantation

(Figure 2). In particular, the unique characteristics and distribu-

tion of the HLA-G antigen at the maternal±fetal interface suggests

that the HLA-G gene may have a critical role during early stages

of reproduction.

Figure 2. HLA expression varies throughout pregnancy and controversy surrounds the distribution and relative importance of HLA antigens during earlydevelopment. This ®gure summarizes published reports on HLA expression from gametogenesis to implantation. HLA-Ia = HLA-A, -B and -C loci. + = expressionof HLA antigen and ± = lack of expression. See text for further details.


139

HLA sharing in outbred individuals

The earliest studies of outbred individuals suggested that HLA

antigen expression in¯uenced fetal development, and subsequent

reproductive outcome (Figure 1). During the 1960±70s, studies

suggested that fetuses inheriting paternal MHC antigens that

differed from maternal antigens (histoincompatible pregnancies)

may have a selective survival advantage compared with

conceptuses inheriting paternal MHC antigens that did not differ

from maternal antigens (histocompatible pregnancies) (Billington,

1964; Kirby, 1970; Beer and Billingham, 1976). A histocompa-

tible pregnancy may be unrecognized by the maternal immune

system, or ellicit an inappropriate maternal response resulting in

reproductive failure. Thus, couples with RSA may be aborting

histocompatible fetuses due to parental HLA similarity

(Christiansen, 1996). Early retrospective studies reported in-

creased HLA sharing amongst couples with RSA compared with

fertile controls (Komlos et al., 1977; Schacter et al., 1979).

However, studies during the last two decades have demonstrated

con¯icting results. Ober and van der Ven (1997) reviewed over 30

studies, from the late 1970s to the mid-1990s, and found that

~50% of these studies reported greater HLA sharing in couples

with RSA than controls. There was little agreement on which

HLA locus or loci were important in RSA. Some studies reported

RSA associations with class I loci, class II loci or increased

sharing over all loci. Discrepancies have arisen since many

studies utilized small sample sizes or different study designs

(Ober, 1999), particularly with respect to the selection and

strati®cation of fertile and RSA couples. These studies also

suffered from limitations inherent in retrospective studies, often

failing to address the effects of HLA sharing in couples not

selected on the basis of reproductive histories (Creus et al., 1998).

Studies also differed in tissue typing methodology and the number

of antigens studied. Older studies utilized serological HLA

typing, which is in¯uenced by both typing reagent quality and

the cell isolation techniques employed. Indeed, a large, multi-

centre study reported that >25% of serologically typed HLA-DR

antigens were incorrect, when compared with DNA±restriction

fragment length polymorphism (RFLP) methods (Lorentzen et al.,

1997). Additionally, it is dif®cult to serologically resolve broad

antigens into potentially important subtypes (Doherty and

Donaldson, 1991). For example, while four HLA-DQ serological

speci®cities can be identi®ed, DNA-based methods can detect 22

DQB and DQA alleles (Bodmer et al., 1995). Thus, differences in

methodology and reliability can confound attempts to compare

data concerning HLA sharing between published studies

(Wagenknecht et al., 1997). The majority of recent studies, using

DNA-based techniques, have not detected increased sharing of

HLA-DR and/or -DQ alleles in couples with RSA (Christiansen et

al., 1989; Ito et al., 1992; Takakuwa et al., 1992; Laitinen et al.,

1993; Wagenknecht et al., 1997). Increased HLA-G (Karhukorpi

et al., 1997; Yamashita et al., 1999) or HLA-C sharing among

couples with recurrent miscarriage has not been observed either

(Christiansen, 1999). Only Ober et al. (1993) reported a

signi®cant increase in HLA-DQ sharing amongst couples with

RSA compared with controls using PCR single-speci®c oligonu-

cleotide probe hybridization (PCR±SSOP). Despite the contro-

versial relationship between HLA sharing and reproductive

failure, there is a general consensus that the degree of HLA

sharing in RSA couples has no impact on the prognosis of

subsequent pregnancies (Smith and Cowchock, 1988;

Christiansen et al., 1994; Recurrent Miscarriage Immunotherapy

Trialist Group, 1994).

Other studies have investigated HLA sharing in couples

experiencing unexplained infertility (for review, see Ober and

van der Ven, 1997), since a subgroup of couples with unexplained

infertility may experience peri-implantational losses (Collins et

al., 1983; Wilcox et al., 1988) due to increased HLA similarity

(Creus et al., 1998). Some studies found no signi®cant differences

in HLA sharing between couples with unexplained infertility and

fertile couples (Nordlander et al., 1983; Persitz et al., 1985).

Others reported signi®cantly higher HLA sharing in couples with

unexplained infertility (Coulam et al., 1987). An alternative

method to investigate peri-implantational loss, involves the study

of parental HLA sharing in infertile couples undergoing assisted

reproductive technology (ART). Although few studies have

utilized this method (Weckstein et al., 1991; Balasch et al.,

1993; Ho et al., 1994; Creus et al., 1998), two larger projects are

notable. Ho et al. (1994) studied HLA-A, -B, -C, -DR and -DQ loci

in 76 couples with unexplained infertility and found a signi®cant

excess of HLA sharing (two or more antigens; P = 0.015). In all,

36 couples who did not achieve pregnancy with IVF had

increased HLA-DQ sharing compared with couples with success-

ful pregnancies (Jin et al., 1995). Recently, Creus et al. (1998)

compared 50 infertile couples who did not achieve a pregnancy

after three or more IVF cycles with at least two transferred

embryos, to 50 infertile couples who achieved a pregnancy after

one IVF treatment. They found that three or more implantation

IVF failures were signi®cantly associated with two or more

antigens shared between partners (P < 0.05).

Other studies have examined HLA antigen frequency among

women with RSA since many maternal autoimmune diseases

involving fetal loss are associated with HLA antigens or

haplotypes (for review, see Ober, 1990). A few studies have

reported statistically signi®cant associations between recurrent

miscarriage and class II HLA-DR antigens (especially for the

serological types DR1 and DR3). Other studies have found

contrary results. Christiansen et al. (1999) conducted a meta-

analysis of 18 case-control studies of the frequency of the DR1

and DR3 antigens among women with recurrent miscarriage. The

authors found that the HLA-DR1 antigen was associated with

increased susceptibility to recurrent miscarriage. Other studies

have detected no signi®cant class I antigen associations with

RSA. For example, no signi®cant differences in HLA-E or -G

allele frequencies, were detected between RSA women and

controls using DNA-based methods (Steffensen et al., 1998;

Penzes et al., 1999). A few studies, focusing on RSA occurrence

amongst the relatives of women with RSA, reported an increased

RSA prevalence amongst ®rst-degree biological relatives and

suggest the existence of a familial predisposition to RSA (Johnson

et al., 1988; Alexander et al., 1988; Christiansen et al., 1990a; Ho

et al., 1991; Christiansen, 1996). Christiansen et al. (1990b)

reported that sisters who shared both parental HLA haplotypes

with probands, experienced signi®cantly higher miscarriage rates

than sisters who shared only one, or no, parental HLA haplotypes

with the proband.

Some prospective studies have followed up subsequent

reproductive outcome in women and couples experiencing RSA.


140

For example, Sbracia et al. (1996) investigated the expression of

HLA-A, -B and -DR alleles using serological typing in 57 and 30

RSA and fertile couples respectively. The researchers documen-

ted reproductive outcome for three years. A signi®cant increase in

HLA-DR sharing was observed in couples that aborted during the

follow-up period, compared with couples and controls who

achieved a livebirth (P < 0.03 and 0.02 respectively). The HLA-

B44/DR5 haplotype frequency was also signi®cantly increased in

the same study (P < 0.03). It is important to note, however, that

couples received different forms of psychological and immuno-

therapy during the follow-up period and reproductive outcome

may be confounded by these factors. Christiansen et al. (1996)

studied 234 women with RSA and 360 controls for HLA-DR and

-DQ antigens using PCR sequence speci®c primers (PCR-SSP)

and RFLP and reported subsequent reproductive outcome.

Signi®cantly greater numbers of women with HLA-DR1 and/or

-DR3 miscarried (62%), compared with those negative for both

allogenotypes (29%; P = 0.025).

Although it has been suggested that parental histocompatibility,

or that HLA homozygosity, may result in preferential abortion of

fetuses, few studies have investigated whether histocompatible or

homozygous fetuses have been preferentially aborted. Wenk and

Boughman (1989) typed HLA-A and -B antigens in 2569 healthy

parents and children and found that parental antigen sharing did

not result in a signi®cant de®cit of histocompatible nor

homozygous children. Ober et al. (1993) identi®ed HLA-DQA1

and -DQB1 alleles in 40 abortuses and 31 liveborn children of 68

couples with RSA. Signi®cantly more couples with RSA shared

two HLA-DQA1 alleles compared with fertile controls (P = 0.031).

Although haplotype frequencies of HLA-DQA1/DQB1 in parents

with RSA and aborted fetuses were not signi®cantly different

from controls (Steck et al., 1995), fewer than expected HLA-

DQA1 compatible fetuses were observed. Assuming the numbers

of compatible and incompatible pregnancies are equal at

conception, the researchers suggested that histocompatible HLA-

DQA1 fetuses may be selectively aborted during early pregnancy

and HLA-DQA1 incompatible fetuses may have better survival

rates (Steck et al., 1995). Thus, HLA molecules may signi®cantly

contribute to reproductive outcome. Modern techniques of HLA

typing, and further molecular studies, should provide more

detailed understanding of the in¯uence of the HLA during

pregnancy.

HLA linkage disequilibrium

These studies suggest that HLA genes are associated with

reproductive outcome in outbred individuals. Importantly, the

genes in¯uencing reproductive outcome may not be HLA genes

per se, but closely linked susceptibility genes, or genetic defects,

which play critical roles in pregnancy maintenance, fetal

development and survival (Creus et al., 1998; Kostyu, 1994).

HLA genes could be markers for these susceptibility genes if they

are retained in preferential allelic combinations over long periods

of time, i.e. linkage disequilibrium. The HLA complex displays

particularly high levels of linkage disequilibrium. Class III genes

BF, C2, C4A, C4 have been found in linkage disequilibrium

(Kostyu, 1994). Class I linkage disequilibrium has been found for

HLA-A and -B, HLA-B and -C or HLA-A extending to HLA-G

(Morales et al., 1993). Interestingly, HLA-G demonstrates striking

levels of linkage disequilibrium with most (but not all) HLA-A

alleles, even among distantly related populations (Le Bouteiller

and Blaschitz, 1999). Furthermore, linkage disequilibrium can

extend from class I (at the telomeric end) to class II (close to the

centrosome), with some ancestral haplotypes extending from

HLA-A to -DR (2000 kb) or to -DP (4000 kb) (Dausset et al.,

1978; Awdeh et al., 1983; Raum et al., 1984; Kay et al., 1988;

van Endert et al., 1992). Linkage disequilibrium increases the

probability that individuals with identical HLA antigens at one

locus will also have identical antigens at a second HLA locus

(Ober, 1990). Therefore, closely related individuals are also more

likely to share haplotypes by virtue of these haplotypes being

identical by descent. Studies on parental HLA sharing between

closely related individuals, rather than outbred couples, may thus

detect the effects of greater HLA homozygosity and MHC-linked

genes in linkage disequilibrium.

HLA sharing in closely-related individuals

Population-based and prospective studies have been conducted in

Hutterite communities (Ober et al., 1983, 1985, 1988, 1992,

1998b; Ober, 1995). The small number of founding Hutterite

HLA haplotypes and reproductive isolation increases the prob-

ability that couples will share HLA antigens and other linked loci.

The Hutterites are amongst the most fertile human population

ever studied (Sheps, 1965; Ober, 1995). Contraception is

prohibited and large families are desired (Ober and van der

Ven, 1997). From their research with Hutterites during the last 16

years, Ober et al. suggested that HLA-DR, or HLA-DR linked

genes, may affect fertilization, pre- or peri-implantational

embryonic survival, and HLA-B, or HLA-B, linked genes may

contribute towards recognized fetal loss. Initial studies demon-

strated longer birth intervals among couples sharing HLA-A or -B

antigens (Ober et al., 1983). Median time intervals to produce 10

children was 13.73 years in couples sharing no antigens, 14.52

years among couples sharing one antigen, and 19 years for those

sharing two or more antigens. Longer median intervals from

marriage to each birth were also observed for Hutterite couples

sharing HLA-A, -B or -DR alleles, compared with Hutterite

couples not sharing alleles (Ober et al., 1985, 1988). When the

effects of individual loci were analysed, signi®cantly longer birth

intervals and smaller completed family sizes (P = 0.041) were

detected with HLA-DR parental sharing (Ober et al., 1985, 1988).

In a subsequent 5 year prospective study of 104 couples and 154

pregnancies, a signi®cant increase in time from menstruation after

birth to the ®rst pregnancy was associated with parental HLA-DR

compatibility (P = 0.015; 5.1 months to pregnancy in couples

sharing one or more DR alleles versus 2 months in couples

sharing no alleles). In the same study, fetal loss was signi®cantly

associated with HLA-B sharing (P = 0.041, with adjustment for

age, gravidity, kinship). Another 10 year prospective study of 111

Hutterite couples with 251 pregnancies, revealed signi®cantly

increased fetal losses in couples sharing the entire 16-locus class

I, II, III, haplotype (P = 0.002). Amongst individual loci,

signi®cant fetal loss rates were detected for sharing of HLA-B

(P = 0.019), HLA-C (P = 0.033) and C4 (P = 0.043), although fetal

losses from HLA-C and C4 sharing may have been due to their

close proximity to HLA-B (Ober et al., 1998b). Laitinen et al.

(1991) also found that RSA was correlated with HLA-A and -B


141

sharing in Finnish populations. Like the Hutterites, the Finns have

been reproductively isolated with reduced genetic diversity.

Laitinen (1993) studied 14 HLA genes in 56 Finnish couples by

serological and DNA-based methods, and found signi®cantly

greater frequencies of 12 HLA haplotypes (most extended from

HLA-A to -DP) in couples with RSA compared with controls (P <

0.001). Interestingly, four of these `risk' haplotypes were also

identi®ed in Ober's (1995) Hutterite study.

Studies among the Hutterites and Finns suggest that particular

HLA haplotypes may have deleterious effects upon fetal viability.

A low frequency of 12 HLA `risk' haplotypes, most extending

from HLA-A to -DP, were reported amongst 29 newborns born to

Finnish couples during a 2 year follow-up period (Laitinen, 1993).

Ober et al. found a 64.4% de®cit of Hutterite homozygous

individuals for a ®ve locus haplotype (HLA-A, -B, -C, -DR and

-DQ), which may re¯ect the additive or interactive effects of

selection against homozygous HLA-B and DR fetuses (Ober,

1995). Needless to say, more studies are required to determine

what HLA haplotypes or loci are commonly associated with

reproductive failure.

Although it has been suggested that parental histocompatibility,

resulting in offspring HLA homozygosity, may lead to prefer-

ential abortion of fetuses, few studies have investigated whether

histocompatible or homozygous fetuses are preferentially aborted.

Studies in outbred populations remain inconclusive. Whereas in

Hutterites, no signi®cant de®cit of homozygous, or histocompa-

tible, children were detected with high resolution typing of over

1000 parents and children for HLA-B, or a 16 loci HLA haplotype

(Ober et al., 1998b). Alternatively, Ober et al. (1998b) has

suggested that fetal/maternal histocompatibility may in¯uence

reproductive outcome. That is, a fetus recognizing maternal

tissues as non-self (a heterozygous compatible pregnancy) may

have survival advantages over a homozygous compatible fetus,

which would not recognize maternal tissues as non-self. This

suggestion may be supported by the observation that Hutterites

experience greater losses of HLA-B homozygous fetal/maternal

compatible pregnancies, than heterozygous compatible pregnan-

cies (Ober et al., 1998b).

HLA regulation

Regulatory mechanisms controlling the expression of HLA, or

HLA-linked genes, may also play a role in reproductive outcome.

Mutations, duplications, translocations or aberrant regulation

could alter HLA expression at the maternal±fetal interface and

lead to reproductive failure. Class I and II expression at the

maternal±fetal interface is regulated by various transcriptional

and post-translational mechanisms, extracellular viruses and

cytokines (Browning and McMichael, 1996). Placental class Ia

transcriptional regulation is modulated by cis- and/or trans-acting

regulatory mechanisms. Cis-regulatory regions comprise CpG-

rich nucleotide sequences (CpG islands) located in the 5¢ region

of many class I coding sequences (Boucraut et al., 1993a; Le

Bouteiller, 1994). Methylation of CpG islands can reduce

transcription. In the HLA class I null trophoblast-derived JAR

cell line, signi®cantly reduced class Ia expression was reprodu-

cibly caused by CpG island methylation. However, only HLA-E

expression appeared unmethylated and transcriptionally active.

Studies on trophoblast cells from human term placenta differed.

Class Ia appeared undermethylated in villous trophoblast cells,

and HLA-E expression remained active (Boucraut et al., 1993b;

Guillaudeux et al., 1995). These results suggest that down-

regulation of HLA class I may be caused by genomic imprinting

as opposed to CpG methylation. Although imprinting of class I

and Pa loci have been observed in rat placenta (Kanbour-Shakir et

al., 1993; Kunz et al., 1996), imprinting has not been detected in

murine (Drezen et al., 1994), equine (Donaldson et al., 1994) or

human trophoblast cells (Hiby et al., 1999).

Transcriptional transacting class I regulation operates through

various nuclear DNA-binding proteins that recognize 5¢ cis-

regulatory elements in the promoter region (Le Bouteiller, 1994).

Mouse studies have demonstrated that the 5¢ upstream region is

largely responsible for speci®c MHC expression patterns

(Oudejans et al., 1989). Promoter elements in this region, such

as CCAAT and TCTAAA boxes, are well conserved in most HLA

class I loci. Others, such as enhancer A, enhancer B and Interferon

Consensus Sequence, are present in HLA-A and -B, and absent, or

deleted, in non-classical loci (Le Bouteiller, 1994). This suggests

that the trans-acting factors which up-regulate transcription differ

between class Ia and Ib genes. Limited data are available

concerning the modulatory effect of transacting factors in human

trophoblast cell subpopulations, although they have been detected

in extravillous cytotrophoblast in human ®rst trimester placenta,

bound to enhancer A of HLA-A and -B (Boucraut et al., 1993b).

Low level membrane-bound HLA-A and -B expression in

trophoblast cell types may be due to the lack, or modi®cations,

of trans-acting factors that normally bind to class I promoters (Le

Bouteiller, 1994). Chiang and Main (1994) detected transacting

nuclear factors bound to a negative regulatory element (NRE) in

class I genes, suggesting that class I regulation may also involve

switching from negative to positive regulation with NREs (IsraeÈl

and Kourilsky, 1996). The accessibility of transacting factors to

speci®c DNA regions is also regulated by chromatin structure.

DNase I hypersensitive sites, located near enhancer A in murine

H-2 genes (Maschek et al., 1989), may be important in chromatin

remodelling as a form of transcriptional regulation. Furthermore,

mutations, or abnormal expression, of both cis- and trans-acting

factors could in¯uence HLA expression.

Post-translational regulation may inhibit membrane trophoblast

expression of HLA-A and -B in the presence of transcripts and

intracellular translated products. This regulation may function

through the availability of speci®c peptides and/or through the

negative regulation of other molecules required for HLA class I

assembly, folding and transport. These molecules include b2m,

LMP2/LMP7, TAP1/TAP2 and chaperone proteins (Jackson et al.,

1994) and the lack, or irregular expression, of these proteins could

result in the absence of HLA trophoblast surface expression.

Little data are available concerning the regulation of HLA-G

expression at the maternal±interface. Abnormal regulation of

HLA-G transcription could affect multiple interactions between

fetal trophoblast cells and maternal immune cells and cytokines.

Class I post-translational mechanisms are likely to regulate HLA-

G (Le Bouteiller et al., 1996), although HLA-G is unique with

transcriptional regulation of alternative splicing, giving rise to

multiple isoforms. The HLA-G promoter is also unique amongst

other class I promoters since all class I regulatory elements

including, enhancer A, enhancer B, IFE, NRE and site a, are

absent in the HLA-G promoter (Le Bouteiller, 1994; Van den


142

Elsen et al., 1998). An interferon (IFN)-g inducibility control

sequence has also been reported in the HLA-G promoter (Hunt et

al., 1998). Recently, Lefebvre et al. (1999) reported that IFN-

responsive regulatory regions outside the HLA-G promoter are

involved in transcriptional and post-transcriptional mechanisms of

IFN-g activation of HLA-G. Moreover, a distal 250 bp regulatory

element, located 1.1 kb upstream from the ®rst HLA-G exon,

directs speci®c HLA-G transcription in mouse placenta (Schmidt

et al., 1995).

Class II regulation is also governed by transcriptional and post-

translational processes. Although modulation primarily occurs on a

transcriptional level involving cis- and trans-acting elements.

Several cis-acting 5¢ elements and proximal promoter regulatory

sequences, containing the X, Y and W/Z box elements, are

important for class II expression (Wassmuth, 1996). Interestingly,

a mutation involving the lack of transcription factor binding to

class II promoters has been linked to some congenital defects

leading to severe combined immunode®ciencies (Wassmuth,

1996).

To escape immune detection by class I cytotoxic lymphocytes,

some viruses have developed mechanisms which regulate HLA

expression in infected cells or transformed cell lines (IsraeÈl and

Kourilsky, 1996). HLA-DR expression can be down-regulated by

the varicella-zoster virus which causes varicella (chickenpox)

(Abendroth et al., 2000). Class I expression can be modulated by

N-myc, c-myc and v-ras oncogenes (Bernads et al., 1986)

adenoviruses, the HIV-1 Tat protein, the herpes simplex virus

(HSV) and the cytomegalovirus (HCMV) (Israel and Kourilsky,

1996). HSV and HCMV infections have also been implicated in

miscarriage and RSA (Kriel et al., 1970; Altshuler, 1974; Gronroos

et al., 1983; Robb et al., 1986; Bujko et al., 1988), suggesting that

viruses may interfere with HLA expression at the maternal±fetal

interface. Schust et al. (1999) reported that JEG3 cell lines,

normally capable of expressing HLA-C and -G, were unable to

express either antigen when infected by HSV. HCMV infection of

trophoblast cell lines have shown con¯icting results. Schust et al.

(1999) reported that trophoblast cell lines did not alter HLA-C and -

G expression when infected with HCMV, but Jun et al. (2000)

reported that HCMV down-regulated HLA-C and -G expression in

human trophoblast cell lines. Tomasec et al. (2000) infected target

cells with a recombinant adenovirus containing the gpUL40 leader

sequence of HCMV and reported that infected target cells up-

regulated HLA-E expression to protect cells from NK lysis. Thus, if

HLA-C and -G, and perhaps HLA-E expression in trophoblasts is

essential for pregnancy maintenance, virally-altered HLA expres-

sion could also result in reproductive failure. The mechanisms by

which viruses alter HLA expression and how this may affect

reproductive failure, is a critical area for new research.

HLA and gene interactions

While HLA, or HLA-linked, genes work independently, they may

also interact with other genes on different chromosomes and

in¯uence pregnancy. For example, the BYSL gene (By the

Ribosomal Protein S6 Gene, Drosophila, Homologue-like), which

maps to 6p21.1 near the HLA, encodes a protein that binds

directly to trophinin and tastin in the endometrium (Pack et al.,

1998; Suzuki et al. 1998). The trophinin protein is encoded by the

TRO gene on the X chromosome (Pack et al., 1997). Trophinin

and tastin form a cell adhesion molecule complex that may be

involved in mediating the initial attachment of the blastocyst to

uterine epithelial cells during implantation (Suzuki et al. 1998).

Using mouse antibodies, tastin was exclusively detected on the

apical side of the human syncytiotrophoblast. Trophinin and

bystin were detected in human syncytiotrophoblast of the

chorionic villi and in endometrial decidual cells at the utero±

placental interface during the sixth week of pregnancy. Tissue

sections, examined by in-situ hybridization using RNA probes

speci®c to each protein, showed that trophoblast and endometrial

epithelial cells expressed trophinin, tastin, and BYSL. After week

10, the protein concentrations decreased and were no longer

detectable in placental villi. These results suggest that BYSL,

trophinin and tastin form a complex in trophectoderm cells during

implantation, mediating adhesion between trophoblast and

endometrial cells (Suzuki et al., 1999). Additionally, HLA alleles

interact with the C3 transferrin allele. A study of 1101 couples

and their 1895 offspring (Weitkamp and Schacter, 1985)

demonstrated that the C3 allele was associated with RSA,

parental HLA sharing and fetal neural tube development in

families. The associations involved interactions between chromo-

somes 3 and 6. Speci®cally, RSA was characterized by low

seminal zinc levels, caused by dysfunction of the C3 iron and zinc

binding transferrin protein encoded on chromosome 3. In these

same couples, RSA and fetal neural tube defects were associated

with increased parental HLA-A and -B sharing.

The interactions of HLA or HLA-linked genes with genes on

different chromosomes may also in¯uence the development of

late-onset diseases. For example, interactions between suscept-

ibility loci for multiple sclerosis have been described for HLA

alleles on chromosome 6 and genes on chromosome 17 (Karpuj et

al., 1997). Additionally, Pandey et al. (1999) demonstrated that

particular homozygous genotypes for tumour necrosis factor

(TNF)-a and immunoglobulin allotypes (GM and KM) interacts

with class II HLA-DQa1, contributing to an increased relative risk

to non-insulin-dependent diabetes mellitus (NIDDM).

Furthermore, HLA alleles are associated with developmental

and malignant diseases and other auto-immune conditions (for

review, see Jin et al., 1995).

It has also been argued that individuals with fewer different

MHC alleles have less resistance to pathogenic infection (Doherty

and Zinkernagel, 1975; Hughes and Nei, 1988). Homozygosity

for HLA class II alleles may increase susceptibility for common

variable immunode®ciency (CVID) (De La Concha et al., 1999).

Considering that HLA alleles are associated with numerous

diseases, mutations within HLA, or HLA-linked, genes could have

pleiotrophic effects, manifesting HLA associations between two

or more diseases. Rheumatoid arthritis has been associated with

decreased fecundity (Nelson et al., 1993) and also RSA (Shelton

et al., 1994). Finally, homozygosity for an as yet unidenti®ed

gene in linkage disequilibrium with the HLA reduces the

susceptibility for both type 1 diabetes and coeliac disease (Lie

et al., 1999), and mutations in this gene may lead to increased

susceptibility for these diseases.

MHC expression and sharing in animal models

Although the relative importance of HLA expression at the

maternal±fetal interface in humans remains unresolved, MHC


143

expression can be studied in many other animals, including

rodents and non-human primates. The mouse MHC region (H-2)

has been studied the most extensively. The H-2 complex spans

4000 kb on chromosome 17 (Klein et al., 1981) and contains more

than 30 genes which are grouped as class I, II, III and class II

region-encoded modi®ers. Of particular interest are the class I

genes, consisting of class Ia (classical) K, D, and L loci and the

class Ib (non-classical) Q, T, and M loci. As in humans, class Ia

products have greater expression, tissue distribution and poly-

morphism than class Ib antigens (Fernandez et al., 1995). Recent

mouse data demonstrated that class Ia, Ib and b2m were

synthesized soon after conception. Class Ia proteins were detected

on unfertilized oocytes, were absent on 1-cell embryos, present on

2-cell embryos and later stages with b2m. Class Ib expression was

detected on secondary oocytes and embryos at all stages of

preimplantation development. During and after implantation,

class Ia and II antigens were switched off in placental tissue but

class Ib antigen expression increased (Fernandez et al., 1999).

MHC distribution at the maternal±fetal interface in other

mammalian species, including sheep, pigs, cows and horses has

also contributed to our general understanding of MHC expression

during pregnancy (for reviews, see MacCluer et al., 1988;

Gautschi and Gaillard, 1990; Amills et al., 1998).

Several MHC homologues have been isolated in non-human

primates, including chimpanzees, the rhesus monkey and the

cotton-top tamarin (Boyson et al., 1995). The rhesus monkey

MHC complex has been extensively studied, since both humans

and the rhesus monkey have similar discoid placentae with villous

morphology, similar fetal membrane organization (Luckett, 1974;

Benirschke and Miller, 1982) and homologous MHC class I loci

(Boyson et al., 1996a). HLA-A, -B, -E, -F, and -G orthologues

(Mamu-A, -B, -E, -F and -G) have been identi®ed in the rhesus

monkey (Miller et al., 1991; Otting and Bontrop, 1993; Boyson et

al., 1995, 1996a,b). Although Mamu-G is an HLA-G orthologue, it

is a pseudogene (Boyson et al., 1996b). Slukvin et al. (1998)

identi®ed another class I gene, Mamu-AG, which shares

characteristics with HLA-G; including a truncated cytoplasmic

domain, alternative splicing, limited polymorphism, multiple

isoelectric isoforms and conserved residues important for NK-

receptor recognition (Boyson et al., 1997). Unlike HLA-G, it is

primarily expressed in ®rst trimester villous syncytiotrophoblasts

(Slukvin et al., 1998). Importantly, both HLA-G and Mamu-AG

are primarily expressed at the maternal±fetal interface with

restricted and low-level expression in other tissues (Slukvin et al.,

1999).

Research among non-human species living under more

controlled conditions, suggests that speci®c MHC genes, or

MHC-linked genetic regions, affect fertility, reproduction and

development. These genes affect fertility, number of offspring,

offspring birth and weaning weights, and offspring viability. For

example, studies on the rhesus monkey, pigs and horses revealed

signi®cant effects of MHC sharing on reproductive outcome. In

macaques, signi®cantly more `unsuccessful' couples shared

MnLA-A antigens than did `successful' couples (Knapp et al.,

1996). In pigs, parental sharing, or particular maternal antigens,

resulted in decreased litters (Renard et al., 1985; Gautschi and

Gaillard, 1990). Particular haplotypes in¯uenced maternal ovula-

tion rates (Rothschild et al., 1984), piglet birth and weaning

weights (Rothschild et al., 1986; Gautschi and Gaillard, 1990) and

piglet mortality (Kristensen et al., 1980). In horses, speci®c dam

MHC alleles, speci®c sire alleles and sire±dam incompatibility

were associated with foaling rate (MacCluer et al., 1988). Swine

and cheetah studies have also indicated MHC-homozygote

offspring de®ciency (Gautschi and Gaillard, 1990; Yuhki and

O'Brien, 1990). Additionally, MHC effects have been correlated

with chicken egg number (Gavora et al., 1986), rabbit ovulation

rates and embryonic loss (Gill, 1996). Thus, it is possible that

MHC or MHC-linked genes could dramatically affect fertility and

reproduction.

The rat grc region, closely linked to the MHC, is ~50 kb and

incorporates several class I genes and pseudogenes. While

wildtype rats are fertile, homozygotes for the deleted region are

small, have increased prenatal mortality, are sterile (males) or

subfertile (females) and are highly susceptible to chemical

carcinogens (Cortese Hassett et al., 1989). An uncharacterized

homologous mouse (grr) region may map distal to H-2D (Vincek

et al., 1990). Although no human homologue has been identi®ed,

Gill (1994) suggests that MHC genes associated with reproductive

disorders in the rat and human map to the same approximate

chromosomal positions. Like HLA gene epistasis in humans, the

grc region interacts epistatically with the tail anomaly lethal gene

(Tal) in rats (Schaid et al., 1982; Gill et al., 1984; Gill, 1987).

Multiple murine genes involved in spermatogenesis and

embryogenesis, known as t haplotypes, map to chromosome 17,

which is homologous to human chromosome 6 (Dobrovolskaia-

Zavadskaia, 1927; Chesley, 1932; Chesley and Dunn, 1936;

Bennett, 1975; Silver, 1985). Phenotypic effects of t haplotype

genes include, tailessness, embryonic homozygote death, hetero-

zygote male sterility and developmental abnormalities caused by

embryonic t complex lethal genes (tcl). The tcl genes, organized

into different complementation groups, disrupt particular em-

bryonic stages of development (Bennett, 1975; Artzt et al., 1982;

Silver, 1985; Klein, 1986). Heterozygous mice for two different

complementation groups are viable but sterile and homozygous

embryos of the same complementation group fail to survive. For

example, homozygosity for the t12 complementation group results

in failure of blastocyst formation and implantation (Shin et al.,

1984). Homozygosity for t0 leads to blastocyst implantation

without ectodermal organization (Ark et al., 1991) and homo-

zygosity for t9 results in primitive streak overgrowth and

duplication of neural tissue during days 8±10 (Bennett and

Dunn, 1960; Ben-Shaul et al., 1983). Interestingly, some t

haplotypes of these complementation groups map to the mouse

MHC H-2 complex (i.e. near H-2Q and H-2K). For example, the

tcl-w5 haplotype, which causes ectoderm degeneration before

embryo differentiation, maps 200 kb of H-2K and is inseparable

by recombination (Shin et al., 1984; Artzt et al., 1988). Kostyu

(1994) suggests that mutations, deletions, duplications, or

abnormal expression of possible HLA-linked t homologues, may

be responsible for some of the HLA associations with various

reproductive or developmental disorders. However, only a few

human t homologous genes have been mapped. TCP11 maps 2

Mb centromeric to the HLA (Ragoussis et al., 1992) while TCP1

and TCP10 are unlinked to the HLA and map to the long arm (q)

of chromosome 6 (Blanche et al., 1992: Masuno et al., 1996).

Thus, human t homologous haplotypes may not exist, since

mapped t homologue genes are separated over several chromo-

somal regions (Bibbins et al., 1989). Although few studies have


144

investigated the existence of human t haplotypes, many

researchers have mapped genes within or near the H-2 region

and HLA, which are involved in gametogenesis, embryogenesis

and fetal development and surivival (Table I). These genes

encode proteins, which either have direct effects upon fertility and

fetal development and survival, or encode transcription factors,

which could regulate expression of MHC and MHC-linked genes.

The functions of these genes have primarily been determined in

rodents, although some human homologues have been found to

play similar roles in human fertility, fetal development and

survival. Considering that the HLA complex displays particularly

high levels of linkage disequilibrium, it is likely that these

homologues are linked to the HLA.

Studies of the Ped gene illustrate the utility of studying the

function of rodent MHC or MHC-linked genes for comparison

with humans. The mouse Ped (pre-implantation embryonic

development) gene of the H-2 region, directly affects embryonic

viability and fetal development. This gene encodes the class Ib,

Qa-2 antigen. The Ped phenotype manifests itself during the ®rst

cleavage division (Goldbard et al., 1982) and is present at the 2-

cell stage (Warner et al., 1987; Xu et al., 1994). The Ped (fast)

allele is dominant and associated with faster in-vitro and in-vivo

cleavage rates. The Ped slow allele shows a 5 h delay in the ®rst

cleavage division (Xu et al., 1993). Notably, the Ped fast allele is

associated with signi®cantly greater cell numbers in the inner cell

mass and the trophectoderm, larger litter sizes and greater birth

weights and weaning weights (Warner et al., 1991; McElhinny et

al., 1998). Moreover, Ped may in¯uence longevity since Ped fast

strains seem to have shorter life spans that Ped slow strains

(TarõÂn, 1997). Ped expression may also be controlled by genes

that regulate apoptosis, since mRNA and protein of the Bcl-2 and

caspase gene families that control apoptosis have been detected in

pre-implantation embryos. A homeostatic mechanism, by genes

that regulate cell survival (e.g. Ped), and those that regulate cell

death (e.g. Bcl-2), may determine the overall viability of pre-

implantation embryos (Warner et al., 1998). A human Ped

homologue may exist since fertilized IVF oocytes have different

cleavage rates and fast-developing human pre-implantation

embryos have better survival rates than slow-developing embryos

(Bolton et al., 1989; Ziebe et al., 1997; Sakkas et al., 1998). Since

Qa-2 is encoded by class Ib genes, the putative human homologue

could be either HLA-E, -F or -G. Subsequently, 108 spare day 3

human preimplantation embryos were examined for class Ib

mRNA by RT±PCR. HLA-E mRNA was detected in 84% of 86

embryos, HLA-G mRNA was detected in 44% of 88 embryos and

HLA-F was detected in 0% of 17 embryos (Cao et al., 1999).

These authors do not believe that HLA-E is the Ped gene since it

differs in tissue distribution from Qa-2. The HLA-E mouse

homologue has also recently been suggested to be Qa-1 (for

review, see Long, 1998). It is also unlikely that HLA-F is the Ped

gene, since it was not detected in day 3 human embryos. HLA-G

has previously been suggested to be the Ped homologue

(Juriscova et al., 1996) since Qa-2 and HLA-G both have similar

peptide ligand repertoires, form truncated membrane-bound and

soluble forms and have been detected in pre-implantation

embryos and trophoblast (Stroynowski and Tabaczewski, 1996).

However, HLA-G and Qa-2 differ in several respects. HLA-G

lacks the interferon response sequence (IRS) in the 5¢ region of

Qa-2. They also differ in tissue distribution (Wei and Orr, 1990;

Schmidt and Orr, 1993) and bind different types of peptides

(O'Callaghan and Bell, 1998). Unlike Qa-2, HLA-G is not linked

to the cell surface by a glycosylphosphatidylinositol (GPI) linkage

(Cao et al., 1999). Moreover, the mouse gene, Blastocyst MHC,

may be the HLA-G homologue since the protein has a similar

peptide-binding region to HLA-G, a truncated cytoplasmic tail and

is transcribed in blastocyst and placenta (Sipes et al., 1996). Cao

et al. (1999) also suggested that the putative human Ped gene is

linked to HLA-G and may be HLA-C. Similar to Qa-2, HLA-C is

also linked by a GPI linkage to the cell surface and may be

involved in reproductive events since it is expressed on

extravillous trophoblast (King et al., 1997b). Although Ped is a

class Ib gene and HLA-C is a class Ia gene, these classi®cations

differentiating between class I and class II genes may be too

speci®c (Hughes et al., 1999). Finally, Qa-2 is encoded by two

class Ib genes, Q7 and Q9, suggesting that several HLA class I

genes (Cao et al., 1999) or perhaps other HLA-linked genes could

contribute to a human Ped phenotype. These studies clearly

demonstrate the value of investigating MHC, or MHC-related,

genetic effects in animal models to determine how human

homologues could have similar functions in human fertility,

reproduction and development.

Interactions: cytokines and HLA expression

Several factors have been associated with one another among

women with recurrent miscarriage. Christiansen et al. (1989)

reported 13% of non-pregnant women with recurrent miscarriage

were positive for two or more serological markers associated with

autoimmune disorders compared with 2% of controls (P = 0.01).

Kwak et al. (1992) found that 16% of women with recurrent

miscarriage were positive for at least two autoantibodies

compared with 8% of controls (P = 0.06). This indicates that

multiple immunological factors may in¯uence pregnancy out-

come, and when combined with immunogenetic factors, they

could form intricate networks and complex feedback loops. In this

context, HLA expression may play a pivotal role by regulating

interactions between factors, e.g. cytokines and natural killer

(NK) cells (Figure 3).

Figure 3. Immunological and immunogenetic factors form networks andfeedback loops which in¯uence human reproduction. See text for details.LIF = leukaemia inhibitory factor; MUC-1 = mucin-1.


145

Tab

leI.

Gen

esw

ithin

or

nea

rth

eH

-2re

gio

nan

dH

LA

,w

hic

hin

¯uen

cefe

rtil

ity,

repro

duct

ion

and

dev

elopm

ent

Hu

man

gen

eF

ull

nam

eH

um

anlo

cati

on

Muri

ne

gen

eM

uri

ne

loca

tion

In-v

itro

or

in-v

ivo

pro

tein

funct

ion

Exp

ress

ion

of

mR

NA

or

pro

tein

Ref

eren

ces

BA

Ts

(BA

T1

-9)

dH

LA

-Bass

oci

ate

d

transc

ripts

dT

elom

eric

end

of

clas

sII

I

d6p21.3

dB

ats

dH

-2re

gio

nd

Mem

ber

of

DE

AD

-box

fam

ily

of

AT

P-d

epen

den

tR

NA

hel

icas

esa,b

dD

EA

D-b

ox

isa

char

acte

rist

ic

Asp

-Glu

-Ala

-Asp

,1

of

8hig

hly

conse

rved

sequen

cem

oti

fsa,b

dIn

volv

edin

tran

slat

ion

init

iati

on,

RN

Asp

lici

ng

and

riboso

me

asse

mbly

a,b

dV

ario

us

tiss

ues

and

cell

lines

aS

pie

set

al.

(1989)

Nunes

etal.

(1994)

Pee

lman

etal.

(1995)

Alc

ock

etal.

(1999)

HK

E4

an

d

HK

E6

dK

E4,

KE

6,

mouse

,

hom

olo

gue

dC

entr

om

eric

end

of

HL

A

d6p21.3

dK

e4,

Ke6

dO

ver

laps

H-2

Kre

gio

n

dH

2K

e1-5

=17p18.4

±18.4

9

dK

e6is

involv

edw

ith

kid

ney

funct

ion

b

dK

e6b-

hydro

xyst

eroid

deh

ydro

gen

ase

regula

tes

oes

trogen

and

andro

gen

conce

ntr

atio

nsb

dK

e6co

ntr

ols

ster

oid

level

s

expose

dto

oocy

te?b

dH

KE

4=

pla

centa

,lu

ng,

kid

ney

,pan

crea

sa

dH

KE

6=

liver

,pan

crea

sa

dK

e6=

ovar

ies

and

test

esb

Abe

etal.

(1988)

Han

son

and

Tro

wsd

ale

(1991)

Janat

ipour

etal.

(1992)

Yeo

met

al.

(1992)

Ando

etal.

(1996)

Kik

uti

etal.

(1997)

Fom

itch

eva

etal.

(1998)

DD

R1

(NE

P,

ED

DR

1,

NT

RK

4,

TR

KE

,

DD

R,C

AK

)

dD

isco

idin

dom

ain

rece

pto

rfa

mil

y,

mem

ber

1

d6p21.3

dN

epd

17C

dP

rote

inty

rosi

ne

kin

aseb

dM

ember

of

gen

efa

mil

yth

atco

ntr

ols

neu

ron

dev

elopm

ent,

dif

fere

nti

atio

n

and

mai

nte

nan

cea

dF

etal

and

adult

tiss

ues

bD

iM

arco

etal.

(1993)

Edel

hoff

etal.

(1995)

NF

YA

dN

ucl

ear

transc

ripti

on

fact

or

Y,

ad

6p21

dN

fya

d17p26

dT

ransc

ripti

on

fact

or?

dE

ssen

tial

for

clas

sII

MH

Cex

pre

ssio

n

dR

ecogniz

esC

CA

AT

T

dM

any

cell

types

a,b

Bec

ker

etal.

(1991)

Li

etal.

(1991)

NO

TC

H4

dN

otc

hd

6p21.3

dIn

t3d

17p18.7

dE

ndoth

elia

ldev

elopm

ent?

bd

Endoth

elia

lce

lls

in

emb

ryonic

and

adult

tiss

ues

b

Uytt

endae

leet

al.

(1996)

Li

etal.

(1998)

PO

U5

FL

(OC

T3

,O

CT

4)

dP

ou

dom

ain

,cl

ass

5,

transc

ripti

on

fact

or

1

d100

kb

telo

mer

icto

HL

A-C

d6p21.3

dP

ou5¯

dB

etw

een

Qan

dT

regio

n

d17p19.2

3

dT

ransc

ripti

on

fact

or

dR

egula

tes

tiss

ue

spec

i®c

expre

ssio

n

inly

mphoid

and

pit

uti

ary

dif

fere

nti

atio

nin

earl

ydev

elopm

entb

dIn

volv

edin

stem

cell

signal

ling

totr

ophec

toder

mb

dO

ocy

tes

bef

ore

and

afte

r

fert

iliz

atio

nb

dM

oru

laan

dbla

stocy

stb

dD

ow

n-r

egula

ted

duri

ng

endoder

man

dm

esoder

m

dif

fere

nti

atio

nb

Okam

oto

etal.

(1990)

Rosn

eret

al.

(1990)

Yeo

met

al.

(1991)

Tak

eda

etal.

(1992)

Guil

laudeu

xet

al.

(1993)

Cro

uau

-Roy

etal.

(1994)

Nic

hols

etal.

(1998)

RX

Rb

dR

etin

oid

X

rece

pto

r,b

d6p21.3

dR

xrb

dH

-2re

gio

n

d17p18.4

9

dB

inds

5¢e

nhan

cers

a

dR

egula

tes

clas

sI

expre

ssio

na

dF

orm

sre

cepto

rsfo

rre

tinoic

acid

,

thyro

idhorm

one

and

vit

amin

D,

enhan

cing

DN

Are

sponse

elem

ent

inte

ract

ionsa

dIn

volv

edin

loco

moto

rbeh

avio

urb

dC

ontr

ols

dopam

iner

gic

mes

oli

mbic

pat

hw

ayb

dS

tria

tum

bH

amad

aet

al.

(1989)

Yu

etal.

(1991)

Hoopes

etal.

(1992)

Nag

ata

etal.

(1992)

Fit

zgib

bon

etal.

(1993)

Alm

asan

etal.

(1994)


146

Tab

leI.

Conti

nued

Hu

man

gen

eF

ull

nam

eH

um

anlo

cati

on

Mu

rin

eg

ene

Muri

ne

loca

tion

In-v

itro

or

in-v

ivo

pro

tein

funct

ion

Expre

ssio

nof

mR

NA

or

pro

tein

Ref

eren

ces

BA

K1

(BC

L2L

7)

dB

CL

2anta

gonis

t

kill

er1

d6p21.3

dB

ak

d17B

dO

nco

gen

ea

dA

ctiv

ates

cell

dea

thby

mit

och

ondri

al

mem

bra

ne

contr

ola

dC

ounte

ract

sB

CL

2pro

tect

ion

from

apopto

sisa

dIn

foll

icula

rly

mphonas

aC

hit

tened

enet

al.

(1995)

Kie

fer

etal.

(1995)

Ulr

ich

etal.

(1997)

Her

ber

get

al.

(1998)

MA

PK

(p3

8-d

)M

ito

gen

-act

iva

ted

pro

tein

kin

ase

d6p21.3

dp38-d

d17A

3-B

dM

ember

of

kin

ase

fam

ily

that

tran

duce

sex

trac

ellu

lar

stim

uli

into

intr

acel

lula

rsi

gnal

s,co

ntr

oll

ing

gen

e

expre

ssio

n

dR

ole

inea

rly

dev

elopm

ent?

b

dA

ctiv

ated

by

envir

onm

enta

lst

ress

and

cyto

kin

esb

dA

uniq

ue

stre

ssre

sponsi

ve

kin

aseb

dA

dult

lung,

test

is,

kid

ney

and

gut

epit

hel

ium

b

dH

igh

expre

ssio

nin

embry

ogut

and

septu

m

tran

sver

sum

in9.5

day

embry

osb

dIn

crea

sed

expre

ssio

nin

oth

erti

ssues

in12.5

day

embry

osb

Cobb

and

Gold

smit

h(1

995)

Hu

etal.

(1999)

TC

P11

dT

-com

ple

x

hom

olo

gue

11

d6p21.3

-p21.2

dT

cp11

dD

ista

lin

ver

sion

of

the

t-co

mple

xd

Clo

sely

linked

toa

Zn

®nger

gen

e,

ZN

F76,

mem

ber

of

GL

IK

ruppel

fam

ily

a,b

dD

evel

opm

enta

lly

regula

ted

b

dT

esti

sb

dZ

nf7

6an

dtc

p11

expre

ssed

sim

ult

aneo

usl

y

inte

stis

b

Rag

ouss

iset

al.

(1992)

TC

TE

1d

T-c

om

ple

x-

ass

oci

ate

d-t

esti

s-

expre

ssed

1

dC

entr

om

eric

toH

LA

d6p21.3

-cen

dT

cte-

1d

17p23.5

0d

Tra

nsc

ripti

onal

regula

tor?

dT

esti

sbS

arvet

nic

ket

al.

(1989)

Sar

vet

nic

ket

al.

(1990)

Kw

iatk

ow

ski

etal.

(1991)

PIM

1d

Onco

gen

eP

IM-1

dC

entr

om

eric

toH

LA

-DP

d6p21.2

dP

im1

dIn

dis

tal

inver

sion

d17p16.4

dP

roto

-onco

gen

ea

dP

rote

in-s

erin

e/th

reonin

ekin

asea

,b

dD

iffe

renti

atin

g

embry

onic

cell

s,ad

ult

sper

mat

ids

and

leukae

mia

sa

dH

igh

expre

ssio

nin

feta

l

liv

eran

dsp

leen

hae

mat

opoie

sisa

Nag

araj

anet

al.

(1986)

Sorr

enti

no

etal.

(1988)

Am

son

etal.

(1989)

Dom

enet

al.

(1987)

Sar

iset

al.

(1991)

CD

KN

IA

(CIP

,W

AF

1,

p2

1)

dC

ycli

n-d

epen

den

t

kinase

inhib

itor

1A

d6p21.2

dW

af1

dP

roxim

alto

H-2

regio

nd

Inhib

its

cycl

inkin

ase

acti

vit

ya

dR

egula

ted

by

p53

a

dp53

effe

ctor

of

cell

cycl

eco

ntr

ol?

a

dIn

most

tiss

ues

bH

uppi

etal.

(1994)

Dem

etri

cket

al.

(1995)

FG

D2

dF

aci

ogen

ital

dys

pla

sia

2

d6p21.2

dF

gd2

d17

ort

holo

gue

dG

uan

ine

nucl

eoti

de

exch

ange

fact

ora

dM

uta

tions

adver

sely

affe

ct

skel

etal

form

atio

na

dR

egula

tes

earl

ydev

elopm

ent?

b

dD

iver

seti

ssues

b

dD

uri

ng

embry

ogen

esis

b

Pas

teri

san

dG

ors

ki

(1999)

Table

I.co

nti

nued

ove

rpage


147

Tab

leI.

Conti

nued

Hu

man

gen

eF

ull

nam

eH

um

anlo

cati

on

Mu

rin

egen

eM

uri

ne

loca

tion

In-v

itro

or

in-v

ivo

pro

tein

funct

ion

Expre

ssio

nof

mR

NA

or

pro

tein

Ref

eren

ces

AE

GL

Id

Aci

dic

Epid

idym

al

Gly

copro

tein

-lik

e1

d6p21.1

-p21.2

(ort

holo

gue)

dA

eg1

dH

-2M

regio

nd

Sper

msu

rfac

epro

tein

involv

edin

gam

ete

fusi

on

b

dC

oat

ssp

erm

ente

ring

epid

idym

isb

dC

ause

ssp

erm

neg

ativ

ech

arge?

b

dR

egula

ted

by

andro

gen

b

dA

egan

tibodie

sblo

ckfe

rtil

izat

ion

b

dP

uri

®ed

Aeg

bin

ds

tooocy

te

and

inhib

its

fusi

on

b

dE

pid

idym

alce

llsb

dE

pid

idym

is,

duct

us

def

eren

s,se

min

alpla

sma,

sper

mat

ozo

aa

Hay

ashi

etal.

(1996)

Yosh

ino

etal.

(1998)

RU

NX

2

(CB

FA

1,

AM

L3

,

PE

BP

2A

,O

SF

2)

dR

unt-

rela

ted

transc

ripti

on

fact

or

2

d6p21

dC

bfa

1d

17p28.0

7d

Ost

eobla

st-s

pec

i®c

tran

scri

pti

on

fact

orb

dC

ontr

ols

ost

eobla

stdif

fere

nti

atio

n

duri

ng

embry

onic

dev

elopm

entb

dC

ontr

ols

bone

mat

rix

dep

osi

tion

of

dif

fere

nti

ated

ost

eobla

sts

post

nat

ally

b

dO

steo

bla

stce

llsa

Lev

anon

etal.

(1994)

Ducy

etal.

(1997)

Ducy

etal.

(1999)

CC

ND

3d

Cyc

lin

D3

d6p21

dcy

clin

d3

dU

nknow

nd

Contr

ols

cell

cycl

ea

dC

ontr

ols

cell

div

isio

nan

d

dif

fere

nti

atio

nin

the

ger

mli

neb

dIm

port

ant

for

dec

idual

izat

ion

b

dE

xpre

ssed

duri

ng

gam

etogen

esis

and

earl

y

embry

ogen

esis

b

dU

p-r

egula

ted

at

bla

stocy

stap

posi

tion

duri

ng

impla

nta

tion

b

dU

p-r

egula

ted

indec

iduab

Inab

aet

al.

(1992)

Moto

kura

etal.

(1992)

Xio

ng

etal.

(1992)

Chap

man

&W

olg

emuth

(1993)

Das

etal.

(1999)

Zhan

get

al.

(1999)

aH

um

anp

rote

in.

bR

od

ent

pro

tein

.


148

Cytokines are small, secreted, membrane-bound proteins that

act through cell-surface receptors and generally induce changes in

gene expression within their target cells (Parham, 2000). They

also play critical roles in pathogen response and regulation of

cellular growth and differentiation. Currently, cytokines are

known to in¯uence several stages of reproduction (Figure 1).

Cytokines e.g. ILs, IFNs and TNFs, can modulate HLA

expression throughout pregnancy (Figure 1). IL-4 can induce

class II expression and IL-10 can increase class I and II

expression (Parham, 2000). Speci®cally, IL-10 enhances HLA-G

transcription in human ®rst trimester cultured trophoblast cells.

IL-10 also up-regulates HLA-G surface-expression in peripheral

blood monocytes (PBMC)s, while other classical class I and class

II genes are simultaneously down-regulated (Moreau et al., 1999).

IFNs can induce class I mRNA and protein expression from

various types of class I negative cells (Wan et al., 1987). In vitro,

IFNs can induce class Ia expression, but not all of the class Ib

antigens (for review, see Singer and Maguire, 1990). IFNs can

induce class I expression from day 8.5 embryonic cells, which

lack class I expression (Ozato et al., 1985; Miyazaki et al., 1986).

Generally, IFN-g is the most potent cytokine since it is an inductor

of Th1 responses and a negative regulator of B cell differentiation

and proliferation (O'Neil et al., 1999). Speci®cally, IFN-g induces

the formation of a nuclear binding protein, which binds to class I

¯anking regulatory sequences, increases class I expression and

decreases class II expression (Brown et al., 1993). IFN-g may also

be an important cytokine involved in maternal±fetal interactions

since IFN-g can induce some cell lines to express HLA-G

(Anderson and Berkowitz, 1985) and it enhances class Ia mRNA

expression in amnion epithelial cells (Kovats et al., 1990; Wei

and Orr, 1990). Interestingly, King et al. (2000a) reported that

IFN-g signi®cantly up-regulated in-vivo surface expression of

HLA-C compared with HLA-G expression in extravillous

trophoblast cells. Furthermore, cytokines are known to in¯uence

each other. For example, IFN-g is regulated by the cytokine IL-

1b, which can suppress IFN-g induced class II expression (Rohn

et al., 1999).

TNFs may also play critical roles in pregnancy maintenance.

For example, TNF enhances T-cell proliferation, modulates T-cell

receptor expression, enhances NK activity and regulates B cell

functions (Aguado et al., 1996). TNF treatment of primary and

established cell lines increased class I cell-surface expression

(Collins et al., 1986). TNF treatment of human endothelial and

dermal cells increased class I protein and mRNA expression 8±

10-fold and 100-fold respectively. Overall, cytokines can have

different, yet simultaneous, effects on HLA expression on cells

involved in reproductive events. These effects may be directly

related to the proliferation and immunotolerance of trophoblast,

or fetal, cells. Unfortunately, precise cytokine/HLA interactions

and timing of interactions during pregnancy remains unresolved.

Interactions: NK cells and HLA expression

NK cells may also contribute to human reproduction during many

stages of pregnancy (Figure 1). Peripheral NK cells are found

throughout the body's circulatory system and a subpopulation of

NK cells, uterine NK cells can be found at the maternal±fetal

interface. Functionally, NK cells are capable of cytotoxic killing

of target cells (Parham, 2000) and studies continue to investigate

whether reproductive outcome is in¯uenced by changes in NK

cell number, function or target-cell recognition (for review, see

Roy Choudhury and Knapp, 2001). NK cell recognition of HLA

antigens may be important during reproduction since NK binding

to HLA antigens can activate, or inhibit, NK target cell killing

(Yokoyama and Seaman, 1993), or alter NK cytokine secretion

(Loke and King, 2000). Three families of peripheral NK

inhibitory and activatory receptors, which recognize HLA

antigens, have been identi®ed. The Killer cell immunoglobulin-

like receptors (KIR), bind HLA-B and -C (Biassoni et al., 1995;

Gumperz et al., 1995; Mandelboim et al., 1996; Gumperz et al.,

1997). The CD94/NKG2 type II membrane glycoprotein of the C-

type lectin superfamily, binds HLA-E, rather than a broad range

of HLA class I antigens as previously suggested (Borrego et al.,

1998; Braud et al., 1998a; Lee et al., 1998; Brooks et al., 1999)

and the Immunoglobulin-like transcript (ILT) family (Samaridis

and Colonna, 1997; Cosman et al., 1997), binds HLA-A, -B, -C

and -G (Borges et al., 1997; Colonna et al., 1997; Cosman et al.,

1997; Fanger et al., 1998; Vitale et al., 1999).

Uterine NK cells also express KIR, CD94/NKG2 and ILT

receptors (Hiby et al., 1997), but assays of peripheral NK and

uterine NK receptors in the same individual, reveal different

proportions in these cells. Moreover, the expression patterns of

these NK receptors vary between women (Verma et al., 1997).

Thus, for these reasons understanding the interactions of uterine

NK cells with HLA expressing cells at the maternal±fetal

interface requires clear identi®cation of uterine NK receptors

(see Roy Choudhury and Knapp, 2001).

Uterine NK cells are in close association with invading

extravillous trophoblast cells in the uterus (King et al., 1998),

which express HLA-C, -E and -G antigens (King et al., 1996b;

King et al., 2000a,b,c). Therefore, HLA expression may regulate

NK activity at the maternal±fetal interface. Peripheral and uterine

NK cell recognition of HLA-C has been demonstrated by several

groups (Colonna et al., 1993; Moretta et al. 1993; Vitale et al.,

1995). Interestingly, uterine NK cells have higher proportions of

KIR receptors for HLA-C and CD94/NKG2 receptors for HLA±E

than peripheral NK cells (King et al., 1997a; King et al., 2000c).

HLA-E binding to CD94/NKG2 uterine NK receptors inhibits

cytotoxity to cell lines. However, uterine NK cells were unable to

kill trophoblast cells when monoclonal antibodies blocked CD94/

NKG2 receptors or HLA class I antigens, or when class I antigens

were removed by acid. This suggests that in-vivo trophoblast

HLA-E binding with CD94/NKG2 receptors on uterine NK cells

may regulate functions other than inhibiting cytotoxicity during

reproduction (King et al., 2000b).

Many studies have focused upon NK receptor recognition of

the HLA-G antigen due to the unique distribution of HLA-G at

the maternal±fetal interface. Uterine NK cell recognition of HLA-

G expressing extravillous trophoblast may inhibit NK trophoblast

lysis. Chumbley et al. (1994) ®rst reported that HLA-G

transfection in a B lymphoblastoid cell line protected ®rst

trimester trophoblasts from decidual NK cell lysis. This has been

con®rmed by many groups (Deniz et al., 1994; Pazmany et al.,

1996; Mandelboim et al., 1997; Rouas-Freiss et al., 1997a,b;

Soderstrom et al., 1997; Colonna et al. 1998; Navarro et al., 1999;

Ponte et al., 1999). HLA-G1 and -G2 isoforms also can inhibit cell

lysis by peripheral NK cells (Perez-Villar et al., 1997; Rouas-

Freiss et al., 1997a,b). In addition, HLA-G1 inhibited NK cell


149

movement across porcine endothelial cells (Durling et al., 2000),

suggests that different HLA-G isoforms may have specialized

functions (Ishitani and Geraghty, 1992).

The type of uterine NK receptors, which bind HLA-G remains

controversial. However, in-vitro studies using .221-HLA-G

transfectants as target cells and peripheral blood NK cells as

effector cells may not parallel in-vivo interactions at the

maternal±fetal interface (Le Bouteiller and Blaschitz, 1999).

Earlier studies demonstrated that CD94/NKG2 recognized HLA-

G. Yet later studies demonstrated that NK cells recognized HLA-

E bound to HLA-G leader peptides, allowing HLA-E surface

expression and recognition by the CD94/NKG2A receptor (Pende

et al., 1997, Perez-Villar et al., 1997; SoderstroÈm et al., 1997;

Llano et al., 1998; Navarro et al., 1999). ILT2 and ILT4,

members of the ILT family, directly bind HLA-G and inhibit

target cell lysis (Colonna et al., 1997; Colonna et al., 1998; Allan

et al., 1999; Navarro et al., 1999; Vitale et al., 1999). However,

the broad speci®city of ILT2 and ILT4 for other HLA class I

antigens suggests that HLA-G binding to these NK receptors does

not provide a unique `fetal' signal at the maternal interface (King

et al., 2000c). Several KIR receptors that bind HLA-G and inhibit

NK cell lysis have been identi®ed, including p49 (Cantoni et al.,

1998) and KIR2DL4 (Ponte et al., 1999, Rajagopalan and Long,

1999). Interestingly, HLA-G1 recognition by NK KIR receptors

inhibited target killing of K562 cell lines. These cell lines lacked

HLA-E expression, demonstrating that HLA-G1 was capable of

inhibiting NK cytotoxicity without NK CD94/NKG2 receptor

recognition of HLA-E (Khalil-Daher et al., 1999). One KIR

receptor, KIR2DL4 may display both inhibitory and activatory

functions (King et al., 2000c), suggesting a unique role for the

interaction of HLA-G and KIR2DL4 at the maternal±fetal

interface. Yet the distribution of this NK receptor remains

controversial since Rajagopalan and Long (1999) reported that all

peripheral cells express KIR2DL4, while Ponte et al. (1999)

reported that only some ®rst trimester uterine NK cells and all

term placental NK cells expressed KIR2DL4.

In summary, studies indicate that NK cells are capable of

recognizing HLA expressing trophoblast during pregnancy and

may inhibit cytotoxic trophoblast killing. King et al. (2000c)

suggest that placentation, or cytokine secretion, may be regulated

by a variety of CD94/NKG2, KIR and ILT uterine NK cell

receptors, which recognize HLA expressing trophoblast cells.

Thus, reproductive failure may occur if trophoblast cells lack or

have aberrant HLA expression or if NK cells lack appropriate

inhibitory receptors (Lopez-Botet et al., 1996).

Interactions: cytokines and NK cells

Both cytokines and NK cells play fundamental roles in pregnancy

maintenance. Not surprisingly, NK cells and cytokines act

together to orchestrate regulatory pathways of pregnancy. For

example, IFN-g, IFN-a, IFN-b, IL-2, IL-7, IL-6, IL-12 and IL-15

exert direct and potent stimulatory effects on NK activity (King

and Loke, 1990; Naume and Espevik, 1994; King et al., 1999).

These activities can include NK proliferation, NK or LAK non-

restricted killing or NK cytokine secretion (Naume and Espevik,

1994). IL-2 and IL-15 are the only cytokines, which stimulate in-

vitro uterine NK cell proliferation (Ferry et al., 1990; Nishikawa

et al., 1991; King et al., 1992; Verma et al., 2000). Since IL-2 is

absent in the decidua and placenta, it is unlikely to be an

important regulatory factor during early pregnancy. However, IL-

15 is found in the endometrium, decidua and placenta and is

particularily expressed by decidual macrophages, suggesting it

may in¯uence uterine NK cell proliferation (Verma et al., 2000).

Some cytokines, e.g. TNF and IL-2, induce NK cells and

cytotoxic lymphocytes to develop into LAK cells, which are

capable of killing trophoblast cells (Drake and Head, 1989) and

may contribute to fetal loss. Although LAK in®ltration and

trophoblast lysis may not have been convincingly demonstrated,

since LAK cells cannot kill freshly isolated human trophoblast

cells without in-vitro tissue culture (Clark, 1991; Ferry et al.,

1991; Clark et al., 1999).

Although various studies have demonstrated that NK cells

secrete a range of cytokines, contaminating cells in isolated NK

populations may have confounded results (Trinchieri, 1989).

Generally though, IFN-a, IFN-g, TNF-a and IL-2 are secreted by

peripheral NK cells in response to target cells and other stimuli

(King and Loke, 1990). Uterine NK cells have also been shown to

secrete a wide variety of cytokines, including granulocyte

macrophage±colony stimulating factor (GM-CSF), CSF-1 and

LIF (Saito et al., 1993; Jokhi et al., 1994a,b,c) and trophoblast

cells express receptors for these cytokines (Jokhi et al., 1993,

1994c; King et al., 1995, 1999). Moreover, the pattern of cytokine

production by uterine NK cells differs from peripheral NK cells,

suggesting that these cytokines may have uterine speci®c

functions. Thus, cytokines secreted by uterine NK cells may be

involved in a cytokine network at the maternal±fetal interface and

in¯uence critical reproductive events such as trophoblast

proliferation and differentiation during implantation (Loke and

King, 2000).

Baines et al. (1997) proposed a model of cytokine and

uterine NK interactions based upon studies of the decidua of

CBA3DBA/2 abortion-prone mice. These observations in-

cluded: (i) signi®cantly increased NK-like and macrophage

in®ltration and increased mRNA expression of nitric oxide,

TNF-a, or IFN-g preceding early fetal resorption; (ii) the

priming of resting macrophages by IFN-g; and (iii) increased

TNF-a and nitric oxide secretion by activated macrophages, if

triggered by a second signal, e.g. TNF-a or lipopolysaccharide

(Ding et al., 1988). Baines et al. (1997) suggested that NK

cells are activated by a physiological stimulus to produce IFN-

g and TNF-a. The release of IFN-g could activate resting

macrophages. These activated macrophages, with the simulta-

neous or subsequent production of TNF-a, could produce

cytotoxic molecules, e.g. nitric oxide, and lead to damage of

placental cells. This model illustrates the important interactions

that may occur at the maternal±fetal interface which involve

several different factors.

Importantly, there may be critical differences in NK and

cytokine pathways between mice and humans. In mouse decidua,

gd T-cells play a dominant role in Th1/Th2 cytokine pro®le

production at the maternal±fetal interface; whereas in human

decidua, CD56+ CD16± NK cells and macrophages seem to

predominate (Clark, 1991; Maruyama et al., 1992; Chernyshov et

al., 1993). Moreover, human decidua have no obvious counterpart

for the mouse's large, NK-lineage metrial gland-type cells, which

secrete nitric oxide and cytokines (Croy, 1990; Parr et al., 1995;

Hunt et al., 1997). However, the smaller CD56+ human cells can


150

also produce cytokines and possibly nitric oxide (Saito et al.,

1993). Thus, different cytokines affect peripheral and uterine NK

cell activity and NK cells are capable of their own cytokine

secretions (Jokhi et al., 1993, 1994c; King et al; 1995). Future

studies should elucidate the type and temporal sequence of

cytokine/NK interactions at the maternal±fetal interface and

determine whether the inappropriate regulation of these pathways

contribute to reproductive failure.

Other interacting factors

In addition to the interactions between NK cells, the HLA

cytokines, there are even more potential interactions between

factors during reproduction. Cytokines, e.g. TNF-a, IL-1a, IL-6,

IL-4 and transforming growth factor (TGF)-1b interact with LIF

(Lorenzo et al., 1994; Wetzler et al., 1994; Arici et al., 1995;

Delage et al., 1995; van Eijk et al., 1996), which may in¯uence

human reproduction from gamete development through implanta-

tion (for review, see Roy Chaudhury and Knapp, 2001). LIF

transcription is induced by IL-1a, IL-1b, TGF-b, and TNF-a(Allan et al., 1990; Arici et al., 1995). Cytokines are also secreted

by many other immune cells, including gd T-cells. Some authors

suggest that these endometrial/decidual cells may be important for

maternal±fetal interactions (Janeway et al., 1988; Mincheva-

Nilsson et al., 1992). In non-pregnant mice, gd T-cells mediate

both Th1 and Th2 responses (Bluestone et al., 1991; Chien et al.,

1996; Fu et al., 1994; Weintraub et al., 1994; Hsieh et al., 1996;

Tanaka et al., 1996); and they can secrete cytotoxic, in¯ammatory

Th1 cytokines, e.g. IL-2, when cultured in vitro with mouse or

human trophoblast cells (Clark et al., 1999). They also can secrete

Th2 cytokines, promoting antibody production, such as IL-10 and

TGF-b2, in the decidua of mice experiencing high rates of fetal

resorptions (Fu et al., 1994; Hsieh et al., 1996; Arck et al., 1997).

Although associating a particular cytokine pro®le with gd T-cells

remains controversial (for review, see Szekeres-Bartho et al.,

1999). In contrast, cytokine expression is correlated with

production of LIF. Piccinni et al. (1998) demonstrated that LIF-

producing T-cells were down-regulated by Th1 cytokines, IL-12,

IFN-g and IFN-a, and up-regulated by Th2 cytokines, IL-4 and

progesterone. Decreased LIF and Th2 (IL-4, IL-10) cytokine

production by decidual T-cells was also detected in women with

recurrent miscarriage compared with fertile controls. Other

factors expressed in the endometrium, e.g. MUC1, may also be

important for NK interactions. Mucins may act as anti-adhesion

molecules on the uterine epithelium and MUC1 down-regulation

may be critical for successful implantation (for review, see

Chaudhury and Knapp, 2001). Zhang et al. (1997) demonstrated

that MUC1 expression can inhibit NK target cell lysis in a dose-

dependent way. Since activated NK (LAK) cells are capable of

killing trophoblast cells (Drake and Head, 1989), MUC1 may

inhibit NK lysis at the maternal±fetal interface. These studies

demonstrate just a limited number of the potential interactions

that may occur between factors during pregnancy. The dis-

turbance of any of these pathways, or the aberrant, or lack of

expression, of factors could thereby, contribute to reproductive

failure. Thus, while the independent effects of each factor is

important, there remains a critical need to investigate the

interactions between factors during pregnancy.

Summary and conclusion

Although the role of HLA during gamete development and

fertilization remains controversial, HLA tissue-speci®c distribu-

tion may directly in¯uence embryo cleavage through post-

parturition (Figure 1). For this reason, HLA distribution at the

maternal±fetal interface is an important ®eld of investigation. The

expression of HLA-G and -C on extravillous trophoblast cells

(King et al., 1998, 2000a,b) suggests that these molecules may

interact with maternal immune cells to inhibit fetal rejection.

HLA-C may be the human homologue of the Ped phenotype,

which directly affects embryonic and fetal development (Cao et

al., 1999). HLA-G exhibits unique characteristics suggesting that

it may be involved in critical maternal±fetal interactions. HLA-G

expression is present throughout gestation. It demonstrates lower

polymorphism than other class I loci and possesses unique

structural properties and transcriptional isoforms (Le Bouteiller et

al., 1999). It has been suggested that HLA-G may prevent

reproductive failure since it could provide fetal protection from

pathogens. HLA-G may also prevent maternal rejection of the

fetus, modulate decidual cytokine shifts which regulate repro-

ductive events and regulate cytotrophoblast differentiation and

trophoblast growth (for review, see Le Bouteiller and Blaschitz,

1999). While reproductive failure could also result from parental

sharing of HLA loci or haplotypes, 30 years of research in outbred

couples has produced little agreement regarding the most

important HLA locus or loci responsible for reproductive failure

(Ober and van der Ven, 1997). Discrepancies in studies of outbred

couples may be due to differences in study sample sizes, design

and methodology. Studies often differ in the selection and

strati®cation of fertile and RSA couples. Interestingly, recent

studies based upon DNA techniques have not detected increased

class II nor class I parental sharing in RSA couples (Christiansen

et al., 1989; Ito et al., 1992; Takakuwa et al., 1992; Laitinen et

al., 1993; Karhukorpi et al., 1997; Wagenknecht et al., 1997;

Yamashita et al., 1999). Intermittent associations between

parental HLA sharing and reproductive failure may re¯ect linkage

disequilibrium in some outbred populations. However, many

confounding variables may complicate studies of outbred

populations. Additionally, HLA antigen frequencies in women

with RSA and subsequent reproductive success remains con-

troversial. Importantly, it has not been demonstrated that parental

histocompatible, or offspring homozygous for particular HLA

loci, results in preferential abortion in outbred families.

In contrast, studies in closely related populations suggest that

some HLA loci, or haplotypes, are associated with fetal loss and

increased birth intervals (Ober et al., 1993; Laitinen, 1993).

Studies in genetically isolated groups also demonstrated a

de®ciency of histocompatible offspring in particular HLA

haplotype pairs (Laitinen 1993; Ober 1995). Thus, immuno-

genetic effects such as linkage disequilibrium, on reproductive

outcome may be more easily detected among isolated, closely

related populations than among outbred groups. The majority of

studies of outbred couples suffer from the limitations of

retrospective studies and most couples were selected on the basis

of reproductive histories (Creus et al., 1998). Reproductive

outcome in outbred groups is confounded by use of contraception,

small family sizes and high genetic diversity. Considering the fact

that MHC genes are the most polymorphic loci known among


151

vertebrates (Klein and Figueroa, 1986), closely related individuals

that exhibit less genetic diversity and similar HLA haplotypes are

more likely to share other homozygous regions in the genome.

Therefore, the HLA may serve as a marker for overall genome

homozygosity. In outbred couples, high amounts of HLA

polymorphism may confound detection of genetic effects upon

reproductive failure since individuals rarely share many HLA

alleles. By studying closely related individuals it may be possible

to detect HLA, or HLA-linked genes and regions, which affect

fertility, reproduction and development. Human homologues of

murine genes that directly affect gametogenesis and embryogen-

esis have also been identi®ed (Table I). Human homologues of the

Ped gene, the grc complex and t haplotypes in¯uencing fertility,

reproduction and development, may also exist. These HLA genes,

or genetic regions, could in¯uence a multitude of networks

between immunological factors involved in reproductive events.

For example, the BYSL gene, which maps to 6p21.1 near the

HLA, encodes a protein that forms a complex with trophinin and

tastin in trophectoderm cells during implantation, mediating

adhesion between trophoblast and endometrial cells (Suzuki et al.,

1999).

Thus, interactions between factors may be crucial to achieve

fertility and to maintain a pregnancy. As described, cytokine

activity may in¯uence HLA expression at the maternal±fetal

interface since IFNs and TNFs cytokines can induce HLA class I

surface expression in class I negative cells (Wan et al., 1987;

Collins et al., 1986) and IFN-g, can control MHC expression via

transcriptional regulatory pathways (Brown et al., 1993).

Cytokines may also affect NK functions at the maternal±fetal

interface since they have direct and potent stimulatory effects

upon NK activity (King and Loke, 1990). Since activated NK

(LAK) cells can kill trophoblast cells (Drake and Head, 1989),

and are capable of secreting cytokines which have appropriate

trophoblast receptors (King et al., 1999), both cytokines and NK

cells could provide important regulatory pathways for pregnancy

maintenance. Moreover, NK cells possess receptors which

recognize HLA antigens, e.g. HLA-G, -C and -E, expressed on

extravillous cytotrophoblast cells (King et al., 1998, 2000a,b,c).

Antigen binding to receptors can alter NK target cell killing and

many groups have demonstrated that HLA-G expression can

inhibit uterine NK target killing of trophoblast cell lines (for

review, see Lanier, 1999). This demonstrates just a few of the

potential factor networks that may function throughout human

reproduction.

Thus, successful reproduction is a dynamic interplay

between immunological and immunogenetic factors, as illu-

strated in Figure 1. Detection and further study of these

factors requires highly speci®c assaying during particular

stages of pregnancy. While only a few interactions have been

reported during, and after, blastocyst formation, fewer studies

have been undertaken during early gamete development,

fertilization and embryo cleavage. Thus, animal models can

serve as reasonable alternatives to human studies, due to the

dif®culties of investigating these processes during human

reproduction. This review demonstrates the importance of

immunogenetic factors and interacting factors and suggests

that the disruption of interacting networks may also contribute

to human reproductive failure.

Acknowledgements

We thank Jenny Underwood, Ashley King and John Aplin for their helpfulcomments during the preparation of this article. S.R.C. was supported by theAssociation of Commonwealth Universities and the British Council.

References

Abe, K., Wei, J.-F., Wei, F.-S. et al. (1988) Searching for coding sequences inthe mammalian genome: the H-2K region of the mouse MHC is repletewith genes expressed in embryos. EMBO J., 7, 3441±3449.

Abendroth, A., Slobedman, B., Lee, E. et al. (2000) Modulation of majorhistocompatibility class II protein expression by varicella-zoster virus. J.Virol., 74, 1900±1907.

Aguado, B., Milner, C.M. and Duncan Campbell, R. (1996) Genes of the MHCclass III region and the functions of the proteins they encode. InBrowning, M.J. and McMichael, A.J. (eds), HLA and MHC: Genes,Molecules and Function. BIOS Scienti®c Publishers, Oxford, pp. 37±75.

Alcock, R.J., Price, P., Gaudieri, S. et al. (1999) Characterization of the humancentral MHC gene, BAT1: genome structure and expression. Exp. Clin.Immunogenet., 16, 98±106.

Alexander, S.A., Latinne, D., Debruyere, M. and et al. (1988) Belgianexperience with repeat immunization in recurrent spontaneous abortions.In Beard, R.W. and Sharp, F. (eds), Early Pregnancy Loss: Mechanismsand Treatment. Springer Verlag, London, UK, pp. 355±363.

Alizadeh, M., Legras, C., Semana, G. et al. (1993) Evidence for apolymorphism of HLA-G gene. Hum. Immunol., 38, 206±212.

Allan, D.S.J., Colonna, M., Lanier, L.L. et al. (1999) Tetrameric complexes ofhuman histocompatibility leukocyte antigen (HLA)-G bind to peripheralblood myelomonocytic cells. J. Exp. Med., 189, 1149±1156.

Allan, E.H., Hilton, D.J., Brown, M.A. et al. (1990) Osteoblasts displayreceptors for and responses to leukemia-inhibitory factor. J. Cell Physiol.,145, 110±119.

Almasan, A., Mangelsdorf, D.J., Ong, E.S. et al. (1994) Chromosomallocalization of the human retinoid X receptors. Genomics, 20, 397±403.

Altshuler, G. (1974) Immunologic competence of the immature human fetus:Morphologic evidence form intrauterine cytomegalovirus infection.Obstet. Gynecol., 43, 811±816.

Amills, M., Ramiya, V., Norimine, J. and Lewin, H.A. (1998) The majorhistocompatibility complex of ruminants. Rev. Scient. Tech. Off. Internat.Epizoot., 17, 108±120.

Amiot, L., Onno, M., Renard, I. et al. (1996) HLA-G transcription studiesduring the different stages of normal and malignant hematopoiesis. Tiss.Ant., 48, 609±614.

Amos, D.B., Bashir, H., Boyle, W. et al. (1969) A simple micro cytotoxicitytest. Transplantation, 7, 220±223.

Amson, R., Siguax, F., Przedborski, S. et al. (1989) The human protooncogeneproduct p33pim is expressed during fetal hematopoiesis and in diverseleukemias. Proc. Natl Acad. Sci. USA, 86, 8857±8861.

Anderson, D.J. and Berkowitz, R.S. (1985) Gamma-interferon enhancesexpression of Class I MHC antigens in the weakly HLA humanchoriocarcinoma cell line BeWo, but does not induce MHC expressionin the HLA choriocarcinoma cell line Jar. J. Immunol., 135, 2498±2501.

Ando, A., Kikuti, Y.Y., Shigenari, A. et al. (1996) cDNA cloning of thehuman homolgues of the mouse Ke4 and Ke6 genes at the centromericend of the human MHC region. Genomics, 35, 600±602.

Apple, R.J. and Erlich, H.A. (1996) HLA class II genes: structure anddiversity. In Browning, M.J. and McMichael, A.J. (eds), HLA and MHC:Genes, Molecules and Function. BIOS Scienti®c Publishers Ltd, Oxford,UK, pp. 97±110.

Arck, P.C., Ferrick, D.A., Steele-Norwood, D. et al. (1997) Murine T celldetermination of pregnancy outcome: I. Effects of strain, ab T cellreceptor, gd T cell receptor, and gd T cell subsets. Am. J. Reprod.Immunol., 37, 492±502.

Arici, A., Engin, O., Attar, E. and et al. (1995) Modulation of leukaemiainhibitory factor gene expression and protein biosynthesis in humanendometrium. J. Clin. Endocrinol. Metab., 80, 1908±1915.

Ark, B., Gummere, G., Bennett, D. and Artzt, K. (1991) Mapping of the Pim-1oncogene in mouse t haplotypes and its use to de®ne the relative mappositions of the tcl loci t0 (t6), tw12 and the marker tf (tufted). Genomics,10, 385±389.

Artzt, K., McCormick, P. and Bennett, D. (1982) Gene mapping within the T/tcomplex of the mouse. I. t-lethal genes are nonallelic. Cell, 28, 463±470.

Artzt, K., Abe, K., Uehara, H. and Bennett, D. (1988) Intra-H-2 recombination


152

in t haplotypes shows a hot spot and close linkage of ltw5 to H-2K.Immunogenetics, 28, 30±37.

Awdeh, Z.L., Raum, D., Yunis, E.J. and Alper, C.A. (1983) Extended HLA/complement allele haplotypes: evidence for T/t -like complex in man.Proc. Natl Acad. Sci. USA, 80, 259±263.

Baines, M.G., Duclos, A.J., Antecka, E. and Haddad, E.K. (1997) Decidualin®ltration and activation of macrophages leads to early embryo loss. Am.J. Reprod. Immunol., 37, 471±477.

Balasch, J., Jove, I., Martorell, J. et al. (1993) Histocompatibility in in vitrofertilization couples. Fertil. Steril., 59, 456±458.

Becker, D.M., Fikes, J.D. and Guarente, L. (1991) A cDNA encoding a humanCCAAT-binding protein cloned by functional complementation in yeast.Proc. Natl Acad. Sci., 88, 1968±1972.

Beer, A.E. and Billingham, R.E. (1976) The Immunobiology of Reproduction.Prentice Hall, New Jersey, USA.

Benirschke, K. and Kaufmann, P. (1995) Pathology of the Human Placenta.Springer-Verlag, New York, USA.

Benirschke, K. and Miller, C.J. (1982) Anatomical and functional differencesin the placenta of primates. Biol. Reprod., 26, 29±53.

Bennett, D. (1975) The T-locus of the mouse. Cell, 6, 441±454.Bennett, D. and Dunn, L.C. (1960) A lethal mutant (tw18) in the house mouse

showing partial duplications. J. Exp. Zoo., 143, 203±219.Ben-Shaul, Y., Artzt, K. and Bennett, D. (1983) Immunoscanning electron

microscopy of antigenic determinants of T/t-complex (tw18) mouseembryos. Cell. Different., 13, 159±170.

Bensussan, A., Mansur, I.G., Mallet, V. et al. (1995) Detection of membrane-bound HLA-G translated products with a speci®c monoclonal antibody.Proc. Natl Acad. Sci. USA, 92, 10292±10296.

Bernads, S.R., Dessain, S.K. and Weinberg, R. (1986) N-myc ampli®cationcauses down-modulation of MHC class I antigen expression inneuroblastoma. Cell, 47, 667±674.

Biassoni, R., Falco, M., Cambiaggi, A. et al. (1995) Amino-acid substitutionscan in¯uence the natural-killer (NK)-mediated recognition of HLA-Cmolecules-pole of serine-77 and lysine-80 in the target-cell protectionfrom lysis mediated by group-2 or group-1 NK clones. J. Exp. Med., 182,605±610.

Bibbins, K., Tsai, J.-Y., Schimenti, J. et al. (1989) Human homologs of twotestes-expressed loci on mouse chrmosome 17 map to opposite arms ofchromosome 6. Genomics, 5, 130.

Billington, W.D. (1964) In¯uence of immunologic dissimilarity of mother andfoetus on size of placenta in mice. Nature, 202, 317±318.

Blanche, H., Wright, L.G., Vergnaud, G. et al. (1992) Genetic mappping ofthree human homologues of murine t complex genes localizes TCP10 to6q27, 15 cM distel to TCP1 and PLG. Genomics, 12, 826±828.

Blaschitz, A., Lenfant, F., Mallet, V. et al. (1997) Endothelial cells inchorionic fetal vessels of ®rst trimester placenta express HLA-G. Eur. J.Immunol., 27, 3380±3388.

Bluestone, J.A., Cron, R.Q., Barrett, T.A. et al. (1991) Repertoire,development and ligand speci®city of murine TCR gd cells. Immunol.Rev., 120, 5±33.

Bodmer, J.G., Marsh, S.G.E., Albert, E.D. et al. (1995) Nomenclature forfactors of the HLA system, 1995. Tiss. Ant., 46, 1±18.

Bolton, V.N., Hawes, S.M., Taylor, C.T. and Parsons, J.H. (1989)Development of spare human preimplantation embryos in vitro: ananalysis of the correlations among gross morphology, cleavage rates, anddevelopment to the blastocyst. J. In Vitro Fertil. Embryo Transfer, 6, 30±35.

Borges, L., Hsu, M.-L., Fanger, N. et al. (1997) A family of human lymphoidand myeloid Ig-like receptors, some of which bind to MHC class Imolecules. J. Immunol., 159, 5192±5196.

Borrego, F., Ulbrecht, M., Weiss, E.H., et al. (1998) Recognition of humanhistocompatibility leukocyte antigen (HLA)-E complexed with HLA classI signal sequence-derived peptides by CD94/NKG2 confers protectionfrom natural killer cell-mediated lysis. J. Exp. Med., 187, 813±818.

Boucraut, J., Guillaudeux, T., Alizadeh, M. et al. (1993a) HLA-E is the onlyclass I gene that escapes CpG methylation and is transcriptionally activein the trophoblast-derived cell line JAR. Immunogenetics, 38, 117±130.

Boucraut, J., Hawley, S., Robertson, K. et al. (1993b) Differential nuclearexpression of enhancer A DNA-binding proteins in human ®rst trimestertrophoblast cells. J. Immunol., 150, 3882±3894.

Boyson, J.E., McAdam, S.N., Gallimore, A. et al. (1995) The MHC-E locus inmacaques is polymorphic and is conserved between macaques andhumans. Immunogenetics, 41, 59±68.

Boyson, J.E., Shuf¯ebotham, C., Cadavid, L.F. et al. (1996a) The MHC class I

genes of the rhesus monkey. Different evolutionary histories of MHCclass I and II genes in primates. J. Immunol., 156, 4656±4665.

Boyson, J.E., Iwanaga, K.I., Golos, T.G. and Watkins, D.I. (1996b)Identi®cation of the rhesus monkey HLA-G ortholog: Mamu-G is apseudogene. J. Immunol., 157, 5428±5437.

Boyson, J.E., Iwanaga, K.K., Golos, T.G. and Watkins, D.I. (1997)Identi®cation of a novel MHC class I gene, Mamu-AG in the placentaof primate with an inactivated G locus. J. Immunol., 159, 3311±3321.

Braud, V.M., Allan, D.S.J., O'Callaghan, C.A. et al. (1998a) HLA-E binds tonatural killer cell receptors CD94/NKG2A, B and C. Nature (London),391, 795±798.

Braud, V.M., Allan, D.S.J., Wilson, D. and McMichael, A.J. (1998b) TAP-and tapasin-dependent HLA-E surface expression correlates with thebinding of an MHC class I leader peptide. Curr. Biol., 8, 1±10.

Brooks, A.G., Borrego, F., Posch, P.E. et al. (1999) Speci®c recognition ofHLA-E, but not classical, HLA class I molecules by soluble CD94/NKG2A and NK cells. J. Immunol., 162, 305±313.

Brown, M.G., Driscoll, J. and Monaco, J.J. (1993) MHC-linked low-molecularmass polypeptide subunits de®ne distinct subsets of proteasomes. J.Immunol., 151, 1193±1204.

Browning, M. and McMichael, A. (1996) Introduction. In Browning, M. andMcMichael, A. (eds), HLA and MHC: Genes, Molecules and Function.BIOS Scienti®c Publishers, Oxford, UK, pp. xiii-xvii.

Bujko, M., Sulovic, V., Zivanovic, V. et al. (1988) Herpes simplex virusinfection in women with previous spontaneous abortion. J. Perinat. Med.,16, 193±196.

Bulmer, J.N., Morrison, L., Longfellow, M. et al. (1991) Granulatedlymphocytes in human endometrium: Histochemical andimmunohistochemical studies. Hum. Reprod., 6, 791±798.

Cantoni, C., Verdiani, S., Falco, M. et al. (1998) p49, a putative HLA class I-speci®c inhibitory NK receptor belonging to the immunoglobulinsuperfamily. Eur. J. Immunol., 28, 1980±1990.

Cao, W., Brenner, C.A., Alikani, M. et al. (1999) Search for a humanhomologue of the mouse Ped gene. Mol. Hum. Reprod., 5, 541±547.

Carosella, E.D., Dausset, J. and Kirszenbaum, M. (1996) HLA-G revisited.Immunol. Today, 17, 407±409.

Chapman, D.L. and Wolgemuth, D.J. (1993) Isolation of the murine cyclin B2cDNA and characterization of the lineage and temporal speci®city ofexpression of the B1 and B2 cyclins during oogenesis, spermatogenesis,and early embryogenesis. Development, 118, 229±240.

Chernyshov, V.P., Slukvin, I.I. and Bondarenko, G.I. (1993) Phenotypiccharacterization of CD7+, CD3+, and CD8+ lymphocytes from ®rsttrimester human decidua using two-color ¯ow cytometry. Am. J. Reprod.Immunol., 29, 5±16.

Chesley, P. (1932) Lethal action of the short-tailed mutation in the housemouse. Proc. Soc. Exp. Biol. Med., 29, 437±438.

Chesley, P. and Dunn, L.C. (1936) The inheritance of tailessness (anury) in thehouse mouse. Genetics, 21, 525±536.

Chiang, M.H. and Main, E.K. (1994) Nuclear regulation of HLA class I genesin human trophoblasts. Am. J. Reprod. Immunol., 32, 167±172.

Chiang, M.H., Steurwald, N., Lambert, H. et al. (1994) Detection of humanleukocyte antigen class I messenger ribonucleic acid transcripts in humanspermatozoa via reverse transcription-polymerase chain reaction. Fertil.Steril., 61, 276±280.

Chien, Y.H., Jores, R. and Crowley, M.P. (1996) Recognition by gd T cells.Ann. Rev. Immunol., 14, 511±532.

Chittenden, T., Harrington, E.A., O'Connor, R. et al. (1995) Induction ofapoptosis by the Bcl-2 homologue Bak. Nature, 374, 733±736.

Christiansen, O.B. (1996) A fresh look at the causes and treatments ofrecurrent miscarriage, especially its immunological aspects. Hum. Reprod.Update, 2, 271±293.

Christiansen, O.B. (1999) The possible role of classical human leukocyteantigens in recurrent miscarriage. Am. J. Reprod. Immunol., 42, 110.

Christiansen, O.B., Riisom, K., Lauritsen, J.G. and Grunnet, N. (1989) Noincreased histocompatibility antigen sharing in couples with idiopathichabitual abortions. Hum. Reprod., 4, 160±162.

Christiansen, O.B., Mathiesen, O., Lauritsen, J.G. and Grunnet, N. (1990a)Idiopathic recurrent spontaneous abortion. Evidence of a familialpredisposition. Acta Obstet. Gynecol. Scand., 69, 597±601.

Christiansen, O.B., Mathiesen, O., Jersild, C. and Grunnet, N. (1990b)Recurrent abortions seem associated with maternal HLA haplotypes.Scand. J. Immunol., 32, 388.

Christiansen, O.B., Mathiesen, O., Husth, M. et al. (1994) Placebo-controlledtrial of active immunization with third party leukocytes in recurrentmiscarriages. Acta Obstet. Gynecol. Scand., 73, 261±268.


153

Christiansen, O.B., Pedersen, B., Mathiesen, O. et al. (1996) Maternal HLAClass II Alleles Predispose to Pregnancy Losses in Danish Women withrecurrent spontaneous abortions and their female relatives. Am. J. Reprod.Immunol., 35, 239±244.

Christiansen, O.B., Ring, M., Rosgaard, A. et al. (1999) Association betweenHLA-DR1 and -DR3 antigens and unexplained repeated miscarriage.Hum. Reprod. Update, 5, 249±255.

Chu, W., Fant, M.E., Geraghty, D.E. and Hunt, J.S. (1998) Soluble HLA-G inhuman placentas: synthesis in trophoblasts and interferon-gamma-activated macrophages but not placental ®broblasts. Hum. Immunol., 59,435±442.

Chumbley, G., King, A., Gardner, L. et al. (1994) Generation of an antibody toHLA-G in transgenic mice and demonstration of the tissue reactivity ofthis antibody. J. Reprod. Immunol., 27, 173±86.

Clark, D.A. (1991) Controversies in reproductive immunology. Crit. Rev.Immunol., 11, 215±247.

Clark, D.A. (1997) HLA-G ®nally does something. Am. J. Reprod. Immunol.,38, 75±78.

Clark, D.A., Arck, P.C. and Chaouat, G. (1999) Why did your mother rejectyou? Immunogenetic determinants of the response to environmentalselective pressure expressed at the uterine level. Am. J. Reprod. Immunol.,41, 5±22.

Cobb, M.H. and Goldsmith, E.J. (1995) How MAP kinases are regulated. J.Biol. Chem. 270, 14843±14846.

Collins, J.A., Wrixon, W., Janes, L.B. and Wilson, E.H. (1983) Treatment-independent pregnancy among infertile couples. N. Engl J. Med., 309,1201±1206.

Collins, T., Lapierre, L., Fiers, W. et al. (1986) Recombinant human TNFincreases mRNA levels and surface expression of HLA-A, -B antigens invascular endothelial cells and dermal ®broblasts in vitro. Proc. Natl Acad.Sci. USA, 83, 446±450.

Colonna, M., Borsellino, G., Falco, M. et al. (1993) HLA-C is the inhibitoryligand that determines dominant resistance to lysis by NK1-speci®c andNK2-speci®c natural-killer-cells. Proc. Natl Acad. Sci. USA, 90, 12000±12004.

Colonna, M., Navarro, F., Bellon, T. et al. (1997) A common inhibitoryreceptor for major histocompatibility complex class I molecules on humanlymphoid and myelomonocytic cells. J. Exp. Med., 186, 1809±1818.

Colonna, M., Samaridis, J., Cella, M. et al. (1998) Human myelomonocyticcells express an inhibitory receptor for classical and nonclassical MHCclass I molecules. J. Immunol., 160, 3096±3100.

Cortese Hassett, A.L., Locker, J., Rupp, G. et al. (1989) Molecular analysis ofthe rat MHC. II. Isolation of genes that map to the RT1.E-grc region. J.Immunol., 142, 2089±2096.

Cosman, D., Fanger, N., Borges, L. et al. (1997) A novel immunoglobulinsuperfamily receptor for cellular and viral MHC class I molecules.Immunity, 7, 273±282.

Coulam, C.B., Moore, S.B. and O'Fallon, W.M. (1987) Association betweenmajor histocompatibility antigen and reproductive performance. Am. J.Reprod. Immunol. Microbiol., 14, 54±58.

Creus, M., Balash, J., FabreÂgues, F. et al. (1998) Parental human leukocyteantigens and implantation failure after in-vitro fertilization. Hum. Reprod.,13, 39±43.

Crisa, L., McMaster, M.T., Ishii, J.K et al. (1997) Identi®cation of a thymicepithelial cell subset sharing expression of the class Ib HLA-G moleculewith fetal trophoblasts. J. Exp. Med., 186, 289±298.

Crouau-Roy, B., Amadou, C., Bouissou, C. et al. (1994) Localization of theOTF3 gene within the human MHC class I region by physical and meioticmapping. Genomics, 21, 241±243.

Croy, B.A. (1990) Granulated metrial gland cells-interesting cells found in thepregnant uterus. Am. J. Reprod. Immunol., 23, 19±21.

Das, S.K., Lim, H., Paria, B.C. and Dey, S.K. (1999) Cyclin D3 in the mouseuterus is associated with the decidualization process during earlypregnancy. J. Mol. Endocrinol., 22, 91±101.

Dausset, J., Legrand, L., Lepage, V. et al. (1978) A haplotype study of HLAcomplex with special reference to the HLA-DR series and to Bf. C2 andglyoxalase I polymorphisms. Tiss. Ant., 12, 297±307.

De La Concha, E.G., Fernandez-Arquero, M., Martinez, A. et al. (1999) HLAclass II homozygosity confers susceptibility to common variableimmunode®ciency (CVID). Clin. Exp. Immunol., 116, 516±520.

Delage, G., Moreau, J.F., Taupin, J.L. et al. (1995) Modulation in vitro de laproduction de la cytokine HILDA/LIF secreteÂ par des explantsd'endometre humain: resultats preÂliminaires. Contracept. Fertil. Sex, 23,622±625.

Demetrick, D.J., Matsumoto, S., Hannon, G.J. et al. (1995) Chromosomal

mapping of the genes for the human cell cycle proteins cyclin C (CCNC),cyclin E (CCNE), p21 (CDKN1) and KAP (CDKN3). Cytogenet. Cell,Genet., 69, 190±192.

Deniz, G., Christmas, S.E., Brew, R. and Johnson, P.M. (1994) Phenotypic andfunctional cellular differences between human CD3(±) decidual andperipheral-blood leukocytes. J. Immunol., 152, 4253±4261.

Di Marco, E., Cutuli, N., Guerra, L. et al. (1993) Molecular cloning of trkE, anovel trk-related putative tyrosine kinase receptor isolated from normalhuman keratinocytes and widely expressed by normal human tissues. J.Biol. Chem., 268, 24290±24295.

Diehl, M., Munz, C., Keilholz, W. et al. (1996) Nonclassical HLA-Gmolecules are classical peptide presenters. Curr. Biol., 6, 305±314.

Ding, A.H., Nathan, C.F. and Stuehr, D.J. (1988) Release of reactive nitrogenintermediates and reactive oxygen intermediates from mouse peritonealmacrophages. Comparison of activating cytokines and evidence forindependent production. J. Immunol., 141, 2407±2412.

Dobrovolskaia-Zavadskaia, N. (1927) Sur la morti®cation sontanee de laqueue chez la souris nouveau-nee et sur l'existence d'un caractere(facteur) hereditaire `non viable'. Comp. Rend. Soc. Biol., 97, 114±116.

Doherty, D.G. and Donaldson, P.T. (1991) HLA-DRB and -DQB typing by acombination of serology, restriction fragment length polymorphismanalysis and oligonucleotide probing. Eur. J. Immunogenet., 18, 111±124.

Doherty, P.C. and Zinkernagel, R.M. (1975) Enhanced immunologicalsurveillance in mice heterozygous at the H-2 gene complex. Nature,256, 50±52.

Dohr, G., Motter, W., Leitinger, S. et al. (1987) A lack of expression ofhistocompatibility leukocyte antigen class I and class II on the humanoocyte. J. Immunol., 138, 3766±3770.

Domen, J., von Lindern, M., Hermans, A. et al. (1987) Comparison of thehuman and mouse PIM-1 cDNAs: nucleotide sequence andimmunological identi®cation of the in vitro synthesized PIM-1 protein.Oncogene Res., 1, 103±112.

Donaldson, W.L., Oriol, J.G., Pelkaus, C.L. and Antczak, D.F. (1994) Paternaland maternal Major Histocompatibility class I antigens are expressed co-dominantly by equine trophoblast. Placenta, 15, 123±135.

Dorling, A., Monk, N.J. and Lechler, R.I. (2000) HLA-G inhibits thetransendothelial migration of human NK cells. Eur J. Immunol., 30, 586±593.

Drake, B.L. and Head, J.R. (1989) Murine trophoblast can be killed bylymphokine-activated killer cells. J. Immunol., 143, 9±17.

Drezen, J.M., Barra, J., Babinet, C. and Morello, D. (1994) MHC class I genesare not imprinted in the mouse placenta. Immunogenetics, 40, 62±65.

Ducy, P., Zhang, R., Geoffroy, V. et al. (1997) Osf2/Cbfa1: a transcriptionalactivator of osteoblast differentiation. Cell, 89, 747±754.

Ducy, P., Starbuck, M., Priemel, M. et al. (1999) A Cbfa1-dependent geneticpathway controls bone formation beyond emryonic development. GenesDev., 13, 1025±1036.

Edelhoff, S., Sweetser, D.A. and Disteche, C.M. (1995) Mapping of the NEPreceptor tyrosine kinase gene to human chromosome 6p21.3 and mousechromosome 17C. Genomics, 25, 309±311.

Ellis, S.A., Palmer, M.S. and McMichael, A.J. (1990) Human trophoblast andthe choriocarcinoma cell line BeWo express a truncated HLA class Imolecule. J. Immunol., 144, 731±735.

Fanger, N.A., Cosman, D., Peterson, L. et al. (1998) The MHC class I bindingproteins LIR-1 and LIR-2 inhibit Fc receptor-mediated signalling inmonocytes. Eur. J. Immunol., 28, 3423±3434.

Faulk, W.P., Hsi, B.L., McIntyre, J.A et al. (1982) Antigens on the humanextra-embryonic membranes. J. Reprod. Fertil. Suppl., 31, 181±189.

Fenichel, P., Donzeau, M., Cervoni, F. et al. (1995) Expression of complementregulatory proteins on human eggs and preimplantation embryos. Am. J.Reprod. Immunol., 33, 155±156.

Fernandez, N., Sprinks, M.T., Cooper, J.C. et al. (1995) Embryonicdevelopment and the major histocompatibility complex. In Kurpisz, M.and Fernandez, N. (eds), Immunology of Human Reproduction. BIOSScienti®c Publishers, Oxford, UK, pp. 185±204.

Fernandez, N., Cooper, J., Sprinks, M. et al. (1999) A critical review of therole of the major histocompatibility complex in fertilization,preimplantation development and feto-maternal interactions. Hum.Reprod. Update, 5, 234±248.

Ferry, B.L., Sargent, I.L., Starkey, P.M. and Redmann, C.W. (1991) Cytotoxicactivity against trophoblast and choriocarcinoma cells of large granularlymphocytes from early pregnancy decidua. Cell. Immunol., 132, 140±149.

Ferry, B.L., Starkey, P.M., Sargent. I.L. et al. (1990) Cell populations in thehuman early pregnancy decidua: natural killer activity and response to


154

interleukin-2 of CD56-positive large granular lymphocytes. Immunology,70, 446±452.

Fiszer, D., Ulbrecht, M., Fernandez, N. et al. (1997) Analysis of HLA class Ibgene expression in male gametogenic cells. Eur. J. Immunol., 27, 1691±1695.

Fitzgibbon, J., Gillett, G.T., Woodward, K.J. et al. (1993) Mapping of RXRBto human chromosome 6p21.3. Ann. Hum. Genet., 57, 203±209.

Fomitcheva, J., Baker, M.E., Anderson, E. et al. (1998) Characterization ofKe, 6, a new 17 beta-hydroxysteroid dehydrogenase, and its expression ingonadal tissues. J. Biol. Chem., 273, 22664±22671.

Fu, Y.X., Roark, C.E., Kelly, K. et al. (1994) Immune Protection and controlof in¯ammatory tissue necrosis by gd T cells. J. Immunol., 153, 3101±3115.

Fujii, T., Ishitani, A. and Geraghty, D.E. (1994) A soluble form of the HLA-Gantigen is encoded by a messenger ribonucleic acid containing intron 4. J.Immunol., 153, 5516±5524.

Galbraith, R.M., Werner, P., Kantor, R.R. and Galbraith, G.M. (1981) Studiesof the interaction between human transferrin and speci®c receptors on thetrophoblast membrane. Placenta Suppl., 3, 49±59.

Gautschi, C. and Gaillard, C. (1990) In¯uence of major histocompatibilitycomplex on reproduction and production traits in swine. Animal Genetics,21, 161±170.

Gavora, J.S., Simonsen, M., Spencer, J.L. et al. (1986) Changes in thefrequency of major histocompatibility haplotypes in chickens underselection for both high egg production and resistance to Marek's disease.J. Animal Breed. Gen., 103, 218±226.

Geraghty, D.E., Koller, B.H. and Orr, H.T. (1987) A human majorhistocompatibility complex class I gene that encodes a protein with ashortened cytoplasmic segment. Proc. Natl Acad. Sci. USA, 84, 9145±9148.

Geraghty, D.E., Koller, B.H., Hansen, J.A. and Orr, H.T. (1992a) The HLAclass I family includes at least six genes and twelve pseudogenes and genefragments. J. Immunol., 149, 1934±1946.

Geraghty, D.E., Koller, B.H., Ji, P. and Hansen, J.A. (1992b) Examination of 4HLA class-1 pseudogenes-common events in the evolution of HLA genesand pseudogenes. J. Immunol., 149, 1947±1956.

Gill, T.J. (1987) Immunological and genetic factors in¯uencing developmentand susceptibility to cancer. Genetica, 73, 15±24.

Gill, T.J. (1994) Reproductive immunology and immunogenetics. In Knobil,E. and Neill, J.D. (eds), The Physiology of Reproduction. Raven Press,New York, USA, pp. 783±812.

Gill, T.J. (1996) Role of the major histocompatibility complex region inreproduction, cancer, and autoimmunity. Am. J. Reprod. Immunol., 35,211±215.

Gill, T.J., Kunz, H.W., Musto, N.A. and Bardin, C.W. (1984) Genetic andmorphometric studies of the heterogeneity in the testicular defect of theHre rat and the interaction between the Hre and grc genes. Biol. Reprod.,31, 595±603.

Goldbard, S.B., Verbanac, K.M. and Warner, C.M. (1982) Role of the H-2complex in preimplantation mouse embryo development. Biol. Reprod.,26, 591±596.

Goldman-Wohl, D.S., Ariel, I., Green®eld, C. et al. (2000) Lack of humanleukocyte antigen-G expression in extravillous trophoblasts as associatedwith pre-eclampsia. Mol. Hum. Reprod., 6, 88±95.

Grob, B., Knapp, L., Martin, R.D. and Anzenberger, G. (1998) The majorhistocompatibilty complex and mate choice: Inbreeding avoidance andselection of good genes. Exp. Clin. Immunogen., 15, 119±129.

Gronroos, M., Honkonen, E., Terho, P. and Punnonen, R. (1983) Cervical andserum IgA and serum IgG antibodies to Chlamydia trachomatis andherpes simplex virus in threatened abortion: a prospective study. Br. J.Obstet. Gynecol., 90, 167±170.

Guillaudeux, T., Mattei, M.G., Depetris, D. et al. (1993) in situ hybridizationlocalizes the human OTF3 to chromosome 6p21.3-p22 and OTF3L to12p13. Cytogenet. Cell. Genet., 63, 212±214.

Guillaudeux, T., Rodriguez, A.M., Girr, M. et al. (1995) Methylation statusand transcriptional expression of the MHC class I loci in humantrophoblast cells from term placenta. J. Immunol., 154, 3283±99.

Guillaudeux, T., Gomez, E., Onno, M. et al. (1996) Expression of HLA-class Igenes in meiotic and postmeiotic human spermatogenic cells. Biol.Reprod., 55, 99±110.

Gumperz, J.E., Litwin, V., Phillips, J.H. et al. (1995) The BW4 public epitopeof HLA-B molecules confers reactivity with natural-killer-cell clones thatexpress NKB1, a putative HLA receptor. J. Exp. Med., 181, 1133±1144.

Gumperz, J.E., Barber, L.D., Valiante, N.M. et al. (1997) Conserved andvariable residues within the BW4 motif of HLA-G make separable

contributions to recognition by the NKB1 killer cell-inhibitory receptor. J.Immunol., 158, 5237±5241.

Hamada, K., Gleason, S.L., Levi, B.-Z. et al. (1989) H-2RIIBP, a member ofthe nuclear hormone receptor superfamily that binds to both the regulatoryelement of major histocompatibility class I genes and the estrogenresponse element. Proc. Natl Acad. Sci. USA, 86, 8289±8293.

Hamai, Y., Fujii, T., Yamashita, T. et al. (1997) Pathogenic implications ofinterleukin-2 expressed in preeclamptic decidual tissues: a possiblemechanism of deranged vasculature of the placenta associated withpreeclampsia. Am. J. Reprod. Immunol., 38, 83±88.

Hamai, Y., Fujii, T., Yamashita, T. et al. (1999) The expression of humanleukocyte antigen-G on trophoblasts abolishes the growth-suppressingeffect of interleukin-2 towards them. Am. J. Reprod. Immunol., 41, 153±158.

Hammer, A., Hutter, H. and Dohr, G. (1997a) HLA class I expression on thematernal±fetal interface. Am. J. Reprod. Immunol., 38, 150±157.

Hammer, A., Hutter, H., Blaschitz, A. et al. (1997b) Amnion epithelial cells,in contrast to trophoblast cells, express all classical HLA class I moleculestogether with HLA-G. Am. J. Reprod. Immunol., 37, 161±171.

Hanson, I.M. and Trowsdale, J. (1991) Colinearity of novel genes in the classII regions of the MHC in mouse and human. Immunogenetics, 34, 5±11.

Hayashi, M., Fujimoto, S., Takano, H. et al. (1996) Characterization of ahuman glycoprotein with a potential role in sperm±egg fusion: cDNAcloning, immunohistochemical localization, and chromosomal assignmentof the gene (AEGL1). Genomics, 32, 367±374.

Herberg, J.A., Phillips, S., Beck, S. et al. (1998) Genomic structure anddomain organisation of the human Bak gene. Gene, 211, 87±94.

Hiby, S.E., King, A., Sharkey, A.M. and Loke, Y.W. (1997) Human uterineNK cells have a similar repertoire of killer inhibitory and activatoryreceptors to those found in blood as demonstrated by RT-PCR andsequencing. Mol. Immunol., 34, 419±430.

Hiby, S.E., King, A., Sharkey, A. and Loke, Y.W. (1999) Molecular studies oftrophoblast HLA-G: polymorphism, isoforms, imprinting and expressionin preimplantation embryo. Tiss. Ant., 53, 1±13.

Ho, H., Gill, T.J., Hsieh, C. et al. (1991) The prevalence of recurrentspontaneous abortions, cancer, and congenital anomalies in the families ofcouples with recurrent spontaneous abortions or gestational trophoblastictumors. Am. J. Obstet. Gynecol., 165, 461±466.

Ho, H.-N., Yang, Y.-S., Hsieh, R.-P. et al. (1994) Sharing of human leukocyteantigens in couples with unexplained infertility affects the success of invitro fertilization and tubal embryo transfer. Am. J. Obstet. Gynecol., 170,63±71.

Hoopes, C.W., Taketo, M., Ozato, K. et al. (1992) Mapping of the mouse Rxrloci encoding nuclear retinoid X receptors RXRa, RXRb, and RXRg.Genomics, 14, 611±617.

Houlihan, J.M., Biro, P.A., Harper, H.M. et al. (1995) The human amnion is asite of MHC class Ib expression: evidence for the expression of HLA-Eand HLA-G. J. Immunol., 154, 5665±74.

Hsieh, B., Schrenzel, M.D., Mulvania, T. et al. (1996) In vivo cytokineproduction in murine listeriosis. J. Immunol., 156, 232±237.

Hu, M.C.-T., You-Ping, W., Mikhail, A. et al. (1999) Murine p38-d mitogen-activated protein kinase, a developmentally regulated protein kinase thatis activated by stress and proin¯ammatory cytokines. J. Biol. Chem., 274,7095±7102.

Hughes, A.L. and Nei, M. (1988) Pattern of nucleotide substitution at majorhistocompatibility complex class I loci reveals overdominant selection.Nature., 335, 167±170.

Hughes, A.L., Yeager, M., Ten Elshof, A.E. and Chorney, M.J. (1999) A newtaxonomy of mammalian MHC class I molecules. Immunol. Today, 20,22±26.

Hunt, J.S. (1992) Immunobiology of pregnancy. Curr. Opin. Immunol., 4,591±596.

Hunt, J.S., Miller, L., Vassmer, D. and Croy, B.A. (1997) Expression of theinducible nitric oxide synthetase gene in mouse uterine leukocytes andpotential relationships with uterine function during pregnancy. Biol.Reprod., 57, 827±836.

Hunt, J.S., Chu, W. and Miller, L. (1998) Nonspeci®c immunity in the uterusand placenta. Trophoblast Res., 12, 125±134.

Huppi, K., Siwarski, D., Dosik, J. et al. (1994) Molecular cloning, seqeuncing,chromosomal localization and expression of mouse p21 (Waf1).Oncogene, 9, 3017±3020.

Hutter, H. and Dohr, G. (1998) HLA expression on immature and maturehuman germ cells. J. Reprod. Immunol., 38, 101±122.

Hutter, H., Hammer, A., Blaschitz, A. et al. (1996) Expression of HLA class I


155

molecules in human ®rst trimester and term placenta trophoblast. CellTissue Res., 286, 439±447.

Hviid, T.V.F., Meldgaard, M., Sùrensen, S. and Morling, N. (1997)Polymorphism of exon 3 of the HLA-G gene. J. Reprod. Immunol., 35,31±42.

Inaba, T., Matsushime, H., Valentine, M. et al. (1992) Genomic organization,chromosomal localization, and independent expression of human cyclin Dgenes. Genomics, 13, 565±574.

Ishitani, A. and Geraghty, D.E. (1992) Alternative splicing of HLA-Gtranscripts yields proteins with primary structures resembling both class Iand class II antigens. Proc. Natl Acad. Sci. USA, 89, 3947±51.

IsraeÈl, A. and Kourilsky, P. (1996) Regulation of MHC class I geneexpression. In Browning, M.J. and McMichael, A.J. (eds), HLA andMHC: Genes, Molecules and Function. BIOS Scienti®c Publishers,Oxford, UK, pp. 139±157.

Ito, K., Obato, F., Tanaka, T. et al. (1992) Analysis of HLA-DR types ofunexplained recurrent spontaneous aborters in the Japanese population byoligonucleotide-DNA typing. Tiss. Ant., 40, 204±209.

Jackson, M.R., Cohen-Doyle, M.F., Peterson, P. and Williams, D.B. (1994)Regulation of MHC class I transport by the molecular chaperone, calnexin(p88, IP90). Science, 263, 384±387.

Janatipour, M., Naumov, Y., Ando, A. et al. (1992) Search for MHC-associated genes in human: ®ve new genes centromeric to HLA-DP withyet unknown functions. Immunogenetics, 35, 272±278.

Janeway, C.A., Jones, B. and Hayday, A. (1988) Speci®city and function of T-cells bearing gamma-delta-receptors. Immunol. Today, 9, 73±76.

Janitz, M., Fiszer, D., Jaruzelska, J. et al. (1993) In-situ localization of HLA-class I messenger-RNA in human testis. Exp. Clin. Immunogenet., 10,202±207.

Janitz, M., Fiszer, D., Michalczak-Janitz, K. et al. (1994) Analysis ofmessenger-RNA for class-I HLA on human gametogenic cells. Mol.Reprod. Dev., 38, 231±237.

Jin, K., Ho, H.-N., Speed, T.P. and Gill, T.J. (1995) Reproductive failure andthe major histocompatibility complex. Am. J. Hum. Genet., 56, 1456±1467.

Johnson, P.M. and Stern, P.L. (1986) Antigen expression at human materno-fetal interfaces. Prog. Immunol., 6, 1056±1069.

Johnson, P.M., Chia, K.V., Risk, J.M. et al. (1988) Immunological andimmunogenetic investigation of recurrent spontaneous abortion. Dis.Markers, 6, 163±171.

Jokhi, P.P., Chumbley, G., Gardner, L. et al. (1993) Expression of the colonystimulating factor-1 receptor (c-fms product) by cells at the humanuteroplacental interface. Lab. Invest., 68, 308±320.

Jokhi, P.P., King, A., Sharkey, A.M. et al. (1994a) Screening for cytokinemessenger ribonucleic acids in puri®ed human decidual lymphocytepopulations by the reverse-transcriptase polymerase chain reaction. J.Immunol., 153, 4427±4435.

Jokhi, P.P., King, A., Sharkey, A.M. et al. (1994b) Production of GM-CSF byhuman trophoblast cells and by decidual large granular lymphocytes.Hum. Reprod., 9, 1660±1669.

Jokhi, P.P., King, A., Jubinsky, P. and Loke, Y.W. (1994c) Demonstration ofthe low af®nity alpha subunit of the granulocyte-macrophage-colony-stimulating factor receptor (GM-CSF-Rb) on human trophoblast anduterine cells. J. Reprod. Immunol., 26, 147±164.

Jun, Y., Kim, E., Jin, M. et al. (2000) Human cytomegalovirus gene productsUS3 and US6 down-regulate trophoblast class I MHC molecules. J.Immunol., 164, 805±811.

Jurisicova, A., Casper, R.F., MacLusky, N.J. et al. (1996) HLA-G expressionduring preimplantation human embryo development. Proc. Natl Acad.Sci., USA, 93, 161±165.

Kanbour-Shakir, A., Kunz, H.W. and Gill, T.J. (1993) Differential genomicimprinting of major histocompatibility complex class I antigens in theplacenta of the rat. Biol. Reprod., 48, 977±986.

Karhukorpi, J., Laitinen, T. and Tiilikainen, A.S. (1997) HLA-Gpolymorphism in Finnish couples with recurrent spontaneousmiscarriage. Br. J. Obstet. Gynaecol., 104, 1212±1214.

Karpuj, M.V., Steinman, L. and Oksenberg, J.R. (1997) Multiple schlerosis: apolygenic disease involving epistatic interactions, germlinerearrangements and environmental effects. Neurogenetics., 1, 21±28.

Kay, P.H., Dawkins, R.L., Williamson, J. et al. (1988) Coexistence of an MHCchromosomal segment marked by HLA B17, BfS, C4A6, B1, DR7, andDQw9 in different ethnic groups. Hum. Immunol., 23, 27±35.

Khalil-Daher, I., Riteau, B., Menier, C. et al. (1999) Role of HLA-G versusHLA-E on NK function: HLA-G is able to inhibit NK cytolysis by itself.J. Reprod. Immunol., 43, 175±182.

Kiefer, M.C., Brauer, M.J., Powers, V.C. et al. (1995) Modulation of apoptosisby the widely distributed Bcl-2 homologue Bak. Nature, 374, 736±739.

Kikuti, Y.Y., Tamiya, G., Ando, A. et al. (1997) Physical mapping 220 Kbcentromeric of the human MHC and DNA sequence analysis of the 43-kbsegment including RING1, HKE6, and HKE4 genes. Genomics, 42, 422±435.

King, A. and Loke, Y.W. (1990) Human trophoblast and JEG choriocarcinomacells are sensitive to lysis by IL-2-stimulated decidual NK cells. Cell.Immunol., 129, 435±448.

King, A., Wheeler, R., Carter, N.P. et al. (1992) The response of humandecidual leukocytes to IL-2. Cell. Immunol., 140, 409±421.

King, A., Jokhi, P.P., Smith, S.K. et al. (1995) Screening for cytokine mRNAexpression in puri®ed human villous and extravillous trophoblastpopulations using the reverse-transcriptase polymerase chain reaction(RT±PCR). Cytokine, 7, 364±371.

King, A., Boocock, C., Sharkey, A.M. et al. (1996a) Evidence for theexpression of HLA-C class I mRNA and protein by human ®rst trimestertrophoblast. J. Immunol., 156, 2068±2076.

King, A., Burrows, T. and Loke, Y.W. (1996b) Human uterine natural killercells. Nat. Immunity, 15, 41±52.

King, A., Hiby, S.E., Verma, S. et al. (1997a) Uterine NK cells and trophoblastHLA class I molecules. Am. J. Reprod. Immunol., 37, 459±462.

King, A., Loke, Y.W. and Chaouat, G. (1997b) NK cells and reproduction.Immunol. Today, 18, 64±66.

King, A., Burrows, T., Verma, S. et al. (1998) Human uterine lymphocytes.Hum. Reprod. Update, 4, 480±485.

King, A., Gardner, L. and Loke, Y.W. (1999) Co-stimulation of humandecidual natural killer cells by interleukin-2 and stromal cells. Hum.Reprod., 14, 656±663.

King, A., Burrows, T.D., Hiby, S.E. et al. (2000a) Surface expression of HLA-C antigen by human extravillous trophoblast. Placenta, 21, 376±387.

King, A., Allan, D.S.J., Bowen, M. et al. (2000b) HLA-E is expressed ontrophoblast and interacts with CD94/NKG2 receptors on decidual NKcells. Eur. J. Immunol., 30, 1623±1631.

King, A., Hiby, S.E., Gardner, L. et al. (2000c) Recognition of trophoblastHLA class I molecules by decidual NK Cell Receptors ± a Review.Placenta, 21, 881±885.

Kirby, D.R. (1970) The egg and immunology. Proc. R. Soc. Med., 63, 59±61.Kirszenbaum, M., Moreau, P., Gluckman, E. et al. (1994) An alternatively

spliced form of HLA-G mRNA in human trophoblasts and evidence forthe presence of HLA-G transcript in adult lymphocytes. Proceed. Nat.Acad. Sci. USA, 91, 4209±4213.

Kirszenbaum, M., Moreau, P., Teyssier, M. et al. (1995) Evidence for thepresence of alternatively spliced HLA-G messenger-RNA forms in humanmononuclear-cells and from peripheral-blood and umbilical-cord blood.Hum. Immunol., 43, 237±241.

Klein, J. (1986) Natural History of the Major Histocompatibility Complex.John Wiley & Sons, New York, USA.

Klein, J. (1990) Antigens and other lymphocyte-activating substances. InKlein, J. (eds), Immunology. Blackwell, Cambridge, MA, USA, pp. 269±293.

Klein, J. and Figueroa, F. (1986) Evolution of the major histocompatibilitycomplex. CRC Crit. Rev. Immunol., 6, 295±387.

Klein, J., Juretic, A., Baxevanis, C.N. and Nagy, Z.A. (1981) The traditionaland a new version of the mouse H-2 complex. Nature, 291, 455±460.

Knapp, L.A., Ha, J.C. and Sackett, G.P. (1996) Parental MHC antigen sharingand pregnancy wastage in captive pigtailed macaques. J. Reprod.Immunol., 32, 73±88.

Komlos, L., Zamir, R., Joshua, H. and Halbrecht, I. (1977) Common HLAantigens in couples with repeated abortions. Clin. Immunol.Immunopathol., 7, 330±335.

Kostyu, D.D. (1994) HLA: Fertile Territory for Developmental Genes? Crit.Rev. Immunol., 14, 29±59.

Kovats, S., Main, E.K., Librach, C. et al. (1990) A class I antigen, HLA-G,expressed in human trophoblasts. Science, 248, 220±223.

Kriel, R.L., Gates, G.A., Wulff, H. et al. (1970) Cytomegalovirus isolationsassociated with pregnancy wastage. Am. J. Obstet. Gynecol., 106, 885±892.

Kristensen, B., WaÈ¯er, P. and de Weck, A.L. (1980) Histocompatibilityantigens (SLA) in swine. Possible correlation between SLA haplotypesand performance of piglets. Animal Blood Groups Biochem. Gen., 11, 58±59.

Kunz, H.W., Dixon-McCarthy, B. and Gill, T.J. (1996) Malignant potentialmajor histocompatibility complex antigen expression, and genomicimprinting of rat trophoblast cell lines. Biol. Reprod., 54, 1288±1299.


156

Kurpisz, M., Fernandez, N., Szymezynski, G. and Kowalik, I. (1986) The non-mature sperm cells: evaluation of surface markers and interaction in invitro cultures. Arch. Immunol. Therap. Exp., 34, 101±109.

Kurpisz, M., Fernandez, N., Witt, M. et al. (1987) HLA expression on humangerminal cells. J. Immunogenet., 14, 23±32.

Kurpisz, M., Fernandez, N. and Fiszer, D. (1995) The expression of MHCgenes in reproductive organs and their role in reproduction. In Kurpisz, M.and Fernandez, N. (eds), Immunology of Human. Reproduction. BIOSScienti®c Publishers, Oxford, UK, pp. 163±184.

Kwak, J.Y.H., Gilman-Sachs, A., Beamen, K.D. and Beer, A.E. (1992)Autoantibodies in women with primary recurrent spontaneous abortion ofunknown etiology. J. Reprod. Immunol., 22, 15±31.

Kwiatkowski, T.J. Jr., Beaudet, A.L., Trask, B.J. and Zoghbi, H.Y. (1991)Linkage mapping and ¯uorescence in situ hybridization of TCTE1 onhuman chromosome 6p: analysis of dinucleotide polymorphisms on nativegels. Genomics, 10, 921±926.

Laitinen, T. (1993) A set of MHC haplotypes found among Finnish couplessuffering from recurrent spontaneous abortions. Am. J. Reprod. Immunol.,29, 148±154.

Laitinen, T., Lokki, M.L., Tulppala, M. et al. (1991) Increased frequency ofcomplement C4 `null' alleles in recurrent spontaneous abortions. Hum.Reprod., 6, 1384±1387.

Laitinen, T., Koskimies, S. and Westman, P. (1993) Foeto-maternalcompatibility in HLA-DR, -DQ and -DP loci in Finnish couplessuffering from recurrent spontaneous abortions. Eur. J. Immunogen., 20,249±258.

Lanier, L.L. (1999) Natural killer cells fertile with receptors for HLA-G?Proc. Natl Acad. Sci. USA, 96, 5343±5345.

Le Bouteiller, P. (1994) HLA class I chromosomal region, genes, products:facts and questions. Crit. Rev. Immunol., 14, 89±129.

Le Bouteiller, P. and Blaschitz, A. (1999) The functionality of HLA-G isemerging. Immunol. Rev., 167, 233±244.

Le Bouteiller, P. and Lenfant, F. (1996) Antigen-presenting function(s) of thenon-classical HLA-E, -F and -G class I molecules: the beginning of astory. Res. Immunol., 147, 301±313.

Le Bouteiller, P., Rodriguez, A.M., Mallet, V. et al. (1996) Placentalexpression of HLA class I genes. Am. J. Reprod. Immunol., 35, 216±225.

Le Bouteiller, P., Solier, C., Proll, J. et al. (1999) Placental HLA-G proteinexpression in vivo: where and what for? Hum. Reprod. Update, 5, 223±233.

Lee, N., Llano, M., Carretero, M. et al. (1998) HLA-E is a major ligand for thenatural killer inhibitory receptor CD94/NKG2A. Proc. Natl Acad. Sci.USA, 95, 5199±5204.

Lee, N., Malacko, A.R., Ishitani, A. et al. (1995) The membrane-bound andsoluble forms of HLA-G bind identical sets of endogenous peptides butdiffer with respect to TAP association. Immunity, 3, 591±600.

Lefebvre, S., Moreau, P., Guiard, V. et al. (1999) Molecular mechanismscontrolling constitutive and IFN-g-inducible HLA-G expression in variouscell types. J. Reprod. Immunol., 43, 213±224.

Levanon, D., Negreanu, V., Bernstein, Y. et al. (1994) AML1, AML2, andAML3, the human members of the runt domain gene-family: cDNAstructure, expression, and chromosomal localization. Genomics, 23, 425±432.

Li, L., Huang, G.M., Banta, A.B. et al. (1998) Cloning, characterization, andthe complete 56.8-kilobase DNA seqeunce of the human NOTCH4 gene.Genomics, 51, 45±48.

Li, X.-Y., Mattei, M.G., Zaleska-Rutczynska, Z. et al. (1991) One subunit ofthe transcription factor NF-Y maps close to the major histocompatibilitycomplex in murine and human chromosomes. Genomics, 11, 630±634.

Lie, B.A., Sollid, L.M., Ascher, H. et al. (1999) A gene telomeric of the HLAclass I region is involved in predisposition to both type 1 diabetes andcoeliac disease. Tiss. Ant., 54, 162±168.

Lim, K.J., Odukoya, O.A., Ajjan, R.A. et al. (1998) Pro®le of cytokine mRNAexpression in peri-implantation human endometrium. Mol. Hum. Reprod.,4, 77±81.

Llano, M., Lee, N., Navarro, F. et al. (1998) HLA-E-bound peptides in¯uencerecognition by inhibitory and triggering CD94/NKG2 receptors:preferential response to an HLA-G derived nonamer. Eur. J. Immunol.,28, 2854±2863.

Loke, Y.W., Hiby, S. and A., K. (1999) Human leukocyte antigen-G andreproduction. J. Reprod. Immunol., 43, 235±242.

Loke, Y.W. and King, A. (2000) Decidual natural-killer-cell interaction withtrophoblast: cytolysis or cytokine production? Biochem. Soc. Transact.,28, 196±198.

Loke, Y.W., King, A., Burrows, T. et al. (1997) Evaluation of trophoblast

HLA-G antigen with a speci®c monoclonal antibody. Tiss. Ant., 50, 135±146.

Long, E.O. (1998) Signal sequences stop killer cells. Nature, 391, 740±743.Lopez-Botet, M., Moretta, L. and Strominger, J. (1996) NK-cell receptors and

recognition of MHC class I molecules. Immunol. Today, 17, 212±214.Lorentzen, D.F., Iwanaga, K.K., Meuer, K.J. et al. (1997) A 25% error rate in

serologic typing of HLA-homozygotes. Tiss. Ant., 50, 359±365.Lorenzo, J.A., Jastrezebski, S.L., Kalinowski, J.F. et al. (1994) Tumor necrosis

factor alpha stimulates production of leukemia inhibitory factor in humandermal ®broblast cultures. Clin. Immunol. Immunopathol., 70, 260±265.

Luckett, W.P. (1974) Comparative development and evolution of the placentain primates. In Luckett, W.P. (ed.), Reproductive Biology of the Primates.Karger, New York, pp. 142±234.

MacCluer, J.W., Bailey, E., Weitkamp, L.R. and Blangero, J. (1988) ELA andfertility in American Standardbred horses. Animal Genetics, 19, 359±372.

Maejima, M., Fujii, T., Kozuma, S. et al. (1997) Presence of HLA-G-expressing cells modulates the ability of peripheral blood mononuclearcells to release cytokines. Am. J. Reprod. Immunol., 38, 79±82.

Mallet, V., Proll, J., Solier, C. et al. (2000) The full length and no otheralternative form of HLA-G is expressed at the cell surface of transfectedcells. Hum. Immunol., 61, 212±224.

Mandelboim, O., Pazmany, L., Davis, D.M. et al. (1997) Multiple receptorsfor HLA-G on human natural killer cells. Proc. Natl Acad. Sci. USA, 94,14666±14670.

Mandelboim, O., Reyburn, H.T., Vales-Gomez, M. et al. (1996) Protectionfrom lysis by natural killer cells of group 1 and 2 speci®city is mediatedby residue 80 in human histocompatibility leukocyte antigen C alleles andalso occurs with empty major histocompatibility molecules. J. Exp. Med.,184, 913±922.

Maruyama, T., Makino, T., Sugi, T. et al. (1992) Flow-cytometric analysis ofimmune cell populations in human decidua from various types of ®rsttrimester pregnancy. Hum. Immunol., 34, 212±218.

Maschek, U., Pulm, W. and Hammerling, G.J. (1989) Altered regulation ofMHC class I genes in different tumor cell lines is re¯ected by distinct setof DNase I hypersensitive sites. EMBO J., 8, 2297±2304.

Masuno, M., Fukao, T., Song, X.-Q. et al. (1996) Assignment of the humancytosolic acetoacetyl-coenzyme A thiolase (ACAT2) gene to chromosome6p25.3-q26. Genomics, 36, 217±218.

McElhinny, A.S., Kadow, N. and Warner, C.M. (1998) The expression patternof the Qa-2 antigen in mouse preimplantation embryos and its correlationwith the Ped gene phenotype. Mol. Hum. Reprod., 4, 966±971.

McMaster, M., Zhou, Y., Shorter, S. et al. (1998) HLA-G isoforms producedby placental cytotrophoblasts and found in amniotic ¯uid are due tounusual glycosylation. J. Immunol., 160, 5922±5928.

McMaster, M.T., Librach, C.L., Zhou, Y. et al. (1995) Human placental HLA-G expression is restricted to differentiated cytotrophoblasts. J. Immunol.,154, 3771±3778.

Messer, G., Zemmour, J., Orr, H.T. et al. (1992) HLA-J, a second inactivatedclass I HLA gene related to HLA-G and HLA-A. Implications for theevolution of the HLA-A-related genes. J. Immunol., 148, 4043±53.

Miller, M.D., Yamamoto, H., Hughes, A.L. et al. (1991) De®nition of anepitope and an MHC class I molecule recognized by gag-speci®c CTL inSIV mac infected rhesus monkeys. J. Immunol., 147, 320±329.

Mincheva-Nilsson, L., HammarstroÈm, S. and HammarstroÈm, M.-L.(1992)Human decidual leukocytes from early pregnancy contain high numbersof gd+ cells and show selective down-regulation of alloreactivity. J.Immunol., 149, 2203±2211.

Miyazaki, J., Appella, E. and Ozato, K. (1986) Negative regulation of theMHC class I gene in undifferentiated embryonal carcinoma cells. Proc.Natl Acad. Sci. USA, 83, 9537±9541.

Morales, P., Corell, A., Martinez-Laso, J. et al. (1993) Three new HLA-Galleles and their linkage disequilibria with HLA-A. Immunogenetics, 38,323±31.

Moreau, P., Carosella, E., Teyssier, M. et al. (1995) Soluble HLA-G molecule.An alternatively spliced HLA-G mRNA form candidate to encode it inperipheral blood mononuclear cells and human trophoblasts. Hum.Immunol., 43, 231±236.

Moreau, P., Adrian-Cabestre, F., Menier, C. et al. (1999) IL-10 selectivelyinduces HLA-G expression in human trophoblasts and monocytes. Int.Immunol., 11, 803±811.

Moretta, A., Vitale, M., Bottino, C. et al. (1993) P58 molecules as putativereceptors for major histocompatibility complex (MHC) class-1 moleculesin human natural-killer (NK) cells-anti P58 antibodies reconstitute lysis ofMHC class-1-protected cells in NK clones displaying differentspeci®cities. J. Exp. Med., 178, 597±604.


157

Motokura, T., Yi, H.F., Kronenberg, H.M. et al. (1992) Assignment of thehuman cyclin D3 gene (CCND3) to chromosome 6p-ql3. Cytogenet. CellGenet., 61, 5±7

Nagarajan, L., Louie, E., Tsujimoto, Y. et al. (1986) Localization of thehuman pim oncogene (PIM) to a region of chromosome 6 involved intranslocations in acute leukemias. Proc. Natl Acad. Sci. USA, 83, 2556±2560.

Nagata, T., Segars, J.H., Levi, B.-Z. and Ozato, K. (1992) Retinoic acid-dependent transactivation of major histocompatibility complex class Ipromoters by the nuclear hormone receptor H-2RIIBP in undifferentiatedembryonal carcinoma cells. Proc. Natl Acad. Sci. USA, 89, 937±941.

Naume, B. and Espevik, T. (1994) Immunoregulatory effects of cytokines onnatural killer cells. Scand. J. Immunol., 40, 128±134.

Navarro, F., Llano, M., Bellon, T. et al. (1999) The ILT2 (LIR1) and CD94/NKG2A NK cell receptors respectively recognize HLA-G1 and HLA-Emolecules co-expressed on target cells. Eur. J. Immunol., 29, 277±283.

Nelson, J.L., Koepsel, T.D., Dugowson, C.E. et al. (1993) Fecundity beforedisease onset in women with rheumatoid arthritis. Arthritis Rheumat., 36,7±14.

Nicols, J., Zevnik, B., Anastassiadis, K. et al. (1998) In situ hybridizationlocalizes the human OTF3 to chromosome 6p21.3-p22 and OTF3L to12p13. Cytogenet. Cell. Genet., 63, 212±214.

Nishikawa, K., Saito, S., Morii, T. et al. (1991) Accumulation of CD16±

CD56+ natural killer cells with high af®nity interleukin 2 receptors inhuman early pregnancy decidua. Int. Immunol., 3, 743±750.

Nordlander, C., Fuchs, T., Hammarstrom, L. and Smith, C.I.E. (1983) Humanleukocyte antigens group A in couples with unexplained infertility. Fertil.Steril., 40, 60±65.

Nunes, M., Peelman, L., Vaiman, M. et al. (1994) Characterization of six newloci within the swine major histocompatibility complex class III region.Mammal. Genet., 5, 616±622.

Ober, C. (1990) The MHC complex in normal pregnancy and pregnancy loss.In Gleicher, E.A. (ed.), Principles of Medical Therapy in Pregnancy.

Ober, C. (1995) Current topic: HLA and reproduction: Lessons from studies inthe Hutterites. Placenta, 16, 569±577.

Ober, C. (1999) Studies of HLA, fertility and mate choice in a human isolate.Hum. Reprod. Update, 5, 103±107.

Ober, C. and Aldrich, C.L. (1997) HLA-G polymorphisms: neutral evolutionor novel function? J. Reprod. Immunol., 36, 1±21.

Ober, C. and van der Ven, K. (1997) Immunogenetics of Reproduction: anOverview. In Olding, L.B. (ed.), Reproductive Immunology (CTIM).Springer, Berlin, pp. 1±23.

Ober, C., Martin, A.O., Simpson, J.L. et al. (1983) Shared HLA antigens andreproductive performance in the Hutterites. Am. J. Hum. Genet., 35, 990±1004.

Ober, C., Hauck, W.W., Kostyu, D.D. et al. (1985) Adverse effects of HLA-DR sharing on fertility: a cohort study in a human isolate. Fertil. Steril.,44, 227±232.

Ober, C., Elias, S., E., O'Brien, E. et al. (1988) HLA sharing and fertility inHutterite couples: evidence for prenatal selection against compatiblefetuses. Am. J. Reprod. Immunol. Microbiol., 18, 111±115.

Ober, C., Elias, S., Kostyu, D.D. and Hauck, W.W. (1992) Decreasedfecundability in Hutterite couples sharing HLA-DR. Am. J. Hum. Genet.,50, 6±14.

Ober, C., Steck, T., van der Ven, K. et al. (1993) MHC Class II compatibilityin aborted fetuses and MHC term infants of couples with recurrentspontaneous abortion. J. Reprod. Immunol., 25, 195±207.

Ober, C., Rosinsky, B., Grimsley, C. et al. (1996) Population genetic studies ofHLA-G: allele frequencies and linkage disequilibrium with HLA-A. J.Reprod. Immunol., 32, 111±123.

Ober, C., Aldrich, C., Rosinsky, B. et al. (1998a) HLA-G1 protein expressionis not essential for fetal survival. Placenta, 19, 127±132.

Ober, C., Hyslop, T., Elias, S. et al. (1998b) Human leukocyte antigenmatching and fetal loss: results of a 10 year prospective study. Hum.Reprod., 13, 33±38.

O'Callaghan, C.A. and Bell, J.I. (1998) Structure and function of the humanMHC class Ib molecules HLA-E, HLA-F, and HLA-G. Immunol. Rev.,163, 129±138.

Okamoto, K., Okazawa, H., Okuda, A. et al. (1990) A novel octamer bindingtranscription factor is differentially expressed in mouse embryonic cells.Cell, 60, 461±472.

O'Neil, D., Swanton, C., Jones, A. et al. (1999) IFN-gamma down-regulatesMHC expression and antigen processing in a human B cell line. J.Immunol., 162, 791±798.

Onno, M., Guillaudeux, T., Amiot, L. et al. (1994) The HLA-G gene is

expressed at a low mRNA levels in different human cells and tissues.Hum. Immunol., 41, 79±86.

Otting, N. and Bontrop, R.E. (1993) Characterization of the rhesus macaque(Macaca mulatta) equivalent of HLA-F. Immunogenetics, 38, 141±145.

Oudejans, C.B.M., Krimpenfort, P., Ploegh, H.L. and Meijer, L.M. (1989)Lack of expression of HLA-B27 gene in transgenic mouse trophoblast. J.Exp. Med., 169, 447±456.

Ozato, K., Wan, Y.J. and Orrison, B. (1985) Mouse major histocompatibilityclass I gene expression begins at midsomite stage and is inducible atearlier-stage embryos by interferon. Proc. Natl Acad. Sci. USA, 82, 2427±2431.

Pack, S.D., Tanigami, A., Ledbetter, D.H., et al. (1997) Assignment oftrophoblast/endometrial epithelium cell adhesion molecule trophinin geneTRO to human chromosome bands Xp11.22®p11.21 by in situhybridization. Cytogenet. Cell. Genet., 79, 123±124.

Pack, S.D., Pak, E., Tanigami, A. et al. (1998) Assignment of the bystin geneBYSL to human chromosome band 6p21.1 by in situ hybridization.Cygenet. Cell. Genet., 83, 76±77.

Pandey, J.P., Zamani, M. and Cassiman, J.J. (1999) Epistatic effects of genesencoding tumor necrosis factor-alpha, immunoglobulin allotypes, andHLA antigens on susceptibility to non-insulin-dependent (type 2) diabetesmellitus. Immunogenetics, 49, 860±864.

Parham, P. (2000) The Immune System. Garland Publishing, London.Parr, E.L., Chen, H.L., Parr, M.B. and Hunt, J.S. (1995) Synthesis and

localization of tumor necrosis factor-a in activated NK cells in thepregnant uterus. J. Reprod. Immunol., 28, 31±40.

Patankar, M.S., Ozgur, K., Oehninger, S. et al. (1997) Expression of glycanslinked to natural killer cell inhibition on the human zona pellucida. Mol.Hum. Reprod., 3, 501±505.

Pasteris, N.G. and Gorski, J.L. (1999) Isolation, characterization, and mappingof the mouse and human Fgd2 genes, faciogenital dysplasia (FGD1;Aarskog syndrome) gene homologues. Genomics, 60, 57±66.

Paul, P., RouasFreiss, N., KhalilDaher, I. et al. (1998) HLA-G expression inmelanoma: a way for tumor cells to escape from immunosurveillance.Proc. Natl Acad. Sci. USA, 95, 4510±4515.

Pazmany, L., Mandelboim, O., Vales-Gomez, M. et al. (1996) Protection Fromnatural killer cell-mediated lysis by HLA-G expression on target cells.Science, 274, 792±795.

Peelman, L.J., Chardon, P., Nunes, M. et al. (1995) The BAT1 gene in theMHC encodes an evolutionarily conserved putative nuclear RNA helicaseof the DEAD family. Genomics, 26, 210±218.

Pende, D., Sivori, S., Accame, L. et al. (1997) HLA-G recognition by humannatural killer cells. Involvement of CD94 both as inhibitory and asactivating receptor complex. Eur. J. Immunol., 27, 1875±1880.

Penzes, M., Rajczy, K., Gyodi, E. et al. (1999) HLA-G gene polymorphism inthe normal population and in recurrent spontaneous abortion in Hungary.Transplant. Proc., 31, 1832±1833.

Perez-Villar, J.J., Melero, I., Navarro, F. et al. (1997) The CD94/NKG2Ainhibitory receptor complex is involved in the natural killer cell-mediatedrecognition of cells expressing HLA-G1. J. Immunol., 158, 5736±5743.

Persitz, E., Oksenber, J.R., Margalioth, E.H. et al. (1985) Histoincompatibilityin couples with unexplained infertility. Fertil. Steril., 43, 733±738.

Piccinni, M.P., Beloni, L., Livi, C. et al. (1998) Defective production of bothleukemia inhibitory factor and type 2 T-helper cytokines by decidual Tcells in unexplained recurrent abortions. Nature Med., 4, 1020±1024.

Ponte, M., Cantoni, C., Biassoni, R. et al. (1999) Inhibitory receptors sensingHLA-G1 molecules in pregnancy: Decidua-associated natural killer cellsexpress LIR-1 and CD94/NKG2A and acquire p49, an HLA-G1 speci®creceptor. Proc. Natl Acad. Sci. USA, 96, 5674±5679.

ProÈll, J., Blaschitz, A., Hutter, H. and Dohr, G. (1999) First trimester humanendovascular trophoblast cells express both HLA-C and HLA-G. Am. J.Reprod. Immunol., 42, 30±36.

Ragoussis, J., Senger, G., Mockridge, I. et al. (1992) A testis-expressed Zn®nger gene (ZNF76) in human 6p21.3 centromeric to the MHC is closelylinked to the human homolog of the t-complex gene tcp-11. Genomics, 14,673±679.

Rajagopalan, S. and Long, E.O. (1999) A human histocompatibility leukocyteantigen (HLA)-G-speci®c receptor expressed on all natural killer cells. J.Exp. Med., 189, 1093±1100.

Raum, D., Awdeh, Z., Yunis, E.J. et al. (1984) Extended majorhistocompatibility complex haplotypes in type I diabetes mellitus. J.Clin. Invest., 74, 449±454.

Recurrent Misscarriage Immunotherapy Trialist Group, T.R.M.I.T. (1994)Worldwide collaborative observational study and meta-analysis on


158

allogeneic leukocyte immunotherapy for recurrent spontaneous abortion.Am. J. Reprod. Immunol., 32, 55±72.

Regan, L. (1997) Miscarriage What Every Woman Needs to Know.Bloomsbury, London, UK.

Reiss, H.E. (1998) Reproductive Medicine: From A to Z, Oxford UniversityPress, Oxford, UK.

Renard, C., Bolet, G., Dando, P. and Vaiman, M. (1985) Relations d'unmarqueur genetique, le complexe majeur d'histocompatibiliteÂ avec laproli®citeÂ des truies et la mortaliteÂ des porcelets. Journees de RecherchePorcine en France, 17, 105±112.

Robb, J., Benirshke, K. and Barmeyer, R. (1986) Intrauterine latent herpessimplex virus infection: Spontaneous abortion. Hum. Pathol., 17, 1196±1209.

Roberts, J.M., Taylor, C.T., Melling, G.C. et al. (1992) Expression of theCD46 antigen, and absence of class-I MHC antigen, on the human oocyteand preimplantation blastocyst. Immunology, 75, 202±205.

Rohn, W., Tang, L.P., Dong, Y.S. and Benveniste, E.N. (1999) IL-1b inhibitsIFN-g-induced class II MHC expression by suppressing transcription ofthe class II transactivator gene. J. Immunol., 162, 886±896.

Rosner, M.H., Vigano, M.A., Ozato, K. et al. (1990) A POU-domaintranscription factor in early stem cells and germ cells of the mammalianembryo. Nature, 345, 686±692.

Rothschild, M.F., Zimmerman, D.R., Johnson, R.K. et al. (1984) SLAhaplotype differences in lines of pigs which differ in ovulation rate.Animal Blood Groups and Biochemical Genetics, 15, 155±158.

Rothschild, M.F., Renard, C., Bolet, G. et al. (1986) Effect of swinelymphocyte antigen haplotypes on birth and weaning weights in pigs.Animal Genetics, 17, 267±272.

Rouas-Freiss, N., Marchal, R.E., Kirszenbaum, M. et al. (1997a) The a1domain of HLA-G1 and HLA-G2 inhibits cytotoxicity induced by naturalkiller cells: Is HLA-G the public ligand for natural killer cell inhibitoryreceptors? Proc. Natl Acad. Sci. USA, 94, 5249±5254.

Rouas-Freiss, N., Marchal-Bras Gonvalves, R., Menier, C., et al. (1997b)Direct evidence to support the role of HLA-G in protecting the fetus frommaternal uterine natural killer cytolysis. Proc. Natl Acad. Sci. USA, 94,11520±11525.

Roy Choudhury, S. and Knapp, L.A. (2001) Human Reproductive Failure I:immunological factors. Hum. Reprod. Update, 7, 113±134.

Saito, S., Nishikawa, K., Mori, T. et al. (1993) Cytokine production by CD16-CD56bright natural killer cells in the human early pregnancy decidua. Int.Immunol., 5, 559±563.

Sakkas, D., Shoukir, Y., Chardonnens, D. et al. (1998) Early cleavage ofhuman embryos to the two-cell stage after intracytoplasmic sperminjection as an indicator of embryo viability. Hum. Reprod., 13, 182±187.

Samaridis, J. and Colonna, M. (1997) Cloning of novel immunoglobulinsuperfamily receptors expressed on human myeloid and lymphoid cells:structural evidence for new stimulatory and inhibitory pathways. Eur. J.Immunol., 27, 660±665.

Saris, C.J.M., Domen, J. and Berns, A. (1991) The pim-1 oncogene encodestwo related-protein-serine/threonine kinases by alternative initiation atAUG and CUG. EMBO J., 10, 655±664.

Sarvetnick, N., Tsai, J.-Y., Fox, H. et al. (1989) A mouse chromosome 17encodes a testes-speci®c transcript with unusual properties.Immunogenetics, 30, 34±41.

Sarvetnick, N., Tsai, J.-Y., Fox, H. et al. (1990) Erratum. A mousechromosome 17 encodes a testes-speci®c transcript with unusualproperties. Immunogenetics, 31, 283.

Sbracia, M., Scarpellini, F., Mastrone, M. and Grasso, J.A. (1996) HLAAntigen Sharing in Italian Couples in Which Women Were Affected byGestational Trophoblastic Tumors. Am. J. Reprod. Immunol., 35, 252±255.

Schacter, B., Muir, A., Gyves, M. and Tasin, M. (1979) HLA-A, Bcompatibility in parents of offspring with neural-tube defects or couplesexperiencing involuntary fetal wastage. Lancet, i, 796±799.

Schaid, D.J., Kunz, H.W. and Gill, T.J.I. (1982) Genetic interaction causingembryonic mortality in the rat: epistasis between Tal and grc genes.Genetics, 100, 615±632.

Schmidt, C.M. and Orr, H.T. (1993) Maternal/Fetal interaction: the role of theMHC class I molecule HLA-G. Crit. Rev. Immunol., 13, 207±224.

Schmidt, C.M., Chen, H.L., Chiu, I. et al. (1995) Temporal and spatialexpression of HLA-G messenger-RNA in extra embryonic tissues oftransgenic mice. J. Immunol., 155, 619±629.

Schust, D.J., Tortorella, D. and Ploegh, H.L. (1999) HLA-G and HLA-C at thefeto-maternal interface: lessons learned from pathogenic viruses. Semin.Can. Biol., 9, 37±46.

Shelton, A.J., Harger, J.H., Dorman, J.S. et al. (1994) Association betweenfamilial autoimmune disease and recurrent spontaneous abortions. Am. J.Reprod. Immunol., 32, 82±87.

Sheps, M.C. (1965) An analysis of reproductive patterns in an Americanisolate. Pop. Stud., 19, 65±80.

Shin, H.-S., Bennett, D. and Artzt, K. (1984) Gene mapping within the T/tcomplex of the mouse. IV. The inverted MHC is intermingled with severalt-lethal genes. Cell, 39, 573±578.

Shorter, S.C., Jackson, M.C., Sargent, I.L. et al. (1990) Puri®cation of humancytotrophoblast from term amniochorion by ¯ow cytometry. Placenta, 11,505±513.

Shukla, H., Swaroop, A., Srivastava, R. and Weissman, S.M. (1990) ThemRNA of a human class I gene HLA G/HLA 6.0 exhibits a restrictedpattern of expression. Nucleic Acids Res., 18, 2189.

Silver, L.M. (1985) Mouse t haplotypes. Ann. Rev. Genet., 19, 179±208.Singer, D.S. and Maguire, J.E. (1990) Regulation of the expression of class I

MHC genes. Crit. Rev. Immunol., 10, 235±256.Sipes, S.L., Medaglia, M.V., Stabley, D.L. et al. (1996) A new major

histocompatibility complex class Ib gene expressed in the mouseblastocyst and placenta. Immunogenetics, 45, 108±120.

Slukvin, I.I., Boyson, J.E., Watkins, D.I. and Golos, T.G. (1998) The rhesusmonkey analog of HLA-G is expressed primarily in villoussyncytiotrophoblasts. Biol. Reprod., 58, 728±738.

Slukvin, I.I., Watkins, D.I. and Golos, T.G. (1999) Tissue distribution of themRNA for a rhesus monkey major histocompatibility class Ib molecule,Mamu-AG. Tiss. Ant., 53, 282±291.

Smith, J.B. and Cowchock, F.S. (1988) Immunological studies in recurrentspontaneous abortion: effects of immunization of women with paternalmononuclear cells on lymphocytotoxic and mixed lymphocyte reactionblocking antibodies and correlation with sharing of HLA antigens andpregnancy outcome. J. Reprod. Immunol., 14, 99±113.

SoÈderstroÈm, K., Corliss, B., Lanier, L.L. and Phillips, J.H. (1997) CD94/NKG2 Is the predominant inhibitory receptor involved in recognition ofHLA-G by decidual and peripheral blood NK cells. J. Immunol., 159,1072±1075.

Sorrentino, V., McKinney, M.D., Giorgi, M. et al. (1988) Expression ofcellular protooncogenes in the mouse male germ line: a distinctive 2.4-kilobase pim-1 transcript is expressed in haploid postmeiotic cells. Proc.Natl Acad. Sci. USA, 85, 2191.

Spies, T., Bresnahan, M. and Strominger, J.L. (1989) Human majorhistocompatibility complex contains a minimum of 19 genes betweenthe complement cluster and HLA-B. Proc. Natl Acad. Sci. USA, 86, 8955±8958.

Steck, T., van der Ven, K., Kwak, J. et al. (1995) HLA-DQA1 and HLA-DQB1 haplotypes in aborted fetuses and couples with recurrentspontaneous abortion. J. Reprod. Immunol., 29, 95±104.

Steffensen, R., Christiansen, O.B., Bennett, E.P. and Jersild, C. (1998) HLA-Epolymorphism in patients with recurrent spontaneous abortion. Tiss. Ant.,52, 569±572.

Stroynowski, I. and Tabaczewski, P. (1996) Multiple products of class Ib Qa-2genes: which ones are functional? Res. Immunol., 147, 290±301.

Sunderland, C.A., Redman, C.W.G. and Stirrat, G.M. (1981a) HLA-A, B, Cantigens are expressed on nonvillous trophoblasts of the early humanplacenta. J. Immunol., 127, 2614±2615.

Sunderland, C.A., Naiem, M., Mason, D.Y. et al. (1981b) The expression ofmajor histocompatibility antigens by human chorionic villi. J. Reprod.Immunol., 3, 323±331.

Suzuki, N., Zara, J., Sato, T. et al. (1998) A cytoplasmic protein, bystin,interacts with trophinin, tastin and cytokeratin and may be involved introphinin-mediated cell and endometrial epithelial cells. Proc. Nat. Acad.Sci. USA, 95, 5027±5032.

Suzuki, N., Nakayama, J., Shih, I.M. et al. (1999) Expression of trophinin,tastin, and bystin by trophoblast and endometrial cells in human placenta.Biol. Reprod., 60, 621±627.

Szekeres-Bartho, J., Barakonyi, A., Polgar, B. et al. (1999) The role of g/d Tcells in progesterone-mediated immunodulation during pregnancy: areview. Am. J. Reprod. Immunol., 42, 44±48.

Takakuwa, K., Higashino, M., Ueda, H. et al. (1992) Signi®cant compatibilitydoes not exist at the HLA-DQB gene locus in couples with unexplainedrecurrent abortions. Am. J. Reprod. Immunol., 28, 12±16.

Takeda, J., Seino, S. and Bell, G.I. (1992) Human Oct3 gene family: cDNAsequences, alternative splicing, gene organization, chromosomal location,and expression at low levels in adult tissues. Nucleic Acids Res., 20, 4613±4620.


159

Tanaka, Y., Brenner, M.B., Bloom, B.R. and Morita, C.T. (1996) Recognitionof nonpeptide ligands by T cells. J. Mol. Med., 74, 223±231.

TarõÂn, J.J. (1997) Do the fastest concepti have a shorter life span? Hum.Reprod., 12, 885±889.

Terasaki, P.I., Mandell, M., Van de Water, J. and Edgington, T.E. (1964)Human blood lymphocyte cytotoxicity reactions with allogeneic antisera.Ann. N.Y. Acad. Sci., 120, 322±334.

Tishkoff, S.A., Dietzsch, E., Speed, W. et al. (1996) Global patterns of linkagedisequilibrium at the CD4 locus and modern human origins. Science, 271,1380±1387.

Tomasec, P., Braud, V.M., Rickards, C. et al. (2000) Surface expression ofHLA-E, an inhibitor of natural killer cells, enhanced by humancytomegalovirus gpUL40. Science, 287, 1031.

Trinchieri, G. (1989) Biology of Natural Killer cells. Adv. Immunol., 47, 187±376.

Ulrich, E., Kauffmann-Zeh, A., Hueber, A.-O. et al. (1997) Gene structure,cDNA sequence, and expression of murine Bak, a proapoptotic Bcl-2family member. Genomics, 44, 195±200.

Uyttendaele, H., Marazzi, G., Wu, G. et al. (1996) Notch4/int-3, a mammaryproto-oncogene, is an endothelial cell-speci®c mammalian Notch gene.Development, 122, 2251±2259.

Van den Elsen, P.J., Peijnenburg, A., van Eggermond, M.C. and Gobin, S.J.(1998) Shared regulatory elements in the promoters of MHC class I andclass II genes. Immunol. Today, 19, 308±312.

van der Ven, K. and Ober, C. (1994) HLA-G polymorphisms in AfricanAmericans. J. Immunol., 153, 5628±5633.

van Eijk, M.J.T., Mandelbaum, J., Salat-Baroux, J. et al. (1996) Expression ofleukemia inhibitory factor receptor subunits LIF-Rb and gp130 in humanoocytes and preimplantation embryos. Mol. Hum. Reprod., 2, 355±360.

van Endert, P.M., Lopez, M.T., Patel, S.D. et al. (1992) Genomicpolymorphism, recombination, and linkage disequilibrium in majorhistocompatibility complex-encoded antigen-processing genes. Proc.Natl Acad. Sci. USA, 89, 11594±11597.

Verma, S., King, A. and Loke, Y.W. (1997) Expression of killer-cell inhibitoryreceptors (KIR) on human uterine NK cells. Eur. J. Immunol., 27, 979±983.

Verma, S., Hiby, S.E., Loke, Y.W. and King, A. (2000) Human decidualnatural killer cells express the receptor for and respond to the cytokineinterleukin 12. Biol. Reprod., 62, 959±968.

Vincek, V., Figueroa, F., Gill, T.J.I. et al. (1990) Mapping in the mouse of theregion homologous to the rat growth and reproduction complex (grc).Immunogenetics, 32, 293±295.

Vitale, M., Sivori, S., Pende, D. et al. (1995) Coexpression of 2 functionallyindependent p58 inhibitory receptors in human natural-killer-cell clonesresults in the inability to kill in all normal allogenic target-cells. Proc.Natl Acad. Sci. USA, 92, 3536±3540.

Vitale, M., Castriconi, R., Parolini, S. et al. (1999) The leukocyte Ig-likereceptor (LIR)-1 for the cytomegalovirus UL18 protein displays a broadspeci®city for different HLA class I alleles analysis of LID-1 (+) NK cellclones. Int. Immunol., 11, 29±36.

Wagenknecht, D.R., Green, K.M. and McIntyre, J. (1997) Analyses of HLA-DQ alleles in recurrent spontaneous abortion (RSA) couples. Am. J.Reprod. Immunol., 37, 1±6.

Wainwright, S.D., Biro, P.A. and Holmes, C.H. (2000) HLA-F is apredominantly empty, intracellular, TAP-associated MHC class Ibprotein with a restricted expression pattern. J. Immunol., 164, 319±328.

Wan, Y.J., Orrison, B.M., Lieberman, R. et al. (1987) Induction of majorhistocompatibility class I antigens by interferons in undifferentiated F9cells. J. Cell. Physiol., 130, 276±283.

Warner, C.M., Gollnick, S.O., Flaherty, L. and Goldberg, S.B. (1987) Analysisof Qa-2 antigen expression by preimplantation mouse embryos: possiblerelationship to the preimplantation embryo development (Ped) geneproduct. Biol. Reprod., 36, 611±616.

Warner, C.M., Brownell, M.S. and Rothschild, M.F. (1991) Analysis of littersize and weight in mice differing in Ped gene phenotype and the Q regionof the H-2 complex. J. Reprod. Immunol., 19, 303±313.

Warner, C.M., Exley, G.E., McElhinny, A.S. and Tang, C. (1998) GeneticRegulation of Preimplantation Mouse Embryo Survival. J. Exp. Zool.,282, 272±279.

Wassmuth, R. (1996) HLA/MHC class II gene regulation. In Browning, M.J.

and McMichael, A.J. (eds), HLA and MHC: Genes, Molecules andFunction. BIOS Scienti®c Publishers Ltd, Oxford, UK, pp. 159±191.

Weckstein, L.N., Patrizio, P., Balmaceda, J.P. et al. (1991) Human leukocyteantigen compatibility and failure to achieve a viable pregnancy withassisted reproductive technology. Acta Eur. Fertil., 22, 103±107.

Wei, X. and Orr, H.T. (1990) Differential expression of HLA-E, HLA-F, andHLA-G transcripts in human tissue. Hum. Immunol., 29, 131±142.

Weintraub, B.C., Jackson, M.R. and Hedrick, S.M. (1994) T cells canrecognize nonclassical MHC in absence of conventional antigenicpeptides. J. Immunol., 153, 3051±3058.

Weitkamp, L.R. and Schacter, B.Z. (1985) Transferrin and HLA: spontaneousabortion, neural tube defects, and natural selection. N. Engl. J. Med., 313,925±932.

Wenk, R.E. and Boughman, J.A. (1989) Inheritance of HLA class I alleles bychildren of parents who share antigens: A reassessment of genetic andinteractive hypotheses. Am. J. Hum. Genet., 45, A251.

Wetzler, M., Estrov, Z., Talpaz, M. et al. (1994) Leukemia inhibitory factor inlong-term adherent layer cultures: Increased levels of bioactive protein inleukemia and modulation by IL-4, IL-1 and TNF-a. Cancer Res., 54,1837±1842.

Wilcox, A.J., Weinberg, C.R., O'Connor, J.F. et al. (1988) Incidence of earlyloss of pregnancy. N. Engl. J. Med., 319, 189±194.

Xiong, Y., Menninger, J., Beach, D. and Ward, D.C. (1992) Molecular cloningand chromosomal mapping of CCND gene encoding human D-typecyclins. Genomics, 13, 575±584.

Xu, Y., Jin, P., Mellor, A. and Warner, C.A. (1994) Identi®cation of the Pedgene at the molecular level: the Q9 MHC Class I transgene converts thePed slow to the Ped fast phenotype. Biol. Reprod., 51, 695±699.

Xu, Y.X., Jin, P. and Warner, C.M. (1993) Modulation of preimplantationembryonic development by antisense oligonucleotides to majorhistocompatibility complex genes. Biol. Reprod., 48, 1042±1046.

Yamashita, T., Fujii, T., Watanabe, Y. et al. (1996) HLA-G genepolymorphism in a Japanese population. Immunogenetics, 44, 186±191.

Yamashita, T., Fujii, T., Tokunaga, K. et al. (1999) Analysis of humanleukocyte antigen-G polymorphism including intron 4 in Japanese coupleswith habitual abortion. Am. J. Reprod. Immunol., 41, 159±163.

Yang, Y., Chu, W., Geraghty, D.E. and Hunt, J.S. (1996) Expression of HLA-G in human mononuclear phagocytes and selective induction by IFN-g. J.Immunol., 156, 4224±4231.

Yelavarthi, K.K., Fishback, J.L. and Hunt, J.S. (1991) Analysis of HLA-GmRNA in human placental and extraplacental membrane cells by in situhybridization. J. Immunol., 146, 2847±2854.

Yeom, Y.I., Ha, H.-S., Balling, R. et al. (1991) Structure, expression andchromosomal location of the Oct-4 gene. Mech. Dev., 35, 171±179.

Yeom, Y.I., Abe, K. and Artzt, K. (1992) Evolution of the mouse H-2K region:a hot spot of mutation associated with genes transcribed in embryos and/orgerm cells. Genetics, 130, 629±638.

Yokoyama, W.M. and Seaman, W.F. (1993) The Ly-49 and NKR-P1 genefamilies encoding lectin-like receptors on natural killer cells: the NK genecomplex. Ann. Rev. Immunol., 11, 613±635.

Yoshino, M., Xiao, H., Amadou, C. et al. (1998) BAC clones and STS markersnear the distal breakpoint of the fourth t-inversion, ln(17)4d, in the H-2Mregion on mouse chromosome 17. Mamm. Genome, 9, 186±192.

Yu, V.C., Delsert, C., Andersen, B. et al. (1991) RXRB: a coregulator thatenhances binding of retinoic acid, thyroid hormone, and vitamin kDreceptors to their cognate response elements. Cell, 67, 1251±1266.

Yuhki, N. and O'Brien, S.J. (1990) DNA variation of the mammalian majorhistocompatibility complex re¯ects genomic diversity and populationhistory. Proc. Natl Acad. Sci. USA, 87, 836±840.

Zhang, K., Sikut, R. and Hansson, G.C. (1997) A MUC1 mucin secreted froma colon carcinoma cell line inhibits target cell lysis by natural killer cells.Cell. Immunol., 176, 158±165.

Zhang, Q., Wang, X. and Wolgemuth, D.J. (1999) Developmentally regulatedexpression of cyclin D3 and its potential in vivo interacting proteinsduring murine gametogenesis. Endocrinology, 140, 2790±2800.

Ziebe, S., Petersen, K., Lindenberg, S. et al. (1997) Embryo morphology orcleavage stage: how to select the best embryos for transfer after in-vitrofertilization. Hum. Reprod., 12, 1545±1549.

Received on June 9, 2000; accepted on December 11, 2000


160

human reproductive failure ii: immunogenetic and interacting factors

Documents