The Role of Different Lymphoid Cell Populations in Preeclampsia Pathophysiology

Abstract

Preeclampsia (PE), new-onset hypertension during pregnancy, affects up to 10% of pregnancies worldwide. Despite being the leading cause of maternal and fetal morbidity and mortality, PE has no cure beyond the delivery of the fetal-placental unit. Although the exact pathogenesis of PE is unclear, there is a strong correlation between chronic immune activation; intrauterine growth restriction; uterine artery resistance; dysregulation of the renin-angiotensin system. Which contributes to renal dysfunction; and the resulting hypertension during pregnancy. The genesis of PE is thought to begin with insufficient trophoblast invasion leading to reduced spiral artery remodeling, resulting in decreased placental perfusion and thereby causing placental ischemia. The ischemic placenta releases factors that shower the endothelium and contribute to peripheral vasoconstriction and chronic immune activation and oxidative stress. Studies have shown imbalances in proinflammatory and anti-inflammatory cell types in women with PE and in animal models used to examine mediators of a PE phenotype during pregnancy. T cells, B cells, and natural killer cells have all emerged as potential mediators contributing to the production of vasoactive factors, renal and endothelial dysfunction, mitochondrial dysfunction, and hypertension during pregnancy. The chronic immune activation seen in PE leads to a higher risk for other diseases, such as cardiovascular disease, CKD, dementia during the postpartum period, and PE during a subsequent pregnancy. The purpose of this review is to highlight studies demonstrating the role that different lymphoid cell populations play in the pathophysiology of PE. Moreover, we will discuss treatments focused on restoring immune balance or targeting specific immune mediators that may be potential strategies to improve maternal and fetal outcomes associated with PE.

Introduction

Preeclampsia (PE) is a hypertensive disorder during pregnancy, affecting 3%–5% of pregnancies in the United States and up to 10% of pregnancies worldwide (1–3). Major characteristics of PE include new-onset hypertension after the 20th week of gestation, multiorgan system dysfunction, fetal growth restriction, endothelial dysfunction, and chronic immune activation (1,3). PE typically worsens as the pregnancy progresses, and delivery of the fetal-placental unit remains the only known treatment. The current standard of care involves treating symptoms of PE with antihypertensive drugs, such as hydralazine and labetalol, magnesium sulfate to reduce risk of seizures and strokes, and steroids for fetal lung development to sustain the pregnancy as long as possible, which is critical for fetal development.

The specific mechanisms that contribute to the development of PE are still being fully clarified, but the current belief is that insufficient trophoblast invasion leads to spiral artery remodeling, compromised blood flow to the fetal-placental unit, and a deficiency of oxygen and nutrient delivery to the placenta and fetus. This results in intrauterine growth restriction and placental ischemia, which appears to be a stimulus for antiangiogenic factors and inflammatory mediators (4–6). In normal pregnancies, trophoblast invasion and spiral artery remodeling rely on maternal immune regulation for adequate vascular transformation (7,8). Due to inadequate vascular remodeling, women with PE have elevated uterine artery resistance, which contributes to activation of the maternal endothelium, chronic inflammation, and oxidative stress systemically (9–11). Women with normal pregnancies experience increased plasma volume that peaks near gestational week 32, continuing until delivery, and a natural upregulation of renin-angiotensin-aldosterone system (RAAS) components, such as angiotensin II (Ang II), which is counterbalanced by the vasodilator angiotensin 1–7 (Figure 1) (12–14). Increased angiotensin 1–7 and increased endothelial nitric oxide production due to estrogen and progesterone may account for increased resistance or decreased sensitivity to Ang II, allowing for BP to remain lower in women who are not pregnant (15,16). In contrast, women with PE have significantly increased sensitivity to Ang II and decreased plasma volumes, the degree of which correlates with the severity of the disease (17). Gallery et al. (18) demonstrated this decreased plasma volume occurs several weeks before elevated BP and other symptoms of PE. In addition to the decreased plasma volume, women with PE have reduced GFR and increased renal vascular resistance (19). PE is associated with glomerular capillary endotheliosis, characterized by swollen glomerular cells and capillary occlusion (20,21). Often, this reverses after delivery (21), however, PE is associated with increased risk for the development of ESKD (22,23). Interestingly, in contrast to normal pregnancy, women with PE have a downregulation of many components of the RAAS that likely contribute to variations in plasma volume load, renal function, and the lack of BP regulation (18,24–26). The compensatory activation of the RAAS in response to decreased plasma volume does not occur. It is unclear whether this phenomenon is a cause or consequence of PE.

Figure 1.

The regulation of the renin-angiotensin-aldosterone system (RAAS) in normal pregnancy (NP) is different than what occurs during preeclampsia (PE). Renin cleaves angiotensin I from angiotensinogen. Angiotensin I is converted to angiotensin II by angiotensin-converting enzyme (ACE). Angiotensin II (Ang II) binds to the Ang II type 1 receptor (AT1-R) which contributes to aldosterone production and vasoconstriction. Angiotensin 1–7 (Ang1–7) binds to the MAS receptor (MAS), which contributes to vasodilation. PE is associated with increased Ang II sensitivity, which is thought to be caused by the antagonistic autoantibody of AT1-R (AT1-AA). This increased sensitivity contributes to increased vasoconstriction that plays a role in placental ischemia, which causes T cell and B cell activation, leading to the production of AT1-AA. AT1-AA bound to healthy tissues also contributes to cytolytic natural killer (NK) cell activation. TH2, T helper 2; Treg, regulatory T cell.

Importantly, women with PE secrete an autoantibody to the Ang II type 1 receptor (AT1-AA) (Figure 1) (27). This AT1-AA is agonistic in nature, activating the AT1 receptor (AT1R), and may account for the exaggerated pressor sensitivity to patients with PE who are infused with Ang II (Figure 1) (28). AT1-AAs increase the production of reactive oxygen species, inflammatory cytokines, and tissue factors through the activation of the AT1R (29–33). Additionally, these AT1-AAs can be recognized as antibodies bound to cells that should be destroyed by natural killer (NK) cells, contributing to NK cell activation (34). In rat models of PE, AT1-AA causes hypertension, oxidative stress, endothelin-1 and cytokine production, and vascular dysfunction (31,33,35,36). Moreover, when the angiotensin II and AT1-AA are infused simultaneously, there is an additive increase in the secondary responses, such as oxidative stress, endothelin-1 production, and renal vascular dysfunction (33). Although the stimulus for the AT1-AA is unknown, the AT1-AA has been shown to be produced by PE women 6–8 years postpartum (37,38). This indicates a long-term memory response is at play, thus indicating a role for lymphoid cells in the pathogenesis of PE.

This review aims to provide insight into mechanics of key lymphoid cells such as T cells, B cells, and NK cells in normal pregnancy and preeclamptic pregnancies, and also discusses potential therapeutics in immune system regulation of PE. T cells are a component of the adaptive immune system and play a central role in modulating the adaptive immune response. B cells are also part of the adaptive immune system and produce antibodies and cytokines to promote immunity and help clear pathogens. NK cells are part of the innate immune system and provide a rapid, cytotoxic response to viral or cancerous cells. Therefore, by summarizing studies examining the role of lymphoid cells in PE, this review will contribute to the collection of research that could lead to the development of new therapeutics for PE. Importantly, it will aid in understanding the molecular mechanisms of lymphoid cells in PE, which is necessary for an adapted approach in diagnosing and treating of PE.

The Role of Lymphoid Cells in Normal Pregnancy and Preeclampsia

Studies demonstrate shifts in immune cell populations and chronic immune activation in women with PE compared with women who have normal pregnancies. During normal pregnancies, populations of anti-inflammatory immune cells are elevated to help the mother tolerate the immunogenic fetus (39,40). In pregnancies complicated by PE, there is a shift toward proinflammatory immune cell types (41,42). The chronic inflammation seen in PE can be sustained postpartum, negatively affecting maternal health and contributing to an increased risk for cardiovascular and metabolic disorders later in life. Studies have shown important roles for multiple cell types in the pathogenesis of PE, including T cells, B cells, and NK cells (Figure 2). Importantly, there is a lack of studies examining the importance of neutrophils, macrophages, and dendritic cells to cause the pathology of PE. This lack of knowledge presents an opportunity for researchers to target specific immune cells with immunomodulating therapeutics in animal models or cell culture systems to further elucidate their role in the pathophysiology of PE and thus potentially find new treatments for PE.

Figure 2.

Lymphoid cell interplay to cause cell death in preeclampsia. Lymphocytes work in tandem to induce inflammation during PE. (1) CD4+ T cells activate B cells through T cell receptor (TCR)–MHCII and CD40 ligand (CD40L)–CD40 crosstalk on T and B cells, respectively. T cell–B cell communication causes B cells to proliferate and transform into memory B cells and plasma cells. Memory B cells retain long-term antigen memory, whereas plasma cells produce antibodies and secrete them into the circulation. (2) Th1 and TH17 cells produce IL-17, TNF-α, and IFN-γ, which polarize NK cells toward the cytolytic NK1 phenotype. In PE, plasma cells produce autoantibodies that can target cells for antibody-dependent cellular cytotoxicity through activation of the CD16 receptor on NK cells. (For example, AT1-AA can bind a cell that expresses the AT1-R. Once the Fc region of the AT1-AA is recognized by CD16 on the NK cell surface, perforin and granzyme granules are released from the NK cell that induce apoptosis of the AT1-AA–marked cell, resulting in death and tissue damage.) (3) Antibody-mediated cell death alongside AT1-R activation leads to vasocontraction and a continuation of the immune response, contributing to chronic inflammation. (4) This leads to vascular dysfunction in the placenta, kidney, and brain and hypertension.

T Cells in Normal Pregnancy and Preeclampsia

T cells play a critical role in regulating and stimulating the adaptive arm of the immune system. CD4+ T cells, also known as T helper (TH) cells, can be further classified as TH1 cells and TH2 cells. In normal pregnancy, the balance between TH1 and TH2 cells is shifted toward TH2 cell types to ensure maternal tolerance toward fetal alloantigens (41,43). This shift is likely due to increased progesterone production and elevated production of IL-4, an anti-inflammatory cytokine (44). Patients with PE produce higher levels of TH1-associated cytokines such as IFN-γ (45). In PE, there is also a shift from TH2 toward predominately TH1 T cells (45). Furthermore, the predominance of either TH1 immunity or TH2 immunity has been implicated in recurrent spontaneous abortion as well as PE (46,47).

TH cells can also differentiate into other subtypes, such as TH17 cells and regulatory T cells (Tregs). TH17 cells selectively produce the proinflammatory cytokine IL-17 and they also produce proinflammatory cytokines IL-21 and IL-22 (48,49). In contrast, Tregs play a role in immunoregulation and tolerance (50–52). During normal pregnancy, the Treg population expands both systemically and at the maternal-fetal interface. These cells have protective functions through cell-to-cell interactions and the production of anti-inflammatory cytokines. Tregs produce IL-10 and TGF-β, which can act to suppress the effector functions of other activated immune cell populations. Anti-inflammatory cytokines, such as IL-10 and IL-4, play a critical role in successful pregnancies and, therefore, Tregs, and TH2 cells, are believed to be critical in maintaining immune balance (53). Moreover, analysis of blood from women with PE shows a significant shift in the populations of circulating TH cell subtypes. Clinical studies demonstrate an increase in TH17 cells alongside a decrease in Tregs in peripheral and umbilical blood samples (51,54). Increased TH17 cells in patients with PE are hypothesized to increase the production of cytokines, recruitment of other immune cells, and the generation of oxidative stress in the placenta (Figure 2) (27,55,56). However, Barnie et al. (57) have shown that increases in IL-17 may also be due to elevations in type 3 innate lymphoid cells. Additionally, Tregs have been shown to lose their suppressive function, and their proliferation is downregulated in autoimmune diseases. This decreased Treg number combined with loss of function contributes to a lack of tolerance toward fetal alloantigens. Interestingly, Tregs adoptively transferred from normal pregnant rats into reduced uterine perfusion pressure (RUPP) rats improves BP and lowers inflammatory mediators (50). The RUPP model is a surgically induced model of PE that recaptures many PE symptoms, such as elevated BP and inflammatory activation, including the predominance of TH1 versus TH2 cells and B cells secreting the AT1-AA. This indicates the downregulation of Tregs in response to placental ischemia plays a role in pathology of the RUPP rat and has a protective role during normal pregnancy to control inflammation.

The laboratory of Zenclussen et al. (58) demonstrated that the adoptive transfer of activated TH1-like cells into pregnant mice induced a PE-like phenotype, including elevated BP, proteinuria, and cytokine production. Our laboratory performed similar studies showing adoptive transfer of CD4+ T cells from the RUPP rat model of PE into normal pregnant rats increased BP; inflammatory cytokines TNF-α, IL-6, and IL-17; and increased production of placental endothelin-1 (ET-1) and circulating soluble fms-like tyrosine kinase 1 (sFlt-1) and AT1-AA (59,60). Studies also demonstrate that inflammatory cytokines, such as TNF-α, IL-6, and IL-17, play crucial roles in immune activation by increasing the production of vasoactive molecules, such as AT1-AA, ET-1, and sFlt-1 (61–63). Importantly we recently showed that adoptive transfer of CD4+ T cells from placentas of patients with PE causes hypertension; intrauterine growth retardation; and increased AT1-AA, sFlt-1, ET-1, and the cytokines TNF-α and IL-17, demonstrating the importance of T cells in causing many characteristics of PE (64).

After a primary antigen response, most activated T cells die off; however, some differentiate into memory T cells that remain to provide quicker responses to a secondary antigenic exposure (65). Memory T cells can fall into subsets, mainly central memory and effector memory cells. Central memory cells differentiate into effector cells upon antigen exposure, and effector memory cells reside in peripheral tissue and conduct the proinflammatory response upon antigen exposure (66,67). There is a delicate balance of these subtypes in a normal pregnancy that is usually higher than in women who are not pregnant (65). During complicated pregnancies like PE, this balance is disrupted and these cells contribute to the chronic inflammation seen in the disorder (65).

Another important role that T cells play in normal immunity is the costimulation of B cells (Figure 2) (68). This interaction occurs between the T cell receptor, CD40 ligand (CD40L), the MHC class II molecule, and CD40 receptor on B cells (69,70). Once this activation takes place, B cells proliferate and differentiate into plasma cells and memory B cells (71,72). Plasma cells provide short-term, immediate protection, whereas memory B cells provide long-term, persistent protection through antibody production (73). We have shown that communication between T cells and B cells is crucial for the development of PE symptoms in an adoptive transfer model of PE. TH cells were isolated from RUPP animals and incubated with anti-CD40L to block communication with B cells before transfer into normal pregnant rats (74). This blockade improved PE characteristics compared with normal pregnant rat recipients of control RUPP T cells. We have also shown that T cells isolated from patients with PE can cause PE symptoms in pregnant nude-athymic rats (75). More recently, we showed that incubation with anti-CD40L before adoptive transfer also improved PE symptoms in the nude-athymic model, further indicating that T cells costimulating B cells is critical to the pathophysiology of PE (64).

B Cells in Preeclampsia

B cells are typically divided into two different populations: B1 and B2 cells. B1 cell development occurs during fetal life and their precursors are present in the fetal liver. These cells are typically found in peripheral circulation in humans, but are found in the peritoneal cavity in murine animals. B1 cells produce “natural antibodies” that do not require an antigenic stimulus. B2 cells are produced from precursors present in the bone marrow and migrate to the spleen where they can then undergo antigen exposure and costimulation from T cells. After this exposure and costimulation, B2 cells undergo Ig class switching and differentiation into plasma cells and memory B cells. During normal pregnancies, B cells are believed to produce protective asymmetric antibodies, and the failure to produce these antibodies results in a failure of the pregnancy (76). Canellada et al. (77) showed that B cells isolated from the placenta and stimulated with CD40L produced a large amount of these protective antibodies. Stimulation with CD40L also supports the hypothesis that T cell–to–B cell communication is key during pregnancy.

B cells have been implicated in the pathology associated with PE, primarily due to the production of autoantibodies (78,79). Anti-phospholipid antibodies have been associated with PE, spontaneous abortion, and intrauterine fetal death (80). Velasquillo et al. (81) have shown a high number of B1a (CD19+CD5+) cells in the peripheral circulation of patients with antiphospholipid syndrome. B1a cells are fetal in origin and have a role in other autoimmune diseases, including type 1 diabetes and lupus erythematosus (82,83). Jensen et al. (84) showed that B1a cells isolated from the blood of patients with PE can produce AT1-AA. Other studies have shown that autoantibody production is persistent for years after a preeclamptic pregnancy, indicating a potential memory mechanism involved in their production (37,38). This would also implicate B2 cells in the production of autoantibodies during PE. B2 cells are classic B cells that produce antibodies after T cell–dependent activation resulting in long-lived memory B cells that continue to produce antibodies (73). This would further implicate B2 cells in chronic inflammation and increased risk of cardiovascular and metabolic disease later in life after a PE pregnancy due to the persistent production of AT1-AA.

NK Cells in Preeclampsia

NK cells are a type of innate lymphoid cell that are important in human pregnancy and systemic regulation due to their high cytolytic potential against tumor-transformed and virus-infected cells (85). NK cells develop in the bone marrow from lymphoid progenitor cells and account for approximately 70% of uterine lymphocytes and 5%–10% of peripheral lymphocytes during the first trimester of pregnancy (85,86). When activated, NK cells release perforin and granzymes that induce cell lysis and they produce proinflammatory cytokines IFN-γ and TNF-α (87,88). However, uterine NK cells promote placental vascular growth rather than cytotoxic activity (89,90). Furthermore, these cells assist with trophoblast invasion and spiral artery remodeling, early processes that are critical for a successful pregnancy, through cytokine production and angiogenic factor secretion (89). The amount and type of NK cells in the decidua during pregnancy appears to be critical for pregnancy success (89,91).

Women with pregnancy disorders such as PE have a shift in NK cell populations away from the regulatory uterine NK cells toward proinflammatory cytolytic NK cells (42,91). Cytolytic NK cell number and activity are elevated in patients with PE (92). Furthermore, clinical studies have demonstrated shifts in the NK cell population in women with PE due to the upregulation of proinflammatory cytokines, such as IL-2, IL-17, TNF-α, and IFN-γ, which also promotes a TH1 cell response (93,94). These elevated cytokines, coupled with increases in other proinflammatory cells and a decrease in anti-inflammatory cells, cause NK cells to shift from regulatory uterine NK cells to cytolytic NK cells (93). Cytolytic NK cells primarily destroy infected or cancer cells through the release of cytolytic enzymes (95). These activated NK cells also produce TNF-α and INF-γ, further contributing to the chronic inflammation in PE (42,96). Recent studies from Travis et al. (56,97) show that NK cells stimulated from the RUPP rat model of PE can cause preeclamptic symptoms in healthy pregnant rats, and that NK cells from PE rats have a five-fold increase in cytolytic activity compared with normal controls. Travis et al. (56) also demonstrated that blockade of IL-17 with its soluble receptor, IL-17 receptor C, reduced NK cell activation and improved PE symptoms in RUPP rats. They postulate that activation of NK cells is one mechanism by which IL-17 contributes to hypertension in PE and that activated NK cells are an important cellular mediator in PE pathophysiology.

On the basis of our previous studies, our overall laboratory hypothesis demonstrates that placental ischemia causes lymphocyte activation, which contributes to hypertension through multiple mechanisms. T cells function to activate B cells, but also secrete inflammatory cytokines that lead to endothelial activation and dysfunction, thus contributing to hypertension. Furthermore, B cells secrete AT1-AA that directly activate the AT1R, resulting in increased vascular resistance but also producing antiangiogenic factors downstream of AT1R activation. AT1-AA and inflammatory cytokines contribute to the activation of NK cells that can kill AT1AA-bound cells, which can contribute to increased multiorgan dysfunction and increased vascular tone, leading to increased vascular resistance and contributing to hypertension.

Experimental Therapeutics Targeting Lymphoid Cells

Due to the critical role that immune cells play in the pathogenesis of PE, numerous therapeutics targeting the immune system have been tested with in vivo or in vitro models of PE. A current preventative for women at high risk of PE is a low-dose course of aspirin that inhibits PG thromboxane A2 (98). Aspirin use early in pregnancy can prevent or delay the onset of PE (98). Aspirin is a cyclooxygenase inhibitor that prevents the production of PGs and thromboxanes. The exact mechanism by which aspirin prevents PE is unknown, but the current hypothesis is that it improves the placentation process, inhibits placental infarct by reducing platelet aggregation through inhibition of thromboxane production, and reduces inflammation and stabilizes the endothelium through inhibition of PG (99,100). Low mol wt heparin has also been used to prevent PE through its anti-inflammatory, anticoagulant, and immunomodulatory effects (101). Patients with SLE continue on hydroxychloroquine, an antimalarial drug that has immunosuppressive effects, during their pregnancy to reduce the risk of PE by reducing T cell activation (102).

Preclinical studies modulating lymphoid cells in the RUPP rat model of PE demonstrate a potential therapy that could be attempted in patients with PE. Rituximab, an anti-CD20 mAb that depletes B cells, improved PE symptoms in the RUPP model of PE (103). B cell depletion correlated with reduced circulating AT1-AA, circulating inflammatory cytokines, and maternal BP. Importantly, rituximab exposure during pregnancy does not increase the rate of preterm delivery or congenital malformations, and there is no evidence the incidence of PE increased in this patient population (104,105). Abatacept, a fusion molecule that inhibits costimulation of T cells by binding to cytotoxic T-lymphocyte–associated antigen 4, has shown beneficial effects in RUPP rats and in another model of PE rats (106,107). Abatacept improves maternal BP, oxidative stress in the placenta and kidney, and anxiety-related behaviors (108). Investigators have also looked for ways to block the actions of IL-17 because TH17 cells seem to play a crucial role in PE pathophysiology. Infusion of soluble IL-17 receptor C into RUPP rats decreased TH17 cell population, AT1-AA production, and BP, and improved pup weight (109). AT1-AA has been targeted directly with a seven amino acid sequence peptide that binds to AT1-AA on its binding region in animal and cell culture models (110,111). This blockade improves BP, NK cell activation, and mitochondrial oxidative stress, and decreases circulating sFlt-1 and ET-1. Etanercept, a soluble receptor for TNF-α, also demonstrated positive effects in RUPP rats by improving maternal BP, circulating inflammatory cytokines, NK cell activation, and ET-1 (112,113). Although there are preclinical studies and some clinical situations in which lymphoid-modulating drugs show promise to alleviate PE phenotypes, the effects and safety of immune-modulating therapeutics on fetal health is still under investigation and require further study before they can become options for clinical use.

Summary

PE is defined as new-onset hypertension with multiorgan dysfunction, including the brain, liver, kidney, cardiovascular system, and the placenta. PE is also associated with abnormal changes in expression of RAAS components that could contribute to the overall pathophysiology of the disorder. Moreover, compelling evidence suggests that PE is also associated with chronic immune activation, characterized by a shift toward proinflammatory cytokines and lymphoid cell types and away from anti-inflammatory cytokines and regulatory lymphoid cells. Although the mechanisms that cause PE are still unclear, the contribution of the immune system to preeclamptic symptoms is unfolding. Multiple cell types contribute to endothelial dysfunction, tissue damage, and further immune activation to cause hypertension in pregnancy. Additionally, autoantibodies play a clear role in contributing to the tissue damage and elevated BP that characterize PE. Better understanding the role of the immune system in the pathophysiology of PE is critical because it allows researchers and clinicians new possibilities for therapy.

Disclosures

B.B. LaMarca reports serving in an advisory or leadership role for Department of Justice for the United States of America and University of Mississippi Medical Center; receiving honoraria from the National Institutes of Health and University of Alabama Birmingham; and serving on a speakers bureau for University of Alabama Birmingham. All remaining authors have nothing to disclose.

Funding

This work was supported by the National Institutes of Health (NIH) (grants RO1HD067541-11 to B.B. LaMarca and P20GM121334 to B.B. LaMarca) and the T32 Trainee (grant T32-HL105324 to E.D. Deer, N.E. Campbel, and O.T. Herrock).

Author Contributions

N.E. Campbell was responsible for project administration; N.E. Campbell, E.M. Deer, O.T. Herrock, and B.B. LaMarca reviewed and edited the manuscript; and all authors conceptualized the study and wrote the original draft.