Fas ligand neutralization attenuates hypertension, endothelin-1, and placental inflammation in an animal model of HELLP syndrome

Abstract

Neutralization of FasL is linked to suppression of hypertension, placental inflammation, and endothelin system activation in an animal model of hemolysis, elevated liver enzymes, low platelets (HELLP) syndrome. During HELLP syndrome the placenta has been reported to serve as the primary source of Fas ligand (FasL), which has an impact on inflammation and hypertension during pregnancy and is dysregulated in women with severe preeclampsia and HELLP syndrome. We hypothesize that neutralization of FasL during pregnancy in an animal model of HELLP syndrome decreases inflammation and placental apoptosis, improves endothelial damage, and improves hypertension. On gestational day (GD) 12, rats were chronically infused with placental antiangiogenic factors sFlt-1 and sEng to induce HELLP syndrome. To neutralize FasL, MFL4 or FasL antibody was infused into a subset of HELLP or normal pregnant rats on GD13. IgG infusion into another group of NP and HELLP rats on GD13 was used as a control for FasL antibody, and all rats were euthanized on GD19 after blood pressure measurement. Plasma and placentas were collected to assess inflammation, apoptosis, and the degree of placental debris activation of endothelial cells. Administration of MFL4 to HELLP rats significantly decreased blood pressure compared with untreated HELLP rats and HELLP rats infused with IgG and improved the biochemistry of HELLP syndrome. Both circulating and placental FasL were significantly attenuated in response to MFL4 infusion, as were levels of placental and circulating TNFα when compared with untreated HELLP rats and HELLP rats infused with IgG. Endothelial cells exposed to placental debris and media from HP + MFL4 rats secreted significantly less endothelin-1 compared with stimulated endothelial cells from HELLP placentas. Neutralization of FasL is associated with decreased MAP and improvement in placental inflammation and endothelial damage in an animal model of HELLP syndrome.

INTRODUCTION

Hemolysis, elevated liver enzymes, low-platelets (HELLP) syndrome is a rare but serious disorder of pregnancy that results in significant maternal and neonatal morbidity and mortality (1, 27). The disorder occurs in 10–20% of patients diagnosed with preeclampsia (sudden onset of hypertension during pregnancy) (14). Although the exact mechanisms causing HELLP syndrome remain undefined, factors associated with inadequate remodeling of the uterine spiral arteries, including antiangiogenic imbalance, endothelial dysfunction, inflammation, and apoptosis, have been proposed to contribute to the development of the disease (14, 39).

The Fas receptor and Fas ligand (FasL) system play a significant role in the normal physiological process of apoptosis and T lymphocyte regulation (6, 37) and an important role in placentation and trophoblastic diseases (22). During normal pregnancy, extravillous trophoblast cells contribute to the remodeling of the uterine spiral arteries by infiltrating the uterine spiral arteries to replace the endothelial cell lining of the lumen (4, 7). One mechanism by which trophoblast cells are thought to disrupt the endothelial lining is through apoptosis of Fas-expressing endothelial and smooth muscle cells (3, 15, 17). FasL and Fas/FasL polymorphisms have been reported to be increased in both clinical and experimental models of HELLP syndrome, suggesting dysregulation of this system in HELLP (10, 28, 29, 37). Women with HELLP and preeclampsia also have been reported to have an increase in activated macrophages that can possibly contribute to the apoptosis of trophoblast cells, leading to inadequate trophoblast invasion of the uterine spiral arteries (30, 31).

Interestingly, a previous study by Strand et al. (37) determined that during HELLP syndrome the placenta serves as the primary source of circulating FasL. Using an animal model of HELLP syndrome, we have also reported that rats with HELLP have an increase in placental and circulating FasL that is blunted (nonsignificant) in response to endothelin receptor A antagonism (10). Based on these previous studies, the goal of the current study was to determine whether FasL plays a role in contributing to the hypertension, increased endothelin-1 secretion, inflammation, and placental apoptosis that occurs in HELLP syndrome.

MATERIALS AND METHODS

All studies were performed using 230- to 250-g timed-pregnant Sprague-Dawley rats (Envigo, Indianapolis, IN). Animals were housed in a temperature-controlled room with a 12:12-h light-dark cycle. All experimental procedures in this study were in accordance with the National Institutes of Health guidelines for use and care of animals and were approved by the Institutional Animal Care and Use Committee at the University of Mississippi Medical Center.

Experimental animal model of HELLP syndrome plus MFL4 infusion.

To induce HELLP syndrome, sEng and sFlt-1 (7 and 4.7, μg/kg respectively; R & D Systems, Minneapolis, MN) were diluted in sterile 0.9% saline, loaded into mini-osmotic pumps (model 2002; Alzet Scientific, Cupertino, CA), and surgically inserted into the peritoneal cavity of pregnant rats (n = 23) for chronic infusion from gestational day (GD) 12 to GD19, as previously described (23, 24, 40). Normal pregnant (NP) rats that did not receive a mini-osmotic pump served as the normal pregnant control group (n = 8 rats). To examine the effects of FasL blockade, we infused anti-FasL antibody MFL4 (500 ng/kg; hamster anti-mouse rat FasL IgG; BD Biosciences cat. no. 555022, RRID:AB 395653) via the jugular vein into either HELLP (n = 7 rats) or NP (n = 7 rats) rats on GD13. Data from preliminary studies indicated that the current MFL4 infusion regimen [i.e., route of delivery, gestational day, and concentration (100, 250, 500, and 750 ng/kg)] significantly decreased circulating and placental FasL and mean arterial pressure and did not have a negative effect on pup size.

An additional group of NP (n = 7 rats) and HELLP (n = 8 rats) rats was infused with an equal amount of hamster IgG (Southern Biotech) on GD13 via the jugular vein. On GD18, carotid artery catheters were placed in each rat, and on GD19, following mean arterial pressure (MAP) measurement, maternal organs and pups were measured and serum and plasma collected. Maternal placentas were immediately frozen, stored in 10% buffered formalin, or used for tissue culture or apoptosis. Lactate dehydrogenase (LDH) and aspartate aminotransferase (AST) and platelet levels were measured as previously described (40).

Placental histopathology.

Placentas collected on GD19 were fixed in 10% buffered formalin before paraffin embedment and cut into 4-µm sections. Consecutive sections were subjected to hematoxylin and eosin (H&E) pathological staining and were photographed using an EVOS XL microscope for morphological evaluation. The decidua (D), basal (B), and labyrinth (L) zones of the placenta were measured under low magnification using Image J software from three placental sections per animal (n = 4 animals/group) (2). The ratios of decidua, basal, and labyrinth zones were calculated per section per animal and averaged to determine the total area of the placenta each zone occupied within treatment groups.

Effect of MFL4 infusion on FasL, TNFα, sFlt-1, and sEng.

Rat FasL, TNFα, sFlt-1, and sEng were assessed in the circulation via ELISA (RayBiotech, Norcross, GA; Millipore, Temecula, CA; R & D Systems, Minneapolis, MN) to determine whether MFL4 infusion decreased these proteins. Individual placentas were homogenized, centrifuged (10, 40), and assayed for these same protein levels. Total placental protein was quantitated using a BCA protein kit from Pierce (ThermoFisher, Rockford, IL).

MFL4 infusion and placental apoptosis.

To assess apoptotic activity, placentas of rats were harvested and weighed. Tissues were minced and digested at 37°C for 45 min in 200 mg/mL collagenase IV solution (collagenase IV; Gibco, Gaithersburg, MD; 1 μg/mL DNase I; Sigma) in Ca²⁺-free PBS (Gibco). After the samples were filtered, the filtrate was labeled with fluorochrome allophycocyanin-conjugated annexin V antibody (BD Biosciences, San Jose, CA) and propidium iodide (Sigma) for flow cytometry analysis. Cells were measured via a Beckman Coulter Gallios Flow Cytometer and data analyzed with Kaluza software.

Effect of FasL neutralization on endothelin-1 secretion from human umbilical vein endothelial cells.

Single placental explants per rat were cultured on six-well plates in 2.5 mL of culture media [50% Dulbecco’s modified Eagle’s medium (Gibco), 50% VascuLife medium (Lifeline Cell Technology, Frederick, MD) with 10% fetal bovine serum (FBS; Hyclone, Logan, UT) and 1% antimycotic-antibiotic (Gibco)]. The cultures were allowed to incubate for 24 h at 37°C in a humidified atmosphere of 5% CO₂, 20% O₂, and 75% N₂. After 24 h, placental explants were crushed in their respective media before collection, as it has been previously reported that the placental debris is required along with secretory factors to stimulate endothelial reactivity (32, 33, 41). This conditioned media was stored at −20°C until the endothelial cell experiments were performed.

To determine whether any responses to MFL4 were pharmacological versus physiological, we incubated placentas from NP and HELLP rats, with and without IgG infusion, with increasing concentrations of MFL4 (0–10 ng) (38) for 24 h and then collected the media, as described above.

Human umbilical vein endothelial cell culture.

Human umbilical vein endothelial cells (HUVECs; passage 2–5; Lifeline Cell Technology, Frederick, MD) were cultured in VascuLife medium, 10% FBS and 1% antimycotic-antibiotic at 37°C in a humidified atmosphere of 5% CO₂, 20% O₂, and 75% N₂. Once cells were 70–80% confluent, they were incubated for 24 h in serum-free media before 50% of the conditioned media collected above was added. HUVEC cells were exposed to conditioned media (50/50 HUVEC media; placental explant media) for 24 h and then removed (24). Fresh serum-free media was added, and the cells were cultured for an additional 24 h. At the end of the experiment, media were collected for endothelin-1 (ET-1) analysis via ELISA (R & D Systems) and normalized to total protein in the media (measured via BCA) (the data are expressed as pg·mg⁻¹·mL⁻¹).

Statistical analysis.

All of the data are expressed as means ± SE. Comparisons between groups were analyzed via one or two-way analysis of variance with Tukey’s multiple-comparisons tests or Student’s t test. P < 0.05 was considered statistically significant.

RESULTS

Infusion of MFL4 attenuates hemolysis, liver injury, thrombocytopenia, and hypertension in HELLP rats.

Hemolysis, assessed by LDH (P < 0.0001) and liver injury and measured by aspartate aminotransferase (AST; P = 0.01), was significantly increased in HELLP rats compared with NP rats, whereas the number of platelets were significantly decreased (P < 0.0001; Fig. 1). There was not a significant difference in hemolysis between NP + IgG and HELLP + IgG rats (P = 0.10; Fig. 1A), but liver enzyme levels were significantly increased in HELLP + IgG compared with NP + IgG rats (P < 0.0001). Similar to uninfused rats, platelets were significantly decreased in HELLP + IgG rats relative to NP + IgG (P = 0.05) rats. There were no statistically significant differences between NP + MFL4 and HELLP + MFL4 rats in LDH (P = 1), AST (P = 1), or platelet levels (P = 1).

Fig. 1.

Infusion of MFL4 into hemolysis, elevated liver enzymes, low platelets (HELLP) rats improves HELLP profile and mean arterial pressure compared with untreated HELLP rats. HELLP rats have increased lactate dehydrogenase (A), increased aspartate aminotransferase (B), decreased platelet count (C), and increased mean arterial pressure (D) compared with normal pregnant (NP) rats. One-way ANOVA with Tukey’s multiple-comparison test was used to determine statistical significance. *P < 0.05, ***P < 0.005, and ****P < 0.00005 compared with the indicated groups; ^aP < 0.05 compared with HELLP rats, ^bP < 0.05 compared with HELLP + IgG rats, ^cP < 0.05 compared with NP + IgG rats, and ^dP < 0.05 compared with NP rats; n = 7–8/group.

Upon multiple comparison analysis, it was noted that infusion of MFL4 in HELLP rats significantly decreased LDH (P < 0.0001, P < 0.0001) and AST (P = 0.0006, P < 0.0001) compared with HELLP and HELLP + IgG rats (Fig. 1, A and B). Platelets in HELLP + MFL4 rats were significantly increased (P = 0.004, P = 0.0003) compared with HELLP and HELLP + IgG rats but did not reach NP platelet levels (P = 0.03; Fig. 1C). HELLP + IgG rats had AST levels that were significantly increased compared with NP (P < 0.0001) and HELLP rats (P = 0.0001; Fig. 1B) and platelets that were significantly decreased compared with NP rats (P < 0.0001). NP + IgG rats had significantly increased LDH compared with NP rats (P = 0.007).

Blood pressure was significantly increased in HELLP rats compared with NP rats (P = 0.01) and in HELLP + IgG versus NP + IgG (P = 0.03; Fig. 1D) rats. There was not a statistically significant difference in blood pressure between HELLP + MFL4 and NP + MFL4 rats (P = 0.74). There was a statistical interaction between blood pressure and treatment group, where HELLP + MFL4 rats had significantly lower pressures compared with HELLP (P = 0.02) and HELLP + IgG (P = 0.04) rats. There was not a statistically significant difference in the number of fetal resorptions (P = 0.67), placental weight (P = 0.78), birth weight (P = 0.24), or fetal/placenta weight (P = 0.09) between the groups (Table 1).

Table 1.

Average birth and organ data per group

	NP (n = 8)	HELLP (n = 8)	NP + IgG (n = 7)	HELLP + IgG (n = 8)	NP + MFL4 (n = 7)	HELLP + MFL4 (n = 7)	P Value
%Fetal resorptions	3.8 ± 2.1	7.6 ± 4.1	4.27 ± 3.18	1.39 ± 1.38	3.07 ± 1.45	4.32 ± 2.27	0.67
Placental weight, g	0.48 ± 0.02	0.53 ± 0.04	0.49 ± 0.03	0.52 ± 0.02	0.52 ± 0.03	0.54 ± 0.03	0.78
Fetal weight, g	2.39 ± 0.07	2.36 ± 0.08	2.03 ± 0.17	2.13 ± 0.23	2.39 ± 0.05	2.39 ± 0.04	0.24
Fetal/placenta, g	4.96 ± 0.15	4.34 ± 0.19	4.05 ± 0.15	4.13 ± 0.4	4.73 ± 0.33	4.62 ± 0.14	0.09

Data are displayed as means ± SE; n, number of rats. HELLP, hemolysis, elevated liver enzymes, low platelets; NP, normal pregnant. Differences between groups were determined by one-way analysis of variance.

H&E staining was used to measure the area of the placenta occupied by the decidual, basal, and labyrinth zones (Fig. 2, A–F). There was a significant zone affect [F_{(2, 45)} = 47.15, P < 0.0001], with the decidua occupying 7.8–23.9% and basal 22.8–38.1%, and the labyrinth zone occupied 45.34–60.35% of the placental area (Fig. 2G). There was not a treatment effect between the groups of animals [F_{(5, 45)} > 0, P > 0.99] or an overall interaction between the different groups and zones of the placenta [F_{(10, 45)} = 1.03, P = 0.43].

Fig. 2.

No significant differences in placental morphology on gestational day 19. Representative photomicrographs of hematoxylin and eosin-stained placental sections showing normal placental morphology consisting of the decidua (de), basal (ba), and labyrinth (lab) zones. A–F: normal pregnant (NP; A), hemolysis, elevated liver enzymes, low platelets (HELLP; B), NP + IgG (C), HELLP + IgG (D), NP + MFL4 (E), and HELLP + MFL4 (F). Scale bar, 400 µM. G: there were no significant differences (2-way ANOVA) in the percent of the placenta occupied by each zone within the different groups; n = 4 rats/group.

Neutralization of FasL decreases systemic TNFα and sFlt-1 in HELLP rats.

To confirm that infusion of MFL4 decreased FasL, we measured FasL in the circulation. There was a significant relationship between FasL and experimental groups, so post hoc analysis was performed (Fig. 3). FasL was significantly increased in HELLP rats compared with NP rats (P = 0.009) and in HELLP + IgG versus NP + IgG rats (P < 0.0001) but not HELLP + MFL4 versus NP + MFL4 rats (P = 0.78). Infusion of MFL4 into HELLP rats attenuated FasL compared with HELLP (P = 0.0003) and HELLP + IgG rats (P < 0.0001). FasL concentrations in NP + IgG and HELLP + IgG were significantly increased compared with untreated NP and HELLP rats. TNFα, which plays an important role in both inflammation and apoptosis, was significantly increased in HELLP versus NP rats (P = 0.001) and in HELLP + IgG versus NP + IgG rats (P = 0.003), but not between HELLP + MFL4 and NP + MFL4 rats (P > 0.99). HELLP + MFL4 rats had significantly decreased TNFα compared with HELLP (P = 0.0007) and HELLP + IgG (P < 0.0001) rats. NP + IgG and HELLP + IgG rats had significantly increased TNFα compared with untreated NP rats and HELLP + IgG versus HELLP rats (Fig. 3).

Fig. 3.

Circulating and placental homogenate ELISA results from experimental groups. Normal pregnant (NP; n = 5–8 rats), NP + IgG (n = 4–7 rats), NP + MFL4 (n = 4–7 rats), hemolysis, elevated liver enzymes, low platelets (HELLP; n = 5–8 rats), HELLP + IgG (n = 5–7 rats), and HELLP + MFL4 (n = 4–7 rats). Data are expressed as means ± SE. P < 0.05 based on one-way ANOVA with Tukey post hoc analysis *P < 0.05, **P < 0.005, ***P <0.0005, and ****P < 0.00005 compared with the indicated groups; ^aP < 0.05 compared with HELLP, ^bP < 0.05 compared with HP + IgG, and ^cP < 0.05 compared with NP + IgG. FasL, Fas ligand.

We next examined antiangiogenic factors sFlt-1 and sEng, which have placental origins and are significantly increased in placental ischemic pregnancies such as HELLP syndrome. Circulating sFlt-1 was significantly increased in HELLP rats compared with NP rats (P = 0.002) and in HELLP + IgG versus NP+IgG rats (P = 0.01) but not between HELLP + MFL4 and NP + MFL4 rats (P = 0.89). Importantly, HELLP + MFL4 rats had significantly lower sFlt-1 compared with HELLP rats (P = 0.02). sEng was significantly increased in HELLP rats compared with NP (P = 0.01) but not in HELLP + IgG rats versus NP + IgG rats (P = 0.67) or between NP + MFL4 and HELLP + MFL4 rats (P = 0.29). sEng was not significantly decreased in HELLP + MFL4 rats compared with untreated HELLP rats (P = 0.40).

Neutralization of FasL decreases placental TNFα and sEng in HELLP rats.

Because the placenta has been suggested to be a primary source of FasL (37), we examined protein expression in placentas from experimental rats. FasL was significantly increased in placental homogenates from HELLP rats compared with NP (P < 0.0001) and in HELLP + IgG rats compared with NP + IgG rats (P < 0.0001; Fig. 3). FasL was not significantly different in HELLP + MFL4 rats compared with NP + MFL4 rats (P = 0.59). FasL was significantly decreased in HELLP + MFL4 rats compared with HELLP rats (P < 0.0001) and HELLP + IgG rats (P < 0.0001). Placental TNFα was significantly increased in HELLP rats compared with NP (P = 0.002) and in HELLP + IgG versus NP + IgG rats (P = 0.0006), but there was no difference between HELLP + MFL4 and NP + MFL4 rats (P = 0.93). HELLP + MFL4 rats had significantly decreased TNFα compared with HELLP (P = 0.0007), NP + IgG (P = 0.004), and HELLP + IgG rats (P < 0.0001). Similar to what was seen in the circulation, infusion of IgG significantly increased TNFα in HELLP rats to concentrations above that in HELLP rats (P = 0.0004; Fig. 3).

HELLP rats had significantly increased placental sFlt-1 compared with NP rats (P = 0.03), as did HELLP + IgG versus NP + IgG rats (P = 0.03). However, there was not a significant difference in sFlt-1 levels between NP + MFL4 and HELLP + MFL4 rats (P = 0.08). Infusion of IgG into NP and HELLP rats significantly increased placental sFlt-1 to concentrations greater than those in uninfused NP (P = 0.02) and HELLP rats (P = 0.006) (Fig. 3). HELLP + MFL4 rats did not have significantly less sFlt-1 compared with HELLP rats (P = 0.51), but they did compared with HELLP + IgG (P = 0.003). sEng levels were increased in HELLP rats compared with NP rats (P = 0.003) and when levels were also increased in HELLP + IgG versus NP + IgG (P = 0.0008). However, there was not a significant difference in placental sEng levels between NP + MFL4 and HELLP + MFL4 rats (P = 0.51). Infusion of IgG into NP rats significantly decreased sEng compared with NP rats (P = 0.005). HELLP + MFL4 rats had significantly less sEng compared with HELLP rats (P = 0.002).

Attenuation of FasL does not decrease placental cell apoptosis in HELLP rats.

Placental tissues were digested and cells stained with annexin V-APC and propridium iodide for flow cytometry to estimate MFL4’s ability to reduce apoptosis in rats with HELLP syndrome. There was a significant overall interaction in the percentages of cells (necrotic vs. apoptotic vs. healthy) between the different groups of animals [F_{(10, 63)} = 27.85, P < 0.0001; Fig. 4, A–F]. In HELLP rats, placental apoptosis was significantly increased (P = 0.001), and the number of healthy cells (P = 0.003) was significantly decreased compared with NP. Infusion of IgG into NP rats led to a significant reduction in apoptosis, which resulted in a significantly higher percentage of healthy cells compared with all other groups. Whereas there were significant differences in the percentage of apoptotic (P < 0.0001) or healthy cells (P < 0.0001) between HELLP + IgG and HELLP rats, treatment with MFL4 did not significantly decrease apoptosis (P = 0.347) or increase the percentage of healthy cells in HELLP rats compared with untreated HELLP rats (P = 0.5; Fig. 4G).

Fig. 4.

Detection of apoptosis by annexin V-propidium iodide (PI) in placenta tissue. Flow cytometric analysis of placental tissue stained with annexin V and PI in placental tissue on gestational day 19. One median weight placenta per animal was used. Necrotic cells (Q1), late apoptotic (Q2), healthy cells (Q3), and early apoptotic (Q4). A–F: normal pregnant (NP; n = 5 rats; A), hemolysis, elevated liver enzymes, low platelets (HELLP; n = 5 rats; B), NP + IgG (n = 4 rats; C), HP + IgG (n = 4 rats; D), NP + MFL4 (n = 5 rats; E), and H⁺MFL4 (n = 6 rats; F). G: the proportion of apoptotic cells gated was combined to represent the total amount of apoptotic cells. H: endothelin-1 secretion decreased in response to MFL4 infusion. One-way ANOVA with Tukey’s multiple-comparison test was used to determine statistical significance. *P < 0.05, **P < 0.005, and ****P < 0.00005 compared with the indicated groups; ^aP < 0.05 compared with HELLP; ^bP < 0.05 compared with HELLP + IgG; #P < 0.00001 compared with all other groups.

FasL neutralization prevents endothelial cell damage.

Endothelin was measured in media and placental debris [conditioned media (CM)] from cultured experimental placentas that were incubated with HUVECs to determine whether neutralization of FasL prevents an increase in ET-1 secretion from endothelial cells. HUVECs exposed to HELLP-CM secreted significantly more ET-1 compared with cells exposed to NP-CM (P = 0.04; Fig. 4H). There were no statistically significant increases in ET-1 secretion between NP + IgG and HELLP + IgG (P = 0.27) or NP + MFL4 and HELLP + MFL4 (P = 0.71) exposed cells. HUVECs exposed to conditioned media from HELLP + MFL4 placentas secreted significantly less ET-1 compared with HUVECs from the untreated HELLP-CM group (P = 0.03) and from HELLP + IgG group (P = 0.006).

There was a significant overall treatment × MFL4 concentration interaction in ET-1 secretion from HUVECs exposed to placentas with and without exogenous MFL4 [F_{(6, 49)} = 3.07, P = 0.01]. HUVECs exposed to CM + 10 ng MFL4 from NP + IgG (0.15 ± 0.14 pg·mg⁻¹·mL⁻¹, P = 0.003) and HP + IgG (0.12 ± 0.05 pg·mg⁻¹·mL⁻¹, P = 0.003) secreted significantly less ET-1 compared with cells exposed to CM + 10 ng MFL4 from HP rats (0.54 ± 0.18 pg·mg⁻¹·mL⁻¹). There were no other significant interactions between any groups regardless of the concentration of MFL4.

Comment.

The current study was performed to examine the role of FasL in contributing to hypertension, placental apoptosis, and inflammation in response to HELLP syndrome. We have previously reported that in our animal model of HELLP syndrome, circulating and placental FasL is significantly increased (10); however, we have not examined the role of increased FasL during pregnancy. As such, the current study was performed to examine the role of FasL in contributing to the manifestations of HELLP syndrome, placental apoptosis, and inflammation in the animal model of HELLP syndrome. The findings from our study indicated that FasL contributes to the symptomatology, hypertension, placental inflammation, and increased endothelin-1 secretion from endothelial cells in HELLP syndrome.

Both in vivo and in vitro studies have reported that administration of an anti-FasL antibody neutralizes the apoptotic actions of FasL and decreases inflammation (18, 37). Whereas administration of MFL4 (anti-FasL antibody) to HELLP rats did attenuate circulating FasL and placental FasL, placental apoptosis was unaffected by the decrease in FasL. FasL neutralization did not significantly decrease placental apoptosis or increase the percent of healthy cells relative to untreated HELLP placentas. Studies have found that women with preeclampsia and HELLP syndrome have significantly more placental and trophoblast apoptosis in late pregnancy compared with normotensive women. Similar findings to our study were reported in an asthma animal model utilizing MFL4 to neutralize the actions of FasL, where there was not a decrease in apoptosis despite a decrease in both FasL and inflammatory cytokines (34). Although the decreased apoptosis in MFL4-treated rats as measured by annexin V was not significant, it is possible that different methods of flow cytometry may detect a significant decrease in other components of the apoptosis pathway. As such, in the future we will extend our studies to include examining Bcl-2, caspase-8, and decoy receptor 1 and 3 via Western blot.

It has previously been reported that ischemic placentas from women with preeclampsia and HELLP syndrome release significantly more trophoblastic debris, which can activate and stimulate the underlying vascular endothelium, possibly contributing to hypertension (19, 21, 35). As such, we examined ET-1 release from HUVECs. Endothelial cells exposed to conditioned media from HELLP + MFL4 placentas had a significantly blunted response, resulting in less ET-1 secretion compared with HUVECs exposed to media from untreated HELLP rats. Interestingly, when we treated placentas with increasing concentrations of MFL4, there was not a difference in ET-1 secretion production due to exogenous MFL4 treatment. These results suggest that the decrease in ET-1 seen from HELLP + MFL4-conditioned media are not due directly to neutralization of FasL. We have previously reported that circulating FasL is blunted in HELLP rats subjected to ET_A receptor antagonism compared with HELLP rats (not significant) (10), which when paired with the current data does seem to support an indirect relationship between FasL and ET-1. Indeed, in tumor cells, both the FasL and ET-1 systems have been reported to be colocalized, resulting in increased concentrations of ET-1, which in turn promoted FasL-meditated apoptosis (26). Placental levels of TNFα and sEng were decreased in response to FasL, which could have contributed to the decrease in ET-1 that was stimulated in response to the conditioned media, as both are important stimulators of ET-1 from HUVECs (13, 20).

Placental development is a highly regulated physiological process in both humans and rodents. When we measured the total area of the placenta occupied by the decidua, basal, and labyrinth zones on GD19, we did not observe any defective growth or development of placentas. Spiral arteries along with invasive trophoblasts and stromal cells can be found in the uterine mesometrial triangle, or metrial gland, in the rat placenta. Cellular activity peaks around GD11–12, with the gland being fully formed by GD14 (8). We did not study spiral artery remodeling or vascularization in the present study, and therefore, we cannot assess whether MFL4 improved the degree of trophoblast cell invasion or spiral artery remodeling. However, as this field has progressed, new methods to study spiral artery remodeling are becoming more available, opening up the possibilities for future investigations in this area (36).

In addition to having a primary role in apoptosis, the Fas/FasL pathway also has a key role in inflammation, primarily through its ability to regulate leukocyte extravasation and contribute to cytokine secretion (6, 12). Circulating sFlt-1 was decreased in response to FasL attenuation, whereas there was no statistically significant difference in sEng expression. The combined circulating and placental changes suggest that attenuation of FasL also leads to a decrease in the antiangiogenic factor sFlt-1 and improves placental inflammation, either of which could contribute to the improvement in mean arterial pressure that was seen in animals treated with MFL4 (11, 25). Although we did not measure lymphocytes in the current study, previous studies have reported that blockade of FasL significantly decreases lymphocytes and inflammatory cytokines, including TNFα secretion (18, 42). Although the placenta is the primary source of sFlt-1, we have previously reported that in this animal model of HELLP syndrome the liver also secretes significantly more sFlt-1 compared with NP rats (40). Interestingly, we did not find these results with sEng, and the placenta was the only organ found to secrete significantly more sEng relative to NP rats. That could potentially explain why circulating sFlt-1 was decreased in HELLP + MFL4 rats despite there being no significant attenuation of placental sFlt-1 compared with HELLP rats. Whereas placental sEng levels were significantly decreased in HELLP + MFL4 rats, there was not a corresponding decrease in circulating sEng compared with HELLP rats.

IgG was infused to determine whether the effects in MFL4 rats are potentially due to IgG rather than inhibition of FasL. IgG infusion led to a significant increase in TNFα, which was significantly higher compared with both uninfused and MFL4-infused rats. To ensure that the increases in IgG infused rats were not possibly due to the surgery or infusion of a foreign object, we measured circulating FasL and TNFα collected on GD19 from NP rats that received a mini-osmotic pump infusing saline on GD12 (24). Neither FasL (P = 0.27) nor TNFα (P = 0.11) concentrations were significantly increased compared with NP rats in the current study, but they were compared with NP + IgG rats (P = 0.03 and P = 0.04, respectively), suggesting that it was not the infusion of a foreign substance leading to an immune response. Our study is not alone in these findings, as it has previously been reported that animals infused with IgG had increases in inflammatory cytokines such as TNFα and IL-6 compared with uninfused rats (16).

Perspectives and Significance

Regulation of the Fas/FasL pathway is vital for remodeling of the uterine spiral arteries during early pregnancy (3). Dysregulation of this pathway is associated with inadequate trophoblast invasion through improper smooth muscle cell and endothelial cell apoptosis, which leads to placental ischemia and potentially preeclampsia and HELLP syndrome (1, 28). It is accepted that the Fas/FasL pathway is dysregulated during preeclampsia and HELLP syndrome and that this pathway plays a substantial role in the pathogenesis of these diseases. Although there are several potential sources of FasL, such as cytotoxic T cells and hepatocytes (5, 9), that can inherently mediate HELLP syndrome, the physiological role of FasL in disease pathogenesis is still not clear.

GRANTS

Flow Cytometry Core Services was supported by the Department of Cell & Molecular Biology and funding from the National Institute of General Medical Sciences of the National Institutes of Health under Award No. P20-GM-121334 and the Mississippi Center of Excellence in Perinatal Research (COBRE). The research reported here was supported by the Office of Sponsored Programs at the University of Mississippi Medical Center.

DISCLOSURES

No conflicts of interest, financial or otherwise, are declared by the authors.

AUTHOR CONTRIBUTIONS

J.G. and K.W. conceived and designed research; J.G., S.-K.S., L.S., T.B., P.B.K., J.S., J.P.D., R.R., and K.W. performed experiments; J.G., S.-K.S., L.S., T.B., and K.W. analyzed data; J.G. and K.W. interpreted results of experiments; J.G., S.-K.S., T.B., and K.W. prepared figures; J.G., S.-K.S., J.S., and R.R. drafted manuscript; J.G., S.-K.S., T.B., P.B.K., J.S., J.P.D., R.R., and K.W. edited and revised manuscript; J.G., S.-K.S., L.S., T.B., P.B.K., J.S., J.P.D., R.R., and K.W. approved final version of manuscript.