Immunology of Pregnancy I and II

Question 1

A young woman had her 9th consecutive miscarriage. Her marriage broke down shortly afterwards. Within months of finding a new partner, she conceived again and the pregnancy went without a hitch.

Why may this have happened?

Answer 1

• The foetus is allogenic; its genes are half paternally and half maternally derived. It has paternal type antigens, which will be foreign to the mother (like a transplanted organ). These may be recognised as foreign by the immune system.
• Woman’s immune system took offence to the first choice of partner by over-reacting to the tissue carrying his genes and expelling the foetus.
• Infertility, recurrent miscarriage, premature delivery and a dangerous complication of pregnancy, pre-eclampsia, may be strongly linked to immunological abnormalities.

Question 2

The foetus is allogenic. It has paternal type antigens, which will be foreign to the mother (like a transplanted organ). These may be recognised as foreign by the immune system. Yet…..350,000 babies are successfully born every day

How does the baby manage to avoid the mothers immune system?

Answer 2

• Half of the fetal genome derives from the father but, unlike a mismatched organ transplant, it isn’t normally rejected.
• The maternal/fetal interface is central to overcoming these problems. This interface occurs at the placenta.

Question 3

What are the immunological problems to solve during pregnancy?

Answer 3

1) Fetal tissue is half foreign – has to be protected from rejection. The foetus is allogenic. It has paternal type antigens, which will be foreign to the mother (like a transplanted organ). These may be recognised as foreign by the immune system. However, in evolutionary terms, the mother’s immune system and her immune defence must be sufficient during pregnancy to ensure survival of the mother. This is the second immunological consideration.
2) Mother’s immune defence must be sufficient during pregnancy to ensure survival
3) Fetus often immunologically immature at birth – must have maternal antibodies to ensure survival

• The maternal/fetal interface is central to overcoming these problems. This interface occurs at the placenta.

Question 4

Describe the structure of the placenta.

Answer 4

• The foetus is attached to the placenta via the umbilical cord
• Note the myometrium and the decidua. There are chorionic villi and foetal blood can be seen. Pools of maternal blood can be seen within the intervillous space. This is filled by spiral arteries coming from the maternal blood flow.
• This interface at the placenta, where it is attached to the uterine wall, is key for considering where the maternal immune system might be able to be in touch with cells from the foetus.

Question 5

What are the three locations where the mother’s and baby’s cells are in direct contact (maternal-foetal interface)?

Answer 5

1. Syncytiotrophoblast layer covering the placenta is bathed in maternal blood
2. Invading trophoblast come into contact with decidual immune cells
3. Invading trophoblast come into contact with decidual blood vessels

• Recapping from the placentation lectures, the cytotrophoblasts can differentiate along two pathways; they can fuse to become syncytiotrophoblasts or differentiate into extravillous trophoblasts.
  SYN = syncytiotrophoblasts
  EVT = extravillous trophoblasts
• The EVT invade into the wall of the decidua and anchor the placenta into here.
• The floating villi (FV) can be seen next to the anchoring villi (AV).
• The chorionic villi are bathed in a large pool of maternal blood, which is supplied from the maternal blood vessels showing the spiral arteries (SA).
• There are three interfaces to consider; the syncytiotrophoblasts lining the chorionic villi come into contact with the maternal blood in the intervillous space, the invading extravillous trophoblasts come into contact with infiltrated decidual maternal immune cells and invading trophoblasts coming into contact with decidual blood vessels (maternal blood in the spiral arteries after they have remodelled those vessels).
• These images also illustrate why it is so difficult to study the immunology of a human pregnancy. Most immune cells in the peripheral blood are easily accessible by taking a blood sample, but it is very difficult to investigate ongoing human pregnancy looking at where the blood is situated in these interfaces. A lot of the research that has been carried out has been done using animal models, where whole pregnancies can be studied. However, the physiological placentation is very different in, for example, the mouse compared to the human. The only way it is really possible to access this maternal-foetal interface in human pregnancies is from studying cells isolated from first trimester surgical terminations of pregnancy. Unfortunately, in these situations, the outcomes of these pregnancies are unknown (would it have developed, would it have been normal, what normal state would the immune system be in?)

Question 6

Thinking about the maternal-foetal interface, describe the syncytiotrophoblast layer.

Answer 6

1. Syncytiotrophoblast layer covering the placenta is bathed in maternal blood
   - The cytotrophoblasts have fused to become a multi-nucleated layer, which is the syncytiotrophoblast. Its role is to form a barrier and it has endocrine functions too. Also, it expresses a lot of specialised transport proteins for gas and nutrient exchange from the maternal blood. There are foetal blood vessels within here (can be seen within the cross section). There is exchange between the mothers blood (shown in brown) and the foetal blood vessels across the layer. The syncytiotrophoblasts are in direct contact with the maternal blood, which will contain immune cells.

Question 7

Thinking about the maternal-foetal interface, describe the extravillous trophoblast that are in contact with decidual immune cells.

Answer 7

2. Invasive extravillous trophoblast are in contact with decidual immune cells
   - The extravillous trophoblast are differentiated fetal cells which invade into the maternal decidua to transform maternal spiral arteries.
   - The second interface to consider is where the invasive extravillous trophoblasts are in contact with the decidua immune cells.
   - Can see the chorionic villi and the invasive trophoblasts are coming down into the decidua and the myometrium. These invasive extravillous trophoblasts are similar to cancer cells in the way they can move and invade through tissue. As they are coming through, they are heading towards the maternal spiral arteries. As they come through the decidua, they encounter this large infiltration of maternal immune cells.
   - The trophoblast cell (T) can be seen coming through.
   S = stromal cell (structural, fibroblast-like cells)
   There are a lot of maternal immune cells that have infiltrated into the decidua. K = natural killer cells, M = macrophage, L = lymphocyte. All of these maternal immune cells could potentially recognise the trophoblast cell, which is foetally derived and will express some of the paternal antigens, as foreign. Trophoblast cells have to successfully navigate their way through this large infiltration of maternal immune cells, such that they are able to reach the spiral arteries where they have important jobs to do.

Question 8

Thinking about the maternal-foetal interface, describe the extravillous trophoblasts that are in contact with decidual vascular cells.

Answer 8

• The extravillous trophoblast are differentiated fetal cells which invade into the maternal decidua to transform maternal spiral arteries.
• The third area where there will be direct contact with the foetal trophoblast cells and the maternal blood cells is when they reach the spiral arteries. As the extravillous trophoblasts invade, they are heading towards these maternal spiral arteries. The spiral arteries in a non pregnant uterus are very tightly coiled (spiral in nature). It is these blood vessels that are supplying all of the maternal blood to the intervillous space. As the foetus and placenta grow, this demand is increased, so a much larger volume of blood needs to be supplied. This is brought about by remodelling the spiral arteries so that they change from a low flow, high resistance vessel to a high flow, low resistance vessel. The initial stages of this occur early in pregnancy alongside decidualisation and are termed the trophoblast-independent remodelling stage. It is important to note that some of the immune cells have an important role to play in this too. This may help to explain why this large infiltration of cells is needed. Part of the trophoblast-independent remodelling will be carried out by signals from some of the immune cells present in the decidua. When the extravillous trophoblast cells reach the vessel, they cause a loss of vascular smooth muscle layer and a temporary loss of the endothelial layer. The lumen of the vessel becomes much larger and loses its contractile properties. This is termed trophoblast-dependent remodelling and allows a much increased blood supply to go to the intervillous space. The trophoblasts start expressing markers of the endothelial cells and actually replace the endothelial cells lining the spiral arteries. These are foetal cells that will then be in direct contact with any of the maternal blood that is travelling through the spiral arteries. This is the third interface.

Question 9

Which immune cells are present at the maternal-fetal interface?

Answer 9

• Large infiltration of immune cells

1) Decidua
- >40% decidual cells are leukocytes in early pregnancy
- Of these, approximately 70% are NK cells (subpopulation of cytotoxic lymphocytes) = function by cell killing or regulatory cytokine production. The predominant type of immune cell is the natural killer cell.
- approximately 20% are macrophages
- T and B cells make up the remaining 10%

2) Intervillous space and spiral arteries
- the immune cells that are present will be the same as the ones circulating in the peripheral maternal blood.

Question 10

Thinking about the maternal immune cells, what are decidual natural killer cells?

Answer 10

• dNK cells are different to peripheral blood (pb)NK cells
• Their pattern of receptor expression is unique and they are identified by CD56hiCD16lo
• They have been identified as being essential to pregnancy in the mouse and they may play a role in human decidual remodelling through the cytokines which they secrete
• The predominant type (70%) is the decidual natural killer cells. They are very different to peripheral blood NK cells. They have a unique pattern of receptor expression and are usually characterised by high expression of a protein CD56 (peripheral blood NK cells have a low expression of this protein on their cell surface).
• The key here is to remember how different they are to peripheral blood natural killer cells. When a natural killer cell is in the decidua, it is no longer a fierce killer cell. Instead, its role is more to produce cytokines that actually encourage some of the important events of placentation, e.g. trophoblast invasion.

Question 11

Thinking about the maternal immune cells, what are macrophages in pregnancy?

Answer 11

• Another immune cell found in the decidua is the macrophage: makes up about 20% of immune cells in the decidua so is the second most abundant immune cell
• dMac have a different phenotype to peripheral blood monocytes
• Broadly, macrophages may be characterised into two phenotypes:
  M1 = pro-inflammatory, secrete TNF-α, IL-6
  M2 = anti-inflammatory, secrete IL-10, VEGF
• Decidual macrophages are more M2-like than M1-like (a bit more anti-inflammatory in their nature)
• The M1 phenotype macrophages are generally proinflammatory in that they secrete cytokines and growth factors that can promote inflammation. M2 macrophages are generally anti-inflammatory in nature and secrete regulatory factors.

Question 12

After understanding how the mother could mount an immune response, what three theories were first suggested by Medford regarding how both of these types of trophoblast (syncytial and extravillous) evade the immune response?

Answer 12

• There have been theories regarding this in scientific literature for almost 70 years. One of the first striking theories was by immunologist PB Medford. His work on acquired immune tolerance was fundamental to the practice of tissue and organ transplantation; went on to win the Nobel prize in 1960.
• Regarding tolerance of the mother’s immune system to the foetus, he proposed three different theories (as listed). His theories still persist today; shown to have some grounding, but he did get some things wrong. Will go through each of these theories to examine the current thinking on the problem.

1) Physical separation of maternal and fetal tissues
2) Antigenic immaturity of fetal tissues
3) Mother is immunologically inert

Question 13

Is there a physical separation of maternal and foetal tissues?

Answer 13

• Firstly, in terms of trophoblasts evading the immune response, a physical separation of maternal and foetal tissues was proposed.
• The foetus is separated from the mother by the foetal trophoblast cells, so only needs to be the foetal trophoblast cells that have some tolerance to the mother’s immune system.
• The foetal and maternal circulation are separated and maternal cells can’t reach the foetus.
• However, in humans, immunoglobulins can cross into the foetal blood via a placental transport mechanism. Therefore, immunoglobulin G directed against foetal antigens could also be transferred. This is necessary for the foetal immunity to bacteria and viruses etc in the first few weeks of neonatal life.
• Why doesn’t the baby get harmed by this? Most foetal blood group and HLA antigens (the histocompatibility antigens) are so widely distributed on both foetal tissues/cells and within the amniotic fluid that any IgG passing in is either mopped up or diluted out, so no tissue damage occurs as there are not enough maternal antibodies left to bind to any of the foetal cells.
• Many fetal antigens are also present as soluble forms in the fetal blood and amniotic fluid - IgG would be mopped up by free soluble antigen.
• The foetus is separated from the mother, but the foetal trophoblast cells are not separated from the maternal immune cells.

Question 14

Is there antigenic immaturity of fetal tissues?

Answer 14

• The second theory, in terms of trophoblasts evading the immune response, was that the foetal tissues are antigenically immature.
• Histocompatibility antigens are targets for rejection
• MHC haplotypes inherited from both parents and are co-dominantly expressed
• To recap on the histocompatibility antigens, which are the targets for rejection, it is important to think about the MHC molecules. There are different classes of these.

1) Class Ia (classical MHC) = HLA-A, B and C.
- They are involved in presenting antigens to CD8+ T cells
- can interact with NK cells
- importantly, these exist in a highly polymorphic form (exist in many forms).
2) Class Ib (non-classical) = HLA-E, HLA-F, HLA-G
- it is important to remember that they are minimally polymorphic.
3) Class II = HLA-DP, HLA-DQ, HLA-DR
- responsible for presenting antigen to CD4+ T cells

• Focussing on MHC expression by trophoblasts, Syncytiotrophoblasts, which can be seen lining the chorionic villi, lack both MHC Class I and II antigens
• Extravillous trophoblasts lack Class II but express an unusual combination of MHC class I antigens = HLA-C, HLA-E and HLA-G (non-classical). They are not antigenically inert; they, particularly the extravillous trophoblasts, are expressing things that could be recognised by maternal immune cells.

Question 15

Is the mother immunologically inert?

Answer 15

• His final theory was that the mother may be immunologically inert.
• The mother has to be able to respond to acute or chronic infections
• Maternal blood in pregnancy is able to respond immunologically to the fetus and fetal cells are detectable in the maternal blood, BUT Pre-sensitisation to paternal antigen does not prevent pregnancy
• There is neither a generalised or specific depression of maternal immune responsiveness (mother has to still continue to fight off infections)
• However, the quality of the maternal immune response may be what differs
• Studies in mice show that pregnant mice tolerate tumours grafted from the father, but not after delivery of the pups (when it is rejected). They will not tolerate tumours grafted from another mouse. This means that something is going on that is modulating the immune response in pregnancy. This is probably caused by some of the hormone levels that are crucial in suppression of these responses.
• A local immune regulation is more likely

Question 16

Moving on from Medford’s theories and thinking of more current immune evasion in the placenta theories, these are things that could be happening at a local level (instead of suppression of the mother’s whole immune response).

What are the more current theories of immune evasion in the placenta?

Answer 16

• Role for natural killer cells in the decidua
• Selective local induction of programmed cell death in maternal immune cells
• Alteration in the cytokine balance (the cytokine balance is important between a pro- and anti-inflammatory environment)
• Local indoleamine 2,3-dioxygenase synthesis (There is evidence that the enzyme 2,3-dioxygenase is important)
• Complement regulatory proteins

Question 17

Why don’t natural killer cells in the decidua attack the foetus?

Answer 17

• dNK cells are different to peripheral blood (pb)NK cells
• Their pattern of receptor expression is unique and they are identified by CD56^hi CD16^lo. DNK cells are very different to (pb)NK cells with their high levels of expression of CD56.
• They have been identified as being essential to pregnancy in the mouse and they may play a role in human decidual remodelling through the cytokines which they secrete
• They are natural killer cells which can attack materials that they recognise as foreign, but they don’t attack the fetus. This is due to the receptors expressed on the surface of NK cells that interact with MHC molecules expressed on the surfaces of the trophoblast cells.

Question 18

How does expression of HLA-C, -E and -G by EVT help immune evasion?

Answer 18

• By binding to receptors on NK cells
• The key to this is the expression of these MHC molecules (HLA-C, -E and -G). These can interact with receptors on the natural killer cells. There are three types of natural killer cell receptors to consider.
• NK cell receptors =
  1) Killer-cell immunoglobulin-like receptors (KIRs)
  2) CD94/NKG2 receptors
  3) Leukocyte immunoglobulin-like receptor (LILRs)
• There are both inhibitory and activating members of these families of receptors

Question 19

What happens when trophoblasts interact with NK cells?

Answer 19

• Binding of HLA class I molecules on trophoblasts to inhibitory NK cell receptors inhibits the cytotoxic action of the NK cell, therefore the trophoblast is not attacked. It dampens the response of the NK cells. Therefore, it inhibits its cytotoxic action and prevents the trophoblast from being attacked.
• The lists show the proteins expressed on the trophoblasts and the receptors on the natural killer cells.
• Inhibitory NK receptor =
  1) CD94/NKG2A
  2) KIR2DL / S1
  3) LILRB
• Trophoblast =
  1) HLA-E
  2) HLA-C
  3) HLA-G
• By the trophoblast expressing these proteins, it quietens down the natural killer cell and makes it less cytotoxic.

Question 20

What evidence supports the role of natural killer cells in the decidua as a current theory of immune evasion in the placenta?

Which HLA has higher affinity for the inhibitory receptor than for any of the activating receptors?

Answer 20

• Inhibitory receptors are expressed at higher levels in uterine natural killer cells than in peripheral blood NK cells. Therefore, they are more predisposed to receiving an inhibitory signal.
  uNK = uterine NK cells
• HLA-E has higher affinity for the inhibitory receptor than for any of the activating receptors
• More uNK cells found in women with a history of recurrent pregnancy loss
• BUT Trophoblast HLA molecules can also bind to activating NK cell receptors. This may be really important to alter the natural killer cell cytokine repertoire that they are producing. Some of the things that they are producing may well contribute to how the trophoblast cells behaves, e.g. producing growth factors that make them invade more. There are two sides; there are the inhibitory receptors that stop them from becoming the killing cell and then the activating receptors which binding can alter the cytokine repertoire.

Question 21

Why has there been particular interest in HLA-G?

Answer 21

• It exists in different isoforms which are alternatively spliced and it has limited polymorphisms (not many forms). The most important form is generally soluble HLA-G (sHLA-G1). It is thought to be particularly important in foetal development as its expression is so restricted throughout the body and, in the placenta, it is only expressed in the extravillous trophoblast.

Question 22

Is there a role for soluble HLA-G?

Answer 22

• Soluble HLA-G can be released from trophoblasts
• In vitro studies have shown that sHLA-G can induce apoptosis in maternal T cells
• This may be an additional way of protecting trophoblasts from attack
• There have been some studies in the IVF field that suggest that there is an association between the presence of soluble human leukocyte antigen G (sHLA-G) in human embryo culture supernatants and the success of implantation.
• This could induce apoptosis in cells and there are other signalling pathways that are thought to be involved in selective local induction of programmed cell death in the maternal immune cells.

Question 23

What experimental evidence supports the theory regarding selective local induction of programmed cell death in maternal immune cells?

Answer 23

• There is experimental evidence that trophoblasts can induce programmed cell death (apoptosis) in maternal immune cells.
• Apoptosis (programmed cell death) is characterised by cell shrinkage, nucleus reorganising, DNA fragments, membranes bleb and cell fragments into membrane bound apoptotic bodies. Apoptosis is also characterised by the formation of apoptotic bodies. The image shows the cell undergoing apoptosis. Apoptosis is a neat method of cell death; it does not generate lysis of the cells and release of lots of proinflammatory factors (as necrosis may induce). Often, apoptotic cells are subsequently phagocytosis by cells such as macrophages.
• Regulation of apoptosis depends on a balance between pro- and anti-apoptotic factors.
• In the context of the maternal immune cells in the decidua, there is evidence that these mechanisms listed are important =
  1) Fas-Fas L
  2) TRAIL-TRAIL R
• This could induce apoptosis in cells and there are other signalling pathways that are thought to be involved in selective local induction of programmed cell death in the maternal immune cells.
• Researchers have found that these are important mainly by looking at mice where they have knocked out one of these killing mechanisms. In these mice, pregnancy is characterised by extensive infiltration of immune cells; they are prone to small litters and resorption of the foetus. The ability of the trophoblast cell to kill off any attacking maternal immune cell seems to be important.

Question 24

Summarise the pathways leading to apoptosis of the maternal immune cell when it comes into contact with the foetal trophoblast cell.

Answer 24

• These are some of the pathways linked to death receptor signalling that are thought to be important.
• Trophoblast cells can either express on their cell surface a protein called Fas ligand or they can secrete a soluble form of this protein (sFasL).
• This can bind to a receptor, Fas, on the surface of maternal immune cells. This triggers a cascade of signalling through Fas-associated death domain protein (FADD). This activates caspases, essentially executioner enzymes of apoptosis. The interaction of this foetal trophoblast through FasL/Fas can lead to cell death in the maternal immune cell, thereby preventing that cell from attacking the foetal trophoblast cell.
• The other pathway that is important is through TNF-related apoptosis-inducing ligand, which can bind to two different death receptors on the surface of immune cells and subsequently triggers the same pathway leading to apoptosis.
• Researchers have found that these are important mainly by looking at mice where they have knocked out one of these killing mechanisms. In these mice, pregnancy is characterised by extensive infiltration of immune cells; they are prone to small litters and resorption of the foetus. The ability of the trophoblast cell to kill off any attacking maternal immune cell seems to be important.

Question 25

Why is the Th1 / Th2 balance in pregnancy a current theory of immune evasion in the placenta?

Answer 25

One of the initial theories behind foetal tolerance was thinking about the balance between Th1/Th2 signalling in pregnancy.

• T cells differentiate into Th1 or Th2 cells in response to signals given during antigen presentation
• A Th1 type reaction in the placenta mainly generates a pro-inflammatory response that activates T cells and natural killer cells. This type of reaction is correlated with miscarriage of pregnancy. E.g. IFNg, IL-2.
• A Th2 type reaction generates non-inflammatory signalling that is consistent with survival of the foetus, e.g. IL-4, IL-6, IL-10, IL-13.
• It is thought that the trophoblast cells are producing cytokines and hormones that act to promote a TH2 balance (they regulate this balance to make the situation more non-inflammatory).
• It is likely a bit more complicated than this; some pro-inflammatory signals are thought to be necessary for the initial implantation and trophoblast invasion stage. It may be that initially in the first few weeks of pregnancy, a more pro-inflammatory environment is needed and then this must be counteracted shortly afterwards by a shift to an anti-inflammatory (more Th2) balance.
• Perhaps in some pregnancy disorders, the shift from the pro-inflammatory to the anti-inflammatory does not take place at the right time.
• It is now thought that the role of the regulatory T cell may be important too. There are not that many T or B cells present (smallest population of maternal immune cells). However, skewing the nature of the T cell response to active tolerance (rather than active rejection response) seems to be important.
• T helpers = Th1, Th2, Th17, Treg
• The T regs seem to be key in this. It has been shown that sufficient T regs are needed in the endometrium for implantation and pregnancy. These cells are more anti-inflammatory and immune suppressive.
• Tregs CD4+CD25+ = anti-inflammatory, immune suppressive
• They can act on other immune cells to produce suppressive cytokines, such as TGFb and IL10.
• This balance of getting sufficient regulatory T cells is regulated by the overall cytokine balance within the endometrium here and signalling from the other cells that are present will regulate this. Also, the correct phenotype of the other endometrial leukocytes allow the stimulation of activating the regulatory T cells. Some of the hormones, such as oestrogen and progesterone, are known to stabilise the phenotype of regulatory T cells and also priming of regulatory T cells by male partner seminal fluid has been shown to stimulate the expansion of the endometrial Treg population.
• It is not known exactly what these cells are doing, but they seem to be required to help with the tolerant environment in the decidua.

Question 26

What is indoleamine 2,3-dioxygenase?

Answer 26

• Enzyme that catabolises tryptophan (an essential amino acid)
• Synthesised and secreted by syncytiotrophoblasts
• Shown to be essential for successful pregnancy (using knockout mice studies)
• IDO may break down tryptophan in maternal T cells in the decidua
• This can reduce or inhibit immune responses

Question 27

Why is local indoleamine 2,3-dioxygenase synthesis in pregnancy a current theory of immune evasion in the placenta?

Answer 27

• Another theory revolves around this enzyme, indoleamine 2,3-dioxygenase.
• Enzyme that catabolises tryptophan (an essential amino acid). Synthesised and secreted by syncytiotrophoblasts.
• T cells are very sensitive to the amino acid tryptophan and any decrease is likely to suppress their proliferation. IDO may break down tryptophan in the maternal T cells in the decidua, essentially starving them of one of their essential amino acids. This can subsequently reduce or inhibit the immune response. This may be a way of the syncytiotrophoblast targeting some of the T cells it could come in contact with.

Question 28

Why is expression of complement regulatory proteins pregnancy a current theory of immune evasion in the placenta?

Answer 28

• The expression of complement regulatory proteins seems to be important.
• The complement system is made up of a large number of different plasma proteins that interact with each other to induce inflammatory responses and that makes the cell more susceptible to opsonisation or recognition by phagocytic cells. There are three different pathways = classical, alternative and lectin; they all converge upon the same enzymatic pathway which leads to the proteins that are the effector functions of complement.
• In normal pregnancies, excessive complement activation is prevented by complement regulatory proteins that are highly expressed on trophoblast membranes(MCP, DAF, and CD59)
• CD46 = MCP – membrane co-factor protein
  CD55 = DAF – decay accelerating factor
  CD59 = MAC-IP- MAC inhibitory protein
• This prevents cell lysis
• It is known components of this system are important again, as mouse models have shown that lacking one of these key complement regulatory proteins means that the embryos do not survive. There are three proteins that have been studied, as listed = membrane co-factor protein, decay accelerating factor and MAC inhibitory protein. The diagram highlights where they interact in the cascade.
• Complement regulatory proteins are particularly highly expressed in trophoblast cells. In humans, there is some evidence that low levels of these proteins are linked with pregnancy failure and also preterm birth. Perhaps, this is because it is creating an inflammatory state in the uterus.

Question 29

Summary of the theories of immune evasion at the maternal – fetal interface

Answer 29

• This summary highlights some of the current thoughts as to how the trophoblast cells invade the maternal immune system at the maternal-foetal interface.

1) Non-classical expression of HLA antigens may help trophoblast evade the immune response
2) Forget what you know about immune cells: DECIDUAL immune cells, particularly the NK cells, are different! NK cells are very different to what is found in the peripheral blood; they are a very specialised type of cell (may actually be helpful rather than having a killing role).
3) Trophoblasts seem to have a few mechanisms by which they may promote death of some immune cells
4) A Th2 balance is promoted. Having the cytokine balance that is secreted by the immune cells and the local environment as generally a more anti-inflammatory (Th2 type balance) is promoted.
5) Production of IDO by syncytiotrophoblasts may inhibit T cell responses
6) It would also appear that some of the complement regulatory proteins are important too

• Some of these pathways will be more important for syncytiotrophoblasts interacting with maternal blood and some more for extravillous trophoblasts interacting with the infiltrated immune cells within the decidua.

Question 30

Is there an immunological basis for disorders of pregnancy?

Answer 30

• This is a controversial and heavily-researched area
• This is an area where there is a lot of research going on at the moment and there is quite a bit of controversy over the best theory. We have touched on a few associations, perhaps excessive complement activation with preterm birth, more of a Th1 (pro-inflammatory) profile being associated with miscarriage.
• Another association is the idea that some incidences of pre-eclampsia and miscarriage may have a basis in maternal-foetal immunological mismatch; a mismatch somehow between how the immune cell and the trophoblasts communicate through their receptors.

Question 31

What happens to the spiral arteries in a pre-eclamptic pregnancy?

Answer 31

• A spiral artery in a non-pregnant state is highly coiled. The trophoblast cells invading through and remodel the spiral artery in a normal pregnancy, as far as into the myometrium.
• There’s a lot of controversy in the pre eclampsia field, but it is agreed that there is a failure to remodel the spiral arteries sufficiently in a pre-eclamptic pregnancy. The remodelling event just about comes down to the decidua-myometrium boundary. This means that there is less blood flow coming into the intervillous space through the spiral arteries (probably a turbulent blood flow coming through), so there is a less steady supply of oxygen and nutrients to be exchanged across the chorionic villi and into the foetal blood vessels. For this reason, it is often associated with foetal growth restriction, where the babies are smaller than they should be for their gestational age.
• Great Obstetric Syndromes (GOS) = pre-eclampsia, stillbirth & fetal growth restriction (FGR).

Question 32

What evidence suggests there is an immunological component to pre-eclampsia?

Answer 32

• Women have a lower risk of pre-eclampsia with a different partner for a second pregnancy.
• The situation with donor egg pregnancies is interesting, because these are pregnancies which are totally non-self to the mother (rather than half-self, like a normal pregnancy) and they are at higher risk of developing pre-eclampsia.
• There is also some evidence that prolonged exposure to paternal semen may lower the risk of pre-eclampsia.
• It is an important condition because it affects about 5 - 8% of pregnancies.

Question 33

What evidence suggests the role of NK cells in pre-eclampsia?

Which combinations of paternal HLA-C and maternal KIR receptors are sub-optimal for implantation?

Answer 33

• Trophoblast cells interact with NK cells through their receptors. The important one on the trophoblast cell is HLA-C. HLA-C is the only one of the three trophoblast MHC molecules that is highly polymorphic (exists in different forms). The receptors on the natural killer cell (the KIR receptors) can be type A or B, while the HLA-C can be type C1 or C2.
• Some studies from a group in Cambridge have looked at an association between which receptors and HLA- C genotypes are linked to miscarriage, reproductive failure and pre-eclampsia. They found that if the match is KIR-A and HLA-C2, then the risk of pre-eclampsia is increased. Those interactions occurring are somehow bad for placentation. One of their theories is that this may be because this KIR-A/HLA-C2 interaction leads to less cytokine production that is needed for the trophoblast cells to invade properly. If the trophoblast cells are not invasive enough, then that means they may not reach the spiral artery in order to bring about enough remodelling. It is possible that some combinations of paternal HLA-C and maternal KIR receptors are sub-optimal for implantation and will contribute to diseases such as pre-eclampsia.
• To summarise, dNK receptors may be Type A or B, and trophoblast receptors may be type C1 or C2
• If the match is KIR-A and HLA-C2, pre-eclampsia risk is increased
• This may be because this interaction leads to less cytokine production which helps the trophoblast invade
• This brings us back to the clinical case of the women who had recurrent miscarriages until she changed partner. Perhaps their combination of HLA-C and KIRs was not the optimal one.

Question 34

How can we study normal remodelling and determine what is going wrong in PE/FGR?

Answer 34

• Work has been carried out at George’s to try and look at what events occur in early human pregnancy.
• As mentioned previously, a lot of the information regarding pregnancy has had to come from animal studies. Studying human pregnancy is tricky, as shown in a timeline of the gestation of human pregnancy. This timeline highlights when studies can be carried out in human pregnancy. Blood samples from pregnant women can be taken, but this would like their peripheral blood immune cells (rather than their decidual immune cells).
• Tissue can be obtained from the placenta and decidua in women who are having surgical terminations of pregnancy (~between 6-14 weeks) to look at the immune cells there. A lot of the studies where they have been investigating, for example, the role of what NK cells, they have isolated cells from the tissue at this stage. The limitations include not having very much tissue and that the outcome of the pregnancy is unknown at this time (e.g. foetal growth restriction, pre-eclampsia, stillbirth) because it is in the first trimester of pregnancy. However, it is a stage where all of the remodelling processes are active. This means that it is possible to look at what the cells could actually be doing at that time.
• It used to be possible to obtain mostly placental and maybe decidual samples from chorionic villus sampling that was carried out for genetic testing (between 10-15 weeks). This is not carried out a great deal anymore. There is very limited tissue obtained here, but these pregnancies are ongoing so the outcome of the pregnancy may be followed up (whether they developed a complication or not).
• When women give birth (~35-40 weeks), the whole placenta can be seen and/or uterine biopsies can be taken (some women will have caesarean sections); can look at the myometrium and decidua. There is a known outcome here (it would be known whether the woman had pre-eclampsia) and there is lots of tissue to be used. However, the problem is the relevance to the events taking place in the first trimester. What happens in the first trimester is what sets up the pathology of conditions like pre-eclampsia.
• The problem is that the clinical symptoms of pre-eclampsia, e.g. proteinuria and maternal hypertension, don’t really begin until beyond 20 weeks. Looking at the stage with tissue from TOP, the next step was finding a way to predict whether a woman might go on to subsequently develop pre-eclampsia.

Question 35

What results were obtained from first trimester studies of pregnancies with spiral artery related pathologies?

Answer 35

• This has been addressed by taking first trimester tissue from pregnancies characterised by uterine artery doppler resistance indices. The clinicians can scan the uterine artery and an event where there is normal remodelling (the remodelled spiral artery comes down and is wider) can be related to a percentile from the uterine artery Doppler resistance index. When there is abnormal remodelling (a higher resistance spiral artery), it will relate to a higher resistance reading in the uterine artery. Various studies have reported on the correlation between uterine artery blood flow measurements obtained by Doppler ultrasound in early pregnancy and the subsequent development of pathologies that relate to poor remodelling. A high resistance index would represent a condition where there was higher resistance uteroplacental circulation, such as that seen with less spiral artery remodelling.
• Based on data from almost 10,000 ongoing pregnancies, it is known that woman with a high resistance index in their Doppler scan have a five fold higher risk of pregnancy complications. PE risk = 2.8% with normal RI, 15% with high RI. PE/FGR/
  stillbirth risk = 4.9% with normal RI, 24% with high RI (Leslie et al. 2015 Am. J. Path). It is not a good clinical predictor because most of those women will not go on to develop complications, but is has been found very useful in first trimester studies as essentially all of the tissue obtained can be classified into a high or lower risk pregnancy of developing these complications.
• High RI group = decreased endovascular trophoblast invasion and artery plugging (Prefumo et al. 2004 Hum. Reprod. 19(1):206-209).
• It has previously been shown that there are less trophoblast cells present in the spiral arteries of the higher risk group. These trophoblasts have impaired invasion and survival mechanisms. Immune cells can also be isolated from these tissues.

Question 36

Are immune cells different in the high RI and normal RI pregnancies?

Answer 36

• Using the decidua of these pregnancies (characterised as normal or high resistance indices to indicate a lower or higher risk of developing pre-eclampsia), these cells have been isolated (depending on what they express on their cell surface) with very high purity. Comparisons can be carried out in the lab with different properties of these cells = comparing those at a higher and a lower risk of developing pre-eclampsia at a time when the pathology of pre-eclampsia is developing.
• Some of the results relating to the immune cells showed that the dNK cells promote trophoblast invasion in a lower risk and normal pregnancy. They secrete factors that help trophoblasts to invade. However, cells isolated from the higher risk pregnancies seemed to lack this ability to be helpful to trophoblast cells.
• When considering the normal spiral artery remodelling, the dNK cells seemed to have a role in promoting spiral artery remodelling in some of the trophoblast-independent remodelling events. However, it was found that they failed to do this when they were isolated from a higher risk pregnancy.
• Some of these differences could be due to the different cell surface receptors or numbers of receptors expressed on these cells.
• Also, what proteins they are secreting could regulate this (have looked at the differences in these areas). dNK cells differ in cytokine secretion between normal and high RI pregnancies.
• Macrophages (dMacrophages) can also be pulled out of this tissue with high purity. They were cultured in the lab with trophoblast cells and it was found that the ones from the higher risk of pre-eclampsia pregnancies actually induced trophoblast apoptosis (appeared to attack the trophoblast). Those from the lower risk pregnancies, however, did not do this.
• Some of these findings are really important because they give us an idea of what might be going on to contribute to the pathology of these obstetrical disorders.
• There is a great deal of extra research that needs to be carried out to look at this.