Hypoxia-Mediated Soluble Fms-Like Tyrosine Kinase 1 Increase Is Not Attenuated in Interleukin 6-Deficient Mice

Animal Procedures and Tissue Preparation

C57BL/6J (wild-type [wt]; Charles River, Germany) and B6.129S2-Il6tm1Kopf/J (interleukin-6 knockout [IL-6^−/−]; Jackson Laboratory, Maine) mice were bred at the animal facility of the University of Cologne (animal facilities of the Pharmaceutical Institute and the Center of Molecular Medicine Cologne) at 22°C on a 12-hour light, 12-hour dark cycle. Water was available ad libitum and the animals were fed a standard laboratory diet (#1310, Altromin, Germany). Female mice were mated at 8 to 12 weeks of age for 24 hours and assumed to be pregnant when a vaginal plug was expelled (g1). Pregnancy was validated by daily monitoring the body weight. At embryonic day 16 (ED 16), the animals were randomly subdivided into a normoxic and a hypoxic group. Mice from the hypoxic group were placed in a hypoxic chamber (O₂ control glove box for tissue culture; COY Laboratory Products Inc, Michigan) and exposed for 6 hours to a 6% oxygen atmosphere as described previously, ²⁶ whereas normoxic mice were kept at regular atmosphere. After 6 hours, mice were sacrificed immediately without reoxygenation. Blood samples were taken by cardiac puncture, incubated at room temperature for 30 minutes, and then centrifuged for 15 minutes at 4°C and 3000g to remove the clotted fraction. Serum was stored at −80°C. Placentas were removed and dissected to detach cord, amnion sac, and connective tissue. Tissue for total RNA and protein isolation was quick frozen on dry ice and tissue for histological analysis was fixated in phosphate-buffered formaldehyde (Roti-Histofix, Roth, Germany).

All procedures in this study were performed according to protocols approved by the German regulations and legal requirements and were also approved by the local government authorities (Bezirksregierung Koeln).

Total RNA Isolation and TaqMan Real-Time Polymerase Chain Reaction Analysis

For reverse transcriptase polymerase chain reaction (PCR), total RNA was prepared using the TRIzol reagent (Invitrogen, Germany) according to the manufacturer’s recommendations. Total RNA of 1 µg was reverse transcribed into complementary DNA (cDNA) with the MMLV Reverse Transcriptase (Promega, Germany) using Oligo-dT oligonucleotides and random hexamer oligonucleotides (Promega). To determine mRNA expression levels of the cytokine genes IL-6, IL-10, and TNF-α, we used a quantitative real-time PCR method. Expression was normalized to housekeeping genes GAPDH and β-actin. TaqMan PCR reagents (Platinum, Quantitative PCR SuperMix-UGD with ROX; Invitrogen) were applied according to the manufacturer’s protocol. Complementary DNA of 2.5 µL was used as template in a total reaction volume of 25 µL. The thermocycler parameters were 50°C for 2 minutes, 95°C for 10 minutes, followed by 40 cycles of 95°C and 60°C for 1 minute, and the 7500 Real Time PCR System from Applied Biosystems (Germany) was used. For IL-10, we used 10× real-time PCR buffer, ROX passive reference, dNTP mix, and enzymes Hot Star Gold and Uracil N-Glycosylase from Eurogentec (Germany). The following oligonucleotides and probes (MWG, Germany) were used: glyceraldehyde 3-phosphate dehydrogenase (GAPDH) for 5’-ATGTGTCCGTCGTGGATCTGA, rev 5’-TGCCTGCTTCACCACCTTCT, probe 5’-CCGCCTGGAGAAACCTGCCAAGTATG; beta-actin for 5’-TGACAGGATGCAGAAGGAGATTACT, rev 5’-GCCACCGATCCACACAGAGT, probe 5’-ATCAAGATCATTGCTCCTCCTGAGCGC ²⁷ ; IL-6 for 5’-ACAAGTCGGAGGCTTAATTACACAT, rev 5’-AATCAGAATTGCCATTGCACAA, probe 5’-TCTTTTCTCATTTCCACGATTTCCCAGAGAA, ²⁸ IL-10 for 5’-GGCGCTGTCATCGATTTCTC, rev 5’-CACCTTGGTCTTGGAGCTTATTAAA, probe 5’-CCTGTGAAAATAAGAGCAAG, ²⁹ TNF-α for 5’-GGCTGCCCCGACTACGT, rev 5’-GACTTTCTCCTGGTATGAGATAGCAA, probe 5’-CCTCACCCACACCGTCAGCCG ²⁸ ; Flt-1 for 5’-GGTGTCTGCTTCTCACAGGATATG, rev 5’-CGTGGGCAGAGACCATGAGT, probe 5’-AGCAGGCCAGACTCTCTTTCTCAAGTGCA; sFlt-1 for 5’-GGGAAGACATCCTTCGGAAGA, rev 5’-TCCGAGAGAAAATGGCCTTTT, probe 5’-CCGCAGTGCTCACCTCTAACGAGAACTTCT. ³⁰ The probes were modified with 6-carboxyfluorescein (FAM) at the 5’ and tetramethylrhodamine (TAMRA) at the 3’ end. The results of the real-time PCR analysis are calculated based on the ΔΔCt method and expressed as fold induction of mRNA expression compared to the corresponding control group (1.0-fold induction).

Isolation and Immunoblot Analysis

For protein isolation, samples were mixed with protein extraction buffer (6.65 mol/L Urea, 10% glycerol, 1% sodium dodecyl sulfate [SDS], 10 mmol/L Tris-HCl pH 6.8, 5 mmol/L Dithiothreitol, and 0.5 mmol/L phenylmethanesulfonylfluoride) and homogenized. Concentration was measured using Bicichoninacid (BCA, Thermo Scientific). Total protein of 30 µg was incubated with sample buffer (0.5 mol/L Tris, 20.6 % SDS, 25 % β-mercaptoethanol, 2.6 % bromphenolblue, and 10 % glycerin), incubated for 10 minutes at 70°C and then run on an SDS polyacrylamide gel electrophoresis. The gel was then transferred onto a nitrocellulose membrane (Whatman, Germany). After blocking the membrane with 5% milk and 2% bovine serum albumin in TBS-Tween (10 mmol/L Tris, 100 mmol/L NaCl, 0.1% Triton-X 100; pH 7.5), it was incubated with the first antibodies diluted in TBS-Tween: the rabbit polyclonal anti-HIF-1α antibody (#NB100-449, Novus Biologicals, Germany; 1:1000 dilution) and the mouse anti-α–tubulin antibody (clone DM1A, Sigma-Aldrich, Germany; 1:100 000 dilution) at 4°C overnight. The horseradish peroxidase (HPO)-linked secondary antibodies goat anti-mouse HPO and goat anti-rabbit HPO (Cell Signaling, Germany) were used at 1:2000 dilution at room temperature for 1 hour. After applying the ECL Prime Western Blotting Detection Reagent (GE Healthcare, United Kingdom) to the membrane, protein bands were acquired with the BIO-RAD Molecular Imager ChemiDOC XRS+ Imaging System (Bio-Rad Laboratories, Germany).

Quantification of Serum sFlt-1 Levels

Serum was analyzed using an enzyme-linked immunosorbent assay (ELISA) kit to determine sFlt-1 (DY471; R&D Systems, Germany) levels as suggested by the manufacturer. Serum was diluted 1:20 for sFlt-1 ELISA. A duplicate analysis was performed and absorption of samples was measured with a Nano Quant M200 (Tecan, Germany).

Papenheim Staining of Placenta Sections

Formaldehyde-fixated placentas were mounted in paraffin and sections of 10 µm using a rotation microtome (Leica, Germany) were prepared. A Pappenheim protocol was used for staining. The sections were first deparaffinated, then stained with a May-Grünwald staining solution (Roth, Germany), and counterstained with a Giemsa solution (VWR, Germany). Stained sections were analyzed using a microscope (Leica DM6000B; Leica) at 40× magnification. For cell count, at least 3 randomized parts of the labyrinth section were chosen in each placenta and leukocytes were counted.

Statistical Analysis

Statistical analysis was performed with the GraphPad Prism Software (Version 5.03, GraphPad Software, Inc., California). All values in the text are mean ± standard error of the mean. Normality distribution was tested using the D’Agostino & Pearson test. Differences were evaluated either using a 2-tailed Student t test (for normally distributed data) or using a Mann-Whitney test (for nonnormally distributed data). Significant differences were taken at the P < .05 level.

Levels of HIF-1α are Increased Upon Hypoxia in the Placenta of wt and IL-6^−/− Mice

Preeclampsia is thought to evolve in the first trimester of pregnancy; however, symptoms like hypertension, proteinuria, and edema usually arise within the second and third trimester. ^31,32 Hence, we used mice at ED 16 in our experiments which corresponds to the last trimester of murine pregnancy. To ensure that the experimental setting induces hypoxic stress in the placenta, we performed immunoblot analyses to determine the level of HIF-1α. This transcription factor is stabilized under hypoxic conditions ²³ and is therefore often used as hypoxic marker. As shown in Figure 1, a 6-hour incubation of the pregnant wt animals under 6% O₂ atmosphere resulted in significant higher HIF-1α levels compared to control animals kept under normoxic conditions. The same was true for IL-6^−/− mice. This confirms that our experimental setting results in severe hypoxia in the murine placenta as has been shown earlier. ²⁶

OPEN IN VIEWER

Figure 1.

Hypoxia-inducible factor 1α (HIF-1α) levels are increased upon hypoxia in the placenta of wild-type (wt) and interleukin 6 (IL-6)^−/− mice. Pregnant mice at embryonic day 16 (ED 16) were exposed to either normoxic (20.9% O₂; nx) or hypoxic (6% O₂; hyx) conditions for 6 hours. Then, mice were sacrificed and placentas were dissected. Total protein was isolated and subsequently examined via an immunoblot. HIF-1α was detected with a specific antibody, and α-tubulin was determined for normalization. A representative set of samples from wt mice (left) and from IL-6^−/− mice (right) is shown. For each mouse type (wt; IL-6^−/−), the samples derive from 2 independent experiments as indicated (exp 1 and exp 2).

Hypoxia Does Not Induce an Inflammatory Response in the Murine Placenta

One feature of the preeclamptic placenta is a proinflammatory state. ^11,12 In order to reveal a potential inflammatory response upon hypoxic stress in the placenta, we performed real-time PCR analyses and measured the expression levels of the proinflammatory cytokines IL-6 and TNF-α normalized to β-actin. In parallel, we also determined the mRNA expression rate of the antiangiogenetic factor IL-10, which is known to be decreased in preeclamptic placentas. ¹³ As can be seen in Figure 2A, there was no physiologically relevant variation in any cytokine measured, neither in wt (left panel) nor in IL-6^−/− (right panel) mice. We also normalized expression levels to GAPDH (data not shown) and found the same trends.

OPEN IN VIEWER

Figure 2.

Hypoxia does not induce an inflammatory response in the murine placenta. A, Placentas from mice exposed to either normoxic (20.9% O₂) or hypoxic (6% O₂) conditions for 6 hours were dissected on embryonic day 16 (ED 16) of pregnancy and total messenger RNA (mRNA) was isolated. A total of n = 10 normoxic (76 placentas) and n = 11 hypoxic (75 placentas) treated wild-type (wt) mice and a total of n = 10 normoxic (74 placentas) and n = 9 hypoxic (73 placentas) treated interleukin 6 (IL-6)^−/− mice were analyzed. The mRNA expression levels of IL-6 (white bar), IL-10 (gray bars), and TNF-α (black bars), normalized to β-actin, were measured and plotted as fold induction compared to normoxic mice (set as 1). ** P value < .005; *** P value < .0005; ns = not significant. Data for wt mice: left graph; data for IL-6^−/− mice: right graph. B, Some of the placentas collected from normoxic and hypoxic mice on ED 16 were analyzed by performing a Papenheim staining of the labyrinth part of placental sections. Leukocytes were counted under ×40 magnification in at least 3 sections of the labyrinth of placentas from wt (left) and IL-6^−/− (right) mice, exposed to either normoxic (20.9% O₂) or hypoxic (6% O₂) conditions and plotted in the graph as (lc/mm)². Left graph: n = 4 normoxic wt mice; n = 4 hypoxic wt mice. Right graph: n = 4 normoxic IL-6^−/− mice; n = 3 hypoxic IL-6^−/− mice. ns indicates not significant.

We quantified the amount of leukocytes in the labyrinth by performing a Papenheim staining of placenta sections, assuming that an increased rate of leukocytes in hypoxic compared to normoxic placentas would indicate an inflammatory state. However, we saw no differences, both in wt (Figure 2B, left panel) and IL-6^−/− (Figure 2B, right panel) mice, between normoxic and hypoxic tissue. This further corroborates our findings from the real-time PCR analyses that hypoxic stress does not contribute to an inflammatory response in the murine placenta.

Maternal Circulating sFlt-1 Levels are Increased Upon Hypoxia

sFlt-1 is one of the best described factors that is upregulated in the human preeclamptic placenta and is then released into the maternal serum where it is responsible for inducing proteinuria and hypertension in the mother. ³³ Release of sFlt-1 can be induced by hypoxia in vivo as has been shown previously. ²⁴ Indeed, when we measured sFlt-1 levels in the maternal serum of normoxic and hypoxic-treated dams at ED 16, we saw a significant increase upon hypoxia in wt animals (see Figure 3A, left panel). However, IL-6 seems not to be involved in this regulatory process, since sFlt-1 release was also induced in IL-6^−/− mice under hypoxic conditions (see Figure 3A, right panel).

OPEN IN VIEWER

Figure 3.

Maternal circulating soluble fms-like tyrosine kinase 1 (sFlt-1) levels are increased upon hypoxia in both wild-type (wt) and interleukin 6 (IL-6)^−/− mice. A, sFlt-1 serum levels of normoxic or hypoxic-treated mice were determined by enzyme-linked immunosorbent assay (ELISA). A total of n = 10 normoxic and n = 11 hypoxic treated wt mice (left graph) and a total of n = 10 normoxic and n = 9 hypoxic-treated IL-6^−/− mice (right graph) were investigated. *P value < .05; **P value < .005. B, Placentas of a total of n = 10 normoxic (76 placentas) and n = 11 hypoxic (75 placentas) treated wt mice (left graph) and placentas of a total of n = 10 normoxic (74 placentas) and n = 9 hypoxic (73 placentas) treated IL-6^−/− mice (right graph) were analyzed for Flt-1 (white columns) and sFlt-1 gray columns) messenger RNA (mRNA) levels by real-time polymerase chain reaction (PCR) analysis. Expression levels in placentas of hypoxic-treated dams, normalized to β-actin, are plotted as fold induction compared to expression levels in placentas of normoxic mice (set as = 1). ns indicates not significant.

In order to find out whether the increased maternal serum sFlt-1 levels are due to an increased sFlt-1 production of the placenta, we analyzed both Flt-1 and sFlt-1 mRNA expression by real-time PCR. The quantification of total Flt-1 mRNA levels (both Flt-1 and sFlt-1 are detected in the assay) revealed no changes in placentas from hypoxic-treated dams compared to control dams, neither in wt nor in IL-6^−/− mice (Figure 3B, white columns). Also, when measuring sFlt-1 levels in a real-time PCR analysis, we saw no differences between hypoxic and normoxic wt or IL-6^−/− mice (Figure 3B, gray columns). We also normalized expression levels to GAPDH (data not shown) and found the same trends. Hence, maternal circulating serum sFlt-1 levels are upregulated under acute hypoxia, but the rise seems not to be due to an increased placental Flt-1 or sFlt-1 gene expression.