Functional autoantibodies against Endothelin-1 receptor type A and Angiotensin II receptor type 1 in patients with preeclampsia

Pregnancy Hypertension 14 (2018) 189–194

Contents lists available at ScienceDirect

Pregnancy Hypertension journal homepage: www.elsevier.com/locate/preghy

Functional autoantibodies against Endothelin-1 receptor type A and Angiotensin II receptor type 1 in patients with preeclampsia

T

Sara Buttrup Larsena,b, Gerd Wallukatc, Ingolf Schimkec, Anna Sandagerb, ⁎ Therese Tvilum Christensena, Niels Uldbjergb, Niels Tørringa, a

Department of Clinical Biochemistry, Aarhus University Hospital, Aarhus, Denmark Department of Obstetrics and Gynecology, Aarhus University Hospital, Aarhus, Denmark c Berlin Cures AG, Berlin, Germany b

A R T I C LE I N FO

A B S T R A C T

Keywords: Autoantibody Angiotensin receptor Endothelin receptor Preeclampsia HELLP

Objectives: Functional autoantibodies against Angiotensin II Receptor type 1 (AT1-AA) and Endothelin-1 Receptor type A (ETA-AA), which belong to the class of functional autoantibodies, have been discovered in patients with preeclampsia and in rodent models of pregnancy-induced hypertension. The aim of the study was to investigate the expression of these autoantibodies in relation to disease progression. Study design: We included 10 controls and 41 cases defined as patients with gestational-induced hypertension, preeclampsia or HELLP syndrome. Main outcome: Serum obtained in the first trimester as well as at the time of disease development were analyzed by means of a biological assay of beating cardiomyocytes. We also measured the protein expression of IL-17A in these samples. Results: 100% of samples from patients with gestational induced hypertension, preeclampsia or HELLP syndrome expressed AT1-AA when they presented with clinical symptoms but not in samples from the first trimester. 44% of samples from patients with severe preeclampsia or HELLP syndrome expressed ETA-AA but only when they presented with clinical symptoms. The controls expressed neither AT1-AA nor ETA-AA. Approximately 40% of patients with severe preeclampsia or HELLP syndrome expressed IL-17A, both at the time of the onset of symptoms and in the first trimester. Conclusion: Autoantibodies against the Angiotensin II receptor 1 and Endothelin receptor are developed in relation to pregnancy-induced hypertension and not present at the start of the pregnancy in these patients. IL-17A is increased in some patients with severe preeclampsia, but the expression is not related to the development of clinical symptoms.

1. Introduction

pathways leading to preeclampsia. Thus, administration of human AT1AA to pregnant mice induces hypertension, proteinuria, and glomerular endotheliosis, the effects of which can be prevented by the AT1 receptor antagonist losartan [11]. Furthermore, the RUPP rat model of preeclampsia, in which placental ischemia is surgically induced, displays increased expression of TNF-a, IL-6 and AT1-AA [12,13]. Also the involvement of ETA-AA is rendered probably, as administration of a specific Endothelin 1 receptor A antagonist to RUPP rats and to the rodent sFlt-1 model normalized their blood pressure [14,15]. However, we know little about the mechanism leading to the expression of autoantibodies in women with preeclampsia. It has recently been shown that infusion of interleukin 17 leads to hypertension and increases the production of AT1-AA in normal pregnant rats [16]; also,

Preeclampsia (PE) affects 2–8% of all pregnancies worldwide [1], and it is a leading factor of maternal mortality, maternal morbidity, fetal growth restriction (FGR), and preterm birth. Although its pathogenesis remains unclear, a number of hallmarks are central. One of these is the development of a pro-inflammatory state [2,3] including the B-cell production of functionally-active autoantibodies directed against the Angiotensin II, type 1 receptor (AT1-AA) [4,5] as well as the endothelin-1 ETA receptor (ETA-AA) [6]. In concordance with endothelin 1 [7–9], these two autoantibodies increase vasoconstriction activation [10]. Animal studies have demonstrated that AT1-AAs are involved in

⁎

Corresponding author at: Department of Clinical Biochemistry, Aarhus University Hospital, Palle Juul-Jensens Boulevard 99, DK-8200 Aarhus N, Denmark. E-mail address: [email protected] (N. Tørring).

https://doi.org/10.1016/j.preghy.2018.10.002 Received 27 June 2018; Received in revised form 5 September 2018; Accepted 12 October 2018 Available online 13 October 2018 2210-7789/ © 2018 International Society for the Study of Hypertension in Pregnancy. Published by Elsevier B.V. All rights reserved.

Pregnancy Hypertension 14 (2018) 189–194

S. Buttrup Larsen et al.

2.4. Sample preparation

the infusion of soluble mouse IL-17 receptor C to inhibit IL-17 signaling decreases blood pressure and AT1-AA production in RUPP rats [17]. Therefore, the objective of this study on women with preeclampsia was to measure AT1-AA, ETA-AA and IL17A in serum taken in the first trimester of their pregnancy and in serum obtained when the diagnosis was made in the third trimester. Furthermore, we wanted to associate the findings with the progression of the disease.

After whole blood collection, serum was prepared according to standardized procedures. For the IgG preparation, 1 ml of serum and 660 μl of saturated ammonium sulfate solution was mixed (final concentration: 40% ammonium sulfate) and incubated for 18 h at 4 °C. After centrifugation for 15 min at 6000g, the pellet was re-suspended in 750 μl of phosphate buffered saline (PBS), mixed with 750 μl of saturated ammonium sulfate solution (final concentration: 50% ammonium sulfate), and centrifuged again. The pellet was then suspended in 700 μl of PBS and dialyzed (VISKING cellulose, type 27/32, MW Cut off 14 kDa; Carl Roth, Germany) against the 100-fold volume of PBS for 3 days at 4 °C. The resulting IgG fraction was aliquotted and stored at −20 °C for at least one month without a loss of activity.

2. Methods 2.1. Study design and participants This was conducted as a case-control study. Cases included women diagnosed with gestational hyper-tension (GH), preeclampsia or HELLP syndrome enrolled in the outpatient clinic or the inpatient ward at Aarhus University Hospital (AUH), Denmark. Controls included healthy, pregnant women enrolled from the midwife’s practice at AUH. Patients and controls were included from July 2014 to August 2017, after written and oral consent to the study were given. The research project was approved by the Regional Committees on Health Research Ethics of the Central Region of Denmark (file no.: 1-10-72-358-14) and the Danish Data Protection Agency (file no.: 1-16-02-625-14). Exclusion criteria was the inability to understand and speak Danish, hence the incapability to secure an informed consent. Controls were excluded in the case of later development of PE.

2.5. Cardiomyocyte preparation and culturing Hearts of approximately twenty 1- to 3-day-old rats were removed under sterile conditions and transferred to PBS (4 °C; without Ca2+, Mg 2+; Biochrom, Berlin, Germany). The ventricle tissue was separated and dissected into small pieces of nearly 1 mm3 for washing twice with 10 ml of solution 1. After decanting the wash solution, the tissue was resuspended in 10 ml of PBS containing 0.2% of crude trypsin and incubated for 15 min at 37 °C under stirring; then, the solution was treated with 10 ml of ice-cold heat-inactivated calf serum to stop the trypsination. The resulting suspension was centrifuged at 130g for 6 min, and the pellet was transferred to 20 ml of SM20-I medium (Biochrom GmbH, Berlin, Germany). For cell counting, 100 μl of this suspension was added to 100 Trypan blue solution. Then, 2.4 × 106 cells in 2.0 ml of SM 20-I medium, which was equilibrated with humid air, were transferred to 12.5 cm2 Falcon flasks and cultured as a monolayer for 4–8 days at 37 °C. The medium was renewed after 2 days. Cardiomyocytes spontaneously started beating after 2 days in culture.

2.2. Clinical and biochemical data Blood samples from cases were obtained at the time of admission with clinical signs of hypertensive disorder, and also at readmission in cases of further clinical development. From 17 cases who developed severe preeclampsia or HELLP syndrome, we also obtained archived serum samples from the first trimester. Blood samples from healthy pregnant controls with a mean gestational age of 35 weeks, were obtained in the midwife clinic. Clinical data of study participants were obtained through the patients’ records (Electronic Patient Journal (EPJ) (Columna Clinical Information System) (Systematic, Aarhus, Denmark)), the obstetrical database 5s (Astraia software GMBH, Germany) and the laboratory database LABKA II. The following clinical data were collected on study participants: maternal characteristics (height, weight, ethnicity, parity, smoking status), pregnancy-data (date of birth, indication of birth/induction of birth, birth weight, outcome), data regarding symptoms, blood pressure and proteinuria at blood sampling, data on risk factors for development of PE (In Vitro Fertilization (IVF) or Intracytoplasmic Sperm Injection (ICSI), egg-donation, autoimmune diseases, severe medical conditions, PE in earlier pregnancy) and available biochemical data at blood sampling (PE serology). Diagnosis at enrollment as well as outcome diagnosis were established via data from EPJ according to the guidelines from The American College of Obstetricians and Gynecologists (ACOG).

2.6. Assay procedure and calculation of the autoantibody activity On the day of measurement, the culture flask was transferred onto a heated stage (37 °C) microscope, and 6 fields with synchronic and rhythmic beating cardiomyocytes were marked on the flask bottom. Next, the basal beating rate of the 6 fields was counted for 15 s and averaged. After the addition of 40 μl of the IgG preparation, the culture flasks were incubated for 40–60 min at 37 °C, and the beating rate in the 6 fields was then counted for 15 sec and averaged. For measurement, 1) the basal beating rate must range between 120 and 200 beats/min and 2) cells stimulated for 5 min with isoprenaline (10 μM) must respond with a frequency increase of more than 45 beats/min. One unit of autoantibody activity corresponds to a 1 beat/min frequency change. The lower limits of detection (LLD) for positive and negative chronotropic activity were calculated as 4.0 U and −4.0 U, respectively. G-protein-coupled receptor (GPCR)- autoantibody positivity was defined using cut-offs based on X ± 3 SD of the GPCR-AA level of more than 100 healthy subjects. The results of ≥8.0 U for positive and ≤−8.0 U for negative chronotropic GPCR-AA activity were calculated.

2.3. Bioassay of spontaneously beating cultured neonatal rat cardiomyocytes To identify and quantify the AT1- and ETA-AA, a bioassay established by Wallukat and Wollenberger and in detail described in [18,19] was used. As the sample material required for the bioassay, IgG isolated from patient serum was used. This bioassay measures the functional activity of the autoantibodies via the cells’ chronotropic response after addition of the IgG. Depending upon either the positive (AT1-AA) or negative chronotropic activity (ETA-AA) of the autoantibodies, the increase and decrease, respectively, of the cells’ beat frequency was monitored. With the use of competitors such as Losartan for the AT1receptor and BQ123 for the ETA receptor, the autoantibodies are specified for their targeted receptors, extracellular binding sites and specific receptor epitopes.

2.7. Autoantibody differentiation related to the AT1- and ETA-receptor For this purpose, the bioassay was performed after successive addition of 1 μmol/l Losartan to block the AT1-receptor and 0.5 μmol/l BQ to block the ETA receptor. The change in the cells’ beating rate, which is the result of positive (AT1-AA) and negative chronotropy (ETA-AA), can be attributed to the individual autoantibodies. 190

Pregnancy Hypertension 14 (2018) 189–194

S. Buttrup Larsen et al.

(Fig. 1B). This chronotropic response below the basal rate indicated the co-appearance of another autoantibody in 15 serum samples that could activate a negative chronotropic receptor. For serum from controls (n = 10) the beat rate of the cardiomyocytes did not change according to the preset cut-off value of Δ 8 beats/minute after the addition of the AS precipitated serum, indicating AT1-AA negativity (Fig. 1A).

2.8. Epitope analysis To identify the binding site of AT1-AA, ammonium sulfate precipitated serum was treated with 5 overlapping epitope-sequences (40 µl serum + 2 µl peptide-sequence): V-F-F-I-E-N, E-N-T-N-I-T, I-T-VC-A-F, A-F-H-Y-E-S-Q- and Q-N-S-T-L-P-I. To identify the binding site of the ETA-R-AA, serum was treated with 5 overlapping epitope-sequences (40 µl AS precipitated serum + 2 µl peptide-sequence): F-E-Y-R-G-E-Q, E-Q-H-K-T-C-M, M-L-N-A-T-S-K, S-K-F-M-E-F-Y and F-Y-Q-D-V-K-D. After 1–2 h this complex was added to the cell cultures and incubated for 1 h; then, the contraction rate was counted.

3.3. ETA-AA in serum from cases The ETA-AA inhibitor BQ123 was added to the 15 samples which displayed a negative chronotropic response to Losartan BQ123 blocked the negative chronotropic response, indicating the presence of ETA-AA in 15 samples (Fig. 1C). The ETA-AA was only present in patients with severe PE or HELLP (n = 15), but only in 44% of the samples from this group of patients. To examine if cases with ETA-AA differed from severe cases without ETAAA, statistical analysis was applied on the available clinical and biochemical data (Table 2). No significant difference between the subgroup of cases with severe PE/HELLP presenting with or without ETAAA was found.

2.9. Il17A Serum levels of interleukin-17A (IL-17A) were measured in single measurements using a high-sensitivity human ELISA, BMS2017HS (ThermoFisher Scientific), according to the manufacturer’s instructions. The assay has a measuring range from 0.23 to 31 ng/l. 2.10. Statistics

3.4. 1st trimester samples contained no AT1-AA or ETA-AA

Data were analyzed by Mann-Whitney or Kruskal Wallis test using GraphPad Prism 7.04. P-values < 0.05 (95% confidence interval) were considered to be statistically significant. Results were depicted as medians (interquantile range).

From each of the 17 cases with severe PE or HELLP, an archived serum sample was obtained in the first trimester and analyzed for the presence of autoantibodies. The bioassay showed no change in the beat rate after the addition of AS precipitated serum from these samples, indicating the absence of AT1-AA and ETA-AA in these samples.

3. Results 3.1. Study population

3.5. Progression of autoantibodies in cases with preeclampsia 41 cases with GH (n = 2), moderate PE (n = 7), severe PE (n = 23) and HELLP-syndrome (n = 9) and 10 healthy pregnant controls were included in the study. The mean gestational age at the time of blood sampling was 34–36 weeks for the cases and 35 weeks for the controls. Cases and controls did not differ in parameters such as age, BMI and GA at blood sampling, but differed according to diastolic and systolic blood pressure, mean arterial pressure and GA at birth (see Table 1).

To investigate if autoantibodies correlate with clinical development of preeclampsia, we examined 5 cases where blood samples were obtained at 3 different time points in the pregnancy covering the first trimester including GH/moderate preeclampsia to severe preeclampsia/ HELLP syndrome. None of the cases expressed autoantibodies in the samples from the first trimester. All five cases had developed AT1-AA when diagnosed with GH or moderate PE. These autoantibodies were present when the cases developed severe PE. One of the five cases had developed ETA-AA in the stage of GH, and four out of five cases were positive for ETA-AA at the stage of severe PE diagnosis. The median number of days between diagnoses of GH/moderate PE to severe PE was 7.5 days.

3.2. AT1-AA in serum from cases and controls All cases (n = 41) with gestational hypertension, preeclampsia or HELLP syndrome presented with autoantibody positivity, significantly increasing the beat rate with a median value of 19.20 beats/minute (Fig. 1A). The increase in beat rate could have been attenuated by the addition of Losartan to the cell cultures, indicating the presence of AT1AA in the samples (Fig. 1B). A subgroup of case samples (n = 15) displayed a negative chronotropic effect after the addition of Losartan

3.6. Epitope analysis The binding epitope of AT1-AA was examined by the use of five

Table 1 Study population according to diagnosis at outcome. Values depicted as median (interquantile range). P-values calculated for nonparametric data using Kruskal Wallis-test. *Mann-Whitney test applied.

Age, years BMI, kg/m2 GA at blood sampling, weeks Systolic BP, mmHg Diastolic BP, mmHg MAP, mmHg Proteinuria, dipstick GA at birth, weeks Birthweight, g Birthweight, g (singleton) Birthweight, g (multiple pregnancies) Birthweight, z-score Birthweight, z- score (singleton) Birthweight, z-score (multiple pregnancies)

Controls (n = 10)

GH + Moderate PE (n = 9)

Severe PE (n = 23)

HELLP (n = 9)

P-value

28 (26–31) 21 (20–22) 35 (30–37) 120 (111–127) (n = 6) 80 (73–85) (n = 6) 93 (86–100) (n = 6) 0 (n = 3) 41 (40–42) 3283 (3042–3950)

30 (26–34) 27 (20–31) 34 (32–38) 142 (135–149) 94 (92–98) 111 (106–114) 2 (0–2.5) 37 (34–38) 2640 (1765–3140) 3133 (2225–3284) 2006 (1640–2636) (n = 3) −1.1 (−2.9–0) −0.3 (−2.3–0) −1.4 (−3.2;−0.8) (n = 3)

28 (26–34) 25 (22–28) 36 (33–38) 165 (159–172) (n = 22) 105 (96–109) (n = 22) 127 (117–128) (n = 22) 2 (2–3) (n = 17) 37 (34–39) 2526 (1888–3005) 2840 (1888–3063) 2455 (1698–2520) (n = 2) −1.1 (−1.9;−0.5) −1.0 (−1.9;−0.5) −1.39 (−3.2;−0.6) (n = 2)

29 (28–32) 21 (19–32) 36 (33–38) 168 (130–175) (n = 8) 107 (96–113) 125 (107–135) (n = 8) 2 (0–2) (n = 7) 36 (33–38) 2040 (1450–2700)

0.8 0.2 0.9 0.0002 0.0002 0.0001 0.06 0.0007 0.002 0.001 0.914* 0.5 0.3* > 0.9*

−0.8 (−1.5–0)

191

−1.1 (−3.2;−0.8)

Pregnancy Hypertension 14 (2018) 189–194

S. Buttrup Larsen et al.

-20

LP EL

G

H

+

G H

+

H

EL LP

ve Se

M od er

H

re

PE

PE at

e

C as es

tr ol s C on

-5 -10

-30

-10

0

re

0

-10

5

Se ve

10

0

M od er at e

20

' beat rate/min

' beat rate/min

' beat rate/min

10

10

30

Fig. 1. A. Effects of the AS precipitated IgGs from controls and PE patients on the beating rate of the cardiomyocytes. The IgGs of the PE patients exert a positive chronotropic effect. B. This effect was blocked by the antagonist of the angiotensin II AT1 receptor. In the presence of Losartan the AT1 receptor autoantibodies of moderate and severe PE patients and those with HELLP syndrome were blocked. Some of the patients’ IgGs developd a negative chronotropic effect realized via the endothelin1 ETA receptor. C. This negative chronotropic response was blocked by the ETA receptor antagonist BQ123. These data show that the sera of the PE patients contain AABs against the AT1 receptor and the patient with a severe PE or HELLP syndrome also AABs against the ETA receptor.

overlapping short peptides from the second extracellular loop of the human AT1 receptor covering residues 169–192 V-F-F-I-E-N, E-N-T-N-IT, I-T-V-C-A-F, A-F-H-Y-E-S-Q- and Q-N-S-T-L-P. Five clinical samples were treated with peptides. The epitope AFHYESQ was the only peptide that could attenuate the positive chronotropic response induced by AT1-AA (Fig. 2A). A similar experiment was performed to estimate the epitope of the autoantibodies on the second extracellular loop of the endothelin receptor, ETA. The peptides F-E-Y-R-G-E-Q, E-Q-H-K-T-C-M, M-L-N-A-T-S-K, S-K-F-M-E-F-Y and F-Y-Q-D-V-K-D covering the sequence 229–254 of the second extracellular loop of human ETA receptor were examined, and only the epitope EQHKTCM was able to abolish the negative chronotropic response induced in 6 samples, suggesting that the peptide corresponds to the binding site of the ETA-AA (Fig. 2B).

Table 2 Biochemical and clinical differences between ETA-AA positive and ETA-AA negative cases with severe PE/HELLP. Results depicted as median (interquantile range). P-values calculated using Mann-Whitney test. *Calculated using Fisher’s exact test.

LDH, U/l ALAT, U/l Platelets, 10^9/l Urat, mmol/l Proteinuria, dipstick Systolic BP, mmHg Diastolic BP, mmHg sFlt-1, pg/mL PlGF, pg/mL sFlt-1/PlGF-ratio ET1 (MoM) GA birth, weeks Birthweight, g Birthweight, z-score Early-onset PE (Before GA 34), n (%)

AT1-AA (n = 18)

AT1-AA + ETA-AA (n = 14)

P-value

247 (197–288) 34 (13–243) 176 (127–235) 0.40 (0.32–0.44) 2 (1–2) 164 (154–169) 103.5 (96–110) 12,025 (9003–13968) 38 (31–80) 294 (203–448) 1.66 (1.14–1.91) 36 (34–39) 2410 (1880–2878) −1.14 (−1.77 to −0.77) 7 (39%)

240 (186–355) 27 (14–79) 193 (95–205) 0.35 (0.30–0.41) 2 (2–3) 166 (162–175) 108 (95–111) 8753 (5766–12198)

0.873 0.55 0.77 0.38 0.21 0.37 0.55 0.14

31 (18–79) 205 (60–634) 1.81 (1.47–2.04) 37 (35–38) 2510 (1740–2940) −1.07 (−2.75 to −0.48) 4 (29%)

0.30 0.82 0.60 0.93 0.71 0.76

3.7. IL-17A IL-17A was measured by ELISA in 28 samples from the first trimester and 46 samples from the third trimester (Fig. 3A). The results show that the mean concentration of IL-17A was low both in the first – (n > LLD) and third trimester (n > LLD) in samples from controls. The mean concentration was also low in samples from patients diagnosed with gestational hypertension (n > LLD) and moderate preeclampsia (n > LLD), as well as in samples from the first trimester in

0.71*

40

' beat rate/min

20 10 0

-10

-20

FE

YR

IP ST L Q N

A FH YE SQ

IT VC A F

N

EN TN IT

FI E

G EQ EQ H LK TC M M LN AT SK SK FM EF Y FY Q D VK D

-30

-10

VF

' beat rate/min

0 30

192

Fig. 2. Epitope analyses of the AABs directed against the second extracellular loop of the AT1 and ETA receptor. A. The AABs against the AT1 receptor were treated with five overlapping peptides corresponding to the AT1 receptor. The identified binding site is represented by the amino acid sequence AFHYESQ. B. The AABs against the ETA receptor were treated with 5 overlapping peptides of the second extracellular loop of this receptor. The sequence EQHLKTCM was able to neutralize the functional activity of the AAB and represent the binding site of the ETA receptor AABs.

Pregnancy Hypertension 14 (2018) 189–194

S. Buttrup Larsen et al.

40

5

30

3

IL17A (ng/l)

IL17A (ng/l)

4

2

20

1

10

Co nt ro M Se ls o ve de re at eP PE E + HE LL P

tr i

tr i

+

m

m

es t

es t

er 1

er 3

0

PE

PE

e ve r Se

ve r

e

Third trimester

Se

First trimester

+

+

H

H

EL

EL

LP

LP

G H

G H

Co + nt M ro od Se ls er ve at re e PE PE + HE LL P

0

Fig. 3. A: Concentration (mean ± SEM) of IL-17A (ng/l) in the serum from patients with gestational hypertension (GH), preeclampsia (PE), HELLP syndrome or controls in samples from the first and the third trimester. B: Concentration of IL-17A in paired samples from the first and third trimester in patients with severe preeclampsia or HELLP syndrome.

drawn at a stage of gestational hypertension. The results from the five cases in which we examined multiple samples at different clinical stages of disease development indicate that the expression of Endothelin-1 Receptor type A autoantibodies is closely linked to the development of severe hypertension in pregnancy-induced hypertensive disorders. However, only approximately 44% of patients with severe preeclampsia or HELLP syndrome displayed Endothelin-1 Receptor type A autoantibodies. The pathophysiological effect of ETA-AA in patients with severe preeclampsia is not known, butexperiments with infusion of ETA-AA into pregnant rats are warranted. Compared to the group which was ETA-AA negative, we did not find any difference related to the severity of the disease, other biomarkers, or time of onset for clinical symptoms. We have recently shown that Endothelin 1 alone and together with sFlt-1 is a strong predictor of the development of severe preeclampsia [11]. Whether expression of ET-1 and ETA-AA are timely correlated in these patients remains to be elucidated. Epitope analyses of AT1-AA and ET-AA indicate that the binding sites of both autoantibodies are located at the second extracelluar loop of both receptors. The specific epitopes for both autoantibodies are similar to what was reported previously in a cohort of patients from Brazil [6] and for AT1-AA also from North America [20], and European patients [5]. In the bioassay the endothelin receptor autoantibodies induce a negative chronotropic effect on cardiomyocytes which is different from the effect of the Angiotensin II receptor I autoantibodies. However, the effect on the in-vitro cultures of cardiomyocytes from newborn rats may differ f'rom the physiological effect of stimulation of the endothelin receptor on vascular smooth muscle cells. Therefore the pathophysiological effects of ETA-AA in preeclampsia are awaiting further studies. Plasma IL-17A [21–23] and Il-17 producing CD3 + CD4 + T lymphocytes [24] are reported to be increased in patients with preeclampsia. Plasma IL-17 is also increased in RUPP rats and in normal pregnant rats supplemented with CD + T cells from RUPP rats [25].

these patients (n > LLD). In patients with symptoms of severe preeclampsia or HELLP syndrome, the mean concentration of IL-17A was high, as with samples from the first trimester in these patients. However approximately 40% of samples from patients with severe preeclampsia or HELLP had a concentration below the detection limit both in samples from the first and third trimesters. The results furthermore show that a subgroup of patients with severe preeclampsia or HELLP syndrome expressed a concentration of IL17A both at the time of their symptoms and also in blood samples from the first trimester (Fig. 3B). There was a tendency towards concentrations of IL-17A above LLD in patients with positive ETA-AA in the third trimester, however not significant (data not shown). 4. Discussion The results from this study show that 100% of patients with pregnancy-induced hypertensive disorders developed autoantibodies against the angiotensin II receptor 1 irrespective of the grade of the disease, and that the antibodies developed during the pregnancy since the antibodies were not present in samples from the first trimester in these patients. It is also evident that development of these antibodies was absolutely related to pregnancy-induced hypertension since none of the controls with normal pregnancy developed AT1 autoantibodies. The percentage of patients with AT1-AA in patients with preeclampsia is similar to the Brazilian population investigated by Velloso et al. [17]. In further comparison to the results shown by Velloso et al. [6], we found that a group of patients with severe preeclampsia or HELLP developed Endothelin-1 Receptor type A autoantibodies, and that these were not present in the first trimester in these patients. Again, none of the controls included in the study developed ETA-AA. Only one out of fourteen samples from patients with gestational hypertension or moderate preeclampsia expressed ETA-AA, and in this particular patient, severe preeclampsia developed nine days after the blood sample was 193

Pregnancy Hypertension 14 (2018) 189–194

S. Buttrup Larsen et al.

Our results show that a subgroup of patients with severe preeclampsia or HELLP syndrome express high concentrations of IL-17A, and that the same patients express high levels in blood samples from the first trimester (gestational week 8–10). Although a high concentration of IL17 was almost exclusively seen in the group of patients which developed severe preeclampsia or HELLP syndrome, a large group of patients with severe PE/HELLP expressed IL-17A below the detection limit of the assay. We did not find any significant difference in clinical or biochemical biomarkers between the two groups. Detection of IL-17A early in blood samples from gestational week 5–16 was also investigated by Salazar et al. [25], who reported that patients who later developed preeclampsia had a significantly higher percentage of CD4 + IL-17 + out of total CD4 + T cells, as compared to patients who did not develop preeclampsia. Together these results suggest that a subgroup of pregnant women displayed a proinflammatory state with an increased concentration of Th17 cells and increased Plasma IL-17A already in the first trimester, and that this is associated with the risk of developing pregnancy-induced hypertensive disorders. In conclusion, according to our results AT1-AA is expressed in patients at all clinical stages of pregnancy-induced hypertensive disorders, and ETA-AA is expressed only in severe preeclampsia and HELLP syndrome. Furthermore, IL-17A is increased in a subset of patients with preeclampsia but also in the first trimester in these patients.

angiotensin AT1 receptor, J. Clin. Invest. 103 (7) (1999) 945–952. [6] E.P. Velloso, R.L. Pimentel, J.F. Braga, A.C. Cabral, Z.S. Reis, M. Bader, et al., Identification of a novel agonist-like autoantibody in preeclamptic patients, Am. J. Hypertens. 29 (3) (2016) 405–412. [7] A.L. Malte, N. Uldbjerg, D. Wright, N. Torring, Prediction of severe pre-eclampsia by a combination of sFlt-1, CT-pro ET1 and blood pressure: exploratory study, Ultrasound Obstet. Gynecol. (2017). [8] K. Wallace, S. Richards, P. Dhillon, A. Weimer, E.S. Edholm, E. Bengten, et al., CD4+ T-helper cells stimulated in response to placental ischemia mediate hypertension during pregnancy, Hypertension 57 (5) (2011) 949–955. [9] S. Wellmann, J. Benzing, S. Fleischlin, N. Morgenthaler, S. Fouzas, C.A. Buhrer, et al., Cardiovascular biomarkers in preeclampsia at triage, Fetal Diagn Ther. 36 (3) (2014) 202–207. [10] K. Wenzel, A. Rajakumar, H. Haase, N. Geusens, N. Hubner, H. Schulz, et al., Angiotensin II type 1 receptor antibodies and increased angiotensin II sensitivity in pregnant rats, Hypertension 58 (1) (2011) 77–84. [11] C.C. Zhou, Y. Zhang, R.A. Irani, H. Zhang, T. Mi, E.J. Popek, et al., Angiotensin receptor agonistic autoantibodies induce pre-eclampsia in pregnant mice, Nat. Med. 14 (8) (2008) 855–862. [12] J. Li, B. LaMarca, J.F. Reckelhoff, A model of preeclampsia in rats: the reduced uterine perfusion pressure (RUPP) model, Am. J. Physiol. Heart Circ. Physiol. 303 (1) (2012) H1–8. [13] B. LaMarca, G. Wallukat, M. Llinas, F. Herse, R. Dechend, J.P. Granger, Autoantibodies to the angiotensin type I receptor in response to placental ischemia and tumor necrosis factor alpha in pregnant rats, Hypertension 52 (6) (2008) 1168–1172. [14] B.T. Alexander, A.N. Rinewalt, K.L. Cockrell, M.B. Massey, W.A. Bennett, J.P. Granger, Endothelin type a receptor blockade attenuates the hypertension in response to chronic reductions in uterine perfusion pressure, Hypertension 37 (2 Pt 2) (2001) 485–489. [15] S.R. Murphy, B.B. LaMarca, K. Cockrell, J.P. Granger, Role of endothelin in mediating soluble fms-like tyrosine kinase 1-induced hypertension in pregnant rats, Hypertension 55 (2) (2010) 394–398. [16] P. Dhillion, K. Wallace, F. Herse, J. Scott, G. Wallukat, J. Heath, et al., IL-17mediated oxidative stress is an important stimulator of AT1-AA and hypertension during pregnancy, Am. J. Physiol. Regul. Integr. Compar. Physiol. 303 (4) (2012) R353–8. [17] D.C. Cornelius, B. Lamarca, TH17- and IL-17- mediated autoantibodies and placental oxidative stress play a role in the pathophysiology of pre-eclampsia, Minerva Ginecol. 66 (3) (2014) 243–249. [18] G. Wallukat, L. Will-Shahab, W. Schulze, A. Wollenberger, Incidence of noradrenaline in the cultures of isolated heart ventricle cells of newborn rats, Acta Biol. Med. Ger. 33 (4) (1974) K57–62. [19] G. Wallukat, A. Wollenberger, Effects of the serum gamma globulin fraction of patients with allergic asthma and dilated cardiomyopathy on chronotropic beta adrenoceptor function in cultured neonatal rat heart myocytes, Biomed. Biochim. Acta 46 (8–9) (1987) S634–9. [20] Y. Xia, H. Wen, S. Bobst, M.C. Day, R.E. Kellems, Maternal autoantibodies from preeclamptic patients activate angiotensin receptors on human trophoblast cells, J. Soc. Gynecol. Investig. 10 (2) (2003 Feb) 82–93. [21] W. Cao, X. Wang, T. Chen, H. Zhu, W. Xu, S. Zhao, et al., The expression of notch/ notch ligand, IL-35, IL-17, and Th17/Treg in preeclampsia, Dis. Mark. 2015 (2015) 316182. [22] A. Molvarec, I. Czegle, J. Szijarto, J. Rigo Jr., Increased circulating interleukin-17 levels in preeclampsia, J. Reprod. Immunol. 112 (2015) 53–57. [23] T. Poordast, F.S. Najib, R. Baharlou, A. Bijani, S.M. Alamdarloo, A. Poordast, Assessment of T helper 17-associated cytokines in third trimester of pregnancy, Iran J. Immunol. 14 (2) (2017) 172–179. [24] D. Darmochwal-Kolarz, M. Kludka-Sternik, J. Tabarkiewicz, B. Kolarz, J. Rolinski, B. Leszczynska-Gorzelak, et al., The predominance of Th17 lymphocytes and decreased number and function of Treg cells in preeclampsia, J. Reprod. Immunol. 93 (2) (2012) 75–81. [25] M.D. Salazar Garcia, Y. Mobley, J. Henson, M. Davies, A. Skariah, S. Dambaeva, et al., Early pregnancy immune biomarkers in peripheral blood may predict preeclampsia, J. Reprod. Immunol. 125 (2018) 25–31.

Competing interests GW and IS, as shareholders and employees of Berlin Cures GmbH, declare that Berlin Cures GmbH only provided support in the form of salaries and research materials but did not have any role in the study design, data collection, analysis and statistical evaluation, decision to publish or preparation of the manuscript. SBL, NU, AS, TTC and NT declare no conflict of interest. Funding The study was supported by a grant from the Independent Research Fund Denmark, and Bagermester August Jensen og Hustrus legat. References [1] L. Duley, The global impact of pre-eclampsia and eclampsia, Semin. Perinatol. 33 (3) (2009) 130–137. [2] A.C. Harmon, D.C. Cornelius, L.M. Amaral, J.L. Faulkner, M.W. Cunningham Jr, K. Wallace, et al., The role of inflammation in the pathology of preeclampsia, Clin. Sci. (Lond.) 130 (6) (2016) 409–419. [3] S.K. Bedoya, B. Lam, K. Lau, J. Larkin III, Th17 cells in immunity and autoimmunity, Clin. Dev. Immunol. 2013 (2013) 986789. [4] B. LaMarca, D.C. Cornelius, A.C. Harmon, L.M. Amaral, M.W. Cunningham, J.L. Faulkner, et al., Identifying immune mechanisms mediating the hypertension during preeclampsia, Am. J. Physiol. Regul. Integr. Compar. Physiol. 311 (1) (2016) R1–9. [5] G. Wallukat, V. Homuth, T. Fischer, C. Lindschau, B. Horstkamp, A. Jupner, et al., Patients with preeclampsia develop agonistic autoantibodies against the

194