麻豆传媒

IT鈥橲 a wonder any of us are born. The womb seems a cosy place, but an unborn
baby is essentially a foreign body inside its mother. Our immune system
ferociously defends us against anything it recognises as 鈥渁lien鈥濃�攚itness
the constant threat of organ rejection faced by transplant patients. And yet we
all made it into the world somehow. So what鈥檚 going on in there?

The developing fetus inherits half its genes from its mother and half from
its father. Together, these genes are the blueprint for every component found in
the developing child鈥檚 body, and many of the components inherited from the
father will be 鈥渁ntigenic鈥� to the mother鈥攖hey should, in theory, trigger
an immune reaction that destroys the fetus.

In 1953, the British immunologist and Nobel Laureate Peter Medawar was the
first to suggest that a fetus is the equivalent of a foreign tissue graft. He
proposed four possible ways it could be evading rejection. First, it may simply
not trigger immune reactions in the mother鈥攖hey are entirely compatible.
We now know this can鈥檛 be the case. Secondly, the uterus might be
immunologically 鈥� privileged鈥� (it turns a blind eye to its occupant).
Thirdly, the placenta could act as a physical barrier protecting the fetus from
its mother. Or fourthly, pregnancy may alter the maternal immune response in
some way.

Almost fifty years on, and none of these explanations has been
proven鈥攖hough there have been plenty of fascinating hints. Meanwhile,
research into the phenomenon has shed light on many branches of the immune
system, and recent findings could give doctors fresh insights into everything
from infertility treatment to organ transplantation and cancer therapy.

To understand the problems faced by the fetus, we need to look at what
happens at the very start of pregnancy, when the fertilised egg is busy getting
a grip on life. The progression from sperm and egg to the birth of a baby is not
only a story of growth and development, but also of a developing relationship
between mother and child.

Fertilisation occurs in one of the fallopian tubes, which connect the
ovaries to the womb. The fertilised egg, or 鈥�zygote鈥�, divides into two,
four, then eight cells as it passes down the tube towards the uterus. The embryo
will have to grow and develop very quickly over the subsequent days and months,
so a plentiful supply of nutrients is essential from the start. To get it, the
tiny ball of cells has to embed itself in the wall of the uterus and tap into
the mother鈥檚 blood supply. Some embryonic cells devote themselves to obtaining
nutrition from the mother by establishing the placenta. Fully developed,
this is a flattened circular organ with a rich blood supply allowing oxygen and
nutrients to pass from mother to the fetus and waste products to pass the other
way.

By the time it reaches the uterus, about four days after fertilisation, the
cluster of embryonic cells has become a blastocyst
(see Figure 1). This
hollow ball of around fifty cells has three main components: the inner mass of
cells that becomes the embryo, the fluid-filled cavity that allows
expansion of the growing inner cell mass, and the trophoblast, or outer
layer, part of which becomes the placenta.

The placenta also includes maternal tissue. It works like this: the
trophoblast cells nearest the inner cell mass land on the lining of the uterus,
the endometrium. They rapidly divide and their outer
layer鈥攖ongue-twistingly called the syncytiotrophoblas迟鈥攑耻蝉丑别蝉
finger-like projections into the endometrium by secreting enzymes that erode it.
In this way, the embryo sinks into the wall of the uterus, where it anchors
itself and enjoys the supply of nutrients leaking from the damaged maternal
tissue.

The embryo becomes completely surrounded by endometrium. Around it appears
the chorion which encloses the space into which the embryo will grow. Out
of the wall of the chorion grows a stalk that joins up with the embryo鈥檚 inner
cell mass. Blood vessels grow into this stalk, which will become the
umbilical cord鈥攑lumbed into the mother via the placenta.

The finger-like projections from the trophoblast cells crowd in among the
damaged, leaky arteries of the endometrium, allowing nutrients to cross from
pools of maternal blood to the fetus. So Medawar鈥檚 suggestion that there might
be a physical barrier between the two genetically different entities, preventing
rejection, seems unlikely. Not only are the embryonic cells buried in the uterus
wall, some are single-mindedly invasive鈥攅xactly the sort of activity you鈥檇
hope your immune system might pick up.

The idea that a mother鈥檚 immune system could attack and destroy her own child
seems shocking, but the real surprise is that most of the time it doesn鈥檛.
Children are clearly not clones of their mothers鈥攖hey are genetically
distinct, with a different set of antigens. So the developing infant really is a
foreign body inside its mother. To understand how the mother鈥檚 immune system
avoids attacking her child, we need to look at how immunity normally operates.
Immune cells patrolling the body have to be 鈥減resented鈥� with foreign material in
a particular way before they can recognise it. To this end, a group of proteins,
the major histocompatibility complex(MHC) molecules, are used by cells to
present foreign material to immune cells, by displaying it on their cell
membrane (see Inside Science Nos. 8 and 127).

There are two main class of MHC molecule: class I (MHC I) and class II (MHC
II). Every cell with a nucleus uses MHC I to present material picked up
within itself鈥攕uch as fragments of a virus that has invaded the
cell鈥攖o immune cells. MHC II is used by specialised,
antigen-presenting cells to display any foreign material they have ingested. MHC
II molecules present these antigens to immune cells known as T helpers
(THs), so called because they 鈥渉elp鈥� other immune cells get on with the job of
fighting infection. They do this by producing a range of chemical messengers
called cytokines (see Inside Science No. 12), which perform a vast array
of functions. These include encouraging another group of immune cells, B cells,
to produce antibodies.

TH cells are classified as TH1 or TH2 depending on the cytokines they
produce. Type 1 cytokines鈥攆rom TH1 cells鈥攖end to be associated with
combating bacterial and viral infection, whereas type 2 are involved in fighting
off infection by multicellular parasites. Type 2 also play a central role in
allergic reactions.

MHC I molecules present antigens to several groups of cells. One is a
specialised group of T cells known as cytotoxic T cells, which attack
infected cells. Another group is the natural killer cells, which are less choosy
than cytotoxic T cells about what cells they target, and provide the earliest
attack on virus-infected cells and 鈥渙dd鈥� host cells (such as tumour cells).
Cytotoxic T cells have a 鈥渕emory鈥�, and will mount a more ferocious attack the
second time they meet a known antigen. Natural killer cells do not, and will
react in the same way whether or not they have encountered a particular antigen
before.

Antigens presented by MHC II molecules are needed for 鈥渉umoral鈥� immune
responses, where antibody molecules are produced. By contrast, antigens
presented by MHC I molecules generate cell-mediated responses, where cells mount
a direct attack without the need for antibody go-betweens. However, the
cell-mediated response has a part to play in most immune reactions
(see Figure 2).

Scientists first encountered MHC molecules because of the role they play in
determining the success of transplants. Tissues that can be transplanted without
being rejected by the recipient鈥檚 immune system are 鈥渉istocompatible鈥� (
histos is Greek for tissue), and the genes that govern compatibility are
located in the MHC on chromosome number 6. This is where the genes that are the
blueprints for MHC I and MHC II molecules are located.

The many genes that make up the complex are tightly grouped on the
chromosome, so they are usually inherited as a complete set, or haplotype
鈥攐ne haplotype from each parent. Each gene exists in numerous different
forms, making unrelated individuals unlikely to share the same haplotype. Of
each parent鈥檚 own two haplotypes, they have one in common with their child, and
the other is mismatched.

All cells with nuclei express MHC molecules on their surface, allowing the
immune system to keep a constant watch on them. In organ transplantation,
failure to take account of the donor and recipient鈥檚 MHC haplotype will trigger
rejection. And yet, sharing haplotypes is not necessary for a successful
pregnancy. Somehow the mother鈥檚 immune system overlooks the fact that the fetus
is incompatible.

And the mystery doesn鈥檛 end there. Cells don鈥檛 need a nucleus to be detected
by the immune system, so MHC compatibility isn鈥檛 the whole story. Red blood
cells鈥攚hich have no nucleus鈥攑resent their own problems. Your blood
is categorised by the antigens displayed on the surface of its red blood cells:
they are covered with either A or B antigens (making you group A or group B),
both (AB) or neither (O). Group A individuals can only accept blood donations
from A or O donors, group B from B or O donors, group AB can accept blood from
anyone, and Os can only accept blood from O donors. In addition, blood is
grouped on the basis of rhesus antigens. One antigen in particular, Rh D, is
either present (rhesus positive) or absent (rhesus negative) on
the surface of red blood cells. And as offspring can inherit their blood group
from either parent, a fetus鈥檚 blood group will not necessarily match its
mother鈥檚 (see Mother and child divided).

Viewed as a 鈥渢ransplant鈥� of foreign tissue, the prospects for the developing
offspring don鈥檛 look good. And yet, somehow, the estimated 150,000 babies born
in the world every day have managed to avoid immune rejection. To find out how
they do it we need to delve a bit deeper.

A defensive 鈥渕oat鈥� of amniotic fluid surrounds the fetus in the
uterus. Both mother and fetus benefit鈥攅ach is protected, more or less,
from the other. By 12 weeks, the fetus has developed its own blood circulation.
It relies completely on its mother to supply oxygen and nutrients, and to remove
carbon dioxide and other wastes. The exchange between the two circulatory
systems must take place without any blood mixing, or the result would be
catastrophic鈥攖he same as a mismatched blood transfusion.

The placenta effectively separates the two circulatory systems. Fetal blood
vessels pass into the placenta along extensions of the chorion known as the
chorionic villi鈥攚hich developed from those finger-like projections
extending from trophoblast into the endometrium. The villi themselves are
embedded in the maternal blood-filled spaces, thelacunae, which make up
the placenta鈥檚 spongy mass (see Figure 3).

Gases, nutrients and wastes are exchanged between the maternal and fetal
blood. But bacteria, viruses and toxins can also pass into the fetus. Luckily,
there鈥檚 a simultaneous uptake of maternal antibodies and other immune
components. Maternal antibodies taken up in this way continue to confer immunity
for a limited period even after the baby鈥檚 birth, when antibodies in the
mother鈥檚 milk start to take over. This gives the baby a vital head start, before
it has built up its own immunity.

But these physical barriers鈥攖he amniotic haven in which the
fetus floats and its separate circulatory system鈥攃annot fully account for
our survival. They are only partially successful. Maternal cells have been
reported to migrate into the fetus, and fetal cells have been detected in the
mother鈥檚 circulation (see All mixed up in the womb).
And how do the fetal cells of the placenta,
which are firmly embedded in maternal tissue, escape attack from the mother鈥檚
immune cells? Come to that, how does the tiny implanted trophoblast get away
with burrowing into maternal tissue in the first place? The mother鈥檚 immune
cells would immediately detect foreign antigens presented by the MHC molecules
on the trophoblast鈥檚 surface. MHC I molecules would provoke the cell-mediated
wing of the immune system (cytotoxic T cells and natural killer cells), and MHC
II would trigger a humoral immune response.

SURVIVAL STRATEGIES

The cloak of anonymity

One way round this might be for the cells of the trophoblast to express no
MHC molecules at all. Indeed, trophoblast cells do not express MHC II molecules.
But a total absence of MHC molecules would backfire, because natural killer
cells also recognise abnormal or absent MHC I, and would immediately destroy the
impostor. It turns out the trophoblast cells express a 鈥渘on-classical鈥� MHC I
molecule called human leucocyte antigen G. This HLA G, unlike other MHC
molecules, exists in few forms. The HLA G in one individual is very similar to
that in another. So the maternal immune system does not recognise fetal HLA G as
foreign鈥攏atural killer cells ignore it.

But even this is not enough to account for the survival of the trophoblast.
Numerous studies suggest that it鈥檚 not simply a case of the developing offspring
avoiding its mother鈥檚 immune response, the mother鈥檚 immune system also manages
to ignore her offspring. Several branches of the immune system appear altered
during pregnancy.

The placenta produces a large number of chemical signals, in the form of
hormones and cytokines, that influence the mother鈥檚 immune response. During
pregnancy, the number of TH2 cells and levels of the cytokines they produce
(type 2 cytokines) are increased, while levels of TH1 cells and type 1 cytokines
are reduced (see Inside Science No. 127).

The causes of this shift are the subject of ongoing research. There is
evidence that when T cells come into contact with the HLA G on the surface of
fetal cells, they are prompted to produce type 2, rather than type 1, cytokines.
Other research has shown that the hormone progesterone (levels of which
are boosted in pregnancy) stimulates the secretion of type 2 cytokines and
inhibits the secretion of type 1 cytokines.

The shift in immune response does have its drawbacks for the mother. Pregnant
women seem more susceptible to some infections, such as malaria. Natural killer
cells target unicellular parasites and are activated by type 1 cytokines, but a
recent study of malarial infection in pregnancy failed to find a single one in
the maternal tissue of the placenta.

Another mechanism may be at work to protect the developing fetus. In pregnant
mice, it has been shown that there is a reduction in the number of T cells that
recognise antigens from the father. Once the mice have given birth, the immune
system returns to normal, and paternal antigens provoke the same response as any
other foreign antigen. This reduction in T cells that target paternal antigens
may have something to do with 鈥減rogrammed cell death鈥濃�攖he T cells could be
committing suicide. Apoptosis, as it鈥檚 known, is highly
regulated鈥攚hich is just as well if our bodies aren鈥檛 to become overrun by
the contents of millions of dead cells. The process is triggered by contact
between a specialised receptor molecule on the surface of one cell, and a
specific molecule on the surface of another cell. One such pair comprises a
receptor called Fas and a molecule called FasL. Once these two
molecules have locked together, the cell displaying the Fas receptor faces
certain death.

It may be that FasL on the surface of trophoblast cells compels any
encroaching activated maternal immune cells displaying the Fas receptor to
鈥渃ommit suicide鈥�. The effect of this would be for the mother鈥檚 immune system to
ignore paternal antigens during pregnancy, but revert to attacking them
afterwards. A similar system is thought to protect other sites where the immune
response is toned down, such as the outer surface of the eyes, where foreign
antigens are constantly present.

But the role that Fas/FasL interactions play in fetal survival is far from
certain. A study published earlier this year showed that genetically modified
mice whose cells never display FasL still manage to have successful
pregnancies鈥攃asting doubt on whether the system really has a central
protective role.

Just the same, there can be no doubt that immunity has a part to play in a
pregnancy鈥檚 success or failure. Indeed, the chances of success are improved if
the prospective mother鈥檚 immune system has become familiar with the prospective
father鈥檚 antigens, training it to ignore them. So women who have cohabited with
their partners for a lengthy period before conceiving are less likely to suffer
鈥�pre-eclampsia鈥溾�攈igh blood pressure during pregnancy, which is
associated with a higher risk of miscarriage. If the couple previously used
condoms, which prevent the paternal antigens in seminal fluid from getting
through, the advantage is lost. But if the couple have practised oral sex there
is an even greater advantage. Immunologists use a similar, more clinical,
approach to treating recurrent miscarriage, by immunising prospective mothers
with paternal antigens.

A better understanding of the remarkable ability of the fetus to escape
immune detection could have benefits in medicine. There may be situations where
doctors want to alert the immune system to a threat it would otherwise
ignore鈥攑erhaps encouraging it to attack a tumour. Conversely, if doctors
could mimic the immune tolerance seen in pregnancy, they could persuade the body
to ignore a transplanted organ.

Let鈥檚 look at the implications for cancer first. It鈥檚 a strange comparison,
but both tumours and fetuses must evade immune surveillance to develop. Perhaps
understanding how the fetus avoids destruction could help explain how cancers do
the same.

The immune system can detect the tiniest differences between antigens, so
that even when a person鈥檚 own antigens change subtly, they too are identified as
foreign. So it does its best to destroy cancer cells鈥攚hich are quite
distinct from normal cells. Patients with AIDS, whose immunity is dramatically
reduced, are almost uniquely susceptible to a type of cancer called 碍补辫辞蝉颈鈥檚
sarcoma. And patients taking immunosuppressive drugs after receiving a
transplant are also at greater risk of developing certain cancers.

In healthy individuals, potentially cancerous cells are destroyed by the
host鈥檚 immune system before a tumour is detectable. There鈥檚 an association
between increasing age and increasing susceptibility to cancer. This may be
because genetic mutations accumulate over a person鈥檚 lifetime. On the other
hand, it may be that the weaker immune system of older people fails to detect
and destroy tumours. If we can understand better how the fetus evades
destruction, we may learn how tumours do the same.

And what about the implications of fetal survival for transplant medicine?
Enormous progress has already been made in transplant techniques鈥攌idney
transplants are almost routine, and heart and lung transplants are usually
highly successful. But the overriding obstacle remains the immune system, for
two reasons. First, the need to find a good tissue match between donor and
recipient limits the already sparse choice of donor organs for a given patient.
Secondly, once the organ has been transplanted, the patient faces a lifetime
taking anti-rejection drugs, and may require a further transplant years later
due to long-term immune rejection of the organ.

If immunosuppressive therapy were improved, matching of donor organs wouldn鈥檛
have to be so precise, boosting availability. Could the right mix of cytokines
somehow switch off the immune system to give a transplanted kidney a better
chance? A fuller understanding of the immune system could also provide clues for
treating conditions in which the immunity plays a significant role, such as
cancer. We all have reason to be grateful for the way our mother tolerated us as
we grew in the womb. If one day you have the misfortune to develop cancer or
need a transplant, you could be more grateful still.