Soluble suppression of tumorigenicity-2 in pregnancy with a small-for-gestational-age fetus and with preeclampsia

Abstract

Objective

Preeclampsia and fetal growth disorders are pregnancy-specific conditions that share common pathophysiological mechanisms. Yet, why some patients develop preeclampsia while others experience fetal growth restriction, or a combination of both clinical presentations, is unknown. We propose that the difference in severity of the maternal inflammatory response can contribute to the clinical phenotypes of preeclampsia vs. small for gestational age (SGA). To assess this hypothesis, we measured maternal plasma concentrations of the soluble isoform of suppression of tumorigenicity-2 (sST2), a member of the interleukin-1 receptor family that buffers proinflammatory responses. Previous reports showed that serum sST2 concentrations rise in the presence of intravascular inflammation and Th1-type immune responses and are significantly higher in patients with preeclampsia compared to those with normal pregnancy. The behavior of sST2 in pregnancies complicated by SGA has not been reported. This study was conducted to compare sST2 plasma concentrations in normal pregnancies, in those with preeclampsia, and in those with an SGA fetus.

Methods

This retrospective cross-sectional study included women with an SGA fetus (n = 52), women with preeclampsia (n = 106), and those with normal pregnancy (n = 131). Maternal plasma concentrations of sST2 were determined by enzyme-linked immunosorbent assay. Doppler velocimetry of the uterine and umbilical arteries was available in a subset of patients with SGA (42 patients and 43 patients, respectively).

Results

(1) Women with an SGA fetus had a significantly higher median plasma concentration of sST2 than normal pregnant women (p = .008); (2) women with preeclampsia had a significantly higher median plasma concentration of sST2 than those with normal pregnancy (p < .001) and those with an SGA fetus (p < .001); (3) patients with SGA and abnormal uterine artery Doppler velocimetry had a higher median plasma concentration of sST2 than controls (p < .01) and those with SGA and normal uterine artery Doppler velocimetry (p = .02); (4) there was no significant difference in the median plasma sST2 concentration between patients with SGA who had normal uterine artery Doppler velocimetry and controls (p = .4); (5) among patients with SGA, those with abnormal and those with normal umbilical artery Doppler velocimetry had higher median plasma sST2 concentrations than controls (p = .001 and p = .02, respectively); and (6) there was no significant difference in the median plasma sST2 concentrations between patients with SGA who did and those who did not have abnormal umbilical artery Doppler velocimetry (p = .06).

Conclusions

Preeclampsia and disorders of fetal growth are conditions characterized by intravascular inflammation, as reflected by maternal plasma concentrations of sST2. The severity of intravascular inflammation is highest in patients with preeclampsia.

Keywords:

• Fetal growth restriction
• interleukin-33 (IL-33)
• placental disease
• suppression of tumorigenesis-2
• systemic inflammation
• umbilical artery Doppler
• uterine artery Doppler

Introduction

Preeclampsia and fetal growth restriction (FGR) share common pathophysiological mechanisms, i.e. an antiangiogenic state [1–19], defective deep placentation [20], uteroplacental ischemia [21,22], exaggerated intravascular inflammation [13,23–34], and endothelial cell dysfunction [35–40]. Despite these similarities, preeclampsia and FGR clearly have different clinical manifestations [39,41,42]. Preeclampsia is characterized by hypertension and proteinuria as well as multiple organ involvement while isolated FGR lacks these clinical findings [43–46]. Elucidating the factors that contribute to the different clinical presentations is of interest to shed light into the mechanisms of disease and can have implications for the prediction, prevention, and treatment of these complications in pregnancy.

Suppression of tumorigenicity-2 (ST2), a member of the interleukin-1 receptor (IL-1R) family, elicits a Th2 response [47] upon binding the cytokine IL-33 [48]. The soluble isoform of this receptor, called soluble ST2 (sST2), can act as a decoy receptor that sequesters IL-33 [49] and induces a shift toward a Th1-like inflammatory response [50,51]. sST2 is released by endothelial cells in intravascular inflammation [52–54] and in proinflammatory Th1 immune responses [55], and its serum concentration is increased in pathologic states such as sepsis [56,57], atherosclerosis [58], and myocardial infarction [59,60]. Plasma sST2 concentrations are higher in women with an uncomplicated pregnancy than in non-pregnant women [61], and previous studies have investigated the behavior of sST2 in different complications of pregnancy, e.g. miscarriage and recurrent pregnancy loss [62,63], preterm labor [64], fetal inflammatory syndrome [65], funisitis [65], and preeclampsia [66,67]. Women with preeclampsia have a higher plasma sST2 concentration than normal pregnant women [61], and plasma sST2 concentration has been proposed to reflect the clinical severity of the disease, the magnitude of intravascular inflammation, and the Th-1 shifted immune response [66,67]. Previous studies also reported an exaggerated intravascular inflammation in pregnancies complicated by FGR or a small for gestational age (SGA) fetus [34,68]; however, there is a lack of data about the behavior of sST2 in these conditions.

Fetal growth restriction is suspected when the estimated fetal weight is below the 10th percentile or when there is a deceleration of fetal growth in serial biometry [69–71]. Doppler velocimetry of the uterine and umbilical arteries reflects increased resistance in the uterine circulation or placental disease and abnormalities in these Doppler parameters are used to distinguish growth-restricted fetuses from those constitutionally small [72–74]. The aim of this study was to examine the differences of plasma sST2 concentration in patients with an SGA fetus, preeclampsia, and uncomplicated pregnancy.

Methods

This is a retrospective cross-sectional study conducted at Wayne State University, the Detroit Medical Center, and the Perinatology Research Branch, NICHD/NIH/DHHS. Patients were enrolled at Hutzel Women’s Hospital of the Detroit Medical Center (Detroit, MI). The study included a total of 289 women classified in the following groups: (1) women with an uncomplicated pregnancy (n = 131); (2) women with preeclampsia (n = 106); and (3) women with an SGA neonate (n = 52). Exclusion criteria were chronic hypertension, multiple gestation, or fetuses with known fetal chromosomal or major structural anomalies. Venipuncture was performed upon diagnosis in patients with preeclampsia or in those with an SGA fetus. Doppler velocimetry of the uterine and umbilical arteries was performed in a subset of patients with an SGA fetus. All women provided written informed consent prior to the collection of plasma samples. The Institutional Review Boards of Wayne State University and NICHD approved the collection and utilization of the samples for research purposes. Many of these samples were also used in other studies at our institution, including our previous studies on preeclampsia [66,67].

Clinical definitions

Preeclampsia was defined in the presence of hypertension (systolic blood pressure ≥140 mmHg or diastolic blood pressure ≥90 mmHg on at least two occasions, 4 h to 1 week apart, after the 20th week of gestation) and proteinuria (≥300 mg in a 24-h urine collection or one dipstick measurement ≥2+) [67,75]. The diagnosis of SGA was based on an estimated fetal weight of less than the 10th percentile by ultrasound and confirmed at the time of birth by using the standard by Alexander et al. [76]. Patients were considered to have a normal pregnancy if they met the following criteria: no medical, obstetrical or surgical complications; absence of labor at the time of venipuncture; and delivery of a term (≥37 weeks) infant with a birthweight between the 10th and 90th percentiles for gestational age.

Doppler velocimetry

Doppler velocimetry investigations of the uterine and umbilical arteries were performed as part of the standard routine in a subset of patients with an SGA fetus (uterine artery Doppler velocimetry, 42 cases; umbilical Doppler velocimetry, 43 cases) as previously described [15,77]. Uterine artery Doppler velocimetry was defined as abnormal when the mean (average of right and left uterine arteries) resistance index was above the 95th percentile for gestational age by using the reference range proposed by Kurmanavicius et al. [78] or in the presence of bilateral diastolic notch. Umbilical artery Doppler velocimetry was defined as abnormal if either the pulsatility index was above the 95th percentile for gestational age, using the reference range proposed by Arduini and Rizzo [79], or in the presence of absent or reversed end-diastolic velocities [80].

Sample collection and human sST2 immunoassay

Blood was collected in tubes containing ethylenediaminetetraacetic acid (EDTA). Samples were centrifuged at 1300×g for 10 min at 4 °C and the fluid phase was stored at −70 °C. Maternal plasma concentrations of sST2 were determined by using a quantitative sandwich enzyme-linked immunoassay (R&D Systems, Minneapolis, MN). The inter- and intra-assay coefficients of variation for sST2 were 4.6% and 3.9%, respectively. The sensitivity of the assay for sST2 was 17.5 pg/mL.

Statistical analysis

We examined the distribution of the data by visual plot inspection and the use of the Kolmogorov–Smirnov or Shapiro–Wilk tests for normality. A Kruskal–Wallis with a post hoc Mann–Whitney’s U-test was used to compare variables not normally distributed among and between groups. Comparison of proportions was performed by using a Chi-square or a Fisher’s exact test, as appropriate. Correlation between two continuous variables was determined by using Spearman’s rank correlation. A general linear model was applied for comparisons of plasma sST2 concentrations between groups, adjusting for gestational age at venipuncture. Analyses were performed with SPSS, version 19 (IBM Corp., Armonk, NY).

Results

Demographic data

Demographic characteristics of the study population are displayed in Table 1. There were no significant differences in maternal age, body mass index (kg/m²), gestational age at venipuncture, and recreational drug use among groups. There were significant differences in gestational age at delivery, birthweight, number of nulliparous women, smokers, and birthweights <5% and <10%. Analysis of the data excluding the SGA fetuses from pregnant patients with preeclampsia indicated similar findings as reported. Clinical characteristics and plasma concentrations of sST2 in preeclampsia and in normal pregnancy have been reported in previous publications [66,67].

Table 1. Demographics and clinical characteristics of the study population.

	Normal pregnancy (n = 131)	SGA (n = 52)	Preeclampsia (n = 106)
Age (years)	25 (21–29)	24 (20–29)	23 (19–29)
BMI (kg/m^2)	26 (21–32)	26 (21–30)	27 (24–30)
Nulliparous	36 (28%)^a,^b	26 (50%)^b	65 (61%)^a
Smoking	23 (18%)	16 (31%)^c	14 (13%)^c
Recreational drug use	14 (7%)	7 (15%)	7 (11%)
GA at blood draw (weeks)	38 (31–39)	37 (33–38)	36 (31–39)
GA at delivery (weeks)	39 (39–40)^a	37 (34–39)^a	36 (32–39)^a
Birthweight (g)	3345 (3110–3630)^a	2060 (1505–2380)^a,d	2280 (1455–2805)^a,d
Birthweight <5%	0 (0%)^a	42 (81%)^a,e	28 (26%)^a,e
Birthweight <10%	0 (0%)^a	52 (100%)^a,e	48 (45%)^a,e

BMI: body mass index; GA: gestational age; SGA: small for gestational age.

Data are presented as median with interquartile range (IQR) or as number of patients and percentage (%). Mann–Whitney’s tests were used for continuous variables and Chi-square tests for categorical variables.

^aSignificant difference between controls and SGA or between controls and preeclampsia (p < .001).

^bSignificant difference between controls and SGA (p = .005).

^cSignificant difference between SGA and preeclampsia (p = .016).

^d Significant difference between SGA and preeclampsia (p = .030).

^eSignificant difference between SGA and preeclampsia (p < .001).

sST2 concentrations were elevated in preeclampsia and in SGA

The median maternal plasma concentration of sST2 was significantly higher in women with preeclampsia and in women with an SGA fetus compared to women with uncomplicated pregnancy (p < .001 and p = .008, respectively). Women with preeclampsia had a significantly higher median plasma concentration of sST2 than those with an SGA fetus (p < .001) (Figure 1). Among patients in the SGA group, none were diagnosed with preeclampsia. Among patients with preeclampsia, there were 48 (45.3%) patients who delivered SGA neonates There was no significant difference in the median (interquartile range) plasma sST2 concentrations between preeclampsia with and without SGA [78.2 (51.1–132.8) ng/mL vs. 73.4 (44.2–130.2) ng/mL; p = .4]. Of note, among patients with SGA, there was no significant correlation between birthweight percentile and maternal plasma sST2 concentration.

Figure 1. Median (interquartile range) plasma concentrations of sST2 in patients with preeclampsia [76 (48–130) ng/mL], patients with an SGA fetus [(38 (25–73) ng/mL], and uncomplicated pregnant women [31 (14–52) ng/mL]. The median plasma concentrations of sST2 were significantly higher in patients with preeclampsia compared to uncomplicated pregnant women (p < .001) and those with an SGA fetus (p < .001). Patients with an SGA fetus had a significantly higher median concentration of sST2 than uncomplicated pregnant women controls (p = .008). SGA: small for gestational age; sST2: soluble suppression of tumorigenicity-2. Y-axis data are presented in logarithmic scale.

sST2 concentrations in women with SGA classified by uterine artery Doppler velocimetry findings

Figure 2 shows plasma sST2 concentrations in normal pregnancy and in patients with SGA classified according to uterine artery Doppler velocimetry findings. Patients with SGA and abnormal uterine artery Doppler velocimetry had a higher median plasma concentration of sST2 than controls [49 (27–79) ng/mL vs. 31 (14–52) ng/mL, p < .01] and those who had an SGA fetus and a normal uterine artery Doppler finding [49 (27–79) ng/mL vs. 37 (18–72) ng/mL, p = .02)]. This difference remained significant after adjusting for gestational age at venipuncture. By contrast, there was no significant difference in the median plasma sST2 concentration between patients with an SGA fetus and normal uterine artery Doppler velocimetry and the controls (p = .4).

Figure 2. Patients with an SGA fetus and abnormal uterine artery Doppler velocimetry had a higher median (interquartile range) maternal plasma concentration of sST2 compared to the controls [49 (27–79) ng/mL vs. 31 (14–52) ng/mL, p < .01] and to the patients with an SGA fetus and normal uterine artery Doppler velocimetry [49 (27–79) ng/mL vs. 37 (18–72) ng/mL, p = .02)], after adjustment for gestational age at venipuncture. SGA: small for gestational age; UT DV: uterine artery Doppler velocimetry. Y-axis data are presented in logarithmic scale.

sST2 concentrations in women with SGA classified by umbilical artery Doppler velocimetry findings

Figure 3 shows plasma sST2 concentrations in controls and in patients with SGA classified according to umbilical artery Doppler velocimetry results. Among patients with an SGA fetus, those with and those without abnormal umbilical artery Doppler velocimetry had higher median sST2 concentrations than uncomplicated pregnant women [38 (23–79) ng/mL vs. 31 (14–52) ng/mL, p = .001 and 48 (24–75) ng/mL vs. 31 (14–52) ng/mL, p = .02, respectively]. This difference remained significant after adjusting for gestational age at venipuncture. There was no significant difference in the median plasma sST2 concentrations between patients with SGA who did and those who did not have abnormal umbilical artery Doppler velocimetry (p = .06).

Figure 3. Patients with an SGA fetus and either abnormal or normal umbilical artery Doppler velocimetry had a higher median (interquartile range) plasma concentration of sST2 than uncomplicated pregnant women [38 (23–79) ng/mL vs. 31 (14–52) ng/mL, p = .001 and 48 (24–75) ng/mL vs. 31 (14–52) ng/mL, p=.022], after adjustment for gestational age at venipuncture. SGA: small for gestational age; UA DV: umbilical artery Doppler velocimetry. Y-axis data are presented in logarithmic scale.

Discussion

Principal findings

(1) Patients with preeclampsia and those with an SGA fetus had significantly higher plasma sST2 concentrations than normal pregnant women; and (2) the subgroup of women with an SGA fetus and abnormal uterine artery Doppler velocimetry had a significantly higher plasma sST2 concentration than those without Doppler abnormalities, supporting the view that sST2 is elevated when fetuses are growth-restricted rather than constitutionally small. Collectively, these findings suggest that intravascular inflammation shifted toward a Th1 response is a feature of preeclampsia and of FGR and that it is more severe in preeclampsia.

What is sST2?

ST2 belongs to the Toll-like/IL-1 receptor super family of cytokines [51]. IL-33 is an alarmin, a member of the IL-1 interleukin family, and structurally related to IL-1β and IL-18 [49]. Binding of IL-33 to the ST2L transmembrane receptor promotes an anti-inflammatory response by inhibition of the Toll-like receptor pathway and by shifting the Th1/Th2 balance toward a Th2 T-cell response [50,51,81]. The sST2 [49] lacks the transmembrane and cytoplasmatic domains and acts as a decoy receptor sequestering IL-33. Therefore, sST2 can modulate the IL-33/ST2 signaling and attenuate the Th2 inflammatory response, favoring a predominant Th1 response [82]. Through these interactions, ST2/IL-33/sST2 signaling has been implicated in immunoregulation, both in health and in disease [82].

ST2 is expressed in a wide range of cells of the immune system including mastocytes [83], macrophages [84], and type 2 T-helper cells [81,85–87] as well as in non-immune cells from different tissues, e.g. the lung (alveolar epithelial cells) [88], kidney [89], heart (cardiomyocytes) [60], gastrointestinal tract [90], and, importantly, endothelial cells [52,53]. In the reproductive tract, ST2 expression has been detected in the ovary, myometrium, and endometrium [91] as well as in specific trophoblast populations within the placenta [91–93] from where sST2 can gain access to the maternal circulation [94]. Elevated concentrations of plasma sST2 have been reported in several disorders characterized by intravascular inflammation, e.g. sepsis [56,57], trauma [56], atherosclerosis [58], Dengue fever [95–97], myocardial infarction [60,98,99], and heart failure [100,101]. Normal pregnancy is characterized by a state of mild systemic inflammatory response [30,102–104]. A previous study reported higher maternal plasma concentrations of sST2 in uncomplicated pregnancies than in non-pregnant controls [61]. However, the same results were not replicated in a study recently published by Zhao et al. [105]. Recently, sST2 has been shown to be elevated in patients with gestational diabetes [106].

Women with an SGA fetus and abnormal uterine artery Doppler velocimetry have a higher plasma sST2 concentration than normal pregnant women

We found that women with an SGA fetus have a higher plasma concentration of sST2 than normal pregnant women. The subgroup of women with an SGA fetus and abnormal uterine artery Doppler velocimetry has a significantly higher plasma sST2 concentration than those without Doppler abnormalities (Figure 2), supporting the view that sST2 is elevated when fetuses are growth-restricted rather than constitutionally small [72–74,107].

Patients with preeclampsia have a higher plasma sST2 concentration than those with a normal pregnancy

We found that patients with preeclampsia have a higher maternal plasma sST2 concentration than normal pregnant women, and this finding is in agreement with previous observations [61,66,67,108–110]. A higher concentration of plasma sST2 has been reported prior to the clinical diagnosis of preeclampsia; therefore, it has been proposed as a biomarker for this syndrome [61,66]. The diagnostic performance of sST2 in identifying preeclampsia is comparable to that of angiogenic/anti-angiogenic factors, and there is a positive correlation between sST2 and plasma soluble vascular endothelial growth factor receptor-1 or soluble fms-like tyrosine kinase-1 and soluble endoglin, whereas there is a negative correlation with placental growth factor (PlGF) [66,67], suggesting a link between inflammation and an abnormal anti-angiogenic state in preeclampsia. Importantly, maternal plasma sST2 concentrations seem to reflect the severity of preeclampsia as the magnitude of the increase is markedly higher in early and severe cases than in late and mild cases [66].

Women with preeclampsia have a higher sST2 concentration than women with SGA

Several lines of evidence suggest that preeclampsia is a condition characterized by a more severe inflammatory state compared to FGR. Previous studies showed that patients with preeclampsia have a more severe involvement of several organ systems than those with SGA as preeclampsia is associated with a higher degree of endothelial cell dysfunction (reflected by higher concentrations of soluble adhesion molecules) [39,68,111], higher plasma concentrations of complement split products [112], abnormal plasma concentrations of angiogenic/anti-angiogenic factors [113], and higher maternal tissue factor activity [114]. Moreover, these conditions differ in the concentration of plasma fetal DNA with higher concentrations in women with preeclampsia compared to women with FGR [115]. The higher median maternal plasma sST2 concentrations in patients with preeclampsia, compared to those with an SGA fetus, may reflect the greater degree of intravascular inflammation and anti-angiogenic status that characterizes preeclampsia. Whether or not plasma sST2 concentrations in patients with SGA could identify patients who subsequently develop preeclampsia remains to be determined.

Strengths and limitations

This is the first study to report plasma concentrations of sST2 in patients with an SGA fetus with and without Doppler abnormalities. Our findings in preeclampsia are consistent with those reported by other investigators. The cross-sectional nature of the study does not allow us to provide information about the temporal relationship between changes in sST2 concentrations and the clinical diagnosis of preeclampsia or SGA. Additionally, not all pregnancies with SGA had Doppler velocimetry results and limited sample size precluded further subgroups analysis.

Conclusions

Plasma sST2 concentrations are higher in women with preeclampsia and in those with SGA compared to normal pregnant women. We propose that the higher concentrations of sST2 in the maternal circulation reflect intravascular inflammation in both conditions and that the inflammation is more severe in preeclampsia than in SGA without preeclampsia.

Disclosure statement

The authors report no conflicts of interest.

Additional information

Funding

This research was supported, in part, by the Perinatology Research Branch, Division of Obstetrics and Maternal-Fetal Medicine, Division of Intramural Research, Eunice Kennedy Shriver National Institute of Child Health and Human Development, National Institutes of Health, U.S. Department of Health and Human Services (NICHD/NIH/DHHS); and, in part, with Federal funds from NICHD/NIH/DHHS under Contract No. HHSN275201300006C. Dr. Romero has contributed to this work as part of his official duties as an employee of the United States Federal Government.