The Role of NK Cells in Recurrent Miscarriage (Abortion)

Abstract

Recurrent miscarriage is an early pregnancy complication that affects about 1–3% of couples. There are specific characteristics of natural killer (NK) cells associated with miscarriage. In patients with recurrent miscarriage, a lack of inhibition of decidua natural killer cells can be observed, which leads to a more activated state and presentation of NK cell dim that is characterized by higher levels of pro-inflammatory cytokines and cytotoxicity effect. In peripheral blood, a dysfunctional cytokine production by natural killer cells has been also reported, with an increase of interferon-γ levels and a decrease of interleukin-4. Accordingly, there are different population of NK cells such as dim and bright. The lack of balance between these populations can lead to miscarriage. Using flow cytometry, we can detect these populations and propose the treatment too.

1. Introduction

Recurrent miscarriage (RM) is an early pregnancy failure that affects about 1–3% of couples who wish to have children [1]. This common complex problem of early pregnancy can be caused etiologically by genetic, anatomical, hormonal, and infectious factors or none etiologic. One underlying cause of unexplained recurrent abortions (RA) may be a maternal immunological factor that causes to interfere with the maternofetal tolerance [2, 3]. Natural killer (NK) cells are important immune cells that participate in innate and adaptive immunity and have roles in recurrent abortion. They include the majority of immune cells present in the maternal-fetal interface. Obviously, uterus and fetus represent immunologically privileged sites and usually, maternal immune system does not attack non-self, paternal inherited, or fetal antigens. The challenge was observed between innate and adaptive immune cells that mediate immune protection at the maternal-fetal interface in immune-privileged sites of the pregnant uterus [4]. Normal fetal growth requires an adequate supply of maternal nutrients and oxygen to the placenta, which is achieved by trophoblast invasion and spiral artery remodeling [5]. NK cells secrete some cytokines and chemokines, such as vasoactive growth factor, placenta growth factor, insulin-like growth factor-binding protein 1 (IGFBP-1), and VEGF to promote this happening [5]. NK cell in peripheral blood and uterus of women with RM have different populations and higher percentage compared with healthy individuals. Indeed, the different expressions of receptors modulate the cytolytic or immunoregulatory condition [6].

2. The receptors of NK cells in abortion

Several factors such as NK cell receptors, cytokines, and hormones, for example, progesterone are involved in successful pregnancy [7]. A successful implantation requires an accurate balance between activation and inhibition of NK cells that are mediated by some receptors including inhibitory and activatory roles. Moreover, a number of receptors are identified as diagnostic markers.

NK cells are lymphocytes that are defined as CD3 − CD56+ population. Peripheral NK (pNK) cells are phenotypically and functionally different from uterine NK (uNK) cells and < 10% of pNK cells are similar to uNK cells. Depending on the intensity of the CD56+ neural cell adhesion molecule, they are classically divided into CD56bright (high intensity, composed of cooperative and tolerogenic cells) and CD56dim (low intensity, more cytotoxic against tumors or virus-infected cells) NK cells.

A subset of the CD56dim NK cell expresses a high level of CD16 (FcγRIII) as a receptor involved in antibody-dependent cell cytotoxicity (ADCC). These are defined by its cytotoxicity function. In contrast, CD56bright NK cells are lack of CD16 and characterized by the production of immune-regulatory cytokines. These are highly proliferative activities. Furthermore, 90% of pNK cells are CD56dim and CD16+, whereas 80% of uNK cells are CD56bright and CD16 negative [8]. CD16 is defined as a unique receptor that functions independently without the help of other receptors of NK cells and acts after the adaptive immune response. The different expressions of receptors modulate the cytolytic or immunoregulatory condition [9]. Some researchers showed a higher percentage of the subpopulation CD56bright NK cells in women who had live births compared with women who miscarried [1]. Furthermore, some studies revealed there are new subpopulations of CD56bright that play a central role in trophoblast interaction. Measuring the ability of NK cytotoxicity or detection of activation markers such as CD69 or CD107a can be identified as predictive value in miscarriage.

In addition, NK activity depends on the activating and inhibitory receptors that are retrieved from several germline-encoded, mainly derived from three families: C-type lectin-like receptors, immunoglobulin (Ig)-like transcripts, and killer cell Ig-like receptors (KIRs). KIRs are classified according to their structure and function. Each KIR molecule consists of two or three extracellular Ig domains (2D and 3D molecules). KIRs are a set of variable gene polymorphisms (each individual inherits 4–20 genes from each parent) that code for inhibitory or activating receptors expressed in a variegated manner on subsets of NK cells and some T cells. Inhibitory and some activating KIR bind to HLA class I molecules, while other activating KIR binds to pathogen components within HLA class I molecules.

3. NK cell and trophoblast

Tolerating the allogeneic fetus by the maternal immune system is essential to a successful pregnancy. In the fetal-maternal interface, the maternal uterine tissue is in direct relation with the fetal-derived trophoblast cells and decidual natural killer (dNK) cells or uterine natural killer (uNK) cells mediate primarily the immune tolerance microenvironment. Uterine natural killer (uNK) cells are a category of NK cells that are located in the uterus and comprise the largest leukocyte population in the endometrium throughout early pregnancy [10]. The maternal tissue contacts with trophoblast cells through paternal antigens [11]. Trophoblast invasion and spiral artery remodeling are responsible for delivering maternal nutrients and oxygen to the placenta. A major cause of related disorders of pregnancy such as recurrent miscarriage, preeclampsia, and fetal growth restriction is related to diminish trophoblast invasion and vascular reconstruction and it leads to defective placentation [6]. Several studies have revealed that upon interaction with the trophoblast, the activating receptors in dNK cells promote trophoblast invasion and endothelial cell angiogenesis. NK cell secretes cytokines and chemokines such as vasoactive growth factor, placenta growth factor, insulin-like growth factor-binding protein 1 (IGFBP-1), and VEGF. Accordingly, some cytokines such as IL-8, VEGF, SDF-1, and IP-10 and IL-10 promote tolerance microenvironment (Figure 1) [7, 12].

Figure 1.

Difference activation between healthy pregnancy and recurrent miscarriage. The population of NK cells have certain receptors that lead to different function. According to these receptors successful pregnancy or abortion have been happened.

4. NK cells are defined as double-edged sword

Peripheral blood NK (pbNK) cells besides playing a substantial role against viral infections and have a tolerogenic role in pregnancy. Unlike most somatic cells that express classical major histocompatibility complex class I (MHC-I) molecules such as HLA-A or HLA-B, trophoblast expresses only some unique HLA class I molecules, including non-classical HLA-E and HLA-G and the only polymorphic class I, for example, HLA-C [13, 14]. Inhibitory dNK cell receptors, such as killer immunoglobulin-like receptors (KIR), interact with specific MHC-I molecules, limiting dNK cell cytotoxicity and helping to maintain the balance of immune tolerance. However, when the intercellular cell-to-cell communication is disrupted at the maternal-fetal interface, insufficient maturation of dNK cells results in decreased immune tolerance and cytotoxic up-regulation (Figure 2).

Figure 2.

Immune response behaves as double-edge sword: During pregnancy, the growth of the fetuses and protection from pathogen should be balanced.

There are numerous mechanisms to influence NK cell immunity by targeting their ligand-receptor, signaling pathways, chemokines, and cytokine secretion in viruses that affect NK cell activation and recruitment. NK cells in turn respond as follows:

1. Recognizing “the missing self” that viruses avoid T-cell responses by downregulating MHC class I.

2. Recognizing host- and pathogen-derived ligands, by activation of receptors such as NKG2C, NCRs, and NKG2D.

3. Antibody-dependent cytotoxicity by FcγRIII (CD16).

4. Responding to some cytokines such as IL-12, IL-15, and IL-18.

The response to infection in pregnancy was influenced by KIR and HLA variants and depending on these variants dNK cells may eliminate cells infected with pathogens. Some studies revealed that pregnant women with KIR2DS1+ dNK cells activated by ligating with fetal HLA-C2 have a lower chance to exposure pre-eclampsia and increasing the development of placenta. KIR2DL4 is a kind of variant to interact with HLA-G and may have a tolerogenic function. HLA-G can also help infected and cancerous cells escape immune control. Although higher HLA-G levels might modulate the immune response and prevent fetal failure, they can also lead to congenital infections [9].

HLA-E is another molecule that is expressed by trophoblast, maternal leukocytes, and stromal cells, where it interacts with activating CD94/NKG2C and inhibitory CD94/NKG2A, expressed by pbNK and dNK cells. While NKG2C may suppress dNK functions, NKG2A may also promote dNK-cell functions by enhancing NK-cell cytotoxicity [15].

Natural cytotoxicity receptors (NCR) on NK cells include NKp46, NKp30, and NKp44, which regulate host NK cell cytotoxicity and cytokine secretion, as well as pathogen-derived ligands such as hemagglutinin neuraminidases, viral hemagglutinins, and other pathogen components. NKp46, as an NCRs, activates NK cells, while splice variants NKp44 and NKp30 either activate or inhibit. Meanwhile, dNK cells express inhibitory isoforms, and pbNKs express activating isoforms. The switch from activation to inhibition of NKp44 and NKp30 is associated with reduced cytotoxicity and is mediated by exposure of pbNK cells to cytokines present in the decidual environment, such as IL-15, IL-18, and TGF-β. The importance of NKp46 was found in the peripheral blood of women with a history of RM and determined with a decreased percentage of NKp46CD56 bright cells [6].

NKG2D is an activating receptor expressed in the majority of pbNK cells and dNK cells too. It binds ligands induced in stressed cells and therefore alerting immune cells’ function. While these ligands are not found on trophoblast, they can be induced on uterine stromal cells and/or produced by trophoblast in soluble forms to interact with both dNK cells and pbNK cells.

T-cell immunoglobulin domain and mucin domain-containing molecule-3 (Tim-3) is a newly defined regulatory factor that can modulate the balance of Th1 cells/Th2 cells and expresses in some dNK cells, and Tim-3+ dNK+ cells displayed decreased cytotoxicity due to producing less perforin [15, 16].

In the non-pregnant endometrium, uterine NK cells (uNK) are largely inactive but have the ability to differentiate and switch to cytotoxic behavior and perform immune defense against intrauterine infection by pathogens [13, 17]. Yan Whua reported the expression of some inhibitory receptors, including KIR2DL4, NKG2A, and KIR2DL1, was significantly lower in women with RM [18, 19]. A high percentage of CD56 + NK cells that expressed KIR specific for HLA-C was observed in healthy women and a decline of this receptor in RM women [14].

5. The cytotoxicity of pNK and uNK cells

The evaluation of uNK cytotoxicity is difficult, rare, and technically challenging. According to a study done by Trundley and Moffett, 2004, we declared that uNK does not possess the same cytotoxic function ability as pNK. In support of this statement, dNK is unable to make activating cytotoxicity signals that trigger perforin release when interacting with classical cancerous target cell lines, such as K562 [20].

The lysis of K562 leukemic cells can be measured after co-incubation with uNK at effector: target (E: T) ratios ranging from 5:1 to 40:1. Percentage of target cell lysis was determined by either a chromium release assay or flow cytometry assay by CFSE method. Indeed, results in uNK cell cytotoxicity case-control studies performed with uNK isolated from first-trimester decidua showed greater target cell lysis in women with RM compared to controls [15]. However, they suggested that there is more pNK in the endometrium of RM patients than in the endometrium of controls. pNK has more ability to kill K562 cancer cells so the accumulation of pNK in the endometrium of RM patients leads to an improvement in the ability of the total number of NK cells to kill K562 cells. In addition, uNK may be more activated in RM patients and thus have increased cytotoxicity function to kill K562 cancer cells, although it cannot result in the same effect also observed in trophoblastic cells.

6. The cytokine expression of pNK and uNK cell

Since uNK derived from women with RM secrete more type 1 cytokine (e.g., IFN-ˠ and TNF-a) compared with type 2 cytokine (e.g., IL-4 and IL-10, TGF-β), the direction of uNK involvement in RM may not be so clear. Moreover, uNK cells are also an important source of angiogenic growth factors, vascular endothelial growth factor (VEGF)-A, VEGF-C, placental growth factor, keratinocyte growth factor, fibroblast growth factor, and platelet-derived growth factor-BB [21].

Although the studies have clinical heterogeneity and differences in the secretory activity of uNK cells, Ledee et al. (2004, 2005, 2008) using a combination of detection assays demonstrated that impairment of vascular remodeling may be caused by either insufficient or excessive NK cell recruitment to the endometrium in addition to dysregulated cytokine signaling. Furthermore, flow cytometry was used as a more objective and robust way which measures the specific antigens CD45 (leucocyte antigen), CD3 (T-cell antigen), CD56, and CD16 to overcome this source of heterogeneity and perform subgroup analysis. Furthermore, the use of a standardized gating strategy would help to overcome heterogeneity between different studies [22]. Li Zhu et al. have reported a high proportion of circulating activated T lymphocytes, high cytotoxic NK cells, and low population of circulating IL-10 CD56bright NK cells and low IL-4, IL-10, and TGF-β in RM patients [12].

As we mentioned before, some studies revealed there are new subpopulations of CD56bright that play a central role in trophoblast interaction. In humans, dNK1 may engage with trophoblast through KIR. A subset of NKG2C+ dNK1 may provide benefits in secondary pregnancy through immunological development, while dNK2 and dNK3 cells may interact with other cell types in the decidua through the production of other factors such as XCL-1, which can attract both DCs and trophoblast. It is clear that uNK undergoes complex interactions with other cells in the immune milieu of the placental bed along with differentiation to subpopulation. For instance, CD56 + CD94-CD117 + CD127+ type 3 ILC produces IL-22 and IL-8, whereas CD56 + CD94+ ILC-1 may contribute to IFN-ˠ production [23]. This new population of immune cells that are present in the decidua (ILCs) is considered to contribute to producing cytokines pool that interferes in pregnancy. Therefore, more consideration needs to be given to the interaction between uNK and trophoblast cells.

7. Immunotherapy for women with abnormal uNK cell numbers

There are several treatment protocols proposed in literature serving as tools for the management of patients diagnosed with RM that targets the immunological background of these diseases. Proposed treatment options for uNK-related RIF and RM include: Glucocorticoids, Intralipid therapy, Immunoglobulin therapy [15].

7.1 Glucocorticoids

Glucocorticoids could act as anti-inflammatory and immunosuppressant actions. Glucocorticoids have been used to treat women with reproductive failure and high numbers of uNK cells that express the glucocorticoid receptor. This treatment can conduct in the reduction of uNK cell numbers and lead to live birth. They can cause hormonal changes in the fetus. uNK cell-mediated cytotoxicity is sensitive to exogenous glucocorticoids.

7.2 Intralipid therapy

Intralipid is a fat emulsion containing soybean oil, glycerin, and egg phospholipids. Intralipid is used for parenteral nutrition in patients who cannot take food by mouth. The ability of intralipid to suppress the activity of NK cells has been shown in various in vitro studies. Animal and human studies illustrate that intravenous intralipid can increase the rate of continued pregnancy by improving implantation in patients with high levels of NK cells. Intralipid may be used as a new effective treatment option for women with RSA and high levels of NK cells [24].

7.3 Immunoglobulin therapy

Intravenous immunoglobulin (IVIg) which is human IgG prepared from pooled plasma administered during pregnancy is used for the treatment of RM. Several theories have been proposed for its action, including the dampening of NK cell number and activity and modification of cytokine production. IVIg therapy in RSA women increases the expression of inhibitory receptor CD94 on NK cells and serum cytokine and shifts the immune system toward Th2 [25]. The underlying mechanism might be associated with the neutralization of autoantibodies, inhibition of natural killer cells, attenuation of cytotoxicity, and increase of regulatory T lymphocytes [26].

7.4 Active immunotherapy (AI) with allogeneic paternal lymphocytes

It appears a feasible treatment option for recurrent abortion (RA). Intradermal immunotherapy with 30–100 million lymphocytes in one to two applications has proven to be effective, associated with low risk, and cost-effective for the treatment of RA. Positive HLA antibody test after AI in RA predicts a successful pregnancy with a probability of 89% making this test suitable as a predictive test [3]. However, the overall response has been followed by severe allergic reactions and painful [27].

8. Conclusion

In this chapter, we reviewed the association between NK cell activity and RM in a number of studies. For instance, a large systematic review published in 2011 was conducted to detect the relationship between NK cell activity and RM. This study showed that the prognostic value of the percentage or activity of peripheral NK cells in patients with idiopathic RSA is still uncertain, and larger studies were recommended. However, a meta-analysis published in 2014 outlined the relationship between peripheral NK cells and RSA by pooling the data from four different studies and found a significant difference in the activity of peripheral NK cells between women with RSA and those in the control group.

Future research should endeavor to reveal single-cell RNA sequencing in first-trimester pregnancies to evaluate pathological pregnancies and immunogenetic screening can be introduced in clinical procedures. Also, the other new population of immune cells that are present in the decidua such as ILCs considered to contribute to the presence of cytokines pool that interfere in pregnancy.

Acknowledgments

We are grateful to the Immunodeficiency Diseases Research Center, and all of those with whom we have had the pleasure to work during this project.

Conflict of interest

We have no conflict of interests to disclose.

Appendices and nomenclature

IDRC

Immunodeficiency Disease Research Center

NK

natural killer cell

RM

recurrent miscarriage

KIR

killer immune receptors

RIF

recurrent implantation failure

ART

assisted reproductive technology