Human Chorionic First Trimester Villous Explant Culture and z-VAD-fmk and z-DEVD-fmk Treatments

Explant cultures were performed as previously described.31 In brief, placental tissues (5 to 8 weeks of gestation) were placed in ice-cold phosphate-buffered saline (PBS) and processed within 2 hours of collection. Tissues were aseptically dissected to remove decidual tissue and fetal membranes. Small fragments of placental villi (25 to 45 mg wet weight) were teased apart, placed on Millicell-CM culture dish inserts (Millipore Corp., Bedford, MA) precoated with 0.15 ml of undiluted Matrigel (Collaborative Biomedical Products, Bedford, MA), and transferred to a 24-well culture dish. Explants were cultured in serum-free Dulbecco’s modified Eagle’s medium/F12 (Life Technologies, Inc., Grand Island, NY) supplemented with 100 μg/ml streptomycin, 100 U/ml penicillin, and incubated overnight at 37°C in 5% CO₂ in air to allow attachment. Explants were maintained in standard condition (5% CO₂ in 95% air) or in an atmosphere of 3% O₂/92% N₂/5% CO₂ for 48 hours at 37°C. The morphological integrity and viability of villous explants under the various oxygenation conditions were monitored daily as previously reported.11,31 Explants from more than 18 different placentae in more than 15 separate experiments were used. A minimum of three explants per experimental condition was used at all times. Explants were exposed to hypoxia-reoxygenation (HR) as previously described11,32 in the presence of 100 μmol/L of the pan-caspase inhibitor z-VAD-fmk (BioMol, Plymouth Meeting, PA) or caspase-3-specific inhibitor z-DEVD-fmk (R&D Systems, Minneapolis, MN) both dissolved in dimethyl sulfoxide (equivalent volume of dimethyl sulfoxide in the absence of inhibitors was used in control conditions).

RNA Analysis

RNA extraction was performed using a RNeasy mini kit (Qiagen, Valencia, CA), reverse-transcribed using a random hexamer approach, and amplified by 40 cycles of quantitative polymerase chain reaction (PCR) (15 minutes at 95°C, cycle: 30 seconds at 95°C, 30 seconds at 60°C, and 30 seconds at 72°C). Southern blot analysis was performed as previously described,11 using a ³²P-labeled full-length Mcl-1L probe. Primer sequences used in reverse transcriptase (RT)-PCR/Southern blot analysis were: Mcl-1 (NM_021960): forward 5′-ATGTTTGGCCTCAAAAGAAACGCG-3′ and reverse 5′-GGCTATCTTATTAGATATGCCAA-3′ (Mcl-1L: predicted size = 1054 bp and Mcl-1S: predicted size = 806 bp) and β-actin (NM_001101): forward 5′-CTTCTACAATGAGCTGCGTG-3′, reverse 5′-TCATGAGGTAGTCAGTCAGG-3′ (predicted size = 304 bp). All amplified and cloned products were confirmed by sequencing. Quantitative PCR was performed using the SYBR Green I dye DyNamo HS kit (MJ Research, Waltham, MA) based on the manufacturer’s protocol using isoform-specific primers for Mtd-L and Mtd-P (Mtd-L: forward 5′-GCCTGGCTGAGGTGTGC-3′, Mtd-P: forward 5′-GCGGGAGAGGCGATGA-3′, reverse (both L and P) 5′-TGCAGAGAAGATGTGGCCA-3′, Mcl-1L: forward 5′-ATGCTTCGGAAACTGGACAT-3′, Mcl-1S: forward 5′-CCTTCCAAGGATGGGTTTG-3′, Mcl-1: reverse (both L and S) 5′-CTAGGTTGCTAGGGTGCAA-3′). For syncytin and cytokeratin 7 analyses, qRT-PCR was performed using Assays-on-Demand TaqMan primers and probe (Applied Biosystems, Foster City, CA). Analysis was done using the DNA Engine Opticon2 System (MJ Research). Data for all qPCR analyses were normalized against expression of 18S ribosomal RNA as previously described.33

Western Blot Analysis

Western blotting was performed as previously described.34 Fifty μg of total protein from placental tissue was subjected to 12% (w/v) sodium dodecyl sulfate-polyacrylamide gel electrophoresis. Membranes were probed at 4°C overnight with a 1:1000 dilution of primary antibodies: rabbit polyclonal Mtd antibody capable of recognizing all isoforms as previously reported11; Mcl-1-specific rabbit polyclonal antibody (SC-819 clone S-19 from Santa Cruz Biotechnology, Santa Cruz, CA); specific cleaved caspase-3 (Asp175) (5A1); cleaved caspase-8 (Asp384) rabbit polyclonal antibodies (Cell Signaling, Beverly, MA), and anti-PARP p85 (active PARP) fragment rabbit polyclonal antibodies (1:300 dilution; Promega, Madison, WI). Rabbit polyclonal antisera against N-terminal amino acids (28 to 41) of syncytin (NM_014590) was generated by injecting rabbits first with 400 μg of KHL-conjugated peptide in Freund’s complete adjuvant followed by three boosts of 100 μg in incomplete adjuvant. Titer was checked by enzyme-linked immunosorbent assay using bovine serum albumin-conjugated peptide as antigen. For anti-Mtd, Mcl-1, and syncytin antibodies, preimmune serum and competing peptides were used as controls. After overnight incubation, membranes were washed with TBS/T and incubated for 60 minutes at room temperature with 1:5000 diluted horseradish peroxidase-conjugated anti-rabbit (Santa Cruz Biotechnology). Blots were exposed to chemiluminescent ECL-plus reagent (Amersham, Piscataway, NJ). All blots were confirmed for equal protein loading and transfer using Ponceau staining.35

Immunohistochemistry

Immunohistochemical analyses were performed using an avidin-biotin-based immunoperoxidase approach, as previously described.32 Nonspecific binding sites were blocked using 5% (v/v) normal goat serum and 1% (w/v) bovine serum albumin in Tris-buffer. Slides were incubated overnight at 4°C with a 1:200 dilution of rabbit polyclonal anti-Mtd or anti-Mcl-1 antibodies. After washing, slides were probed with a 300-fold dilution of biotinylated goat anti-rabbit or goat anti-mouse IgG (Vector Laboratories, Burlingame, CA) for 1 hour at 4°C. Avidin-biotin complex was applied for 1 hour. Slides were developed in 0.075% (w/v) 3,3-diaminobenzidine in PBS (pH 7.6) containing 0.002% (v/v) H₂O₂, counterstained with hematoxylin, dehydrated in an ascending ethanol series, cleared in xylene, and mounted. In control experiments, primary antibodies were replaced with blocking solution [5% (v/v) normal goat serum and 1% (w/v) bovine serum albumin].

Transfection Experiments

Human choriocarcinoma JEG-3 cells (75% confluence) were transfected with 1.5 μg/35-mm dish of either empty pcDNA3.1 vector or pcDNA-3-Mcl-1L vector (kindly provided by Dr. Ruth Craig, Dartmouth Medical School, Hanover, NH) using Lipofectamine 2000 (Invitrogen, Carlsbad, CA). Twenty-four hours after transfection, cells were treated with 10 mmol/L of the NO donor sodium nitroprusside (SNP; Sigma, St. Louis, MO) or exposed to HR for 5 hours (3 hours at 3% O₂ followed by 2 hours at 20% O₂) as previously described.11 Cells were then harvested, and the percentage of dead cells was assessed by trypan blue staining.

Mcl-1 Cleavage in Human Villous Explants Is Regulated by Caspase Activation

The appearance of several specific Mcl-1-immunoreactive bands suggested the presence of various Mcl-1 isoforms. Taking into account our transcript data, the band migrating at 26 kd is consistent with the predicted molecular weight of Mcl-1S, originally cloned from a placental cDNA library.19,20 To investigate whether the p29 Mcl-1-immunoreactive band can be a result of caspase-mediated cleavage, we exposed first trimester placental explants to conditions of HR known to induce caspase activation and trophoblast apoptosis in vitro.11,36 Explants were exposed to HR in the presence or absence of z-VAD-fmk, a broad-based inhibitor of caspase activity and caspase-mediated Mcl-1L cleavage.21,22,23,26 Relative to untreated controls, tissues exposed to HR demonstrated a notable Mcl-1 isoform switch; ie, Mcl-1L protein decreased with a concomitant increase of p29 immunoreactive Mcl-1 protein (Figure 2a). Exposure of HR-treated explants to z-VAD-fmk abrogated the appearance of the immunoreactive p29 Mcl-1 band, restored Mcl-1L content, and increased Mcl-1S amount relative to HR and untreated tissues (Figure 2, a and c). To investigate whether Mcl-1 cleavage under HR was mediated by executioner caspases, explants under HR were incubated in the presence and absence of a caspase-3/7-specific inhibitor. Similar to z-VAD-fmk treatment, z-DEVD-fmk exposure under HR conditions prevented caspase-mediated Mcl-1 p29 formation and increased both Mcl-1L and Mcl-1S levels (Figure 2b). Thus, both Mcl-1L and Mcl-1S seem to be regulated by caspase cleavage under HR stress and p29 Mcl-1 is most likely the previously described proapoptotic cleavage product of Mcl-1L.21,22,23,24,25

Open in a separate window

Figure 2

Inhibition of caspase activity and its effect on Mcl-1 cleavage. a: Representative immunoblot of Mcl-1 isoforms in control (untreated explant) and explants exposed to HR in the presence of 100 μmol/L concentration of pan-caspase inhibitor z-VAD-fmk relative to vehicle treatment (control). b: Representative Mcl-1 immunoblot of untreated control explants and explants exposed to HR (3 hours at 3% O₂ followed by 2 hours at 20% O₂) in the presence or absence of caspase-3 inhibitor z-DEVD-fmk. c: Densitometric analysis of Mcl-1-specific protein isoforms between HR (open bar)- and HR + z-VAD-fmk (black bar)-treated explants. *P < 0.05. n = 5 separate experiments performed in triplicate.

Oxygen Regulation of Mcl-1 in First Trimester Human Placental Tissues

Because preeclampsia is associated with placental hypoxia/oxidative stress, the effects of varying oxygenation on Mcl-1 expression were assessed in vitro using villous explants exposed to 20% oxygen, 3% oxygen, and HR. In 3% oxygen (normal oxygen tension for <8 weeks) relative to standard 20% O₂ conditions, mRNA expression of Mcl-1L significantly increased (more than 2.5-fold, P = 0.0095), whereas that of Mcl-1S decreased (0.2-fold, P = 0.01) (Figure 3a). The opposite expression profile was observed under HR conditions in which Mcl-1L decreased (twofold, P = 0.01) and Mcl-1S increased (2.5-fold, P = 0.01) relative to 20% control (Figure 3a). The protein profile of Mcl-1 under 3% oxygen showed increased Mcl-1L expression relative to the 20% O₂ controls (Figure 3b). Importantly, HR resulted in markedly decreased levels of anti-apoptotic Mcl-1L and a concomitant increased formation of proapoptotic Mcl-1c (p29 Mcl-1) relative to 20% O₂ (Figure 3b). Although the expression of Mcl-1S at the transcript level was markedly increased in HR, this pattern was not evident at the protein level.

Open in a separate window

Figure 3

Effect of varying oxygenation on Mcl-1 expression. a: qRT-PCR analyses of Mcl-1L (black bars) and Mcl-1S (open bars) in first trimester villous explants exposed to 20% O₂, 3% O₂, and HR. b: Representative immunoblot of Mcl-1 isoforms in first trimester villous explants exposed to 20% O₂, 3% O₂, and HR. c: Left: qRT-PCR analyses of Mcl-1L (black bars) and Mcl-1S (open bars) in placental tissue from 7 to 20 weeks of gestation; right: Mcl-1 immunoblot of normal first and second trimester placental tissues. d: Percentage of cell death measured by trypan blue exclusion in JEG-3 cells transfected with Mcl-1L and subsequently subjected to HR (3 hours at 3% O₂ followed by 2 hours at 20% O₂) and SNP treatments. All immunoblots were confirmed for equal protein loading using Ponceau staining. *P < 0.05 versus C; **P < 0.05 versus HR and SNP.

During first trimester, the human placenta experiences a surge in oxygenation ~10 to 12 weeks of gestation when the intervillous space opens to maternal circulation. Rapid placental oxygenation, similar to a reperfusion injury, may hence result in a transient state of oxidative stress.37 Therefore, we hypothesized that Mcl-1 cleavage would occur at that period. Indeed, there was increased mRNA expression of both Mcl-1L and Mcl-1S (Figure 3c, left) and formation of Mcl-1c (p29 Mcl-1) at 10 to 13 weeks (coinciding with the placental surge of oxidative stress), which declined at later gestational ages (Figure 3c, right).

Next, we assessed the function of Mcl-1L under conditions of oxidative stress (HR) because this condition has been shown to be the strongest inducer of Mtd-L and -P expression.11 Because SNP, a nitric oxide donor, is a powerful inducer of cell death,38 SNP treatment was used as positive control. Both HR and SNP treatments of JEG-3 cells significantly increased cell death when compared with cells maintained at 20% O₂ (Figure 3d). Mcl-1L overexpression in JEG-3 cells significantly decreased HR or SNP-induced cell death when compared with cells transfected with empty vector (Figure 3d).

Mcl-1 Expression in in Vivo Chronic Placental Hypoxia

As previously demonstrated, preeclamptic placentae have a global gene expression similarity to placental tissues obtained from HA pregnancies because both conditions may be affected by aberrant placental oxygenation.39 As such, Mcl-1 expression was next examined under physiologically reduced placental oxygenation using HA placentae from normal pregnancies. Placental Mcl-1 isoform mRNA expression was different between HA and MA relative to SL control placentae (SL). Although Mcl-1L transcript was significantly increased in HA (~2.3-fold, P = 0.005) and MA (~2-fold, P = 0.022) relative to SL, Mcl-1S transcript levels were significantly decreased in HA and MA relative to SL samples (HA versus SL, ~0.4-fold, P < 0.05; MA versus SL, ~0.5-fold, P = 0.016) (Figure 4, a and b).

Open in a separate window

Figure 4

Transcript expression of Mcl-1 and Mtd isoforms in placental tissue from SL, MA, and HA pregnancies. a–d: qRT-PCR analysis of Mcl-1L (anti-apoptotic), Mcl-1S (proapoptotic), Mtd-L (proapoptotic), and Mtd-P (proapoptotic) transcript expression in SL (n = 10), MA (n = 10), and HA (n = 16) placentae. *P < 0.05. e and f: Mcl-1 and Mtd immunoblots of protein lysates obtained from SL, MA, and HA placentae. All immunoblots were confirmed for equal protein loading using Ponceau staining (not depicted). g: Immunohistochemical localization of Mcl-1 (top) and Mtd (bottom) in SL, MA, and HA placentae. S, stroma; ST, syncytium.

Because Mtd is one of the interacting partners of Mcl-113,14 and Mtd-L and Mtd-P expressions have been shown to increase in preeclamptic placentae as well as under conditions of oxidative stress,11 their expression was herein examined in conditions of chronic hypoxia. Although Mtd-L mRNA expression was not significantly different between the various groups, the number of Mtd-P transcripts was significantly decreased in HA samples relative to MA (4-fold, P < 0.0001) and SL tissues (0.5-fold, P = 0.04) (Figure 4d).

Similar to Mcl-1 mRNA expression, Mcl-1L protein expression was increased in HA and to a lesser extent in MA when compared with SL samples (Figure 4e). Both Mcl-1Lc and Mcl-1S molecules were hardly detectable at the protein level in SL, MA, and HA samples with no apparent change in expression (Figure 4e). Although expression of all known Mtd proteins (L: ~28 kd; S: ~18 kd; and P: ~15 kd) could be observed, no apparent difference in Mtd expression was noted between the conditions tested (Figure 4f).

Mcl-1 immunoreactivity in placental sections from SL, MA, and HA samples was predominantly observed in trophoblast cell layers (Figure 4g, top) with low to absent staining in stromal cells. Mtd immunoreactivity was generally low and also restricted to trophoblast cells. Positive immunoreactivity for Mtd was greater in SL tissue compared with HA and MA tissues (Figure 4g, bottom). No positive staining was observed in control sections in which Mtd and McL-1 antibodies were replaced with nonimmune IgG (data not shown).

Caspase Activation in in Vivo Chronic and Pathological Placental Hypoxia

Excessive cell death via death receptor is a well-known phenomenon occurring frequently in trophoblast cells of preeclamptic placentae.3,40,41 Although this death pathway was described previously, downstream events related to caspase-8 activation have not been explored in PE tissues. Relative to AMC tissues, cleaved caspase-8 was increased in severe early-onset preeclampsia (Figure 5a, left), consistent with previous reports showing increased activation of caspase-3.42

Open in a separate window

Figure 5

Markers of cell death in conditions of chronic and pathological placental hypoxia. a: Representative cleaved caspase-8 and p85 PARP immunoblots in AMC and PE tissues. b: Top: Representative immunoblots of cleaved caspase-3 and -8 in HA and control tissues (SL and MA); bottom: densitometric analysis of cleaved caspase-3 and -8 protein in HA (n = 12) relative to control tissues (SL and MA, n = 18). SL samples were set at 1 and all other (MA and HA) samples were normalized to the SL samples. Values of MA and SL samples were combined together as controls to which HA values were compared. c: Representative immunoblot of p85 PARP in HA and control tissues (SL and MA). All immunoblots were confirmed for equal protein loading using Ponceau staining. *P < 0.05.

Because we observed a shift in the Mtd/Mcl-1 apoptotic rheostat toward increased prosurvival and decreased death-inducing isoforms in HA, we also measured markers of trophoblast cell death in HA and control placentae. Expression of cleaved caspase-3, a known marker of trophoblast apoptosis,11,32 was reduced in HA by 50% (P = 0.0297) relative to controls (MA and SL) (Figure 5b, right). Similar to cleaved caspase-3, expression of activated caspase-8 in HA was decreased by 40% relative to controls (MA and SL) (P = 0.0335) (Figure 5b, left). We validated our apoptosis findings by measuring the p85 fragment of cleaved poly-ADP-ribose polymerase (PARP), another classical marker of apoptosis. As anticipated, p85 PARP was increased in preeclamptic tissue relative to age-matched normotensive controls (Figure 5a, right), whereas a marked decrease in p85 PARP content was observed in HA samples relative to SL and MA tissues (Figure 5c).

We thank Dr. Ljiljiana Petkovic for placental collection.