Expression of NKp46 and other activating inhibitory receptors on uterine endometrial NK cells in females with various reproductive failures: A review.

Since KIR2DL1 (CD158a) expression is not related to individual HLA type 130 ; KIR2DL1 may be appropriate for representing the NK cell status. KIR2DL1 binds to the HLA‐C2 ligand. Moreover, KIR2DL2 (CD158b1) and KIR2DL3 (CD158b2) possess alleles of the same gene and are present in all individuals. In contrast, the highest KIR2DL1 expression in uNK cells has been observed in the proliferative phase in infertile females. 131 Moreover, in normal pregnancy, a 4‐fold increase has been reported in KIR2DL1 + /(KIR2DL2/3) + dNK cell percentage in the first‐trimester decidua compared to that in the peripheral blood. 132 Uterine endometrial NK cells secrete chemokines (IL‐8 and IP‐10) that directly promote trophoblast invasion and VEGF, placental growth factor, and Ang2 19 , 32 , 133 ; however, the production of these factors by uNK cell clones is reduced when KIR2DL1 binds to HLA‐C2. 19 One study demonstrated that the percentage of KIR2DL1 + /(KIR2DL2/3) + dNK cells during an anembryonic pregnancy was markedly lower than that during a normal pregnancy. 134 Moreover, in patients with miscarriages who possess normal fetal chromosome karyotype, the accumulation of cytolytic granules in both dNK cells and T cells was reported to be inversely correlated with CD94 and KIR2DL1 expression in NK cells. 135 These studies indicated that KIR2DL1 expressed by uNK/dNK cells may serve as a protective factor of normal pregnancy. Additionally, a previous study showed that patients with RPL exhibited higher absolute numbers of KIR2DL1, KIR2DL2/3, and/or NKG2A receptor‐expressing pNK cells than seen in the healthy female controls. 136 Upregulation of KIR2DL2/3 in dNK cells has also been observed in females with RPL. 108 In regard to IVF failures, elevated expression of KIR2DL1 in pNK cells and enhanced NK cell cytotoxicity have also been observed. 137 Conversely, significantly reduced KIR2DL1 and KIR2DL2/3 expression in pNK cells in patients with implantation failure, 138 and reduced KIR2DL1 expression in pNK cells in patients with RPL compared to healthy fertile controls have been reported. 139 Similar studies have indicated that increased and decreased expression of KIR2DL1 in pNK cells could be considered immune deviations that are potentially predictive of failure in IVF, 140 implantation, and reproduction. 141 Therefore, dysregulation of KIR2DL1 and KIR2DL2/3 expression is common in the various forms of reproductive failure. However, whether this is the cause or result of reproductive failure remains unknown. As EB6 mAb (anti‐KIR2DL1) can recognize the inhibitory KIR2DL1 receptor as well as activate the p50 form of the KIR2DS1 receptor, KIR2DL1 expression may be associated with increased or decreased cytotoxic activity, assuming that the target cells are HLA‐C + . 6 , 132 , 142 , 143 , 144 Therefore, phenotypic analysis using this mAb may not appropriately discriminate between these two forms. However, the affinity of the HLA‐C ligand for the KIR2DS1 receptor is incredibly lower than that for the KIR2DL1 receptor. Therefore, most molecules detected by the EB6 mAb could be inhibitory KIR2DL1 molecules. KIR2DL5 is represented in the ImmunoPolymorphism Database (v2.4.0) as 15 KIR2DL5A and 25 KIR2DL5B alleles. 53 Most healthy individuals express less than 10% KIR2DL5‐expressing pNK cells; accordingly, KIR2DL5 remains a mysterious KIR in the uterine endometrium in terms of its ligand and function. Cross‐linking of naturally expressed KIR2DL5 inhibits NK‐cell cytotoxicity against mAb‐coated P815 target cells through the SHP‐2 phosphatase, which to an extent, is comparable to the classical KIR3DL1 and exhibits a clonal distribution on NK cells. 145 , 146 , 147 KIR3DL1 and KIR3DS1 are currently regarded as different alleles of the same gene. KIR2DS4 is considered as an ancestral KIR, originating from a gene conversion event with KIR3DL2, a KIR that is exquisitely sensitive to HLA‐A allotypes and their bound peptides. 148 Since HLA‐A and‐B are not expressed by trophoblasts, the roles of KIR3DL1 and KIR3DL2 in the uterine endometrium remain to be elucidated. KIR3DL3 is highly polymorphic and present in every human, which is uncommon for KIR; it is also conserved across catarrhine primate species. Furthermore, Leaton et al. suggested that the KIR3DL3 polymorphism is critical for human survival and may be involved in reproduction, likely during placentation. 149 Trundley et al. detected KIR3DL3 mRNA only in CD56 bright NK cells from the peripheral blood and decidua, inferring that the KIR3DL3 gene encodes a protein that is not expressed in healthy individuals but could be induced under certain developmental or pathological conditions. 150 CD94/NKG2A receptor presents on nearly all uNK cells. NKG2A and NKG2B are different isoforms derived from the same gene through alternative splicing. Accordingly, NKG2B differs from NKG2A in the absence of a membrane‐proximal fragment that is located in the extracellular region of the molecule. 151 Since both proteins have immunoreceptor ITIM motifs in their cytoplasmic tails, they can transduce inhibitory downstream signals. The combination of CD94/NKG2A or CD94/NKG2B and the HLA‐E‐peptide induces an inhibitory signal, consequently preventing lysis of the target cell; thus, lymphocytes remain tolerant to autologous cells. 152 , 153 In the absence of MHC classical (Ia) protein expression, HLA‐E is downregulated, as no peptides are present. 154 In this case, NK and T cells do not receive adequate inhibitory signals and thus attack the target cell via a number of activating receptors that recognize a set of ligands expressed during infection, cell stress, or cell malignization. Therefore, HLA‐E recognition is considered a process whereby the immune system monitors the capacity of a cell to process and present antigens associated with other classical Ia molecules. 154 Furthermore, a study on KIR and NKG2 expression, which included more than 100 pNK cell clones from two individuals, showed differences in pNK cell receptor repertoire expression. 155 Interestingly, these receptors tended to complement each other at the clonal level, that is, clones that did not express any of the major inhibitory KIR receptors, such as 2DL1, 2DL2/3, 3DL1, or 3DL2, expressed the CD94/NKG2A receptor. Moreover, the donors who utilized KIRs predominantly had three different KIR ligands, whereas those who utilized the CD94/NKG2A receptor mostly had only a single KIR ligand. ILT‐2 is expressed by less than 25% of dNK cells in all females, and HLA‐G is a ligand of ILT‐2. Reportedly, the expression of KIR2DL1/S1 and KIR2DL2/L3/S2, NKp30, NKp44, and ILT‐2 in dNK cells decreases, whereas that of NKG2D increases in the first trimester of normal pregnancy, indicating that the phenotypic evolution of dNK cells may conserve their activation/inhibition balance during the first trimester. 156 However, NKp46 expression in dNK cells was not altered in that study. Studies on ILT‐2 are limited; therefore, it was difficult to reach any conclusion regarding its function in reproductive outcomes. Uterine endometrial NK cells also express other inhibitory receptors such as P75, inhibitory receptor protein 60 (IRp60), and leukocyte‐associated immunoglobulin‐like receptor 1 (LAIR‐1). Decidual NK cells expressing both p55 and p75 have receptors exhibiting high affinity to IL‐2. 157 IRp60, an inhibitory receptor expressed by all human NK cells, binds to phosphatidylserine (PS) and phosphatidylethanolamine (PE), which are expressed on the plasma membranes of activated, infected, transformed, or apoptotic cells 158 and possibly play roles in controlling effector functions and cell‐mediated inflammatory responses. 159 The LAIR‐1 expressed on dNK cells, whose ligand is collagen, 160 may play a crucial role in maternal‐fetal interface tolerance through the JAK–STAT pathway 161 ; additionally, the abnormal expression of LAIR‐1 and collagen disrupts the function of dNK cells. 162 However, the roles of p75, IRp60, and LAIR‐1 in uNK cells are rarely reported.

By reviewing the current information regarding the activating and inhibitory receptors expressed on NK cells in healthy human pregnancies and various types of reproductive failures, we are convinced that these receptors (mainly NCRs, KIRs, and C‐type lectin receptors), especially those expressed on uNK cells, play crucial roles in successful pregnancies. Dysregulation of these factors can lead to adverse outcomes. In patients with abnormal levels and/or activity of NK cells, immunotherapies such as intravenous immunoglobulin (IVIG), intralipid, prednisolone, granulocyte colony‐stimulating factor, TNF‐α inhibitors, and lymphocyte immunotherapy have generally been used as a treatment clinically, however, conflicting views of immunotherapy still exist regarding its effectiveness. 163 Some past studies support the opinion that immunotherapy improves pregnancy outcomes in patients with reproductive failures. 164 , 165 , 166 , 167 , 168 Among these immunotherapies, only IVIG has been reported to be associated with the modulation of NK cell receptor expression, while others mainly play a role in the reduction of NK cell frequency, type 1 cytokines, and NK cell cytotoxicity activity. 169 , 170 , 171 , 172 , 173 IVIG can upregulate inhibitory receptors (KIR2DL1, KIR2DL2, KIR2DL3, and NKG2A) and downregulate activating receptor ((KIR2DS1, KIR2DS4, and NKG2C) expression on pNK cells, thus contributing to improved pregnancy outcomes. 174 Shimada et al. have reported a higher percentage of inhibitory CD94 + pNK cells after IVIG treatment. 175 Antiinflammation, suppression of autoantibodies and complements, blockade of FcRn and FcγRn, up‐regulation of inhibitory FcγRIIB, modulation of monocytes, macrophages, dendritic cells, NK cells, T cells, B cells, and endothelial cells account for the immunomodulatory effects of IVIG treatment. 176 , 177 However, the precise mechanism through which IVIG alters NK cell receptor expression remains unclear. Others hold the conservative opinion that more large‐scale randomized controlled trials are needed to confirm the effectiveness of immunotherapy. 163 , 178 , 179 , 180 There is no consensus regarding the reference range of NK cells, and divergence in the definition of reproductive failure and immunotherapy protocols may account for controversial conclusions. Vitamin D deficiency has been suggested as a contributing factor to infertility, 181 and pNK cell levels and cytotoxicity were found to be significantly higher in females with low vitamin D levels than those with a normal vitamin D level. 182 Therefore, vitamin D supplementation is a promising immunomodulating treatment in clinical practice. The expressions of KIR2DL1 and KIR2DL2/3 were significantly upregulated, whereas the expressions of NKp30, NKp44, as well as the degranulation marker CD107a, and the production of TNF‐α and IFN‐γ were significantly downregulated in pNK cells whereas no change was observed in the expression of NKp46 in pNK cells following treatment with 1,25(OH) 2 D3 (the active form of vitamin D), consequently, NK cell cytotoxicity and degranulation capability were decreased. 183 Interestingly, the pNK cells of healthy fertile females did not respond to 1,25(OH) 2 D3 treatment, although they expressed the vitamin D receptor, which may be attributed to their inactivated pNK cell status. 183 , 184 Long‐term, high‐dose vitamin D supplementation did not increase the risk of adverse events. 185 To date, the functional mechanisms of most NK cell receptors remain unclear or controversial, and the benefits and cost‐effectiveness of immunotherapies require further investigation. More efforts are required to confirm the effectiveness of immunotherapy in restoring NK cell activity to normal levels and to reveal the underlying mechanisms to improve the reproductive outcomes of patients with abnormal levels and/or activity of NK cells.

The innate immune system plays a crucial role in successful pregnancy. Uterine endometrial natural killer (uNK) cells, which constitute 60%–80% of lymphocytes in the endometrium from the secretory phase to the early pregnancy decidua stage, are the key player of the immune system. uNK cells express various receptors, including activating and inhibitory ones, and corresponding ligands bind to the receptors. The balance between the activating and inhibitory receptors determines NK cell cytotoxicity, that is, whether the NK cell would attack the target cell or not depends on the type of receptor it binds to. However, some of these ligands remain unknown (Table  1 ). For example, if the target cell binds to both the activating and inhibitory receptors expressed on the NK cell, the response of the NK cell to the target cell depends on the balance between the expression of activating and inhibitory receptors. Since the chorion, including syncytiotrophoblasts and cytotrophoblasts, expresses human leukocyte antigen (HLA)‐C, HLA‐E, HLA‐F, and HLA‐G but does not express classical class I (HLA‐A and HLA‐B) or class II (HLA‐DP, HLA‐DQ, and HLA‐DR), allorecognition by maternal T lymphocytes is prevented. 1 However, these chorionic cells are vulnerable to the cytotoxicity of peripheral NK (pNK) cells. 2 On the contrary, both normal trophoblast cells and choriocarcinoma cell lines are tolerant to lysis by freshly isolated decidual NK (dNK) cells. 3 NK cells exist in the endometrium and decidua where dNK cells reside close to the trophoblast cells; therefore, it is possible that uNK or dNK cells play a role in establishing and maintaining pregnancy. Specific ligands for most NK cell receptors are the molecules expressed on extravillous trophoblasts (EVTs), such as HLA‐C, HLA‐E, and HLA‐G (Table  1 ). 4 , 5 , 6 The activating and inhibitory receptors on uterine endometrial NK cells and their ligands. Since studies on human uNK cells are limited by the low availability of fertile samples and extrapolating from animal models carries limitations, the in vivo role of uNK cells has not yet been fully elucidated. Therefore, this review summarizes the most recent information regarding the activating and inhibitory receptors on uNK cells in healthy human pregnancies, along with the evidence indicating their significance in various reproductive failures. The following sections demonstrate both activating receptors, including natural cytotoxic receptors (NCRs), NK22 receptors, killer cell immunoglobulin‐like receptors (KIRs), CD16, C‐type lectin receptors, some other activating receptors, and inhibitory receptors, including KIRs, C‐type lectin receptors, immunoglobulin‐like transcripts (ILTs), and some other inhibitory receptors.

The authors declare no conflict of interest for this article.