Estrous cycle-dependent changes of Fas expression in the bovine corpus luteum: influence of keratin 8/18 intermediate filaments and cytokines

All animal studies described herein were approved by the UNH Institutional Animal Care and Use Committee (IACUC# 090205). Estrous cycles of Holstein dairy cows were monitored using transrectal ultrasonography, and corpora lutea (CL) were removed by colpotomy at days 5 (early stage; n=6 cows) and 16–18 (late stage; n=9 cows) postovulation (ovulation = day 0). Luteal cells obtained from CL at these two stages of luteal function express relatively high and low amounts of keratin intermediate filaments, respectively, based upon previous findings [16, 23]. Prior to CL removal, blood samples were obtained by coccygeal venipuncture using heparinized tubes to measure plasma progesterone concentration and verify the relative stage of the estrous cycle. Corpora lutea and blood samples were transported to the laboratory on ice where the CL were extracted for total RNA (described below) and enzymatically dissociated using collagenase type I (Worthington, Lakewood, NJ) as described previously by others [24]. Following enzymatic dissociation, the viability of the luteal cells was estimated to be 88-93% as determined by trypan blue exclusion. The dissociated luteal cells were then either freshly-fixed in paraformaldehyde for flow cytometric analysis, or placed in serum-free culture for further experimentation (described below). The heparinized blood samples from the cows were centrifuged at 2056xg for 20 min at 4°C to obtain plasma, which was then frozen at −20°C until assayed for progesterone by radioimmunoassay (RIA) as described previously [25].

Total RNA was isolated from the two stages of bovine CL (n= 5–7 CL/stage) using a Quick-RNA^TM Mini Prep kit (Zymo Research, Irvine, CA). The total RNA was then purified from genomic DNA contamination using RQ1 RNase-Free DNase (Promega, Madison,WI). The purified total RNA was reverse-transcribed to synthesize cDNA using the qScript™ cDNA Synthesis Kit (Quanta Biosciences, Gaithersburg, MD). The cDNA was then used for subsequent quantitative real time polymerase chain reaction (Q-RTPCR) with SyBr Green detection (Quanta Biosciences, Gaithersburg, MD). Sequence-specific primers for bovine Fas and β-actin (an internal control gene), validated previously by Vickers et al.[26] and Taniguchi et al.[5], respectively, were as follows:

Forward and reverse primers, respectively for bovine Fas were: 5^′-ATGGGCTAGAAGTGGAACAAAAC-3^′ and 5^′- TTCTTCCCATGACTTTGATACC-3^′. Forward and reverse primers, respectively, for bovine β-actin were: 5^′- GAGGATCTTCATGAGGTAGTCTGTCAGGTC-3^′ 5^′-CAACTGGGACGACATGGAGAAGATCTGGCA-3^′.

A thermal cycler was used to conduct the Q-RTPCR with the cyclic conditions as follows: an initial Taq activation at 95°C for 2 min, followed by 40 cycles of 95°C for 1 second, 55°C for 30 seconds and 72°C for 30 seconds. All reactions were carried out on a 7500 Fast Real-Time PCR System. The data were collected during the last 30 seconds of cycling and the amplification signals of Fas transcripts were quantified using a standard curve based upon an absolute quantitation method. The results were expressed as a ratio of Fas relative to β-actin transcripts as the reference (i.e., internal control gene). Melting curve analysis was performed with conditions as follows: 95°C for 15 seconds, 60°C for 1 min, and 95°C for 15 seconds.

Freshly dissociated luteal cells were seeded in T25 flasks at a density of 2×10⁶ viable cells/flask and in 8-well microchamber slides at 2×10⁴ viable cells/well. The cells were cultured in serum-free Ham’s F12 culture medium (Invitrogen, Carlsbad, CA) supplemented with insulin, transferrin, selenium (ITS; 5μg/5μg/5ng/mL; Sigma Aldrich, St. Louis, MO) and gentamicin (20μg/mL; Invitrogen, Carlsbad, CA) and incubated at 37°C, 5% CO₂ in air and 95% humidity overnight. The purity of the cultures under these serum-free conditions is estimated to be 70-75% steroidogenic cells because other types of cells (e.g., endothelial cells, fibroblasts, etc.) are unable to persist. The day after seeding, the flasks and chamber slides were rinsed and the conditioned medium replaced with fresh culture medium prior to treatments. Initial treatments consisted of flasks and chamber slides treated with either culture medium (control) or 5mM acrylamide (Fisher Scientific, Pittsburgh, PA) for 4 h to disrupt K8/K18 filaments and potentially increase the cell surface expression of Fas. Acrylamide is a selective, reversible, disrupter of K8/K18 filaments in mammalian cells [27] that under short-term culture conditions does not adversely affect microtubules [28, 29], organelles (e.g., mitochondria, [30]), steroid synthesis [31], or cell viability [11]. After the initial 4 h treatment period, all flasks and chamber slides were rinsed twice and the medium replaced. Cells from several flasks were immediately prepared for flow cytometric analysis of Fas and K8/K18 expression as described below. The remaining flasks were treated with a cytokine cocktail containing bovine interferon-γ (IFN, 200 IU/mL; R&D Systems, Minneapolis, MN), murine tumor necrosis factor-α (TNF, 10ng/mL; US Biological, Swampscott, MA), and human recombinant soluble Fas ligand (FasL, 50ng/mL; R&D Systems, Minneapolis, MN) with a murine monoclonal anti-6x histidine cross-linking antibody (1mg/mL; R&D Systems, Minneapolis, MN) for 24 h to induce cell death. Others have previously shown this mixture of cytokines is appropriate, and necessary, to induce Fas-mediated death of bovine ovarian steroidogenic cells in vitro[5, 6, 26, 32, 33]. After 24 h incubation, the flasks were re-treated with the cytokine cocktail for an additional 24 h, prior to assessment of cytokine-induced cell death.

Cytokine-induced cell death in the cultured luteal cells was assessed at three different times during the experiment. The number of attached cells in five random microscopic fields of view was counted in all of the flasks prior to cytokine treatment using a 0.25 mm² grid (initial cell counts). At 24 and 48 h after treatment, the number of attached cells in the flasks was again counted to estimate cell loss (post-treatment cell counts). All five fields of view per flask were averaged and the percent cell death was determined using the following equation:

Murine hepatocytes were among the first cells used to demonstrate that disruption of K8/K18 filaments enhances Fas trafficking to the cell surface [20]. Here we utilized human hepatocyte carcinoma cells (HepG2 cells) to corroborate this finding under the experimental conditions used to disrupt K8/K18 filaments in bovine luteal cells with acrylamide. Briefly, HepG2 cells were seeded into T150 flasks at 2×10⁶ cells/flask. The cells were cultured in Eagle’s Minimal Essential Medium (Sigma Aldrich, St. Louis, MO) supplemented with 10% fetal bovine serum (JRH Biosciences, Lenexa, KS) and incubated at 37°C, 5% CO₂ and 95% humidity. At approximately 70% confluency, the HepG2 cells were subcultured in T25 flasks using approximately 1x10⁶ cells/flask. The following day, the medium was changed and the cultures were exposed to vehicle (control) or 5mM acrylamide for 4 h. Following treatment, the cultures were prepared for flow cytometry to assess cell surface expression of Fas.

Luteal cells from freshly dissociated CL and from serum-free culture were used to analyze Fas and K8/K18 filament expression by flow cytometry. For cells obtained through dissociation of CL, approximately 1.5x10⁶ cells/tube in 0.4mL of Ham’s F12 culture medium were centrifuged using screen-capped tubes (Ref # 352235, BD Falcon, San Jose, CA) for 5 min at 276xg, 4°C. The screened cells were then fixed for 2 h on ice by adding 0.4mL 2% paraformaldehyde to the cell suspension for a final concentration of 1% paraformaldehyde. After fixation, the cells either remained in fixative (for detection of cell surface Fas) or were rinsed twice with PBS and then permeabilized using 70% ethanol (for detection of total Fas and K8/K18 filament expression). Both the fixed and permeabilized cells were stored at 4°C and −20°C, respectively, until further processed for flow cytometry.

Luteal cells in serum-free culture and the HepG2 cells cultured in serum-containing conditions were fixed in a similar manner to the freshly isolated luteal cells described above. Briefly, the flasks of cells were rinsed twice (5 min each) with Hank’s Balanced Salt Solution (Sigma Aldrich, St. Louis, MO), followed by two quick washes with trypsin-EDTA (Cell Gro Mediatech, Manassas, VA). After the second trypsin-EDTA rinse, the remaining trypsin was removed and the flasks were left for 10 min. The trypsinized cells were then collected in Ham’s F12 culture medium containing 10% fetal bovine serum (JRH Biosciences, Lenexa, KS), centrifuged for 5 min at 276xg, 4°C and resuspended in Ham’s F12 culture medium without serum. As above, approximately 1.5x10⁶ cells/tube were centrifuged using screen-capped tubes for 5 min at 276xg, 4°C. The filtered cells were fixed for 2 h on ice in 1% paraformaldehyde and either remained in fixative (detection of cell surface Fas; luteal and HepG2 cells) or were permeabilized with 70% ethanol (detection of total Fas and K8/K18 filaments; luteal cells only). Both the fixed and the permeabilized cells were stored at 4°C and −20°C, respectively, until analyzed by flow cytometry.

Fixed cells (i.e., luteal and HepG2) were washed twice (5 min each) with phosphate buffered saline with 0.1% bovine serum albumin (PBS-BSA) and centrifuged at 276xg for 5 min at 4°C between each wash. Following the second wash, the cells were stained for Fas using a mouse anti-human Fas antibody (clone CH11; Millipore, Billerica, MA; diluted 1:25 with PBS with 10% normal goat serum) or an identical concentration of nonspecific, IgG1 isotype (clone MOPC-21; Sigma) as a control. The cells were incubated in primary antibody overnight at 4°C and then washed twice (5 min each) with PBS-BSA with spins at 276xg for 5 min at 4°C between each wash. Detection of the primary antibody was achieved fluorescently using a goat anti-mouse Alexa 488-conjugated IgG secondary antibody (Invitrogen, Carlsbad, CA) diluted 1:200 with PBS-BSA with 10% normal goat serum. For detection of K8/K18, luteal cells from CL dissociation and from culture were washed twice (5 min each) with PBS-BSA and spun at 276xg for 5 min at 4°C between each wash. The cells were then incubated for 1 h at 37°C with a mouse anti-human K18 FITC-conjugated antibody (clone CY-90; Sigma Aldrich, St. Louis, MO; diluted 1:100 with PBS- BSA). Previously we have shown K18 dimerizes with K8 such that targeting of K18 is sufficient for the detection of K8/K18 filaments in bovine luteal cells [16]. Quantification of cells expressing Fas and K8/K18 was accomplished using a 4 color, dual laser FACScalibur flow cytometer (Becton Dickinson Biosciences, San Jose, CA) with a 488nm argon laser for FITC/Alexa 488 excitation. The negative controls, either IgG1-FITC (for K18 detection) or Alexa-488 secondary antibody only (for Fas detection), were used to set the fluorescence gating to 1% positive controls prior to analysis. The cells were recorded on the FL-1 filter at no more than 800 events/second with a total of 10,000 recorded events. Data were collected using Cell Quest (Becton Dickinson Biosciences, San Jose, CA) and graphs of the results were generated using WinMDI 2.9 software (Scripps Institute, La Jolla, CA). Mean fluorescence intensity (MFI), a measure of staining intensity for each cell, was calculated using the following equation:

Bovine luteal cells cultured in microchamber slides were used to evaluate microscopically the efficacy and specificity of acrylamide as a disrupter of K8/K18 filaments. The cells were rinsed twice with PBS, fixed using 4% paraformaldehyde in PBS for 20 min on ice, and then stored in PBS at 4°C until permeabilized with methanol and analyzed for K8/K18 expression and microtubule expression (negative control) by fluorescent microscopy. Briefly, the previously-fixed luteal cells were rinsed twice with PBS-BSA followed by a 1 h block/permeabilization step with 0.3% triton x-100 in PBS containing 10% normal goat serum (Vector Labs, Burlingame, CA) and 3% BSA. The slides were rinsed 3 × 5 min with PBS-BSA and incubated overnight at 4°C with either a mouse anti-human K18 monoclonal antibody (clone CY-90; Sigma Aldrich, St. Louis, MO; diluted 1:800 in PBS-BSA with 10% normal goat serum), or a mouse anti-bovine alpha-tubulin monoclonal antibody (clone 236–10501; Invitrogen, Carlsbad, CA; diluted 1:200 in PBS-BSA with 10% normal goat serum). The following day, after 3 × 5 min washes with PBS-BSA, fluorescent detection of the K18-containing filaments or tubulin-containing microtubules was achieved by incubating the slides with a goat anti-mouse Alexa 488-conjugated IgG antibody (K18; Invitrogen, Carlsbad, CA) or a goat anti-mouse Texas Red-conjugated antibody (microtubules; Santa Cruz, Santa Cruz, CA). Both secondary antibodies were diluted 1:200 in PBS-BSA with 10% normal goat serum (Vector Labs, Burlingame, CA). The slides were counterstained with 4',6-diamidino-2-phenylindole (DAPI) mounting medium (Vector, Burlingame, CA) and then coverslipped.

The data were analyzed by 1-way or 2-way ANOVA followed by Tukey’s multiple comparison test using the general linear model of Systat 12.0 (Point Richmond, CA). Results are expressed as mean ± SEM, with each experiment repeated three to nine times (i.e., n= 3–9). For experiments requiring cultured cells, the cells were cultured in triplicate for a given experiment and were derived from individual CL or from a frozen stock of cells (HepG2 cells). Thus, the total number of experiments (n=) is equivalent to the total number of CL or frozen aliquots (HepG2) used to establish the cultures; shown in figure legends). Differences among means at a value of P<0.05 were considered statistically significant.

Freshly dissociated luteal cells from early and late stage CL were characterized for total Fas expression (Figure 1) and expression on the cell surface (Figure 2) relative to a non-specific IgG control. Measurement of plasma progesterone revealed the cows used to obtain early stage CL had lower systemic progesterone than cows used for late stage CL (1.8 ± 0.2 versus 5.9 ± 0.7 ng/ml, respectively; P<0.05, n=6-9 CL/stage). However, a higher percentage of luteal cells expressed total Fas in early stage CL compared to late stage CL (Figure 1A, P<0.05). Mean fluorescence intensity (MFI), a measure of staining intensity for each cell, was also higher among cells from early stage CL compared to late stage CL (Figure 1C,P<0.05). Similarly, the expression of Fas on the cell surface was greater for cells of early stage CL compared to late stage CL (Figure 2B, P<0.05), as was MFI (Figure 2C, P<0.05). Overall, quantification of the percentage of cells expressing Fas on the cell surface relative to total Fas expression revealed cells from early stage CL express the majority of Fas on the cell surface (76%), whereas less than half these cells from late stage CL do so (47%). In terms of relative steady-state concentrations of Fas mRNA in the luteal tissue, Q-RTPCR indicated there was no difference between early versus late stage CL (Figure 3; P>0.05, n=5-7 CL/stage).

Interestingly, a comparison of Fas expression for freshly dissociated luteal cells versus luteal cells placed in culture for 24 h revealed that culture alone substantially increased the relative cell surface expression of Fas for cells of both early and late stage CL. Cell surface expression of Fas increased from ~65% to ~97% as a result of culture for cells of early stage CL (P<0.05, n=4 expts.), and from ~18% to ~66% for cells of late stage CL (P<0.05, n=4 expts.).

A higher percentage of freshly dissociated luteal cells from early stage CL expressed K8/K18 filaments than late stage CL (Figure 4, P<0.05). Average number of cells expressing K8/K18 filaments in early stage CL was 46% compared to 26% for late stage CL (Figure 4B). In contrast to what was observed for cell surface expression of Fas, culture of luteal cells for 24 h did not enhance K8/K18 expression in cells of early or late stage CL. Relative percentage of K8/K18-positive cells was 46% vs. 49% for freshly dissociated vs. cultured cells, respectively, in early stage CL, and was 26% vs. 23% for freshly dissociated vs. cultured cells, respectively, in late stage CL (P>0.05, n=4 expts., data not shown).

Exposure of cultured bovine luteal cells to acrylamide disrupted K8/K18 filaments without adversely affecting microtubule organization (Figure 5). Cells in control cultures exhibited extensive, filamentous networks of K8/K18 staining (Figure 5A) that became aggregated around the perinuclear region of the cells following acrylamide exposure (Figure 5C). Conversely, microtubule organization when compared between control and acrylamide-treated cultures remained unaffected (Figure 5B and D, respectively). In addition, there was no observable effect of stage of CL on these outcomes, and the acrylamide treatment overall had no effect on the number of cells expressing K8/K18 filaments, luteal cell viability or progesterone secretion (P>0.05; n=2-4 CL/stage, data not shown).

Although acrylamide disrupted K8/K18 filaments, no increase in the cell surface expression of Fas was observed for luteal cells of either stage of CL (Figure 6A-C; P>0.05). Moreover, K8/K18 filament disruption failed to enhance Fas cell surface expression on specific cells, as reflected by the lack of change in relative MFI (Figure 6D; P>0.05). Consistent with the observations of freshly isolated luteal cells, cultured luteal cells of early stage CL expressed higher amounts of Fas on the surface than cultured cells of late stage CL (Figure 6C and D; P<0.05). In contrast, disruption of K8/K18 filaments in HepG2 cells, using identical experimental conditions to those for bovine luteal cells, increased the number of cells expressing Fas on the cell surface (Figure 7; P<0.05).

Exposure of the cultured bovine luteal cells for 48 h to a cytokine cocktail consisting of IFN, TNF, and FasL induced cell death, as expected, but there was no effect of K8/K18 disruption by acrylamide (P>0.05) or stage of CL (P>0.05, n= 4 CL/stage) on this outcome (Figure 8). Similar results were observed when the luteal cells were exposed to cytokines and acrylamide for only 24 h (data not shown).