Fractalkine receptor is expressed in mature ovarian teratomas and required for epidermal lineage differentiation

Tissue microarray (TMA) slides containing specimens of ovarian teratoma (Cat# OV805) and specimens of human fetal tissues (Cat# BE01014) were obtained from US Biomax (Rockville, MD). Because these specimens were commercially available and were deidentified, no approvals by the Institutional Review Board were required. GFP-tagged CX₃CR1 shRNA and scrambled shRNA constructs were obtained from Origene Technologies (Rockville, MD). DharmaFECT was obtained from Dharmacon (Lafayette, CO). Matrigel was obtained from BD Biosciences (Bedford, MA). Mouse monoclonal CK14 and CK18 and rabbit polyclonal CX₃CR1 (Cat# ab49747, ab49824, ab8021 and ab8020, respectively) antibodies were obtained from Abcam (Cambridge, MA), and mouse monoclonal β-tubulin antibody was obtained from Developmental Studies Hybridoma Bank (Iowa City, IA), goat polyclonal CX₃CL1 (Cat# AF365) was purchased from R&D systems (Minneapolis, MN), and mouse monoclonal actin antibody was purchased from Santa Cruz Biotechnology (Dallas, TX). Vectastain ABC and DAB kits were obtained from Vector Laboratories (Burlingame, CA). BMP-4 was purchased from Invitrogen (Carlsbad, CA).

Ovarian teratocarcinoma cell line PA-1 and ovarian carcinoma cell line SKOV-3 were obtained from American Tissue Culture Collection (Manassas, VA). PA-1 cells were plated on Matrigel-coated (at 1/100 dilution) tissue culture plates or glass cover slips at 5% confluence and cultured in minimal essential media supplemented with sodium pyruvate and non-essential amino acids for 2 weeks. Cells subjected to differentiation into skin lineage were cultured in the above described media containing 10 μg/ml BMP-4.

TMA slides were deparaffinized by baking at 60°C for 2 h and rehydrated by incubation in xylene, 100% ethanol, 95% ethanol, 70% ethanol, and phosphate buffered saline, pH 7.4, for 5 min each. Peroxidase activity was inhibited with hydrogen peroxide. Antigen retrieval was performed by 15 min incubation in 1 mM ethylene diamine tetraacetic acid (EDTA; pH 8.0) at 95°C. Prior to primary antibody staining (1:50 dilution, rabbit anti-human CX₃CR1, Abcam Cat# ab8021, 1 h at room temperature (RT)) non-specific binding was blocked by incubation with 10% goat serum for 1 h. The biotin-conjugated goat anti-rabbit secondary antibody was used at a 1:200 dilution for 30 min at RT. The Vectashield ABC kit was used as directed by the manufacturer, and tissues were incubated for 45 min at RT. The DAB reagent was prepared as instructed by the manufacturer and applied to tissues on TMA slides for 2 – 10 min until brown color developed. TMAs were stained with hematoxylin, dehydrated in 50%, 70%, 95%, and 100% ethanol, and mounted with Permount. Pancreatic cancer tissue was used as a positive control. Staining was evaluated by A.K.-B. and L.R., who were both blinded to the experimental outcomes of the study. Staining was scored based on the intensity and percentage of positive cells. Intensity of staining was “0” for negative samples, “1” for weakly positive samples, “2” for moderately positive samples, and “3” for highly positive samples. Overall scores were derived as the intensity score multiplied by the percentage. Staining was assessed separately in the cytoplasm and the membrane.

PA-1 cells were cultured to 80% confluence and transfected with either CX₃CR1-specific GFP-tagged shRNA construct or scrambled GFP-tagged shRNA construct (Origene Technologies) using DharmaFECT according to the manufacturer’s instructions. Selection of the clones was performed with puromycin and green fluorescence as suggested by the manufacturer.

Cells were cultured on glass coverslips to 50-70% confluence, fixed, permeabilized with 0.01% Triton X-100, and blocked in goat serum. Mouse monoclonal CK-14 (Cat# ab49747) antibody was used at a dilution of 1:50. Cells were incubated with the primary antibodies for 1 h at room temperature (22°C). The cells were then incubated with secondary Alexa555-conjugated anti-mouse antibodies (1:500) for 1 h at RT in the dark. 4’,6-Diamidino-2-phenylindole (DAPI) was added to the secondary antibody solution to a final concentration of 10 μg/ml 10 min prior to the end of the incubation period. Cells were washed, air dried, and mounted on glass slides using ProlongGold (Invitrogen, Carlsbad, CA). Fluorescent imaging was performed using a Zeiss AxioObserverD.1 fluorescent microscope.

Western blotting analysis was used to detect the expression of CK14, CK18, CX₃CR1, CX₃CL1, actin, and β-tubulin in PA-1. This procedure was performed as previously described [8, 14, 15] Antibodies were used at the following dilutions: 1:500 mouse anti-human-CK14 or CK18, or CX₃CL1, rabbit anti-human CX₃CR1 (Abcam ab8020) and 1:200 mouse anti-human-β-tubulin or goat anti-human actin in 3% BSA in a solution of 50 mM tris-buffered saline, pH 7.4, 150 mM NaCl, and 0.05% Tween-20 (TBST). Immunoreactive bands were visualized with anti-(mouse-IgG)-peroxidase, anti-(goat-IgG)-peroxidase, and anti-(rabbit-IgG)-peroxidase (Sigma, St. Louis, MO) (1:1000 in 3% BSA in TBST), and enhanced chemiluminescence was read using Chemidoc (Bio-Rad) and Bio-Rad Chemidoc ImageReader software.

According to our previous observations CX₃CR1 was expressed in cells comprising ovarian teratoma tissue specimens [8], although the number of specimens was very limited in order to be able to draw any conclusions. In the present study we aimed to expand this observation on a larger number of ovarian teratoma specimens and determine the predominant cell type(s) positive for CX₃CR1. Thus, we tested expression of CX₃CR1 using a tissue microarray containing 68 specimens of mature ovarian teratoma, 10 specimens of immature teratoma, and 2 cases of monodermal teratoma. Interestingly, we have found that skin keratinocytes represent the most predominant CX₃CR1-positive cell type, which was present in 38% of tested cases of mature ovarian teratoma (Table 1, Figure 1). The next most common CX₃CR1-positive types of cells in the tested cases of dermoid cyst were those comprising the pilosebaceous unit of the skin assembled of sebaceous gland and hair follicle (Table 1, Figure 1). Other CX₃CR1-positive cell types in the order of diminution included lymphatic, apocrine, adipose, thyroid, respiratory epithelial, peripheral nervous, endothelial, and eccrine (Table 1). Cases of tested immature ovarian teratoma were characterized by the presence of CX₃CR1-positive brain, lymphatic, respiratory epithelial, adipose, and sebaceous cells (Table 1). Two tested monodermal teratomas consisted of CX₃CR1-positive thyroid and lymphatic tissues, respectively (Table 1). Thus, our data suggest that skin keratinocytes are the predominant type of CX₃CR1-positive cells in the tested specimens of mature ovarian teratomas.

Further we wished to determine CX₃CR1-positive cells undergoing the normal development. We tested both male and female fetal tissues aged 5 months in utero. We found that the majority of epithelial cells, including skin keratinocytes, were CX₃CR1-positive (Table 2). Hence, CX₃CR1-positive skin cells are present not only within abnormally differentiated tissues, such as mature ovarian teratomas, but also in the normally developing skin.

Based on our observations regarding CX₃CR1 expression in both normally and abnormally developing keratinocytes, we hypothesized that CX₃CR1 may be important for cell differentiation into the epidermal lineage. Thus, we tested the requirement for CX₃CR1 in keratinocyte differentiation using a cell culture model. Ovarian teratocarcinoma cell line PA-1 has been established from the ascites of a patient with malignant ovarian teratoma [16]. Chromosomal analysis had suggested that these cells have a heterozygous origin and derived at a stage prior to the first meiotic division [17]. Several studies have shown that PA-1 is capable of differentiation into diverse cell lineages at the appropriate culture conditions [16, 18, 19]. Based on many similarities between the teratocarcinoma cells and early embryonic cells, the former has been used to study processes of differentiation [19, 20]. Using gene specific shRNA constructs, we downregulated CX₃CR1 in PA-1 cell line (Figure 2A). Further we cultured both CX₃CR1sh-PA-1 and the control cells transfected with a scrambled shRNA construct on Matrigel in the presence and absence of 0.5 nM BMP-4 for 2 weeks. Expression of cytokeratin 14 (CK14) and cytokeratin 18 (CK18) was analysed with immunofluorescence microscopy and Western blotting (Figure 2B,C). We observed that only PA-1 transfected with scrambled shRNA and cultured in BMP-4-supplemented media expressed significantly higher levels of CK14 compared to scrambled shRNA PA-1 or CX₃CR1 shRNA PA-1cultured in the absence of BMP-4, as well as CX₃CR1 shRNA PA-1 cultured in the presence of BMP-4 (Figure 2B,C). In opposite, scrambled shRNA PA-1 or CX₃CR1 shRNA PA-1cultured in the absence of BMP-4, as well as CX₃CR1 shRNA PA-1 cultured in the presence of BMP-4 expressed higher levels of CK18 compared to scrambled shRNA PA-1 cultured in the presence of BMP-4 (Figure 2C). PA-1 expressed fractalkine ligand (CX₃CL1) necessary for the receptor (CX₃CR1) activation (Figure 2D). These data suggest that CX₃CR1-positive PA-1 cells are able to proceed toward differentiation into skin keratinocytes in the presence of BMP-4 and begin expressing a marker of stratified epithelium CK14, while the loss of CX₃CR1 restricts differentiation and the cells remain CK18-positive simple-epithelial cells.