Dendritic Cells and the Establishment of Fetomaternal Tolerance for Successful Human Pregnancy

Many cytokines are involved in the development of DCs from the progenitor cells. FAM-like tyrosine kinase 3 ligand (FLT3L) is a hematopoietic cytokine and a key regulator of these developmental processes, which are also regulated by several TFs (Poltorak and Schraml 2015). FLT3L activates hematopoietic progenitors to increase the number of immune cells. It works by binding and activating the FAM-like tyrosine kinase 3 (FLT3) receptors, found on multipotent progenitor (MPP) and common lymphoid progenitor cells, as observed in mice (Shortman and Naik 2007). In fact, steady-state DC subsets were profoundly absent in mice lacking the Flt3L (McKenna et al. 2000) or Flt3 gene (Waskow et al. 2008), but FLT3L exogenous supply increased both their diversity and number in mice (Ding et al. 2014) and humans (Anandasabapathy et al. 2015; Bhardwaj et al. 2016). Administration of FLT3L in humanized mice (mice reconstituted with human HSCs) showed an enhanced number of human CD141⁺ DC, CD1c⁺ DC, and to some extent pDC in the blood, spleen, and BM of humanized mice (Ding et al. 2014). In contrast to this, a study with Flt-3 ligand knockout mice showed the development of DCs independent of the endogenous Flt-3 ligand, and they exhibited the same phenotype as that of the steady-state spleen DCs (Fancke et al. 2008; O’Keeffe et al. 2010). FLT3L is essential for cDC1 and cDC2 subset development in both humans and mice (Merad et al. 2013). The TFs IRF8, Batf3, ID2, and Nfil3 are involved in the cDC1 development (Anderson et al. 2018), whereas IRF4 is a crucial TF for the development of the cDC2 sub-type (Bajana et al. 2016). The monocyte-derived dendritic cells (MoDCs) are derived from monocyte precursors and the macrophage-colony stimulating factor (M-CSF) induce their development (Greter et al. 2012). The important regulators of MoDCs development (from monocytes) are PU.1, IRF4, aryl hydrocarbon receptor, NR4A3, and NCOR2 proteins; and the C–C motif chemokine receptor 2 (CCR2) protein is essential for the migration of monocytes from BM to the inflamed tissue regions and to the LNs (Yin et al. 2021). The TF TCF4 is essential for the development of pDCs, and other TFs such as IRF8, Bcl11a, Zeb2, and SpiB are also involved in the pDCs development (Reizis 2019). The development of Langerhans cells (LCs) is promoted by IL-34, M-CSF, and transforming growth factor (TGF)-β proteins, and FLT3L has no role in their development. The important regulators of LCs development are RUNX3, PU.1, and ID2 proteins (Kashem et al. 2017).

Granulocyte-macrophage colony-stimulating factor (GM-CSF) also plays a crucial role in supporting the survival of DCs. Overexpression of GM-CSF in the spleen, thymus, and LNs enhanced the frequency of DCs, which highlights the role of GM-CSF in DCs proliferation (Vremec et al. 1997; Balan et al. 2019). In addition to this, interferon (IFN)-γ and tumor necrosis factor (TNF)-α are the other proinflammatory cytokines that can promote DC maturation by inducing the upregulation of major histocompatibility complex class II (MHC-II) and co-stimulatory molecules on DCs, which in turn enhances their ability to present antigens to T cells (Banchereau et al. 2000; Schroder et al. 2004).

MiRNAs play a crucial role in the regulation of DC development (Table 1). Differential miRNA expression has been observed throughout the murine DC developmental process, starting from HSC to matured DC (mDC) formation. Georgantas et al. identified the human HSC miRNA expression profile to better comprehend their function in hematopoiesis and DCs development. In CD34⁺ HSCs, they discovered 33 miRNAs that regulated several mRNAs involved in hematopoietic differentiation (Georgantas et al. 2007). Another study revealed that miR-155 expression is upregulated in human monocyte-derived DCs upon activation, which in turn targets and downregulates the expression of the suppressor of cytokine signaling (SOCS)-1, a negative regulator of the IL-1 signaling pathway. This regulation has direct implications for the immune response as it leads to the increased production of proinflammatory cytokines and the ability of DCs to communicate with other immune cells, suggesting intricate regulatory mechanisms involving miRNAs in DCs and their impact on immune function (Ceppi et al. 2009). Lu et al. (2011) have reported the increased expression of miR-221 and miR-155 in human DCs, suggesting their role in the regulation of different physiological functions of DCs. In addition, miR-221 is found to target and downregulate the expression of p27kip1 (a protein that inhibits cell cycle progression), which in turn promotes DC development and maturation. On the other hand, miR-155 targets and downregulates KPC1 (a protein associated with apoptosis regulation) and inhibits apoptosis, promotes survival, and potentially enhances the function of DCs. A study has also revealed that miR-155 downregulates SOCS-1 and leads to increased IL-12 (an important cytokine involved in immune responses) production (Lu et al. 2011). The miR-155 also inhibits the pathogen-binding ability of DC-specific intracellular adhesion molecule-3 grabbing nonintegrin during maturation (Martinez-Nunez et al. 2009). In another human study, it showed that a slight increase in miR-34a and miR-21 has been observed during the differentiation of monocytes into mature DCs, which in turn leads to the differentiation of DCs by targeting JAG1 and WNT1 (Hashimi et al. 2009). A differential miRNA expression study was performed in CD34⁺ CD38⁻ HSCs, where miR-520h and miR-129 revealed varying degrees of enhanced and decreased expression, respectively (Liao et al. 2008). The miR-520h was found to inhibit the expression of the gene for ATP-binding cassette subfamily G member 2 (ABCG2), which controls the differentiation of HSCs. On the other hand, genes encoding eukaryotic translation initiation factor 2C3 (EIF2C3, a crucial component of miRNA biogenesis) and calmodulin-binding transcription activator 1 (CAMTA1) (a TF involved in cell development) were found to be regulated by the miR-129 (Liao et al. 2008). Furthermore, a study showed differential expression of 391 miRNAs through Illumina sequencing during different stages of DC development (Su et al. 2013). This study identified substantial expression of miR-132 and miR-147 in immature and mDC, but not in HSCs (Su et al. 2013). In another study, miR-125b was found to be down-regulated in committed progenitors despite being abundantly expressed in healthy HSCs. miR-125b can encourage HSC survival and growth because of its anti-apoptotic action. Bcl2 modifying factor (BMF) and Krueppel-like factor 13 (KLF13) are two pro-apoptotic targets whose mRNA expression levels are decreased, which causes this antiapoptotic action (Ooi et al. 2010).

CD4⁺ DC formation and maintenance have been linked to miR-142 expression (Mildner et al. 2013a). It is observed that miR-142 is strongly expressed in CD4⁺ DCs that are FLT3 dependent, but not in CD8a⁺ or CD4⁻ CD8a⁻ DCs. It is also observed that a higher rate of CD4⁺ DC death caused a 60% decrease in Class II CD11c^hi DC in mice lacking miR-142. Moreover, in the presence of FLT3L, miR-142-deficient BM cells failed to differentiate into CD4⁺ DCs in vitro; however, there was no suppression of CD8a⁺ DC formation, which indicates that miR-142 plays an important role in DCs development and maintenance (Mildner et al. 2013a). The overexpression of miR-146a, another miRNA that has been demonstrated to alter pDC survival, led to apoptosis in a pDC cell line. This may be due to the fact that miR-146a down-regulates anti-apoptotic genes by targeting the IL-1 receptor-associated kinase 1 (IRAK1), which suppresses TLR-induced nuclear factor-κB activity (Karrich et al. 2013).

A study in murine models has revealed that DC development from the HSC is promoted by the double-stranded RNA-specific endoribonuclease (DROSHA)-dependent cleavage of two mRNAs such as Myl9 and Todr1 mRNAs (Johanson et al. 2015). In this study, it was shown that the overexpression of Myl9 and Todr1 blocked the DC development, and their knockdown restored the DC development. Consistent with murine HSCs, DROSHA knockdown also hampered the differentiation of human HSCs, which proved that DROSHA plays a significant role in DC development, besides their involvement in miRNA biogenesis function (Gu et al. 2022). miRNAs have been identified in human hematopoietic lineage cells and thus their importance in DC development.

According to their lineage position, conventional DCs (cDCs, also known as myeloid DCs), pDCs (also known as lymphoid DCs), MoDCs, and LCs are the four subsets of DCs (Wei et al. 2021) (Table 2). Steinman and Cohn (1973) found the cDCs to constitute the vast majority of DCs. They are primarily responsible for capturing, processing, and delivering antigens to T cells, inducing immunological reactions against invasive infections, or enforcing self-tolerance (Pakalniškytė and Schraml 2017). The cDCs are classified into 2 classes, i.e., cDC1 (effective in presenting antigens to CD8⁺ cytotoxic T lymphocytes [CTLs]) and cDC2 (with diverse functions like activation of naive CD4⁺ T cells and their polarization to Th2, Th17, Treg, and T follicular helper cells depending on the inflammatory state) (Tian et al. 2017; Anderson et al. 2021). The classification of cDCs into cDC1 and cDC2 is based on unique TFs responsible for their development. Both cDC1 and cDC2 encompass migratory and resident subsets; in other words, there are migratory cDC1 and cDC2, as well as resident cDC1 and cDC2. The distinction in their antigen-processing capabilities is instead rooted in the intrinsic differences between cDC1 and cDC2 rather than solely based on their migratory or resident subsets (Collin and Bigley 2018; Eisenbarth 2019; Ferris et al. 2020; Liu et al. 2022; Heger et al. 2023). In mice, cDC1s are typically identified by their characteristics, including being Lin⁻MHC-II⁺CD11c⁺CD8⁺ (referred to as resident cDC1s) or expressing CD103 (referred to as migratory cDC1s), while in humans, they are recognized as Lin⁻CD64⁻HLA-DR⁺CD141⁺ cells. Additionally, both human and mouse cDC1s commonly express XCR1, Clec9A, and CADM1 but in terms of their development, cDC1s rely on specific TFs, namely IRF8, Batf3, ID2, and Nfil3 (Yin et al. 2021). On the other hand, murine type 2 conventional DCs (cDC2s) are characterized as Lin⁻ MHCII⁺CD11c⁺CD11b⁺SIRPa⁺, while in humans, they are identified as Lin⁻HLA-DR⁺CD1c⁺SIRPa⁺ DCs (Yin et al. 2021). In addition to this, cDC2s show the heterogeneous expression of TFs. For example, fetal and adult gut cDC2 showed little to no expression of CD2 and FcR1 markers, while lung cDC2 from both the fetal and adult stages had increased levels of these two markers (McGovern et al. 2017). In response to viral infections, a small subset of DCs exhibit modest levels of MHC-II and costimulatory molecule expression, while secreting large amounts of IFN-α, and are known as pDCs (Swiecki and Colonna 2015; Bird 2017). Human pDCs are HLA-DR⁺CD11cCD4⁺BDCA2⁺BDCA4⁺CD123⁺ cells, while murine pDCs are MHC-II^intCD11c^intB220⁺Ly6C⁺BST2⁺Sigl ecH⁺ cells (Yin et al. 2021). The MoDCs are extremely rare in homeostasis and are produced from blood monocytes under some inflammatory conditions like autoinflammatory diseases or infection-induced inflammation due to bacterial and viral infections (León et al. 2007; Domínguez and Ardavín 2010; Tang-Huau and Segura 2019; Marzaioli et al. 2020). As compared to the other three categories, LCs have a distinct embryonic genesis. They can come from either yolk sac macrophages or fetal liver monocytes. The LCs continue to serve as resident sentinels in the mucosa and epidermis, where they can gather antigen, travel to LNs, and mature into powerful immune-stimulatory cells (Kaplan 2010; Collin and Milne 2016).

The cDCs and pDCs that were discovered in the PB and decidua have received the most attention in studies on immunological changes during pregnancy (Darmochwal-Kolarz et al. 2003). It was observed that greater number of cDCs exist in the decidua of the pregnant women, as compared to that of pDCs (Miyazaki et al. 2003). Researchers have varied conclusions for cDCs and pDCs changes as they have chosen different stages of pregnancy and different surface markers (like MHC class II, CD11c, CD80, CD86 and programmed death-ligand 1 (PD-L1) for cDCs and CD123, CD303, CD304, TLR7 and TLR9 for pDCs). Most of the studies have concluded that there is a higher cDC/pDC ratio during pregnancy because of the increase in cDCs number, either with decrease in or stable number of the pDCs (Darmochwal-Kolarz et al. 2003; Gardner and Moffett 2003; Shin et al. 2009; Ehrentraut et al. 2019). The balance of cDC and pDC number is maintained by the increased secretion of human chorionic gonadotropin (hCG) hormone during pregnancy (Sauss et al. 2018). In contrast to this, mouse studies have reported decrease in the cDC/pDC ratio throughout pregnancy (Zarnani et al. 2007; Li et al. 2018). Additionally, during pregnancy, there are differences in the surface molecule expression on cDCs and pDCs. There is decreased MHC-II expression to induce immune tolerance, as it can dampen T cell activation and upregulation of tolerogenic markers such as PD-L1 and indoleamine 2,3-dioxygenase (IDO) that in turn inhibit T cell responses and promote immune tolerance. There is reduction of IFN-α production and altered TLR expression in pDCs, suggesting that DC activities are modified in response to their tissue microenvironment (Saito et al. 2010; Svensson-Arvelund et al. 2017). As the pregnancy progressed, Darmochwal-Kolarz et al. (2012, 2013) found that the expression of CD200, CD200R, B7-H1, and B7-H4 on both the cDCs and pDCs was higher in the first trimester than in the luteal phase of the ovarian cycle, peaked in the second trimester, and then decreased in the third trimester. These changes in the cDCs lead to the reduction in T cell specific antigen response along with increase in the number of CD4⁺CD25^highFoxp3⁺ Treg cells (Shah et al. 2017; Ehrentraut et al. 2019). However, the mechanism underlying these modulations needs further study.

Only a few studies have been done on MoDCs and LCs as compared to cDCs and pDCs during pregnancy. A preliminary analysis during the first trimester of pregnancy has shown that there is accumulation of MoDCs in the decidua but that a very less number is found in the blood (Ivanova et al. 2005). Monocytes from pregnant women differentiated into DCs that were less phenotypically developed and had lower levels of the molecules CD80, CD86, and HLA-DR than those from nonpregnant women. Additionally, when exposed to inflammatory stimuli, pregnant women’s monocyte-derived DCs responded by upregulating CD86 more than CD80 and secreting more IL-10 rather than IL-12p70 compared to non-pregnant controls (Bachy et al. 2008). Human decidual stromal cells, which are present in the uterine lining during early pregnancy, were found to have a significant impact on the function of monocyte-derived DCs. Shao et al. (2020) have reported that the interaction between decidual stromal cells and DCs leads to functional re-programming of the DCs, resulting in altered DC behavior and function. It is suggested that the crosstalk between GM-CSF and IL-1β is critical in mediating the functional changes observed in DCs, which in turn induces immune responses that are conducive to maternal–fetal tolerance and crucial for a successful pregnancy (Shao et al. 2020). On the other hand, the role of LCs during pregnancy has been ignored although they are present in the decidua because of their defensive function against the pathogens, which induce the immune responses rather than the immune tolerance (Puts et al. 1986; Morelli et al. 1992; de Jong and Geijtenbeek 2010).

DCs are classified as immature DCs (imDCs) or mDCs depending on their maturity (Table 2). These two groupings were initially recognized as populations at two unique maturational stages in the context of cDCs (O’doherty et al. 1994). In the BM, HSCs produce imDCs, from where they migrate to the lymphoid organs through the blood and get matured (mDCs) when encounter the antigens. The mDCs show higher expression of CD80, CD86, and CD40 (key costimulatory molecules for T cells) along with the MHC-II, and can be distinguished from monocytes and imDCs by differential marker expression patterns (Peters et al. 1993; Lutz and Schuler 2002; Jeras et al. 2005; Hopkins and Connolly 2012).

During pregnancy, both types of DCs are present in the decidua along with DC-SIGN⁺CD209⁺DCs, which show similar features of imDCs (Gardner and Moffett 2003; Kämmerer et al. 2003). It is observed that a huge number of imDCs are present in pregnant women decidua than the mDCs in comparison to the nonpregnant women (Aldebert et al. 2007; Bartmann et al. 2014). Similar observations were obtained in pregnant mice (Blois et al. 2004). Although imDCs in the uterus of mice fluctuated upward, downward, upward, mDCs fluctuated downward, upward, and downward during the 1st, 2nd, and 3rd trimesters of pregnancy (Blois et al. 2004; Gu et al. 2019). Human decidual DCs are more specifically divided into three classes by Kwan et al. (2014): imDCs, mDCs, and intermediate DCs, and they also observed that the number of intermediate DCs decreased in the second trimester as compared to the first trimester of pregnancy. The increased level of imDCs during pregnancy could be due to the halt in the developmental process of imDCs to mDCs (Kwan et al. 2014). The imDCs in the decidua have been found to promote an angiogenic and tolerogenic microenvironment. The mDCs in the decidua produce fewer cytokines including IFN-γ, TNF-α, and IL-15 and have a lower proliferative effect on NK cells than the imDCs (Bachy et al. 2008). In contrast to PB mDCs, decidual mDCs have a lower ability to generate IL-12 and elicit a Th2 response (type 2 immune responses secrete cytokines such as IL-4, IL-5, IL-10, and IL-13 to activate and maintain humoral or antibody-mediated immune responses against extracellular parasites, bacteria, allergens, and toxins) (Miyazaki et al. 2003). All these findings suggest that DCs’ immune-stimulating activities are diminished in the decidua, which helps to maintain the fetomaternal tolerance.

DCs are divided into three types based on their distribution: circulatory DCs in peripheral circulation/blood, migratory DCs in nonlymphoid tissues, and resident DCs, which live their entire lives in the lymphoid tissues (Table 2) (Merad et al. 2013; Boltjes and Van Wijk 2014; Wei et al. 2021). Circulatory DCs, including mDCs and pDCs, are typically characterized by surface markers such as CD11c, CD11b, CD33, CD1c for mDCs, and CD303 (BDCA-2) and CD304 (BDCA-4) for pDCs. They express higher levels of MHC class II molecules (which are essential for antigen presentation to T cells) and co-stimulatory molecules like CD40, CD80, CD86, and CD83. In addition, they often express chemokine receptor CCR7, which in turn enables their migration to LNs (Banchereau and Steinman 1998). In contrast to circulatory DCs, migratory DCs express tissue-specific markers depending on their location, such as langerin (CD207) or CD103 for DCs in the skin and gut, respectively. They also express CCR7 to facilitate migration to LNs and MHC class II and co-stimulatory molecules for antigen presentation (Merad and Manz 2009; Ginhoux and Jung 2014). However, resident DCs within tissues express tissue-specific markers such as LCs in the skin expressing langerin (CD207), and intestinal DCs may express CD103. They are specialized for maintaining tolerance and may promote regulatory T cell responses. Expression of pattern recognition receptors (PRRs), like TLRs to detect local pathogens, has also been observed (Coombes and Powrie 2008; Reizis et al. 2011). It is found that the migratory DCs come in contact with the antigen, get activated, and travel to the LNs, where they present the antigens to the T cells (Broggi et al. 2013). While resident DCs in the secondary lymphoid organs carry out the antigen presentation function to CD4⁺ T cells, resident DCs in the central lymphoid organ carry out the negative selection of T cells that kills T cells with T cell receptors (TCRs), which in turn bind very strongly to MHC complexes and are likely to be self-reactive (Allenspach et al. 2008; Zhou and Wu 2017). Also, in contrast to circulating DCs and migrating DCs, variations in resident DCs throughout pregnancy have received less attention, despite the lymphoid organs being a key “hidden” maternal–fetal interface, and detailed study is required in this area (Taglauer et al. 2010).

Based on how they function, DCs can be classified as tolerogenic (also known as regulatory) or inflammatory DCs (Table 2). Since they limit a particular immune response, tolerogenic DCs, which are found in tissues, promote tolerance (Steinman et al. 2003; Smits et al. 2005; Hubo et al. 2013; Domogalla et al. 2017; Takenaka and Quintana 2017). On the other hand, inflammatory DCs show their activity at the time of inflammation and enhance the immunity, by secreting various inflammatory cytokines, which help in clearing pathogens and tumor cells (Shortman and Naik 2007; Segura and Amigorena 2013).

It is hypothesized that during pregnancy, a huge number of tolerogenic DCs are present in the decidua, which help to maintain tolerance at the maternal–fetal interface. Various evidences are found in this favor including the fact that progesterone treatment in a dose-dependent way inhibits DCs’ ability to release the pro-inflammatory cytokines TNF-α and IL-1β. IL-10, a cytokine that suppresses an inflammatory response, however, was unaffected. In addition to this, progesterone treatment reduced the expression of the costimulatory molecule CD80 as well as the MHC class II molecule RT1B, and stopped proliferation of T cells, in response to DC stimulation (Butts et al. 2007). Some fetal components also promote tolerogenic DC functions by the inhibition of TNF-α, C-X-C Motif Chemokine Ligand (CXCL) 10, CXCL9, and C-C Motif Chemokine Ligand 5 (CCL5) (inflammatory cytokines) secretion by the amniotic mesenchymal tissue cells (Magatti et al. 2009; Abomaray et al. 2015). On the contrary, studies also indicate the presence of inflammatory DCs in the decidua during normal pregnancy, and they play a vital role in the implantation of the embryo (Krey et al. 2008; Plaks et al. 2008). Hence, pregnancy is a dynamic event with delicate regulation of both tolerogenic and inflammatory DCs for successful pregnancy (Segerer et al. 2012).

Normal human decidua has abundant CD45⁺ immunocompetent leukocyte subpopulations in the first trimester of pregnancy, which include primarily CD3⁻ CD16⁻ CD56^bright++ NK cells, T cells, and APCs, allowing for appropriate cellular interactions (Rukavina et al. 1995; Juretic et al. 2004). Throughout pregnancy, there is continual reciprocal crosstalk between DCs and NK cells in the decidua, either directly through cell-to-cell contact or indirectly through cytokine production. In a study, the majority (about 60%) of decidual imDCs have been found to be in intimate interaction with uterine NK cells at the maternal–fetal interface during pregnancy, which suggests the pregnancy-specific interactions of imDCs with NK cells (Kämmerer et al. 2003; Tirado-González et al. 2012). A delicate balance is maintained between the activating and inhibitory signals by the DC–NK cell interaction, to achieve a tolerogenic environment for the semi-allogenic fetus and maintaining the immune-activation state required for protection against pathogens. The mDCs are found to produce the IL-15 cytokine, which recruits the NK cells in decidua and further induce their proliferation and IFN-γ production. Both the IL-15 and IL-15Rα (present in the cell surface of imDCs) are involved in NK cell recruitment and proliferation (Ferlazzo et al. 2004). The imDCs induce the release of IFN-γ from decidual NK cells, and these IFN-γ further promote the production of IDO from DCs. The IDO molecules in turn induce the production of Treg cells and thus create a tolerogenic environment (Terness et al. 2007; Vacca et al. 2010). Moreover, the IL-10 produced by decidual natural killer (dNK) cells blocks the maturation of imDCs and thus protects the trophoblast cells from the mDC-driven cytotoxic T-cell responses (Dietl et al. 2006). Besides, dNK cell-produced IL-10 also inhibits the DC function and induces the Th-2 cytokine production. This physiological state prevents the NK cell-driven cytotoxic effects on trophoblast cells (Leno-Durán et al. 2014). An in vitro study has revealed that when CD56⁺ NK cells were co-cultured with imDCs, only one-third of them grew, which suggests that imDCs control the proliferation of NK cells (Juretic et al. 2004; Laškarin et al. 2008). Decidual CD56^bright cells had lesser number of NKG2D-activating receptors on their surface after being co-cultured with autologous mDCs, which binds the MICA/B and HLA class I chain-related proteins A and B produced on trophoblast cells (Mincheva-Nilsson et al. 2006). The expression of NKG2A, an inhibitory receptor that is expressed on both T and NK cells and forms a heterodimer with CD94, increased in the decidual CD56^bright cell subpopulation under the same conditions. While NKG2C, an activating receptor that CD94 interacts with and forms dimers with, decreased in the CD56^dim cell subpopulation, and can bind to HLA-E to cause NK cells to activate (Laškarin et al. 2007). A study has reported that the DC–NK conjugation leads the imDCs to display an apoptotic cell like shape in human decidua, suggesting that NK cells influence the phenotype of DC population (Tirado-González et al. 2012). Karsten et al. (2009) discovered that the implantation sites lacking DCs have lower mRNA and protein levels of IL-15 and IL-12, due to the indirect interaction between DCs and NK cells that is mediated by cytokines, resulting in decreased NK cell frequency, size, and IFN-γ expression. This suggests that DCs in the pregnant uterus serve as a source of IL-15 and IL-12 (Karsten et al. 2009). Altogether, these evidences suggest that DCs play an important role during pregnancy in regulating NK cell function and vice versa, for successful pregnancy.

The immune system’s most potent APCs, DCs can cause naive T cells to grow into a variety of subsets, including Th1 (secrete IFN-γ and TNF-α), Th2 (secrete IL-4, IL-5, and IL-13), Th17 (secrete IL-17), and Treg cells (secrete IL-10 and TGF-β) (Zhu and Paul 2008). Saito et al. (2010) revealed that maternal–fetal tolerance is due to a shift from the immunologically active Th1/Th17 predominance to the immune tolerant Th2/Treg predominance, with DCs playing a crucial role in appropriately driving the T cell subpopulation (Saito et al. 2010). It is observed that during pregnancy, DCs promote the T cell polarization toward Tregs and Th2 cells by secreting cytokines like IL-10 and TGF-β, and express co-stimulatory molecules like CD80 and CD86, which, when engaged with TCRs, can guide T cell differentiation. The Th2 response generates an anti-inflammatory microenvironment by promoting the production of cytokines like IL-4, IL-5, and IL-10. This anti-inflammatory environment helps prevent harmful inflammation that could jeopardize the pregnancy (Saito et al. 2010). A study in mouse showed that the Th1/Th2 balance is linked to the specific type of DC subtypes, i.e., DEC-205⁺ DCs that are responsible for the polarization of Th1 and 33D1⁺ DCs responsible for the induction of Th2 dominance (Negishi et al. 2012). The alteration in DC subtypes indicated that Th1 was diminished throughout pregnancy unless they were required for delivery. On the other hand, it has been observed that in comparison to peripheral cDCs, cDCs in human decidua generated a higher percentage of Th2 cells and secreted less IL-12, suggesting the influence of the microenvironment on cDCs to elicit either a Th1 or a Th2 response (Miyazaki et al. 2003). Decidual CD4⁺CD25⁺ T cells are encouraged to multiply and differentiate into CD4⁺CD25⁺FOXP3⁺ regulatory T cells (Tregs) by thymic stromal lymphopoietin-activated decidual DCs through the action of TGF-1. By secreting cytokines of the Th2 subtype, promoting the invasiveness of trophoblasts and the expression of HLA-G, and reducing the cytotoxicity of CD56^bright CD16 NK cells, these newly produced Treg cells demonstrate immunosuppressive features (Du et al. 2014). DCs can upregulate αβ⁺ T cells in addition to their impact on CD4⁺αβ⁺ T cells, which have been demonstrated to inhibit the immune response during pregnancy (Szekeres-Bartho et al. 1999; Miranda et al. 2006). In addition, it has been reported that fetal DCs release the enzyme arginase-2, which effectively suppresses the activation and proliferation of T cells by depleting the level of essential amino acid arginine. This immune suppression maintains the tolerogenic environment in the developing fetus (McGovern et al. 2017).

The DCs and Mϕs bear several phenotypic characteristics in common; hence, it is difficult to characterize them with distinct phenotypes. DCs and Mϕs share several morphological and functional traits as well as monocytic origin (Le Gars et al. 2016). The main APCs at the maternal–fetal interface are Mϕs and DCs, which have complementary immune regulatory functions, due to the lack of B cells in decidua as APCs (Iijima et al. 2008). They are essential for decidualization and implantation as they secrete various cytokines, chemokines, and enzymes, which target the luminal epithelium and play a role in tissue remodeling, angiogenesis, and the development of endometrial receptivity (Jena et al. 2019). The cytokines IL-1β and TNF-α stimulate the synthesis of chemokines, which recruit Mϕs and DCs in the first trimester via the NF-κB and MAPK pathways, resulting in the buildup of DCs and Mϕs in decidua. The chemokine CCL2 is the major chemoattractant for Mϕs, whereas CCL5 is the predominant chemoattractant for imDCs (Li et al. 2011). Besides, M-CSF affects Mϕs and DC-cell migration into decidua in a CCR2-independent way (Tagliani et al. 2011). Intriguingly, decidual Mϕs were found to differentiate into DC-like cells with immunosuppressive properties in the second trimester of pregnancy, but into DC-like cells with immunostimulatory properties in the third trimester, indicating a switch from maternal–fetal immune tolerance to the maternal–fetal immune rejection state (Wang et al. 2016). The two subsets of Mϕs could be immunosuppressive M-2 and immune active M-1, based on their cytokine repertoire and functions (Mills et al. 2000). The M-1 Mϕs are proinflammatory in nature, which is triggered by pathogen (lipopolysaccharide [LPS]) exposure and tissue damage caused by IFN and TNF proteins (Martinez et al. 2008; Jena et al. 2019). These Mϕs upregulate the level of enzyme-inducible nitric oxide synthase, which subsequently produce the nitric oxide from arginine, besides their role in reactive oxygen species production. In contrast to this, Th2 cytokines like IL-4 and IL-13, as well as anti-inflammatory IL-10, apoptotic cells, and M-CSF, all induce M-2 Mϕs formation, which are anti-inflammatory in nature (Figure 2) (Mantovani et al. 2004).

Fig 2.

The uterine DCs (uDCs) fine-tune decidual angiogenesis by supplying two key factors, soluble Flt1 (sFlt1) and TGF-β1, which enhance coordinated blood vessel formation. Regardless of their projected role in immunological tolerance, uDCs appear to regulate tissue remodeling and angiogenesis and promote uterine receptivity (Lee et al. 2011).

The intricate interplay between DCs and trophoblast cells stands as a critical component of the immune tolerance mechanism that underpins the success of implantation and pregnancy. The interaction of DCs with invading trophoblast cells are multifaceted and dynamic, with the goal of establishing maternal immune tolerance to the semi-allogenic fetal tissues. Key molecules and mechanisms have been elucidated in this context. Salamone et al. (2012) studied the effect of trophoblast cells on the functional profile of DCs and observed that trophoblasts induce a tolerogenic DC population at the maternal–fetal interface. The trophoblasts significantly reduced the production of IL-12p70 and TNF-α, while they enhanced the IL-10 production. The co-culture of DCs and trophoblast cells revealed the suppression of the LPS-stimulated allogenic response. The conditioned DCs enhanced the frequency of CD4⁺ CD25⁺ Foxp3 cells, which further induced enhanced IDO expression in the DCs and thus a tolerogenic environment is created (Salamone et al. 2012). In addition, other studies have further worked on the immune regulatory functions of DCs, emphasizing their capacity to organize a microenvironment conducive to fetal development and immune homeostasis. These interactions extend beyond mere immune suppression, involving cytokines, antigen presentation, and immunomodulatory molecules (Wei et al. 2021). The crosstalk between DCs and trophoblast cells is a dynamic process, integral to the intricate balancing act that allows the maternal immune system to protect against threats while simultaneously permitting the growth and development of the fetus (Kammerer et al. 2008). A comprehensive understanding of these interactions is pivotal not only for advancing our knowledge of the immunological aspects of pregnancy but also for devising novel therapeutic strategies for complications arising from immune dysregulation during gestation. A study has revealed that a significant number of DCs become trapped within the uterine tissues during pregnancy, particularly in the myometrium (muscular wall of the uterus) and decidua. This entrapment is a unique feature of pregnancy and hinders DCs’ ability to effectively patrol and monitor the maternal–fetal interface. As a result, there is a decreased capacity for immune surveillance in this critical area during pregnancy. This altered presence and function of DCs at the maternal–fetal interface influence the activation and regulation of T cell responses, which has implications for maintaining immune tolerance and preventing the rejection of the developing fetus (Collins et al. 2009).

“Recurrent spontaneous abortion (RSA) is defined as three or more consecutive pregnancy losses of clinically recognized pregnancies” (El Hachem et al. 2017). Although aberrant chromosomes, endocrinological diseases, and uterine abnormalities were known causes in around half of the instances, the etiology of the other half remains unexplained (Garrido-Gimenez and Alijotas-Reig 2015). In RSA mice, the proportion of cDCs (CD11c⁺B220⁻) was significantly higher, whereas the proportion of pDCs (CD11c⁺B220⁺) and the pDC/cDC ratio were much lower than in the normal pregnant group. Similar findings were observed in humans as well suggesting that the cDCs may play role in RSA, but the mechanism is unknown (Lai et al. 2022). Askelund et al. (2004) were the first to show that mDCs in decidua from women with RSA at 8 weeks’ gestation are higher than in normal controls of the same gestational age, implying that DCs are implicated in the etiology of RSA (Askelund et al. 2004). In addition, Qian et al. (2015) reported a decrease in imDCs in RSA, as well as an increase in mDCs. These data supported the idea that DC immaturity promotes pregnancy and that DC maturation at an unsuitable time may play a role in RSA pathogenesis (Qian et al. 2015).

Additionally, the uDCs have been linked to a critical developmental function in modulating interactions between the uterus and the embryo during the peri-implantation period (Krey et al. 2008; Plaks et al. 2008). These studies utilized CD11c-DTR transgenic mice, where the CD11c promoter controls the expression of the diphtheria toxin receptor (DTR). This genetic setup makes cells expressing CD11c susceptible to acute removal when exposed to diphtheria toxin (Jung et al. 2002). In the study by Plaks et al. (2008), they conducted a controlled removal of uDCs just before the process of embryo implantation on day 3.5 of embryonic development (E3.5). This removal led to pregnancy failure due to its impact on both the attachment of the embryo to the uterine wall (implantation) and the formation of the decidual tissue (Plaks et al. 2008). The latter effect resulted from impaired growth and transformation of the cells in the uterine lining (endometrial stromal cells). It also involved problems with the development of blood vessels in the uterine tissue, marked by increased permeability of blood vessels and incomplete vessel maturation. These observations were linked to reduced levels of two important factors in the uterus: TGF-β1, a growth factor, and sFlt1, a protein that acts as a decoy receptor for vascular endothelial growth factor-A (VEGF-A), counteracting the ill-effects of the excess proangiogenic factor VEGF-A. The uDCs themselves contained measurable amounts of sFlt-1, suggesting that these cells might have a role in regulating local concentrations of VEGF-A. Additionally, the authors found that removing DCs after the implantation phase on day 5.5 of murine embryonic development (E5.5) had no impact on the formation of decidual tissue or the survival of the developing embryos. This indicates that uDCs were specifically necessary during the critical period when the embryo attaches to the uterine lining in the implantation process.

Maternal hormones were also sought to identify the processes underlying the DC aberrations in RSA. According to the findings of the studies on spontaneous abortion, progesterone had greater impact on abortion than hCG (Negishi et al. 2012; Ehrentraut et al. 2019). Progesterone inhibited cDCs’ ability to induce NK cell proliferation and secrete IL-15, resulting in human abortion. It was hypothesized that progesterone-shaped DCs cannot efficiently multiply or equip NK cells with the cytotoxic mediators perforin and granulysin during DC/NK cell interaction, which may harm tropho-blasts and induce abortion (Laskarin et al. 2018). However, in DBA/2 J-mated CBA/J abortion-prone mice, a substantial rise in the uterine Treg cell pool was induced by progesterone administration, and the rate of abortion was unaffected. Furthermore, following the application of progesterone, there were no distinct changes in peripheral Treg cell numbers or DC counts. It implied that in this impaired fetal tolerance, progesterone-induced local Treg cell number increase is not adequate to overcome fetal rejection (Schumacher et al. 2017). Hence, progesterone and its influence on DCs might not be enough to account for fetal rejections in RSA. Hence, this aspect needs to be explored further.

High blood pressure, proteinuria, and symptoms of damage to organ system (most commonly the liver and kidneys) describe the PE, which is a serious pregnancy complication (Mol et al. 2016). Assessment of the blood vessels in the fetal membranes and the basal plate extravillous trophoblasts (EVTs) (located in the decidua basalis, which is the maternal surface of the placental disc) has unveiled indications that PE may be linked to irregularities in the differentiation, restructuring of blood vessels, and invasion by EVTs. These abnormalities appear to be more pronounced in cases of pre-term births and severe forms of pre-eclampsia (Benirschke et al. 2012; Redline et al. 2018). There can be decidual vasculopathy, which is marked by the enlargement and excess growth of the muscular layer in the arterioles of the decidua, as well as the presence of chronic inflammation around these blood vessels, fibrinoid necrosis, and/or the infiltration of foamy macrophages. When decidual vasculopathy is observed alongside the occurrence of multifocal infarctions in the placentae of pregnancies affected by PE or fetal growth restriction (FGR), it strongly indicates a malfunction in the endovascular function of EVTs. This malfunction leads to an insufficient maternal blood supply to the placenta (Redline et al. 2018). Further understanding of the differentiation and performance of EVTs has been gained through placental bed biopsies conducted for research after childbirth. These investigations have indicated that there is restricted alteration of blood vessel structure in both PE and FGR placenta (Farah et al. 2020). More recently, there have been advances in identifying and characterizing specific markers that distinguish between different subtypes of differentiated EVTs in the context of PE. One such example is PLAC8, which has been pinpointed as a marker specific to interstitial EVT. PLAC8 is a protein associated with actin that enhances the migration and invasion abilities of these cells by activating RAC1 and CDC42. Interestingly, in PE placentae, there was an observed increase in the expression of PLAC8 in interstitial EVT. This upregulation is speculated to be a compensatory response to the shallower invasion of these cells in cases of PE (Chang et al. 2018). Another marker is leukocyte-associated immunoglobulin-like receptor 2 (LAIR2), which appears to be specific to endovascular and vascular plug EVT (Founds et al. 2013). Notably, LAIR2 levels were found to be reduced in chorionic villus samples from placentae associated with the development of PE later in pregnancy (Founds et al. 2009).

Research findings have revealed that PE is a Th1/Th2 immune disorder characterized by a strong Th1 response leading to the inflammatory state, and DCs are considered to play a crucial role in PE by regulating the Th1/Th2 response (Darmochwal-Kolarz 2005). Moreover, it has been observed that after phagocytosis of necrotic and aponecrotic tropho-blasts, DCs release type-I cytokines such as TNF-α, IL-12, and IFN-γ, which increase inflammation, leading to apoptosis of extra villous trophoblasts and impaired placentation, as seen in PE (Huppertz et al. 2003; Jena et al. 2019). Some studies have observed the elevated level of DC-recruiting chemokine in the decidua of PE patients, such as GM-CSF, suggesting the accumulation of more DCs in the uterus of PE patients than in controls (Huang et al. 2008, 2010). Also, in PE patients, higher percentage of cDCs, a higher mDC/pDC ratio, and a lower percentage of pDCs have been observed than in normal pregnant individuals (Darmochwal-Kolarz et al. 2003; Wang et al. 2013; Li et al. 2019). These alterations in DC population are linked to a higher number of Th1 cells and a lower number of Th17 cells (Wang et al. 2013). In vitro studies confirmed that DCs generated from PE patients’ PB mononuclear cells had a greater potential to trigger CD4⁺ T cells to differentiate into Th1/Th17 cells (Wang et al. 2014). Altogether, these findings suggest that altered DCs differentiation and function have a potential role in PE pathogenesis.

PTB is also known as premature birth that occurs before the 37th week of pregnancy and refers to the termination of pregnancy before term (Goldenberg et al. 2008). Infections or physiological factors may be responsible for the inflammation that occurs in PTB (Goldenberg et al. 2008; Cappelletti et al. 2016). In addition to this, genetic predisposition and environmental factors such as stress can influence immune responses during pregnancy. Variations in immune-related gene expression have been associated with increased susceptibility to PTB (Wadhwa et al. 2001). Systemic immune factors, such as maternal infections and inflammation, can impact the risk of PTB. For instance, infections, including urinary tract infections and periodontal disease, have been linked to an increased risk of PTB (Goldenberg et al. 2008). Systemic inflammation, as evidenced by elevated levels of C-reactive protein and other markers, has also been associated with PTB. Research has shown that an imbalance in local immune factors, increased inflammation, and altered immune cell populations can contribute to the onset of PTB (Menon 2016). Aberrant cytokine profiles, including elevated levels of proinflammatory cytokines like IL-1β and TNF-α, have been associated with PTB (Keelan et al. 2003). The lowered expression of the anti-inflammatory cytokine IL-10 (a putative early indicator of preterm delivery) in immature DCs suggests that these cells may play a role in the genesis of premature labor (Ruiz et al. 2012). In addition to this, uDC activation was also observed in the LPS-treated PTB mouse model, a paradigm that mimics infections, suggesting that DCs are implicated in the induction of labor (Bizargity et al. 2009). In TLR4-deficient female mice, longer gestation length was linked to fewer DCs and more Treg cells in the myometrium, leading some to hypothesize that DC activation in infection-related PTB was a downstream effect of TLR4 (Wahid et al. 2015). Another prevalent cause of PTB is acute chorioamnionitis (ACAM), a physiological state of inflammation. A human pregnancy study has revealed that DCs derived from ACAM women selectively promote the growth of the T cell subset invariant natural killer T cells (iNKT) cells (Negishi et al. 2017). Further in vitro research showed that decidual iNKT cells co-cultured with LPS-pulsed DCs exhibited considerably decreased extracellular and intracellular IFN-γ production, as well as lower surface expression of CD69 (Li et al. 2015). These results demonstrated that PTB affected the interaction between DCs and NK cells. In addition to this, numerous studies have shown that PTB is associated with heightened inflammatory responses in the maternal–fetal interface. Increased levels of proinflammatory cytokines, such as IL-6 and TNF-α, have been observed in the cervicovaginal fluid and placental tissues of women at risk for PTB, suggesting that PTB is due to a loss of tolerance (including DC) or inability to mount a required immune response (Romero et al. 1994).