View - Human Reproduction Update
Transcript of View - Human Reproduction Update
Cytokine knockouts in reproduction: the use of gene ablationto dissect roles of cytokines in reproductive biology
Wendy V. Ingman1,3 and Rebecca L. Jones2
1Discipline of Obstetrics and Gynaecology, Research Centre for Reproductive Health, University of Adelaide, South Australia 5005,
Australia; 2Maternal and Fetal Health Research Group, University of Manchester, St Mary’s Hospital, Manchester, UK
3Correspondence address. E-mail: [email protected]
Cytokines play many diverse and important roles in reproductive biology, and dissecting the complex interactionsbetween these proteins and the different reproductive organs is a difficult task. One approach is to use gene ablation,or ‘knockout’, to analyse the effect of deletion of a single cytokine on mouse reproductive function. This review sum-marizes the essential roles of cytokines in reproductive biology that have been revealed by gene knockout studies,including development and regulation of the hypothalamo-pituitary-gondal axis, ovarian folliculogenesis, implan-tation and immune system modulation during pregnancy. However, successful utilization of this approach must con-sider the caveats associated with gene ablation studies, e.g. embryonic lethality, systemic effects of cytokine ablation onlocal reproductive processes and the limited exposure to pathogens in mice housed in laboratory conditions. New soph-isticated technology that temporally or spatially regulates gene ablation can overcome some of these limitations. Dis-coveries on the roles of cytokines in reproductive function uncovered by gene ablation studies can now be applied toimprove in vitro fertilization for infertile couples and in the development of contraceptive therapies.
Keywords: cytokines; gene ablation; animal model; IVF; contraception
Introduction
Cytokines are soluble protein signalling molecules that interact
with cells of the immune system. These include the interleukins
(ILs), colony stimulating factors (CSFs) and tumour necrosis
factor (TNF) families. Growing understanding of the cytokine-like
actions of many growth factors [e.g. transforming growth factors
(TGFs), activin] and endocrine hormones (e.g. prolactin, growth
hormone), has broadened the range of factors we now consider
as cytokines. The functions of cytokines are far from restricted
to immune cell regulation. Cytokines are also critical to the
success of the reproductive process, through direct actions on
reproductive cells including germ cells, the embryo, non-
haematopoietic cells in the gonads and uterus, and indirectly
through the promotion of an immunologically receptive environ-
ment for the production of gametes, implantation and development
of the conceptus and parturition. Understanding which cytokines
are involved in the reproductive process, and how, is critical to
the design of therapeutics which aim to correct reproductive path-
ologies of immune origin, including some forms of infertility,
endometriosis, recurrent spontaneous abortion, pre-eclampsia
and preterm labour.
Without highly regulated immune responses, the ability to
defend the organism against dangerous pathogens, while tolerating
self and non-harmful pathogens and allowing reproduction of a
genetically dissimilar offspring, is in jeopardy. Decisions regard-
ing when to mount an immune response to a pathogen and what
type of response is required are mediated by a complex array of
cytokines, many with overlapping functions. It is generally not
the presence or absence of any one cytokine that is responsible
for the resultant immune response, rather the collective effect of
what has become known as the ‘cytokine milieu’, or cytokine
microenvironment surrounding the effector cell. Redundancy is
therefore common amongst cytokines, and discovering which
cytokines are most functionally significant in any given process
is not a simple task.
Studies describing in vivo expression of cytokines and their
receptors tell us which cytokines we should be investigating,
and in vitro cell culture experiments inform on what role the cyto-
kine might have. However, in vivo experimentation involving the
addition or removal of the cytokine is necessary to understand the
impact this cytokine has on physiology and pathology. Com-
monly, mice are used for initial in vivo experimentation, for a
variety of reasons, including the ease at which the mouse
genome can be manipulated by the selective deletion or addition
of genes. This review examines the impact of ‘knockout’ technol-
ogy on our understanding of the role of cytokines in reproductive
biology, highlighting the problems which are encountered by this
type of approach, how these problems can be overcome and oppor-
tunities for future applications of this technology.
# The Author 2007. Published by Oxford University Press on behalf of the European Society of Human Reproduction and Embryology. All rights reserved.
For Permissions, please email: [email protected] 179
Human Reproduction Update, Vol.14, No.2 pp. 179–192, 2008 doi:10.1093/humupd/dmm042
Advance Access publication December 6, 2007
Dow
nloaded from https://academ
ic.oup.com/hum
upd/article-abstract/14/2/179/609519 by guest on 18 Novem
ber 2018
Materials and Methods
The current literature on the roles of cytokines in reproductive biology
was searched on Entrez Pubmed (http://www.ncbi.nlm.nih.gov/sites/entrez) using combinations of the following keywords: cytokine, repro-
duction, fertility, null, knockout, mouse, immune, IL, transforming-
growth-factor, colony-stimulating-factor, tumour-necrosis-factor, leu-
kaemia-inhibitory-factor, activin, suppressors-of-cytokine-signalling,
hormone, ovary, testis, pregnancy, implantation, endometrium, decidua-
lization and parturition.
Introduction to the roles of cytokines in reproductive function
The following is a summary of the major roles of cytokines within
reproductive tissues. It is not intended to be a comprehensive review
of all cytokines involved in reproductive function, but aims to
provide a framework for understanding the most significant pathways
and processes currently recognized to be affected by cytokines.
Gonads
An array of cytokines are secreted by testicular cells. Through precise
spatial and temporal expression, these cytokines facilitate many inter-
actions within the testis, between Leydig cells, Sertoli cells and germ
cells, regulating immune privilege, steroidogenesis and spermatogen-
esis (Hedger and Meinhardt, 2003).
The blood testis barrier divides the seminiferous tubule epithelium
into a basal and adluminal compartment, and prevents direct contact
between post-meiotic male germ cells and the systemic immune
system (Xia et al., 2005). The blood testis barrier must physically
restructure to allow the migration of maturing spermatocytes, and
this is facilitated by TGF-b3 and TNF-a (Lui et al., 2001; Siu
et al., 2003). However, immune privilege in the testis is not just phys-
ical isolation of germ cells, but a more complex system involving tes-
ticular macrophages and anti-inflammatory cytokines produced by
non-immune cells in the testis (Fijak and Meinhardt, 2006).
The testis contains many immune cells. Macrophages comprise the
majority of leukocytes in the testis, with mast cells and T lymphocytes
also present (Wang et al., 1994). Testicular macrophages produce cyto-
kines including IL-1 and -6, TNF-a and granulocyte macrophage-CSF
(GM-CSF) (Kern et al., 1995). Pro-inflammatory cytokines, TNF-a and
IL-1 produced by activated macrophages have inhibitory effects on
Leydig cell testosterone synthesis (Calkins et al., 1990; Mauduit
et al., 1992; Xiong and Hales, 1993) and are believed to repress repro-
ductive behaviour in immune challenged rodents (Hales, 2002). Macro-
phages have also been shown to be critical for spermatocyte
development, presumably by the production of cytokines that act on
Leydig cells to promote testosterone synthesis, on Sertoli cells, and
on the gamete itself (Hedger, 2002).
Less is known about how immune privilege manifests in the ovary.
However, the mechanism is likely to involve an array of cytokines
mediating different forms of immune tolerance and suppression.
Many cytokines, including members of the TGF-b superfamily,
IL-1b and TNF-a, are an important part of cell–cell signalling in
the ovary, between the oocyte and its surrounding somatic cells in
the follicle, and in regulation of follicle survival and apoptosis
(Kaipia and Hsuen, 1997; Bornstein et al., 2004; Knight and Glister,
2006). Macrophages in the ovary also secrete cytokines, including
interferon-g (IFN-g), TNF-a and GM-CSF, which promote oocyte
development, the production of progesterone by the corpus luteum
and ovulation (Wu et al., 2004).
Endometrium
Menstruation is an inflammatory event involving the infiltration of
large populations of neutrophils and macrophages, the activation of
resident mast cells and up-regulation of matrix degrading proteases
(Finn, 1986). Inflammatory chemokines, including IL-8, fractalkine
and macrophage derived chemokine (MDC), are up-regulated in the
endometrium premenstrually (Jones et al., 1997; Hannan et al.,
2004; Jones et al., 2004). Leukocyte products (pro-inflammatory cyto-
kines IL-1, TNF-a, chemokines, prostaglandins and proteases) are
likely to be instrumental in the focal initiation and amplification of
endometrial inflammation and breakdown (Salamonsen and Lathbury,
2000). Re-epithelialization is complete within 48 h of the onset of
bleeding, and occurs in the absence of estrogen (Ferenczy, 1976).
To date, much of our understanding of endometrial repair is from
extrapolation of wound healing studies, and many repair-promoting
cytokines are expressed perimenstrually (activin A and B, TGF-b)
(Jones et al., 2006). The endometrium repairs without scarring,
similar to fetal wound healing, which has been postulated to involve
subtle alterations in cytokine expression (including ILs: IL-6, IL-8,
IL-10) (Salamonsen, 2003).
Decidualization is a necessary preparatory event for pregnancy and
involves the differentiation of endometrial stromal cells and wide-
spread tissue remodelling. An array of cytokines are produced and
secreted from decidualizing cells, and indeed a number of decidual
cell products facilitate decidualization of human endometrial
stromal cells in vitro. IL-11, G-CSF and activin A promote deciduali-
zation, whereas TNF-a, M-CSF and IL-1b inhibit this process (Inoue
et al., 1994; Dimitriadis et al., 2005).
Pregnancy
Embryo development and implantation. Pre-implantation deve-
lopment occurs in a cytokine and growth factor-rich fluid, secreted
by the epithelial cells of the Fallopian tube and uterus, many of
which promote embryo development in vitro (Sargent et al., 1998;
Sjoblom et al., 1999). Uterine gland products, including leukaemia
inhibitory factor (LIF) and TGF-b have continued roles in early pla-
cental development (Hempstock et al., 2004). Development of the pla-
cental villous structure occurs in conjunction with invasion of the
decidua by placental cytotrophoblast cells, which remodel decidual
arteries to establish utero-placental circulation. There is accumulating
evidence for critical interactions between immune cells, including
uterine natural killer (uNK) cells, and cytotrophoblasts during vascu-
lar remodelling (Leonard et al., 2006). Suboptimal placentation con-
tributes to pre-eclampsia, intrauterine growth retardation and
pregnancy failure. Cytotrophoblast invasion of the decidua is regu-
lated by a host of paracrine factors—including cytokines that act to
either promote (IL-1, IL-6, IL-15, activin) or limit invasive potential
(IL-10, TNF-a, TGF-b) (Bischof et al., 2000, Dimitriadis et al., 2005).
Maternal immune responses during pregnancy. Pregnancy poses a
significant challenge to the maternal immune system, which must
protect both mother and fetus from pathogenic assault, and yet
prevent immune-mediated rejection of the semi-allogeneic (geneti-
cally disparate) fetus. This has been the topic of much debate,
notably since the description of the fetus as an allograft by Medawar
(1953), which formulated the commonly held view of a generalized
suppression or alteration of the maternal immune system to enable tol-
erance to be acquired to the fetus. The field was advanced by Wegmann
in 1993, who proposed that pregnancy required a switch from an
inflammatory T helper cell 1 (Th1) immune profile (e.g. TNF-a,
IFN-g, IL-12, IL-2), towards a protective Th2 (e.g. IL-10, IL-4)
Ingman and Jones
180
Dow
nloaded from https://academ
ic.oup.com/hum
upd/article-abstract/14/2/179/609519 by guest on 18 Novem
ber 2018
profile (Wegmann et al., 1993). Despite some supporting evidence of a
Th2 bias in pregnancy, over a decade of research has failed to confirm
this simplistic hypothesis (reviewed by Chaouat et al., 2004). Indeed
there is strong evidence for both a local intrauterine and systemic
inflammatory response to pregnancy, involving increased immune
trafficking and cytokine production at the implantation site in mice
and humans (Robertson et al., 1997; Sharkey, 1998) and elevated
levels of circulating pro-inflammatory cytokines (IL-6, IL-12 and
TNF-a) in sera of pregnant women (Sargent et al., 2006).
Despite waning support for a global Th2 bias in pregnancy, evi-
dence for a role of T cells in maternal tolerance is gaining momentum.
Regulatory T cells, involved in suppression of T cell responses, are
found in increasing quantity in peripheral blood during pregnancy in
mice and humans (Aluvihare et al., 2004; Somerset et al., 2004) and
are necessary for allogeneic fetal survival (Aluvihare et al., 2004).
At the maternal–fetal interface, T cell responses are suppressed by
local secretion of indoleamine 2,3-dioxygenase, an enzyme that cata-
bolizes the amino acid tryptophan, necessary for T cell proliferation
(Munn et al., 1998). Again, this mechanism of T cell regulation is
necessary for allogeneic but not syngeneic fetal survival.
There is a growing body of evidence to suggest that the specialized
innate immune system within the uterus is particularly important for
pregnancy success (Sargent et al., 2006). At the maternal–fetal inter-
face, a number of placental and maternal strategies have been
identified that appear to contribute to maternal tolerance to the concep-
tus. Importantly, from the moment of implantation, the embryo is pro-
tected from direct contact with maternal cells by the placenta.
Placental cells evade immune surveillance by the maternal adaptive
immune system, either by lacking cell surface expression of MHC
class I antigens (villous syncytiotrophoblast) or by expression of a
unique repertoire of non-classical non-polymorphic human leukocyte
antigens (HLA), HLA-C, -E and -G (extravillous cytotrophoblast
cells) (Ellis, 1990; King et al., 2000). The latter cell type invades
the maternal decidua, and come into contact with the unique immuno-
logical environment comprising mainly uNK cells and macrophages
(King et al., 1998). Communication between uNK cells and HLA-G,
in particular, has been proposed to be a critical component of maternal
acceptance of the semi-allogeneic embryo (Moffett-King, 2002).
At the same time as averting immune-mediated rejection, the
maternal immune system must act to protect both the mother and
the developing fetus against pathogens (bacterial, viral, fungal).
Intrauterine infections (sexually transmitted and ascending vaginal
infections) are potentially harmful to mother and fetus, and are a
major cause of premature labour and maternal and neonatal morbidity.
During pregnancy, the primary uterine mucosal defences are those of
the innate immune system: NK cells, macrophages, dendritic cells and
neutrophils, with a major supporting role by endometrial luminal epi-
thelial cells (Wira et al., 2005). These secrete antimicrobials
[b-defensins, secretory leukocyte protease inhibitor (SLPI)] that
create a sterile environment (King et al., 2003), express Toll-like
receptors (TLRs) that recognize foreign antigens, and are primed to
secrete an array of cytokines (e.g. IL-6, TNF-a, GM-CSF) and chemo-
kines (e.g. IL-8, MCP-1), which serve to activate cells of the innate
immune system (Wira et al., 2005).
There has been recent speculation concerning a more widespread
role for the innate immune system, specifically for peripheral NK
cells in pregnancy-specific immune responses. NK cells have been
demonstrated to exhibit similar cytokine profiles to Th cells, with a
proposed shift towards Th2-type responses in pregnancy (Sargent
et al., 2006). This intriguing development suggests that despite
growing knowledge of the complex cytokine networks at the
maternal–fetal interfaces of pregnancy, we are far from understanding
the intricate immune system balance required for pregnancy success.
Parturition. Myometrial quiescence and cervical competence are
essential for maintaining pregnancy, and a number of anti-
inflammatory mechanisms [e.g. progesterone suppression of inflam-
matory cytokines and mediators (Kelly, 1994)] are in place to
prevent premature labour. These cytokine networks are reversed at
term, allowing activation of a cascade of events inducing cervical
ripening, myometrial contraction and parturition. Intrauterine bac-
terial infection is a major cause of preterm labour, stimulating local
(fetal membrane and decidual) and systemic expression and release
of pro-inflammatory cytokines (e.g. IL-1 and TNF-a) (Park et al.,
2005). There is evidence however for inherent differences in suscep-
tibility of individuals to infection during pregnancy. A delicate
balance between hypo- and hyper-responsiveness in the immune
response to infection has been hypothesized (Simhan et al., 2003).
Women with low capacity to respond to vaginal infection through
the production of pro-inflammatory cytokines, IL-1b, IL-6 and IL-8,
might have a more permissive environment for pathogens to flourish,
and are at risk of ascending uterine infection and chorioamnionitis
(Simhan et al., 2003). However, hyper-responsiveness to vaginal
infection is suggested to also be detrimental to pregnancy, and elev-
ated levels of IL-6 have been found to be a predictor of preterm
labour (Wenstrom et al., 1998; Goepfert et al., 2001).
Gene knockout technology
Conventionally, gene targeting experiments are performed in mice, by
deletion of a specific gene in embryonic stem cells using homologous
recombination. This approach has been widely used to ‘knockout’
genes encoding cytokine ligands, specific receptors and constituents
of signalling pathways. Random gene ablation technology can also
be used to discover genes important in reproductive function.
Random disruptions to the genome are created by a process known
as gene trapping, whereby a reporter gene is inserted into the
genome, randomly disrupting genes, or by chemical-based mutagen-
esis using N-ethyl-N-nitrosourea. The resultant offspring are analysed
for interesting phenotypes, and the expression pattern of the disrupted
gene can be discovered by tracking reporter gene expression.
Use of cytokine knockout mice to study roles in reproduction
A wealth of information on the roles of cytokines in reproductive
biology has been revealed through study of knockout mice (Table I).
A useful framework for studying reproductive function in mouse
models is provided in Fig. 1. These studies have revealed essential
roles for cytokines in regulation of the hypothalamo-pituitary-gonadal
(HPG) axis of both male and female mice, as well as production of
oocytes from the ovary (Fig. 2). Roles of cytokines in pregnancy,
including embryo development, endometrial tissue remodelling and
implantation, placental development, protection of the conceptus
from external pathogens, and the timing of parturition have also been
confirmed and explored using this technology (Fig. 3).
TGF-b superfamily
Transforming growth factor-b. The three mammalian isoforms of
TGF-b (1, 2 and 3) influence a wide variety of biological functions,
including cell viability, apoptosis, proliferation, differentiation,
adhesion and migration. The three isoforms are encoded by individual
genes; however, they share many overlapping functions, and bind
the same receptor complex. The synthesis of TGF-b ligand and its
Cytokine knockouts in reproduction
181
Dow
nloaded from https://academ
ic.oup.com/hum
upd/article-abstract/14/2/179/609519 by guest on 18 Novem
ber 2018
receptors appears to be almost ubiquitous in reproductive tissues and
consequently TGF-b has many diverse roles in reproductive biology
(Ingman and Robertson, 2002).
Before the application of knockout technology, it was widely
believed that the three isoforms were functionally redundant.
However, the discovery that knockouts for the different TGF-b iso-
forms display quite distinct phenotypes clearly showed that each
isoform is critical to normal development, and serve different
roles. Without TGF-b1, 50% of embryos do not survive past the
pre-implantation stage (Kallapur et al., 1999), or have defective
yolk sac vasculogenesis and hematopeioesis (Dickson et al.,
1995). The embryos which do survive probably do so because of
the presence of TGF-b2 and 3, as 100% of mice carrying a domi-
nant negative receptor transgene, whereby none of the TGF-b iso-
forms can signal transduce, die in utero (Goumans et al., 1999).
Mice deficient in TGF-b2 and 3 do not survive longer than 24 h
after birth due to defects in cardiovascular development (Sanford
et al., 1997), and lung and palate development (Proetzel et al.,
1995), respectively.
Approximately half of TGF-b1 knockout mice develop normally
in utero, but die from multi-organ inflammatory lesions around the
age of weaning (Shull et al., 1992; Kulkarni et al., 1993), preventing
analysis of adult reproductive function. This can be overcome by ren-
dering the mice immunocompromized, which prevents inflammatory
pathology and increases the lifespan of TGF-b1 knockout mice
(Diebold et al., 1995; Letterio et al., 1996; Kobayashi et al., 1999).
Using TGF-b1 knockout mice on a severe combined immunodefi-
ciency background, TGF-b1 was shown to be critical for both male
and female fertility. Female mice deficient in TGF-b1 have impaired
ovarian function, due to dysfunctional LH secretion and a requirement
for TGF-b1 in the ovarian follicle, leading to disrupted estrous cycles,
reduced number of ovulated oocytes and reduced developmental com-
petence of these oocytes (Ingman et al., 2006). The result of these
lesions is greatly reduced fertility, however, some TGF-b1 knockout
female mice are able to carry pregnancies to term and deliver
overtly normal pups.
Deficiency of TGF-b1 in male mice also causes severe infertility
(Ingman and Robertson, 2007). Reduced LH secretion leads to
Table I. Summary of fertility phenotypes in cytokine ablated mice
Cytokine Method of gene ablation Reproductive phenotype Reference
TGF-b TGF-b1 null mutation 50% embryo lethality, impaired ovarian function, reduced
oocyte competence (female), sexual dysfunction (male),
dysfunctional LH secretion (male and female)
Kallapur et al. (1999), Ingman et al. (2006) and
Ingman and Robertson (2007)
TGF-b2 null mutation Peri-natal lethality Sanford et al. (1997)
TGF-b3 null mutation Peri-natal lethality Proetzel et al. (1995)
TGF-b dominant negative
receptor targeted to prostate
Increased number of prostate cells and decreased apoptosis Kundu et al. (2000)
Activin Activin bA null mutation Neonatal lethality Matzuk et al. (1995b)
Activin bB null mutation Increased gestation length, peri-natal lethality of offspring Vassalli et al. (1994)
bB ‘knockin’ in bA null
mouse
Viable, reduced female fertility, delayed male fertility Brown et al. (2000)
ActRII Female infertility, thin uteri and small ovaries, normal
folliculogenesis but no ovulation
Matzuk et al. (1995a)
CSF-1 Null mutation Dysfunctional LH secretion, reduced antral follicles and
ovulation, reduced immune protection of conceptus against
Listeria infection
Pollard et al. (1994), Araki et al. (1996), Cohen
et al. (1996, 1997a,b), Gouon Evans et al.
(2000) and Guleria et al. (2000)
GM-CSF Null mutation Reduced litter size, reduced fetal and neonate weight,
impaired ovarian luteinization, reduced blastocyst cell
number
Robertson et al. (1999, 2001) and Jasper et al.
(2000)
LIF Null mutation Implantation failure Stewart et al. (1992)
Injection of antisense
oligonucleotides to 2-cell
embryos
Decreased rate of blastocyst development Cheng et al. (2004)
LIF receptor null mutation Peri-natal lethality Ware et al. (1995)
IL-1 IL-1 receptor null mutation Reduced litter size in females, no male fertility defect
detected
Abbondanzo et al. (1996) and Cohen et al.
(1998)
IL-5 IL-5 null mutation Reduced placental weight Robertson et al. (2000)
IL-10 IL-10 null mutation Increased size of placental labyrinth Roberts et al. (2003)
Increased sensitivity to LPS Murphy et al. (2005) and Robertson et al.
(2006)
IL-11 IL-11 receptor null mutation Impaired decidualization Robb et al. (1998)
IL-15 Null mutation Absence of uNK cells at implantation sites, no fertility
defect detected
Ashkar et al. (2003)
IL-4, IL-5,
IL-9, IL-13
Quadruple null mutation No fertility defect detected Fallon et al. (2002)
TNFa Null mutation No fertility defect detected Taniguchi et al. (1997)
TNF-RI null mutation Increased prepubertal ovarian responsiveness to
gonadotropins, impaired ovarian cycling in aged females,
no male fertility defect detected
Roby et al. (1999)
TNF-RII null mutation No fertility defect detected Roby et al. (1999)
Ingman and Jones
182
Dow
nloaded from https://academ
ic.oup.com/hum
upd/article-abstract/14/2/179/609519 by guest on 18 Novem
ber 2018
diminished testosterone output by the testes. The null mutation also
leads to sexual dysfunction, which appears to be independent of testos-
terone secretion. Male mice show outward signs of sexual interest in
receptive female mice, however are incapable of performing intromis-
sion, possibly due to erectile dysfunction. TGF-b has also been shown
to be a regulator of prostate epithelial cells, as a dominant negative
receptor targeted to the prostate had increased number of cells and
reduced apoptosis (Kundu et al., 2000).
Activin. Activins (A, B and AB) are closely related to TGF-bs, and
signal through a common pathway (via Smads). Activins act as
locally produced paracrine factors in the ovary, uterus, placenta and
testis. In females, activins are associated with ovarian biology
(Findlay, 1993), placental development and function (Qu and
Thomas, 1995; Caniggia et al., 1997) and endometrial decidualization
(Jones et al., 2002; Tierney and Giudice, 2004).
Attempts to decipher the exact reproductive roles of activins have
been confounded by the existence of multiple highly homologous
related subunits, which form closely related dimers. Ablation of
activin bA results in neonatal lethality due to major malformation
of the jaw, preventing feeding (Matzuk et al., 1995b), whereas mice
lacking activin bB subunit survive to adulthood with minor defects,
including prolonged gestation and impaired lactation (Vassalli et al.,
1994). Double knockouts have additive effects (Matzuk et al.,
1995b). The apparent distinct functions for bA and bB have been
attributed to differences in spatio-temporal expression or affinity for
the common receptors. This was substantiated by the rescue of the
neonatal lethality in bA deficient mice by ‘knocking-in’ the bB
gene into the bA locus in the knockout mice, such that bB expression
Figure 1: Framework for studies on reproductive function in mouse models
Knockout mice mated together are compared to wild-type breeders of the same genetic background. The flowchart enquires whether the parameters of fertility are the
same between knockout and control breeders. If the answer is yes for all parameters, the knockout mice have normal fertility, however does not preclude a role for the
cytokine in allogeneic matings or under pathogenic challenge. To evaluate reproductive function in male and female knockout mice caged together, the female is
checked each morning for a mating plug to indicate a mating event. To determine the success of the mating event, the females are sacrificed during late pregnancy
(e.g. day 17.5 post-coitus) and the uterus dissected. Number of implantation sites and resorptions are quantified, and fetal and placental weights can be determined to
evaluate placental function. Pups are monitored post-partum for developmental and lactational defects. If these fertility parameters deviate from wild-type controls,
knockout mice can be mated with wild-type mice to evaluate whether the defect is due to male or female reproductive disorders, or a fetal requirement of the cyto-
kine. Male knockout mice mated with wild-type controls may have defects in hormone synthesis, sexual behaviour dysfunction, spermatogenesis or altered seminal
plasma content causing downstream effects on embryo quality or maternal immune response to pregnancy (Johansson et al., 2004). Female knockout mice may have
defects in hormone synthesis, oocyte development, ovulation, implantation, placental development, parturition or lactation
Cytokine knockouts in reproduction
183
Dow
nloaded from https://academ
ic.oup.com/hum
upd/article-abstract/14/2/179/609519 by guest on 18 Novem
ber 2018
is driven by the bA promoter (Brown et al., 2000). These mice survive
to early adulthood, but with reproductive abnormalities, including
hypogonadism, delayed male onset fertility, defective folliculogenesis
and reduced fertility in females. This demonstrates a unique approach
in a multigene family where redundancy can be used as an advantage
to rescue lethality, enabling investigation of more subtle differences in
function of family members. Components of the signalling pathway,
cell surface receptors (Alk-4, ActRIIB) and intracellular mediators
Smads (2, 3 and 4) have been knocked out—generally resulting in
embryonic lethality (reviewed in Ethier and Findlay, 2001; Jones
et al., 2006). An exception is the activin type II receptor (ActRII);
the null females are viable but infertile due to a block in folliculogen-
esis and suppressed FSH levels (Matzuk et al., 1995a).
Colony-stimulating factors
Colony-stimulating factor-1. A spontaneously occurring genetic
mutation which caused severe osteopetrosis led to the discovery of
CSF-1 as a critical cytokine in osteoclast and macrophage develop-
ment (Wiktor-Jedrzejczak et al., 1990; Yoshida et al., 1990), and in
reproductive function in both male and female mice (Pollard, 1997).
CSF-1 regulates mononuclear phagocyte and macrophage growth, via-
bility and differentiation, and is a major chemoattractant for these cells
(Tushinski et al., 1982; Webb et al., 1996). Studies using these mice
have been instrumental in defining critical roles for CSF-1, and
CSF-1-regulated macrophages, in reproductive function.
Male mice lacking CSF-1 have severe infertility, caused primarily
by defective HPG axis signalling. Levels of LH are low due to
defects in the hypothalamus, which leads to low serum testosterone
and a cascade of defects including low sperm number and reduced
libido (Cohen et al., 1996; Cohen et al., 1997a; Cohen et al., 2002).
Targeted disruption of the CSF-1 receptor revealed a similar pheno-
type (Dai et al., 2002).
Female mice lacking CSF-1 also exhibit defects in the HPG axis as
well as other defects that compromise reproductive function. The
number of antral follicles and oocytes ovulated in CSF-1 deficient
mice is reduced, and CSF-1 deficient mice are unresponsive to
exogenous gonadotropins, suggesting that CSF-1 acts locally in the
ovary to promote ovulation (Araki et al., 1996; Cohen et al.,
1997b). Once established, pregnancy proceeds normally in CSF-1
deficient mice, however, they have impaired innate immune defences
when challenged with Listeria infection, revealing a further role for
CSF-1 in protection of the conceptus against infection (Guleria and
Pollard, 2000).
Granulocyte macrophage-CSF. The survival, proliferation, differen-
tiation and activation of cells of the myeloid leukocyte lineage (i.e.
monocytes, macrophages and granulocytes) are regulated by
GM-CSF. This cytokine has important functions in establishment of
maternal tolerance to fetal-placental tissues, and promotion of early
pregnancy events (Robertson, 2007). The ovaries of female
GM-CSF knockout mice contain normal numbers of leukocytes
however the activation state of these leukocytes is altered, with
increased nitric oxide production and reduced expression of class II
major histocompatibility complex on ovarian macrophages and den-
dritic cells (Jasper et al., 2000). The mice have a modest increase in
estrous cycle length and normal ovulation rate, however luteinization
following ovulation is compromised, with lower progesterone output
during early pregnancy, potentially caused by altered leukocyte
activation.
Litter size at weaning was found to be reduced in GM-CSF
knockout breeding pairs, due to late gestational and peri-natal loss
(Robertson et al., 1999). Comparisons between fetal and pup
weights between offspring of GM-CSF knockout females mated
with GM-CSF knockout or heterozygote males revealed that both
maternal and fetal sources of GM-CSF are important for normal fetal
development, and that defects are likely to be due to insufficiencies in
placental growth (Robertson et al., 1999). This may be due to impaired
pre-implantation embryo development, as GM-CSF deficient blasto-
cysts have reduced total cell number, which can be rescued by culturing
embryos in the presence of GM-CSF (Robertson et al., 2001).
Figure 2: Cytokines identified as important in functioning of the
hypothalamo-pituitary-gonadal axis, and the production of mature gametes,
as determined by knockout mouse studies
CSF-1, colony-stimulating factor
Figure 3: Cytokines identified as important in pregnancy, as determined by
knockout mouse studies
Ingman and Jones
184
Dow
nloaded from https://academ
ic.oup.com/hum
upd/article-abstract/14/2/179/609519 by guest on 18 Novem
ber 2018
Leukaemia inhibitory factor. LIF is one of the few genes whose
expression is obligatory for embryo implantation. Knockout of LIF
results in total female infertility, caused by failure of the blastocyst
to attach to the endometrial epithelium (Stewart et al., 1992). In the
mouse uterus, LIF is produced by the luminal epithelium and epithelial
glands, and is up-regulated from the time of ovulation until implan-
tation (Bhatt et al., 1991). Both luminal epithelium and blastocysts
express LIF receptors, suggesting LIF is required for maternal recep-
tivity and/or embryo development (Chen et al., 1999; Ni et al., 2002).
Indeed, there is evidence that the epithelial development and gene
expression is defective in LIF knockout mice (Sherwin et al., 2004;
Fouladi-Nashta et al., 2005). Furthermore, immune cell populations
are markedly altered in the peri-implantation phase in LIF knockout
mice, suggesting that LIF has roles, directly or indirectly, in regulating
leukocyte trafficking or distribution in the uterus (Schofield and
Kimber, 2005). Pregnancy can be rescued in the LIF null mouse by
LIF injection on the day of implantation and LIF deficient embryos
implant successfully in a wild-type recipient female, indicating that
maternal, but not embryonic, LIF production is essential in the peri-
implantation period (Stewart et al., 1992). However, reduction of
LIF expression in 2-cell embryos by microinjection of antisense oligo-
nucleotides, decreases rate of development to the blastocyst stage
(Cheng et al., 2004), suggesting that LIF promotes, although is not
essential for, pre-implantation embryo development.
Interestingly, a more severe phenotype is observed when the
specific LIF receptor is knocked out; null progeny have widespread
abnormalities including severely abnormal placentae, and die shortly
after birth (Ware et al., 1995). This suggests that the LIF receptor
binds additional ligands.
ILs and their receptors
Interleukin-10. Trophoblast invasion of maternal decidua and placen-
tal development is a tightly regulated event involving intricate cross-
talk between fetal and maternal cells, to promote yet regulate invasion.
In vitro studies have implicated roles for a number of cytokines, but
verification in vivo has been troublesome. An exception is IL-10, a
Th2 cytokine proposed to be a major player in suppressing endo-
metrial inflammatory responses during pregnancy. Although no per-
turbation of uterine immunological environment was observed in
IL-10 knockouts housed in specific pathogen free conditions (White
et al., 2004), differences in the structure and function of the placenta
were observed (Roberts et al., 2003). Placentae showed a relative
expansion of the labyrinth zone, leading to increased efficiency in
nutrient transfer to the developing fetus.
However, the primary role of IL-10 in pregnancy appears to be pro-
tection of the developing fetus against pathogens. IL-10 knockout
mice have a greater susceptibility to some pregnancy pathologies
when challenged with an infection. Lipopolysaccaride (LPS) causes
fetal resorption (Murphy et al., 2005) and preterm labour (Robertson
et al., 2006) when given in low doses to pregnant IL-10 knockout mice
during early or late gestation, respectively. No effect on pregnancy
was observed when wild-type mice were given the same dose. IL-10
provides this protection via uNK cells and possibly macrophages,
through inhibition of inflammatory cytokines including TNFa,
IFN-g and IL-6 (Murphy et al., 2005; Robertson et al., 2007).
Interleukin-11. IL-11 is a cytokine closely related to LIF and is
expressed in abundance in the mouse decidua after implantation. As
with LIF knock out mice, IL-11 receptor (IL-11R)-null progeny are
viable and essentially normal, except female mice are infertile
(Robb et al., 1998). In this case, blastocyst implantation occurs and
is followed by the primary decidual response in the anti-mesometrial
subepithelial stroma. Between days 5 and 6 of pregnancy marked
differences are evident between wild-type and knockout implantation
sites: although the decidualization response progresses in wild-type
mice with the formation of the secondary and mesometrial decidual
zone, decidualization is stalled in the IL-11R knockout. Secondary
decidual formation is dramatically retarded, eventually leading to
overgrowth of embryonic trophoblast tissues and implantation
failure. A number of genes encoding extracellular matrix components
are differentially expressed between wild-type and IL-11 knockout
mice early in the decidualization response, suggesting that primary
decidualization is defective (White et al., 2004).
Interleukin-15. Uterine NK cells constitute a major component of the
decidua in both humans and mice. A number of decidual-derived che-
mokines [macrophage inflammatory protein-1b (MIP-1b), secondary
lymphoid chemokine (SLC), monocyte chemoattractant protein-3
(MCP-3)] (Jones et al., 2004) are able to stimulate migration of
human uNK cell precursors in vitro (Campbell et al., 2001), although
knockout of murine homologues fails to identify any requirement for a
single chemokine for their endometrial recruitment (Chantakru et al.,
2002). In contrast, knockout of the cytokine IL-15, a cytokine critical
for lymphoid NK cell differentiation, demonstrates an essential role in
uNK cell differentiation (Ashkar et al., 2003). In its absence, uNK
cells are completely ablated from implantation sites, and decidual
integrity and vascular development are compromised. Transplantation
studies verified that autocrine production of IL-15 is not required, and
that peripheral NK cells require a uterine signal. However, despite
their abundance and apparent roles in decidual formation, absence
of NK cells does not adversely affect pregnancy in mice, although a
small but significant reduction in birthweight was observed,
suggesting a suboptimal uterine environment (Barber and Pollard,
2003). The only other cytokine definitively shown to be critical in
recruitment of immune cells to the endometrium is eotaxin; eosino-
phils are absent from the uterus in the eotaxin knockout mouse, but
again no uterine functional deficiency was found (Gouon-Evans and
Pollard, 2001).
Multiple IL ligand knockouts. In addition to IL-10, a number of other
Th2 cytokines have been selectively mutated to explore the role for
Th2 cytokines at the maternal–fetal interface. Individually these
knockouts have no reproductive phenotype, potentially due to a high
degree of compensatory pathways that protect fertility. To verify
this, quadruple knockouts of Th2 cytokine genes were generated,
through intercrossing of double, then triple knockout F2 mice
(Fallon et al., 2002). Despite lacking IL-4, IL-5, IL-9 and IL-13
genes and having compromised Th2 responses as determined
through parasitic challenge, female mice reproduced successfully.
Litter sizes were not affected when female quadruple knockouts
were crossed with either males of the same background strain
(known as a syngeneic mating) or of a different background strain
(known as an allogeneic mating). This suggests that allogeneic preg-
nancy is not dependent on a maternal Th2 bias. However, there may
be more subtle effects which are dependent on the specific allogeneic
cross. An IL-5 only knockout on a C57/Bl6 background had a 9%
reduction in placenta weight in late pregnancy when mated with a
wild-type CBA male, but no reduction when mated with a C57/BL6
or BalbC male (Robertson et al., 2000).
Additionally, the quadruple knockout females are likely to be
unable to respond appropriately to an immunological or inflammatory
challenge during pregnancy, which may lead to increased fetal loss or
preterm labour as observed in IL-10 knockout mice (Murphy et al.,
2005; Robertson et al., 2006).
Cytokine knockouts in reproduction
185
Dow
nloaded from https://academ
ic.oup.com/hum
upd/article-abstract/14/2/179/609519 by guest on 18 Novem
ber 2018
IL receptors. Despite its abundance in reproductive tract tissues, no
deficiency in fertility was observed with genetic knockout of the IL-1
receptor (type I) in male (Cohen and Pollard, 1998) or female mice
(Abbondanzo et al., 1996), apart from slightly reduced litter sizes in
female IL-1R knockout mice. This is in contrast to the drastically
reduced implantation rate when IL-1 biological activity was blocked
by administration of IL-1 receptor antagonist in the peri-implantation
period (Simon et al., 1994). The absence of a placental phenotype is
undoubtedly related to redundancy in this system, as many other cyto-
kines and growth factors could compensate for lack of IL-1 in stimulat-
ing trophoblast migration (e.g. activin, IL-6, IL-15) (Caniggia et al.,
1997; Zygmunt et al., 1998; Meisser et al., 1999).
The problem of redundancy among family members can be over-
come by targeting a common receptor. The IL-2 receptor g chain is
shared amongst IL-2, -4, -7, -9 and -15. Ovarian cycle regularity is
affected in mutants lacking IL-2Rg (Miyazaki et al., 2002),
however, they are fertile and no abnormalities were noted throughout
pregnancy, with the exception of an absence of uNK cells and associ-
ated decidual phenotype, later verified to be due to lack of IL-15 action.
Tumour necrosis factor-a
Similar discrepancies in phenotype to IL-1 and its receptor exist
between knockouts of TNF-a ligand and its receptor. Although no
reproductive abnormalities were reported in the absence of the
ligand (Taniguchi et al., 1997), detailed analyses of reproductive per-
formance in female TNF-RI null mice identified disruption of the
estrus cycle, leading to premature fertility loss, and increased sensi-
tivity of the prepubertal ovary to gonadotropin stimuli (Roby et al.,
1999). These effects were not seen in TNF-RII null females, thus
demonstrating divergent roles for the receptor isoforms in ovarian
function and steroidogenesis.
Manipulation of negative regulators of cytokine action
In addition to gene ablation of ligand, receptor or signalling com-
ponents, cytokine action can be explored by knockout or overexpres-
sion of genes encoding negative regulators. This is a useful approach
for the initial understanding of a gene family involvement in a process,
bypassing redundancy issues. Suppressors of cytokine signalling
(SOCS) are a family of negative regulators of cytokine signalling.
Their expression is acutely stimulated at the transcription level by
ligand-induced activation of the cytokine pathways, producing a self-
regulated fine-tuning mechanism for cytokine action. There are eight
SOCS family members, each differing in terms of cytokine pathway
specificity, and functioning in a variety of ways, e.g. through targeting
of Janus Kinase (JAK) proteins for degradation (Alexander, 2002;
Larsen and Ropke, 2002). SOCS3 is up-regulated in response to
IL-11, LIF and IL-6, amongst others factors. SOCS3 null homozyo-
gotes die in midgestation (E11.5–12.5), due to defective placental for-
mation resulting in increased numbers of invasive giant trophoblast
cells and reduced spongiotrophoblast (Takahashi et al., 2003). As
LIF promotes the differentiation of trophoblast giant cells, a double
knockout of LIFR and SOCS3 was created, and this rescued the
SOCS3 null phenotype (Takahashi et al., 2003). Thus suppression
of LIF signalling is clearly fundamental for normal placental develop-
ment. However, many cytokines and growth factors, including IL-10
which is known to be important for placental development, are also
regulated by SOCS3, and thus may be involved.
Cytokine action can also be inhibited by the overexpression of
endogenous binding proteins or negative regulators. Lefty is a
member of the TGF-b superfamily, and can bind to TGF-b type II
receptors and inhibit the action of TGF-b (and possibly other
ligands) (Ulloa and Tabibzade, 2001). Lefty A was overexpressed
locally in the mouse uterus by retroviral transfection (Tang et al.,
2005), resulting in implantation failure. Similarly, follistatin, a nega-
tive binding protein for activin, has been transgenically overexpressed
(Guo et al., 1998), creating a functional activin knockout. This results
in reproductive abnormalities, including decreased testis size and
defective Leydig cell function and spermatogenesis in males, while
females have small ovaries and become infertile with advancing
age. The neonatal lethality of the follistatin knockout (Matzuk et al.,
1995c) was overcome by a conditional knockout to examine ovarian
phenotype (Jorgez et al., 2004). The granulosa cell specific promoter
Amhr2 driving cre recombinase expression was used to locally inacti-
vate the follistatin gene, disrupting fertility through reducing numbers
of ovarian follicles, fertilization defects and elevated gonadotropins,
eventually leading to infertility. This phenotype is strikingly similar
to symptoms of premature ovarian failure, supporting a role for
tightly regulated activin bioactivity in maintaining normal ovarian
function. A somewhat different phenotype was seen when activin
action was enhanced by knockout of the inhibin a subunit (Draper
et al., 1998), thus removing the competitive antagonism of activin
action. Mice lacking inhibin were normal at birth, but both males
and females developed severe gonadal tumours (reviewed in Chang
et al., 2001).
Caveats associated with the use of knockout technology
Despite the large amount of knowledge that can be gained by knockout
studies, there are some important considerations to be taken when
investigating the role of cytokines in reproductive biology using this
approach.
Redundancy
Depletion of a single cytokine may not have any noticeable effect,
despite other evidence suggesting it to have an important role in the
reproductive process. As most cytokines work together to create a
cytokine milieu that will overall influence how a cell will grow and
differentiate, the loss of one cytokine can, in some circumstances, be
made up for by altered production of others. This problem can be over-
come to some extent by the use of multiple knockouts. In situations
where multiple ligands bind the same receptor, a dominant negative
receptor approach can knockout the effects of all ligands at once.
Lethality
Cytokines play many roles during the life of an organism, during
development, in haematopoiesis, in the maintenance of homeostasis
and health, and the initiation of cell death or differentiation. Whole
organism knockout of a particular cytokine can have dramatic conse-
quences on specific organ development, to the extent that the organism
does not function, in which case we see a lethality phenotype. Lethal-
ity phenotypes severely compromise the ability to analyse reproduc-
tive function. However, some analysis can still be achieved by
transplanting the organ of interest into a healthy wild-type host.
Delineating systemic versus local effects
Another potential problem is developmental defects that cause sec-
ondary effects on reproductive function. The reproductive system is
particularly susceptible to secondary effects of cytokine depletion
because of its large dependence on sex hormone production.
Hormone levels are regulated at the level of the hypothalamus, pitu-
itary and gonads by positive and negative feedback mechanisms,
and can be influenced by a large array of other endocrine factors
Ingman and Jones
186
Dow
nloaded from https://academ
ic.oup.com/hum
upd/article-abstract/14/2/179/609519 by guest on 18 Novem
ber 2018
including leptin, cortisol and thyroid hormone. Therefore, altered
function in a wide variety of organs can lead to imbalances in sex hor-
mones leading to infertility. Therefore, when a fertility phenotype is
observed in a knockout animal, the question must be asked: is it
caused by a primary effect on the reproductive organ of interest, or
as a secondary consequence of altered functioning of another tissue?
The general approach to this question in the past has been to
analyse sex hormone production and ovarian cyclicity, and conduct
organ transplant studies or hormone replacement studies.
Subtle effects
As we begin to understand more about the importance of fetal growth
for adult health, subtle effects on fetal development take on greater
significance. Even small changes in nutrient delivery to the fetus
can program post-natal and adult metabolic status and lead to
increased susceptibility to a range of adult onset disease, including
stroke, hypertension and non-insulin dependent diabetes (reviewed
by McMillen and Robinson, 2005; Barker, 2006). Although a particu-
lar cytokine may not be necessary for pregnancy to proceed or for
normal litter sizes, it is important to consider more subtle effects on
fetal and placental development, and growth of the neonate, to under-
stand the full impact of gene ablation on reproductive physiology.
Specific pathogen free conditions
Laboratory mice in research institutions are generally housed in a
specific pathogen free environment, and are therefore not challenged
with the array of pathogens most mice and humans are exposed to.
Disrupting cytokine networks may lead to reduced capacity to
respond to these challenges, which would not be observed under
normal laboratory conditions.
Strategies to improve the use of gene knockout technology
A number of alternative strategies can be used in place of, or to comp-
lement, conventional gene knockout that will overcome problems of
embryonic lethality, complications through actions in other organs,
and developmental defects caused by the long-term absence of a gene.
Regulatable gene ablation
Ablation of a gene can be spatially or temporally regulated, so that
only the organ of interest is affected, or the mutation is induced
only at a particular stage of development. A useful method is the
cre/lox system, whereby 2 loxP sites are inserted into the gene, flank-
ing a critical section (Gu et al., 1994). Gene expression occurs nor-
mally unless the cre recombinase protein is co-expressed, as cre
inverts the genetic material between the two sites back to front, effec-
tively disabling the gene. Cre expression is then controlled by a trans-
gene, whereby expression is limited by a cell lineage specific
promoter, or is inducible, e.g. by using a promoter regulated by tetra-
cycline or RU486. Alternatively, a dominant negative receptor
approach can be used, whereby the dominant negative receptor is
regulated by a cell lineage specific or inducible promoter. The use
of conditional knockouts in reproductive biology research is limited
by the availability of cell or tissue specific promoters. Lack of a tissue-
specific promotor can be overcome by using adenovirus expressing
cre-recombinase (Baba et al., 2005) and targeted delivery (e.g. injec-
tion into the mouse uterine lumen; Beauparlant et al., 2004).
Temporary gene ablation: gene knock downs
Genes can be transiently down-regulated or ‘knocked down’ by tar-
geted blockade of gene expression, either systemically or within a
specific tissue. Transient knockout approaches offer a more rapid
and more economical method to examine the specific actions of a
gene product at a particular time point.
One approach is to use antisense technology, the efficacy of which
has been improved dramatically by the development of morpholino
antisense oligonucletides (Summerton and Weller, 1997). These are
short (around 25 bp) modified DNA sequences complementary to
the mRNA of the gene of interest, designed to span the 50UTR and
the beginning of the coding sequence. Binding of the antisense pre-
vents translation initiation by blocking ribosomal interaction with
the ATG start codon. Alternatively, morpholino oligonucleotides
can target splice junctions and modify pre-RNA, resulting in abnormal
protein synthesis. Introduction of the oligonucleotides into cells is
achieved through microinjection, cell scraping or by specialized
delivery reagents such as EPEI (ethoxylated polyethylenimine) or
Endo-porter, that aid endocytotic uptake (Morcos, 2001). Temporary
knockdown of genes has been widely achieved by microinjection
into embryos, and recently in cell culture systems (Heasman, 2002).
An alternative, more recently developed technology for gene
knockdown involves the use of short interfering RNA (siRNA)
(reviewed by Hannon, 2002). This approach exploits evolutionarily
conserved cellular mechanisms to protect against parasitic and
pathogenic nucleic acids. Small lengths (19–22 bp) of double
stranded RNA designed against the gene of interest are introduced
to the cell where they interact with intracellular machinery to
form RNA-induced silencing complexes (RISCs). These unwind the
siRNA strands, enabling specific binding to the complementary
mRNA sequence, followed by cleavage and destruction of the now
doubled stranded mRNA, thus preventing subsequent translation.
How will cytokine knockout mice improve reproductive medicine?
Gene ablation studies in mice can provide clues about the role of cyto-
kines in human reproduction. Some genes associated with human
reproduction have been identified using screening methods such as
gene arrays, to be differentially expressed between fertile and infertile
individuals or populations. However, the absolute requirement
of a particular gene for fertility is virtually impossible to verify in
humans, and enormous heterogeneity between individuals and the
multifactorial nature of infertility add further complexity. Thus appli-
cation of findings from gene ablation studies in mice can be invaluable
in shedding light upon reproductive processes in humans, in combi-
nation with comparison of expression patterns in human tissues and
in vitro functional studies.
Improving in vitro fertilization
Maternal-derived GM-CSF was shown by mouse knockout studies to
improve pre-implantation embryo development, fetal and placental
growth, and pup weight and survival (Robertson et al., 1999, 2001).
During in vitro fertilization and culture for infertile couples, this cyto-
kine is absent from traditional culture media. Culturing normal mouse
embryos in the presence of exogenous GM-CSF has been found to
improve fetal and post-natal growth, with growth trajectories closer
to in vivo developed offspring compared with embryos cultured
in vitro in the absence of GM-CSF (Sjoblom et al., 2005). Importantly,
human cultured embryos also benefit from exogenous GM-CSF, with
improved development to blastocyst stage, increased hatching, and
increased inner cell mass due to reduced apoptosis (Sjoblom et al.,
1999, 2002). This cytokine may become a routine additive to in
vitro fertilization culture protocols in the future.
Cytokine knockouts in reproduction
187
Dow
nloaded from https://academ
ic.oup.com/hum
upd/article-abstract/14/2/179/609519 by guest on 18 Novem
ber 2018
Improving oocyte development
Successful pregnancy following IVF positively correlates with intrafol-
licular content of TGF-b1 at the time of oocyte retrieval (Fried and
Wramsby, 1998). In the complete absence of TGF-b1, infertility was
found to be caused by oocyte incompetence leading to early embryo
arrest in knockout mice (Ingman et al., 2006). This suggests that some
human infertility or subfertility may result from impaired oocyte devel-
opment secondary to low levels of ovarian TGF-b1. We can now inves-
tigate interventions to increase intrafollicular levels of TGF-b1 using
this mouse model, including dietary supplementation, as orally delivered
TGF-b1 is known to cross the gastro-intestinal wall and rescue the
TGF-b1 knockout mouse autoimmunity phenotype (Letterio et al.,
1994).
Improving implantation
Similarities between mice and humans in the early stages of embryo
implantation have made knockout mouse models useful in identifying
those genes which may be important for the establishment of human
pregnancy. As described earlier, gene ablation of LIF results in a
failure of embryo implantation (Stewart et al., 1992). LIF is therefore
an exciting candidate for either enhancing or preventing pregnancy in
humans. In humans, LIF is expressed by surface and glandular epi-
thelium, and is elevated at the time of implantation as it is in the
mouse (Bhatt et al., 1991; Charnock-Jones et al., 1994). LIF is detect-
able in uterine fluid, and there is evidence that reduced levels are
present in women experiencing unexplained infertility (Laird et al.,
1997), correlating to lower levels of expression by endometrial cells
(Tsai et al., 2000; Dimitriadis et al., 2006). However, low LIF
levels in uterine fluid were found to be predictive of implantation
success following IVF (Ledee-Bataille et al., 2002), indicating a deli-
cate balance is required for optimal embryonic/uterine stimulation.
The clinical application of these findings is being explored. Measure-
ment of LIF in uterine flushing is currently being investigated as a
potential tool for diagnosis of infertility (Mikolajczyk et al., 2003).
Furthermore, mutations in the LIF gene have been identified in
women with unexplained infertility and spontaneous abortion (Giess
et al., 1999; Steck et al., 2004; Kralickova et al., 2006).
LIF treatment increases rate of fertilization of sheep oocytes, and
enhances embryo quality during pre-implantation embryo culture in
sheep, mice and humans (Cheung et al., 2003; Ptak et al., 2006).
A preliminary study demonstrated improved murine implantation,
pregnancy and viability rates when LIF was administered during
transcervical blastocyst transfer, whereas a neutralizing antibody
exerted negative effects on implantation rates (Mitchell et al.,
2002). Similar experiments in rhesus monkeys prevented pregnancy,
strongly supporting the potential for translation to human biology
(Sengupta et al., 2006). However, human implantation is likely to
be more complex, with additional layers of redundancy evolved to
protect fertility. Closely related IL-11 is limited to a role in decidua-
lization in the mouse, but broader actions are indicated in the
human endometrium, where it shares an identical spatial and temporal
expression pattern with LIF in the peri-implantation phase (Dimitria-
dis et al., 2000). This hints at a possible redundancy between the
ligands in the human endometrium, which may necessitate the target-
ing of multiple cytokines for contraceptive development.
A mouse model of menstruation
Menstruation and the associated endometrial repair is a challenging
problem to study, as it occurs in a very small number of species,
not including mice. For this reason, a mouse model mimicking
menstruation has been developed (Finn and Pope, 1984; Brasted
et al., 2003), where the endometrium is artificially decidualized,
prior to progesterone withdrawal to trigger the onset of endometrial
breakdown and subsequent repair. This simulates menstruation in the
human and there are broad similarities in the patterns of inflammatory
leukocyte influx, production and activation of matrix metalloprotei-
nases (MMPs) and rapid regeneration and repair (Kaitu’u et al.,
2005). Importantly endometrial breakdown and repair is retarded by
treatment with antibodies that effectively knock out neutrophils
(Kaitu’u-Lino et al., 2007), confirming that the neutrophil infiltrate
observed in the premenstrual phase in human endometrium is involved
in the initiation of menstruation. This model can now be applied to
knockout mice, to test the involvement of individual cytokines in the
inflammatory breakdown and post-menstrual repair, information that
will assist in the future development of treatments for abnormal men-
strual bleeding.
Conclusion
Both the male and female reproductive tracts undergo extensive
development and remodelling over the course of an individual’s
life. Cytokines and cells of the immune system are intimately
involved in many of the processes critical to production of male
and female gametes, in the endometrium during menstruation and
implantation, during placental and embryonic growth, and during
parturition. Knockout mouse models have greatly assisted in reveal-
ing non-redundant essential roles that particular cytokines serve.
Sophisticated new approaches, including regulatable and transient
gene silencing, are now being utilized to further explore the role
of cytokines in particular reproductive processes. These technol-
ogies are expected to challenge, support and provide novel concepts
on the fundamental roles of cytokines in reproduction in mice, and
will form the basis of translational research into a better understand-
ing of reproductive processes and pathologies in humans.
Acknowledgements
The authors thank A/Prof Sarah Robertson (University of Adelaide,SA, Australia) for critical review of the manuscript and Sue Panck-ridge (Prince Henry’s Institute for Medical Research, Vic, Australia)for assistance in creating figures. This review aims to discuss howknockout technology has provided insight on the role of cytokinesin reproductive biology. The authors acknowledge that many import-ant papers on the roles of cytokines in reproductive biology discoveredby other means have been omitted from this review due to spacerestrictions. For more general information on cytokines in specificreproductive tissues or processes, excellent reviews are identified inthe reference list by an asterisk.
References
Abbondanzo SJ, Cullinan EB, McIntyre K, Labow MA, Stewart CL.Reproduction in mice lacking a functional type 1 IL-1 receptor.Endocrinology 1996;137:3598–3601.
Alexander WS. Suppressors of cytokine signalling (SOCS) in the immunesystem. Nat Rev Immunol 2002;2:410–416.
Aluvihare VR, Kallikourdis M, Betz AG. Regulatory T cells mediate maternaltolerance to the fetus. Nat Immunol 2004;5:266–271.
Araki M, Fukumatsu Y, Katabuchi H, Shultz LD, Takahashi K, Okamura H.Follicular development and ovulation in macrophage colony-stimulatingfactor-deficient mice homozygous for the osteopetrosis (op) mutation.Biol Reprod 1996;54:478–484.
Ashkar AA, Black GP, Wei Q, He H, Liang L, Head JR, Croy BA. Assessmentof requirements for IL-15 and IFN regulatory factors in uterine NK cell
Ingman and Jones
188
Dow
nloaded from https://academ
ic.oup.com/hum
upd/article-abstract/14/2/179/609519 by guest on 18 Novem
ber 2018
differentiation and function during pregnancy. J Immunol2003;171:2937–2944.
Baba Y, Nakano M, Yamada Y, Saito I, Kanegae Y. Practical range of effectivedose for Cre recombinase-expressing recombinant adenovirus without celltoxicity in mammalian cells. Microbiol Immunol 2005;49:559–570.
Barber EM, Pollard JW. The uterine NK cell population requires IL-15 butthese cells are not required for pregnancy nor the resolution of aListeria monocytogenes infection. J Immunol 2003;171:37–46.
Barker DJ. Adult consequences of fetal growth restriction. Clin Obstet Gynecol2006;49:270–283.
Beauparlant SL, Read PW, Di Cristofano A. In vivo adenovirus-mediated genetransduction into mouse endometrial glands: a novel tool to modelendometrial cancer in the mouse. Gynecol Oncol 2004;94:713–718.
Bhatt H, Brunet LJ, Stewart CL. Uterine expression of leukemia inhibitoryfactor coincides with the onset of blastocyst implantation. Proc NatlAcad Sci USA 1991;88:11408–11412.
*Bischof P, Meisser A, Campana A. Mechanisms of endometrial control oftrophoblast invasion. J Reprod Fertil Suppl 2000;55:65–71.
*Bornstein SR, Rutkowski H, Vrezas I. Cytokines and steroidogenesis. MolCell Endocrinol 2004;215:135–141.
Brasted M, White CA, Kennedy TG, Salamonsen LA. Mimicking the events ofmenstruation in the murine uterus. Biol Reprod 2003;69:1273–1280.
Brown CW, Houston-Hawkins DE, Woodruff TK, Matzuk MM. Insertion ofInhbb into the Inhba locus rescues the Inhba-null phenotype and revealsnew activin functions. Nat Genet 2000;25:453–457.
Calkins JH, Guo H, Sigel MM, Lin T. Tumor necrosis factor-alpha enhancesinhibitory effects of interleukin-1 beta on Leydig cell steroidogenesis.Biochem Biophys Res Commun 1990;166:1313–1318.
Campbell JJ, Qin S, Unutmaz D, Soler D, Murphy KE, Hodge MR,Wu L, Butcher EC. Unique subpopulations of CD56þ NK and NK-Tperipheral blood lymphocytes identified by chemokine receptorexpression repertoire. J Immunol 2001;166:6477–6482.
Caniggia I, Lye SJ, Cross JC. Activin is a local regulator of humancytotrophoblast cell differentiation. Endocrinology 1997;138:3976–3986.
*Chang H, Lau AL, Matzuk MM. Studying TGF-beta superfamily signaling byknockouts and knockins. Mol Cell Endocrinol 2001;180:39–46.
Chantakru S, Miller C, Roach LE, Kuziel WA, Maeda N, Wang WC,Evans SS, Croy BA. Contributions from self-renewal and trafficking tothe uterine NK cell population of early pregnancy. J Immunol2002;168:22–28.
Chaouat G, Ledee-Bataille N, Dubanchet S, Zourbas S, Sandra O, Martal J.TH1/TH2 paradigm in pregnancy: paradigm lost? Cytokines inpregnancy/early abortion: reexamining the TH1/TH2 paradigm. IntArch Allergy Immunol 2004;134:93–119.
Charnock-Jones DS, Sharkey AM, Fenwick P, Smith SK. Leukaemia inhibitoryfactor mRNA concentration peaks in human endometrium at the time ofimplantation and the blastocyst contains mRNA for the receptor at thistime. J Reprod Fertil 1994;101:421–426.
Chen HF, Shew JY, Ho HN, Hsu WL, Yang YS. Expression of leukemiainhibitory factor and its receptor in preimplantation embryos. FertilSteril 1999;72:713–719.
Cheng TC, Huang CC, Chen CI, Liu CH, Hsieh YS, Huang CY, Lee MS, LiuJY. Leukemia inhibitory factor antisense oligonucleotide inhibits thedevelopment of murine embryos at preimplantation stages. Biol Reprod2004;70:1270–1276.
Cheung LP, Leung HY, Bongso A. Effect of supplementation of leukemiainhibitory factor and epidermal growth factor on murine embryonicdevelopment in vitro, implantation, and outcome of offspring. FertilSteril 2003;80(Suppl 2):727–735.
Cohen PE, Pollard JW. Normal sexual function in male mice lacking afunctional type I interleukin-1 (IL-1) receptor. Endocrinology1998;139:815–818.
Cohen PE, Chisholm O, Arceci RJ, Stanley ER, Pollard JW. Absence ofcolony-stimulating factor-1 in osteopetrotic (csfmop/csfmop) miceresults in male fertility defects. Biol Reprod 1996;55:310–317.
Cohen PE, Hardy MP, Pollard JW. Colony-stimulating factor-1 plays a majorrole in the development of reproductive function in male mice. MolEndocrinol 1997a;11:1636–1650.
Cohen PE, Zhu L, Pollard JW. Absence of colony stimulating factor-1 inosteopetrotic (csfmop/csfmop) mice disrupts estrous cycles andovulation. 1997b;56:110–118.
Cohen PE, Zhu L, Nishimura K, Pollard JW. Colony-stimulating factor 1regulation of neuroendocrine pathways that control gonadal function inmice. Endocrinology 2002;143:1413–1422.
Dai XM, Ryan GR, Hapel AJ, Dominguez MG, Russell RG, Kapp S, SylvestreV, Stanley ER. Targeted disruption of the mouse colony-stimulating factor1 receptor gene results in osteopetrosis, mononuclear phagocytedeficiency, increased primitive progenitor cell frequencies, andreproductive defects. Blood 2002;99:111–120.
Dickson MC, Martin JS, Cousins FM, Kulkarni AB, Karlsson S, Akhurst RJ.Defective haematopoiesis and vasculogenesis in transforming growthfactor-beta 1 knock out mice. Development 1995;121:1845–1854.
Diebold RJ, Eis MJ, Yin M, Ormsby I, Boivin GP, Darrow BJ, SaffitzJE, Doetschman T. Early-onset multifocal inflammation in thetransforming growth factor beta 1-null mouse is lymphocyte mediated.Proc Natl Acad Sci USA 1995;92:12215–12219.
Dimitriadis E, Salamonsen LA, Robb L. Expression of interleukin-11 during thehuman menstrual cycle: coincidence with stromal cell decidualization andrelationship to leukaemia inhibitory factor and prolactin. Mol HumReprod 2000;6:907–914.
*Dimitriadis E, White CA, Jones RL, Salamonsen LA. Cytokines, chemokinesand growth factors in endometrium related to implantation. Hum ReprodUpdate 2005;11:613–630.
Dimitriadis E, Stoikos C, Stafford-Bell M, Clark I, Paiva P, KovacsG, Salamonsen LA. Interleukin-11, IL-11 receptoralpha and leukemiainhibitory factor are dysregulated in endometrium of infertile womenwith endometriosis during the implantation window. J Reprod Immunol2006;69:53–64.
Draper LB, Matzuk MM, Roberts VJ, Cox E, Weiss J, Mather JP, WoodruffTK. Identification of an inhibin receptor in gonadal tumors from inhibinalpha-subunit knockout mice. J Biol Chem 1998;273:398–403.
Ellis S. HLA G: at the interface. Am J Reprod Immunol 1990;23:84–86.*Ethier JF, Findlay JK. Roles of activin and its signal transduction mechanisms
in reproductive tissues. Reproduction 2001;121:667–675.Fallon PG, Jolin HE, Smith P, Emson CL, Townsend MJ, Fallon R, McKenzie
AN. IL-4 induces characteristic Th2 responses even in the combinedabsence of IL-5, IL-9, and IL-13. Immunity 2002;17:7–17.
Ferenczy A. Studies on the cytodynamics of human endometrial regeneration.II. Transmission electron microscopy and histochemistry. Am J ObstetGynecol 1976;124:582–595.
*Fijak M, Meinhardt A. The testis in immune privilege. Immunol Rev2006;213:66–81.
*Findlay JK. An update on the roles of inhibin, activin, and follistatin as localregulators of folliculogenesis. Biol Reprod 1993;48:15–23.
*Finn CA. Implantation, menstruation and inflammation. Biol Rev Camb PhilosSoc 1986;61:313–328.
Finn CA, Pope M. Vascular and cellular changes in the decidualizedendometrium of the ovariectomized mouse following cessation ofhormone treatment: a possible model for menstruation. J Endocrinol1984;100:295–300.
Fouladi-Nashta AA, Jones CJ, Nijjar N, Mohamet L, Smith A, ChambersI, Kimber SJ. Characterization of the uterine phenotype during theperi-implantation period for LIF-null, MF1 strain mice. Dev Biol2005;281:1–21.
Fried G, Wramsby H. Increase in transforming growth factor beta1 in ovarianfollicular fluid following ovarian stimulation and in-vitro fertilizationcorrelates to pregnancy. 1998;13:656–659.
Giess R, Tanasescu I, Steck T, Sendtner M. Leukaemia inhibitory factor genemutations in infertile women. Mol Hum Reprod 1999;5:581–586.
Goepfert AR, Goldenberg RL, Andrews WW, Hauth JC, Mercer B, Iams J,Meis P, Moawad A, Thom E, VanDorsten JP et al. The PretermPrediction Study: association between cervical interleukin 6concentration and spontaneous preterm birth. National Institute of ChildHealth and Human Development Maternal-Fetal Medicine UnitsNetwork. Am J Obstet Gynecol 2001;184:483–488.
Goumans MJ, Zwijsen A, van Rooijen MA, Huylebroeck D, RoelenBA, Mummery CL. Transforming growth factor-beta signalling inextraembryonic mesoderm is required for yolk sac vasculogenesis inmice. Development 1999;126:3473–3483.
Gouon Evans V, Rothenberg ME, Pollard JW. Postnatal mammary glanddevelopment requires macrophages and eosinophils. Development2000;127:2269–2282.
Gouon-Evans V, Pollard JW. Eotaxin is required for eosinophil homing into thestroma of the pubertal and cycling uterus. Endocrinology 2001;142:4515–4521.
Gu H, Marth JD, Orban PC, Mossmann H, Rajewsky K. Deletion of a DNApolymerase beta gene segment in T cells using cell type-specific genetargeting. Science 1994;265:103–106.
Cytokine knockouts in reproduction
189
Dow
nloaded from https://academ
ic.oup.com/hum
upd/article-abstract/14/2/179/609519 by guest on 18 Novem
ber 2018
Guleria I, Pollard JW. The trophoblast is a component of the innate immunesystem during pregnancy. Nat Med 2000;6:589–593.
Guo Q, Kumar TR, Woodruff T, Hadsell LA, DeMayo FJ, Matzuk MM.Overexpression of mouse follistatin causes reproductive defects intransgenic mice. Mol Endocrinol 1998;12:96–106.
*Hales DB. Testicular macrophage modulation of Leydig cell steroidogenesis.J Reprod Immunol 2002;57:3–18.
Hannan NJ, Jones RL, Critchley HO, Kovacs GJ, Rogers PA, AffandiB, Salamonsen LA. Coexpression of fractalkine and its receptor innormal human endometrium and in endometrium from users ofprogestin-only contraception supports a role for fractalkine in leukocyterecruitment and endometrial remodeling. J Clin Endocrinol Metab2004;89:6119–6129.
Hannon GJ. RNA interference. Nature 2002;418:244–251.Heasman J. Morpholino oligos: making sense of antisense? Dev Biol
2002;243:209–214.*Hedger MP. Macrophages and the immune responsiveness of the testis. J
Reprod Immunol 2002;57:19.*Hedger MP, Meinhardt A. Cytokines and the immune-testicular axis. J
Reprod Immunol 2003;58:1–26.Hempstock J, Cindrova-Davies T, Jauniaux E, Burton GJ. Endometrial glands as
a source of nutrients, growth factors and cytokines during the first trimesterof human pregnancy: a morphological and immunohistochemical study.Reprod Biol Endocrinol 2004;2:58.
*Ingman WV, Robertson SA. Defining the actions of transforming growthfactor beta in reproduction. Bioessays 2002;24:904–914.
Ingman WV, Robertson SA. TGF{beta}1 null mutation causes infertility inmale mice associated with testosterone deficiency and sexualdysfunction. Endocrinology 2007;148:4032–4043.
Ingman WV, Robker RL, Woittiez K, Robertson SA. Null mutation intransforming growth factor beta1 disrupts ovarian function and causesoocyte incompetence and early embryo arrest. Endocrinology2006;147:835–845.
Inoue T, Kanzaki H, Iwai M, Imai K, Narukawa S, Higuchi T, KatsuragawaH, Mori T. Tumour necrosis factor alpha inhibits in-vitro decidualizationof human endometrial stromal cells. Hum Reprod 1994;9:2411–2417.
Jasper MJ, Robertson SA, Van der Hoek KH, Bonello N, BrannstromM, Norman RJ. Characterization of ovarian function ingranulocyte-macrophage colony-stimulating factor-deficient mice. BiolReprod 2000;62:704–713.
Johansson M, Bromfield JJ, Jasper MJ, Robertson SA. Semen activates thefemale immune response during early pregnancy in mice. Immunology2004;112:290–300.
Jones RL, Kelly RW, Critchley HO. Chemokine and cyclooxygenase-2expression in human endometrium coincides with leukocyteaccumulation. Hum Reprod 1997;12:1300–1306.
Jones RL, Salamonsen LA, Findlay JK. Activin A promotes human endometrialstromal cell decidualization in vitro. J Clin Endocrinol Metab2002;87:4001–4004.
Jones RL, Hannan NJ, Kaitu’u TJ, Zhang J, Salamonsen LA. Identification ofchemokines important for leukocyte recruitment to the humanendometrium at the times of embryo implantation and menstruation. JClin Endocrinol Metab 2004;89:6155–6167.
*Jones RL, Stoikos C, Findlay JK, Salamonsen LA. TGF-beta superfamilyexpression and actions in the endometrium and placenta. Reproduction2006;132:217–232.
Jorgez CJ, Klysik M, Jamin SP, Behringer RR, Matzuk MM. Granulosacell-specific inactivation of follistatin causes female fertility defects.Mol Endocrinol 2004;18:953–967.
*Kaipia A, Hsueh AJ. Regulation of ovarian follicle atresia. Annu Rev Physiol1997;59:349–363.
Kaitu’u TJ, Shen J, Zhang J, Morison NB, Salamonsen LA. Matrixmetalloproteinases in endometrial breakdown and repair: functionalsignificance in a mouse model. Biol Reprod 2005;73:672–680.
Kaitu’u-Lino TJ, Morison NB, Salamonsen LA. Neutrophil depletion retardsendometrial repair in a mouse model. Cell Tissue Res 2007;328:197–206.
Kallapur S, Ormsby I, Doetschman T. Strain dependency of TGFbeta1 functionduring embryogenesis. Mol Reprod Dev 1999;52:341–349.
Kelly RW. Pregnancy maintenance and parturition: the role of prostaglandin inmanipulating the immune and inflammatory response. Endocr Rev1994;15:684–706.
Kern S, Robertson SA, Mau VJ, Maddocks S. Cytokine secretion bymacrophages in the rat testis. Biol Reprod 1995;53:1407–1416.
*King A, Burrows T, Verma S, Hiby S, Loke YW. Human uterinelymphocytes. Hum Reprod Update 1998;4:480–485.
King A, Burrows TD, Hiby SE, Bowen JM, Joseph S, Verma S, Lim PB, GardnerL, Le Bouteiller P, Ziegler A et al. Surface expression of HLA-C antigen byhuman extravillous trophoblast. Placenta 2000;21:376–387.
King AE, Critchley HO, Kelly RW. Innate immune defences in the humanendometrium. Reprod Biol Endocrinol 2003;1:116.
*Knight PG, Glister C. TGF-beta superfamily members and ovarian follicledevelopment. Reproduction 2006;132:191–206.
Kobayashi S, Yoshida K, Ward JM, Letterio JJ, Longenecker G, Yaswen L,Mittleman B, Mozes E, Roberts AB, Karlsson S et al. Beta2-microglobulin-deficient background ameliorates lethal phenotype ofthe TGF-beta 1 null mouse. J Immunol 1999;163:4013–4019.
Kralickova M, Sima R, Vanecek T, Sima P, Rokyta Z, Ulcova-Gallova Z,Sucha R, Uher P, Hes O. Leukemia inhibitory factor gene mutations inthe population of infertile women are not restricted to nulligravidpatients. Eur J Obstet Gynecol Reprod Biol 2006;127:231–235.
Kulkarni AB, Huh C, Becker D, Geiser A, Lyght M, Flanders KC, Roberts AB,Sporn MB, Ward JM, Karlsson S. Transforming growth factor b1 nullmutation in mice causes excessive inflammatory response and earlydeath. Proc Natl Acad Sci USA 1993;90:770–774.
Kundu S-D, Kim I-Y, Yang T, Doglio L, Lang S, Zhang X, Buttyan R, Kim S-J,Chang J, Cai X et al. Absence of proximal duct apoptosis in the ventralprostate of transgenic mice carrying the C3(1)-TGF-beta type IIdominant negative receptor. Prostate 2000;43:118–124.
Laird SM, Tuckerman EM, Dalton CF, Dunphy BC, Li TC, Zhang X. Theproduction of leukaemia inhibitory factor by human endometrium:presence in uterine flushings and production by cells in culture. HumReprod 1997;12:569–574.
Larsen L, Ropke C. Suppressors of cytokine signalling: SOCS. APMIS2002;110:833–844.
Ledee-Bataille N, Lapree-Delage G, Taupin JL, Dubanchet S, FrydmanR, Chaouat G. Concentration of leukaemia inhibitory factor (LIF) inuterine flushing fluid is highly predictive of embryo implantation. HumReprod 2002;17:213–218.
*Leonard S, Murrant C, Tayade C, van den Heuvel M, Watering R, Croy BA.Mechanisms regulating immune cell contributions to spiral arterymodification—facts and hypotheses—a review. Placenta 2006;27(SupplA):S40–S46.
Letterio JJ, Geiser AG, Kulkarni AB, Roche NS, Sporn MB, Roberts AB.Maternal rescue of transforming growth factor-beta 1 null mice. Science1994;264:1936–1938.
Letterio JJ, Geiser AG, Kulkarni AB, Dang H, Kong L, Nakabayashi T,Mackall CL, Gress RE, Roberts AB. Autoimmunity associated withTGF-beta1-deficiency in mice is dependent on MHC class II antigenexpression. J Clin Invest 1996;98:2109–2119.
Lui W-Y, Lee W-M, Cheng Y. Transforming growth factor-beta3 perturbs theinter-Sertoli tight junction permeability barrier in vitro possibly mediatedvia its effects on occludin, zonula occludens-1, and claudin-11.Endocrinology 2001;142:1865–1877.
Matzuk MM, Kumar TR, Bradley A. Different phenotypes for mice deficient ineither activins or activin receptor type II. Nature 1995a;374:356–360.
Matzuk MM, Kumar TR, Vassalli A, Bickenbach JR, Roop DR, JaenischR, Bradley A. Functional analysis of activins during mammaliandevelopment. Nature 1995b;374:354–356.
Matzuk MM, Lu N, Vogel H, Sellheyer K, Roop DR, Bradley A. Multipledefects and perinatal death in mice deficient in follistatin. Nature1995c;374:360–363.
Mauduit C, Chauvin MA, Hartmann DJ, Revol A, Morera AM, Benahmed M.Interleukin-1 alpha as a potent inhibitor of gonadotropin action in porcineLeydig cells: site(s) of action. Biol Reprod 1992;46:1119–1126.
McMillen IC, Robinson JS. Developmental origins of the metabolic syndrome:prediction, plasticity, and programming. Physiol Rev 2005;85:571–633.
Medawar PB. Some immunological and endocrinological problems raised bythe evolution of viviparity in vertebrates. Symp Soc Exp Biol1953;7:320–338.
Meisser A, Cameo P, Islami D, Campana A, Bischof P. Effects of interleukin-6(IL-6) on cytotrophoblastic cells. Mol Hum Reprod 1999;5:1055–1058.
Mikolajczyk M, Skrzypczak J, Szymanowski K, Wirstlein P. The assessment ofLIF in uterine flushing—a possible new diagnostic tool in states ofimpaired fertility. Reprod Biol 2003;3:259–270.
Mitchell MH, Swanson RJ, Oehninger S. In vivo effect of leukemia inhibitoryfactor (LIF) and an anti-LIF polyclonal antibody on murine embryo andfetal development following exposure at the time of transcervicalblastocyst transfer. Biol Reprod 2002;67:460–464.
Ingman and Jones
190
Dow
nloaded from https://academ
ic.oup.com/hum
upd/article-abstract/14/2/179/609519 by guest on 18 Novem
ber 2018
Miyazaki S, Tanebe K, Sakai M, Michimata T, Tsuda H, Fujimura M,Nakamura M, Kiso Y, Saito S. Interleukin 2 receptor gamma chain(gamma(c)) knockout mice show less regularity in estrous cycle butachieve normal pregnancy without fetal compromise. Am J ReprodImmunol 2002;47:222–230.
*Moffett-King A. Natural killer cells and pregnancy. Nat Rev Immunol2002;2:656–663.
Morcos PA. Achieving efficient delivery of morpholino oligos in cultured cells.Genesis 2001;30:94–102.
Munn DH, Zhou M, Attwood JT, Bondarev I, Conway SJ, Marshall B, BrownC, Mellor AL. Prevention of allogeneic fetal rejection by tryptophancatabolism. Science 1998;281:1191–1193.
Murphy SP, Fast LD, Hanna NN, Sharma S. Uterine NK cells mediateinflammation-induced fetal demise in IL-10-null mice. J Immunol2005;175:4084–4090.
Ni H, Ding NZ, Harper MJ, Yang ZM. Expression of leukemia inhibitory factorreceptor and gp130 in mouse uterus during early pregnancy. Mol ReprodDev 2002;63:143–150.
Park JS, Park CW, Lockwood CJ, Norwitz ER. Role of cytokines in pretermlabor and birth. Minerva Ginecol 2005;57:349–366.
*Pollard JW. Role of colony-stimulating factor-1 in reproduction anddevelopment. Mol Reprod Dev 1997;46:54–60.
Pollard JW, Hennighausen L. Colony stimulating factor 1 is required formammary gland development during pregnancy. Proc Natl Acad SciUSA 1994;91:9312–9316.
Proetzel G, Pawlowski SA, Wiles MV, Yin M, Boivin GP, Howles PN, Ding J,Ferguson MW, Doetschman T. Transforming growth factor-beta 3 isrequired for secondary palate fusion. Nat Genet 1995;11:409–414.
Ptak G, Lopes F, Matsukawa K, Tischner M, Loi P. Leukaemia inhibitoryfactor enhances sheep fertilization in vitro via an influence on theoocyte. Theriogenology 2006;65:1891–1899.
Qu J, Thomas K. Inhibin and activin production in human placenta. Endocr Rev1995;16:485–507.
Robb L, Li R, Hartley L, Nandurkar HH, Koentgen F, Begley CG. Infertility infemale mice lacking the receptor for interleukin 11 is due to a defectiveuterine response to implantation. Nat Med 1998;4:303–308.
Roberts CT, White CA, Wiemer NG, Ramsay A, Robertson SA. Alteredplacental development in interleukin-10 null mutant mice. Placenta2003;24(Suppl A):S94–S99.
*Robertson SA. GM-CSF regulation of embryo development and pregnancy.Cytokine Growth Factor Rev 2007;18:287–298.
*Robertson SA, Mau VJ, Hudson SN, Tremellen KP. Cytokine-LeukocyteNetworks and the Establishment of Pregnancy. Am J Reprod Immunol1997;37:438–442.
Robertson SA, Roberts CT, Farr KL, Dunn AR, Seamark RF. Fertilityimpairment in granulocyte-macrophage colony-stimulatingfactor-deficient mice. Biol Reprod 1999;60:251–261.
Robertson SA, Mau VJ, Young IG, Matthaei KI. Uterine eosinophils andreproductive performance in interleukin 5-deficient mice. J ReprodFertil 2000;120:423–432.
Robertson SA, Sjoblom C, Jasper MJ, Norman RJ, Seamark RF.Granulocyte-macrophage colony-stimulating factor promotes glucosetransport and blastomere viability in murine preimplantation embryos.Biol Reprod 2001;64:1206–1215.
Robertson SA, Skinner RJ, Care AS. Essential role for IL-10 in resistance tolipopolysaccharide-induced preterm labor in mice. J Immunol2006;177:4888–4896.
Robertson SA, Care AS, Skinner RJ. Interleukin 10 regulates inflammatorycytokine synthesis to protect against lipopolysaccharide-induced abortionand fetal growth restriction in mice. Biol Reprod 2007;76:738–748.
Roby KF, Son DS, Terranova PF. Alterations of events related to ovarianfunction in tumor necrosis factor receptor type I knockout mice. BiolReprod 1999;61:1616–1621.
Salamonsen LA. Tissue injury and repair in the female human reproductivetract. Reproduction 2003;125:301–311.
*Salamonsen LA, Lathbury LJ. Endometrial leukocytes and menstruation. HumReprod Update 2000;6:16–27.
Sanford LP, Ormsby I, Gittenberger de Groot AC, Sariola H, Friedman R,Boivin GP, Cardell EL, Doetschman T. TGFbeta2 knockout mice havemultiple developmental defects that are non-overlapping with otherTGFbeta knockout phenotypes. Development 1997;124:2659–2670.
Sargent IL, Martin KL, Barlow DH. The use of recombinant growth factors topromote human embryo development in serum-free medium. Hum Reprod1998;13(Suppl 4):239–248.
*Sargent IL, Borzychowski AM, Redman CW. NK cells and humanpregnancy—an inflammatory view. Trends Immunol 2006;27:399–404.
Schofield G, Kimber SJ. Leukocyte subpopulations in the uteri of leukemiainhibitory factor knockout mice during early pregnancy. Biol Reprod2005;72:872–878.
Sengupta J, Lalitkumar PG, Najwa AR, Ghosh D. Monoclonal anti-leukemiainhibitory factor antibody inhibits blastocyst implantation in the rhesusmonkey. Contraception 2006;74:419–425.
*Sharkey A. Cytokines and implantation. Rev Reprod 1998;3:52–61.Sherwin JR, Freeman TC, Stephens RJ, Kimber S, Smith AG, Chambers I,
Smith SK, Sharkey AM. Identification of genes regulated byleukemia-inhibitory factor in the mouse uterus at the time ofimplantation. Mol Endocrinol 2004;18:2185–2195.
Shull MM, Ormsby I, Kier AB, Pawlowski SA, Diebold RJ, Yin M, Allen R,Sidman C, Proetzel G, Calvin D et al. Targeted disruption of the mousetransforming growth factor a1 gene results in multifocal inflammatorydisease. Nature 1992;359:693–699.
Simhan HN, Caritis SN, Krohn MA, Martinez de Tejada B, Landers DV, HillierSL. Decreased cervical proinflammatory cytokines permit subsequentupper genital tract infection during pregnancy. Am J Obstet Gynecol2003;189:560–567.
Simon C, Frances A, Piquette GN, el Danasouri I, Zurawski G, Dang W, PolanML. Embryonic implantation in mice is blocked by interleukin-1 receptorantagonist. Endocrinology 1994;134:521–528.
Siu MK, Lee WM, Cheng CY. The interplay of collagen IV, tumor necrosisfactor-alpha, gelatinase B (matrix metalloprotease-9), and tissueinhibitor of metalloproteases-1 in the basal lamina regulates Sertolicell-tight junction dynamics in the rat testis. Endocrinology2003;144:371–387.
Sjoblom C, Wikland M, Robertson SA. Granulocyte-macrophagecolony-stimulating factor promotes human blastocyst development invitro. Hum Reprod 1999;14:3069–3076.
Sjoblom C, Wikland M, Robertson SA. Granulocyte-macrophagecolony-stimulating factor (GM-CSF) acts independently of the betacommon subunit of the GM-CSF receptor to prevent inner cell massapoptosis in human embryos. Biol Reprod 2002;67:1817–1823.
Sjoblom C, Roberts CT, Wikland M, Robertson SA. Granulocyte-macrophagecolony-stimulating factor alleviates adverse consequences of embryoculture on fetal growth trajectory and placental morphogenesis.Endocrinology 2005;146:2142–2153.
Somerset DA, Zheng Y, Kilby MD, Sansom DM, Drayson MT. Normal humanpregnancy is associated with an elevation in the immune suppressiveCD25þ CD4þ regulatory T-cell subset. Immunology 2004;112:38–43.
Steck T, Giess R, Suetterlin MW, Bolland M, Wiest S, Poehls UG, Dietl J.Leukaemia inhibitory factor (LIF) gene mutations in women withunexplained infertility and recurrent failure of implantation after IVFand embryo transfer. Eur J Obstet Gynecol Reprod Biol 2004;112:69–73.
Stewart CL, Kaspar P, Brunet LJ, Bhatt H, Gadi I, Kontgen F, Abbondanzo SJ.Blastocyst implantation depends on maternal expression of leukaemiainhibitory factor. Nature 1992;359:76–79.
Summerton J, Weller D. Morpholino antisense oligomers: design, preparation,and properties. Antisense Nucleic Acid Drug Dev 1997;7:187–195.
Takahashi Y, Carpino N, Cross JC, Torres M, Parganas E, Ihle JN. SOCS3: anessential regulator of LIF receptor signaling in trophoblast giant celldifferentiation. Embo J 2003;22:372–384.
Tang M, Taylor HS, Tabibzadeh S. In vivo gene transfer of lefty leads toimplantation failure in mice. Hum Reprod 2005;20:1772–1778.
Taniguchi T, Takata M, Ikeda A, Momotani E, Sekikawa K. Failure of germinalcenter formation and impairment of response to endotoxin in tumornecrosis factor alpha-deficient mice. Lab Invest 1997;77:647–658.
Tierney EP, Giudice LC. Role of activin A as a mediator of in vitro endometrialstromal cell decidualization via the cyclic adenosine monophosphatepathway. Fertil Steril 2004;81(Suppl 1):899–903.
Tsai HD, Chang CC, Hsieh YY, Lo HY. Leukemia inhibitory factor expressionin different endometrial locations between fertile and infertile womenthroughout different menstrual phases. J Assist Reprod Genet2000;17:415–418.
Tushinski RJ, Oliver IT, Guilbert LJ, Tynan PW, Warner JR, Stanley ER.Survival of mononuclear phagocytes depends on a lineage-specificgrowth factor that the differentiated cells selectively destroy. Cell1982;28:71–81.
Ulloa L, Tabibzadeh S. Lefty inhibits receptor-regulated Smad phosphorylationinduced by the activated transforming growth factor-beta receptor. J BiolChem 2001;276:21397–21404.
Cytokine knockouts in reproduction
191
Dow
nloaded from https://academ
ic.oup.com/hum
upd/article-abstract/14/2/179/609519 by guest on 18 Novem
ber 2018
Vassalli A, Matzuk MM, Gardner HA, Lee KF, Jaenisch R. Activin/inhibinbeta B subunit gene disruption leads to defects in eyelid developmentand female reproduction. Genes Dev 1994;8:414–427.
Wang J, Wreford NG, Lan HY, Atkins R, Hedger MP. Leukocyte populationsof the adult rat testis following removal of the Leydig cells by treatmentwith ethane dimethane sulfonate and subcutaneous testosteroneimplants. Biol Reprod 1994;51:551–561.
Ware CB, Horowitz MC, Renshaw BR, Hunt JS, Liggitt D, Koblar SA, GliniakBC, McKenna HJ, Papayannopoulou T, Thoma B et al. Targeteddisruption of the low-affinity leukemia inhibitory factor receptor genecauses placental, skeletal, neural and metabolic defects and results inperinatal death. Development 1995;121:1283–1299.
Webb SE, Pollard JW, Jones GE. Direct observation and quantification ofmacrophage chemoattraction to the growth factor CSF-1. J Cell Sci1996;109(Pt 4):793–803.
Wegmann TG, Lin H, Guilbert L, Mosmann TR. Bidirectional cytokineinteractions in the maternal-fetal relationship: is successful pregnancy aTH2 phenomenon? Immunol Today 1993;14:353–356.
Wenstrom KD, Andrews WW, Hauth JC, Goldenberg RL, DuBard MB, CliverSP. Elevated second-trimester amniotic fluid interleukin-6 levels predictpreterm delivery. Am J Obstet Gynecol 1998;178:546–550.
White CA, Robb L, Salamonsen LA. Uterine extracellular matrix componentsare altered during defective decidualization in interleukin-11 receptoralpha deficient mice. Reprod Biol Endocrinol 2004;2:76.
Wiktor-Jedrzejczak W, Bartocci A, Ferrante AW, Jr, Ahmed-Ansari A, SellKW, Pollard JW, Stanley ER. Total absence of colony-stimulatingfactor 1 in the macrophage-deficient osteopetrotic (op/op) mouse. ProcNatl Acad Sci USA 1990;87:4828–4832.
Wira CR, Fahey JV, Sentman CL, Pioli PA, Shen L. Innate and adaptiveimmunity in female genital tract: cellular responses and interactions.Immunol Rev 2005;206:306–335.
*Wu R, Van der Hoek KH, Ryan NK, Norman RJ, Robker RL. Macrophagecontributions to ovarian function. Hum Reprod Update 2004;10:119–133.
*Xia W, Mruk DD, Lee WM, Cheng CY. Cytokines and junction restructuringduring spermatogenesis–a lesson to learn from the testis. CytokineGrowth Factor Rev 2005;16:469–493.
Xiong Y, Hales DB. The role of tumor necrosis factor-alpha in the regulation ofmouse Leydig cell steroidogenesis. Endocrinology 1993;132:2438–2444.
Yoshida H, Hayashi S, Kunisada T, Ogawa M, Nishikawa S, Okamura H, SudoT, Shultz LD. The murine mutation osteopetrosis is in the coding region ofthe macrophage colony stimulating factor gene. Nature 1990;345:442–444.
Zygmunt M, Hahn D, Kiesenbauer N, Munstedt K, Lang U. Invasion ofcytotrophoblastic (JEG-3) cells is up-regulated by IL-15 in vitro. Am JReprod Immunol 1998;40:326–331.
Submitted on September 13, 2007; resubmitted on October 3, 2007; acceptedon October 25, 2007
Ingman and Jones
192
Dow
nloaded from https://academ
ic.oup.com/hum
upd/article-abstract/14/2/179/609519 by guest on 18 Novem
ber 2018