Human pancreatic cancer cells lines, Panc-1 (CRL-1469), MiaPaCa-2 (CRL-1420), and Capan-2 (HTB-80), were obtained from ATCC and used as an in vitro model in this study. All cell lines were cultured at 37°C under 5% v/v CO₂ using DMEM-F12 media (Thermo Fisher) containing 10% v/v fetal calf serum with 2 mM l-glutamine, and penicillin (100 U/ml)/streptomycin (100 µg/ml) (Thermo Fisher) until 80–90% confluency was reached.

Plasmid pUK-D1 (containing cDNA sequences for 8 Gly-X-Y repeats, neck, and CRD region of human SP-D), transformed into Escherichia coli BL21 (λDE3) pLysS (Invitrogen), was used to express rfhSP-D, as described earlier (15, 16). The expression cassette included a short stretch of eight N-terminal Gly–X–Y triplets with substitution of S for P in position 2 (residue 180), followed by the α-helical coiled-coil neck region (residues 203–235) and the globular CRD region (residues 236–355). Endotoxin levels were determined using the QCL-1000 Limulus amebocyte lysate system (Lonza) and the assay was found to be linear over a range of 0.1–1.0 EU/ml (10 EU = 1 ng of endotoxin). The amount of endotoxin levels were <4 pg/μg of the rfhSP-D. Full length native SP-D (FL-SP-D) was purified form lung washings of alveolar proteinosis patients using methods previously described by Strong et al. (24).

All cell lines used in this study (Panc-1, MiaPaCa-2, and Capan-2) were grown on coverslips using 0.5 × 10⁵ cells overnight. Next day, cells were washed three times with PBS before being incubated with rfhSP-D (20 µg/ml) in a serum-free DMEM-F12 medium. For rfhSP-D and FL-SP-D binding analysis, the coverslips were incubated for 1 h with mouse anti-human SP-D (rfhSP-D) and rabbit anti-human SP-D (FL-SP-D) (MRC Immunochemistry Unit, Oxford; 1:200), followed by goat anti-mouse IgG H&L (Cy5) and Goat anti-Rabbit IgG H&L Alexa Fluor 488 (1:500; Abcam), respectively, and Hoechst (1:10,000; Thermo Fisher) for fluorescence microscopy analysis. For apoptosis analysis via fluorescence microscopy using an FITC annexin V apoptosis detection kit with propidium iodide (PI) (BioLegend), the cells were incubated with rfhSP-D (20 µg/ml) for 48 h. After 48 h, the cells were incubated with annexin V binding buffer containing FITC annexin V (1:200), PI (1:200), and Hoechst (1:10,000) for 15 min, and washed twice with PBS before mounting on the slides to visualize under a HF14 Leica DM4000 microscope.

Cell lines were plated in a 6-well plate (0.1 × 10⁷) and incubated with rfhSP-D (20 µg/ml), FL-SP-D (10 and 20 µg/ml), and an untreated control, for 24 and 48 h, followed by cell detachment using 5 mM EDTA, pH 8, and centrifugation at 1,200 × g for 5 min. For cell cycle analysis, the cells were fixed in 70% v/v ethanol for 30 min at 4°C, followed by PBS wash twice at 850 × g. The cells were then treated with ribonuclease (100 µg/ml) to ensure DNA staining without RNA contamination before staining with PI (50 µg/ml). 10,000 cells were then acquired for both treated and untreated samples and the PI histograms were plotted using the set markers within the analysis program of Novocyte Flow Cytometer. For apoptosis analysis via FACS, FITC annexin V apoptosis detection kit with PI (BioLegend) was used, as per manufacturer’s instructions. Compensation parameters were acquired using unstained, untreated FITC stained, and untreated PI stained cells.

MTT (3-[4,5-dimethylthiazol-2-yl]-2,5-diphenyltetrazolium bromide) (Thermo Fisher) assay was performed by incubating pancreatic cancer cells (0.1 × 10⁵) in a 96-well microtiter plate with rfhSP-D, FL-SP-D (10 and 20 µg/ml), and an untreated control in serum-free DMEM-F12 medium for 48 h, followed by incubation with 50 µg/µl MTT (5 mg/ml stock) per well for 4 h at 37°C. Majority of the media was removed leaving behind 25 µl per well, which was mixed thoroughly with 50 µl of dimethyl sulfoxide and incubated for another 10 min at 37°C. The absorbance was read at 570 nm using a plate reader.

Cell lines (0.1 × 10⁷ cells) were seeded in a 6-well plate (Nunc) and incubated with rfhSP-D (20 µg/ml), together with an untreated control, in a serum-free DMEM-F12 medium. The cells were lysed within the wells using treatment buffer (50 mM Tris-HCl pH 6.8, 2% v/v β-mercaptoethanol, 2% v/v SDS, 0.1% w/v bromophenol blue, and 10% v/v glycerol) and transferred to pre-cooled microcentrifuge tubes followed by sonication for 15 s. The samples were heated at 100°C for 10 min and subjected to SDS-PAGE (12% w/v) for 90 min at 120 V. The SDS-PAGE separated proteins were then electrophoretically transferred onto a nitrocellulose membrane (Thermo Fisher) using an iBLOT (Thermo Fisher). The membrane was then blocked using 5% w/v dried milk powder (Sigma) in 100 ml PBS for 2 h on a rotatory shaker at room temperature. The membrane was incubated with rabbit anti-human caspase primary antibodies (anti-cleaved caspase 3; anti-cleaved caspase 8; Cell Signaling) at 4°C overnight, followed by incubation with secondary Goat anti-rabbit IgG HRP-conjugate (1:1,000; Promega) for 1 h at room temperature. The membrane was washed with PBST (PBS + 0.05% Tween 20) three times, 10 min each time. The color was developed using 3,3′-diaminobenzidine substrate kit (Thermo Fisher).

Panc-1, MiaPaCa-2, and Capan-2 cells were incubated with and without rfhSP-D (20 µg/ml) for various time points. The cell pellet for each time-point was centrifuged and stored at −80°C. RNA was extracted using GenElute Mammalian Total RNA Purification Kit (Sigma-Aldrich, UK), as per manufacturer’s instructions, followed by treatment with DNase I (Sigma-Aldrich, UK). The absorbance at 260 and 260:280 nm ratio was used to determine the concentration and purity of total RNA, respectively, using NanoDrop 2000/2000c (Thermo-Fisher Scientific). Total RNA (2 µg) was used for cDNA synthesis using High Capacity RNA to cDNA Kit (Applied Biosystems). The forward and reverse primers used in this study were designed using the web based Basic Local Alignment Search Tool and Primer-BLAST (http://blast.ncbi.nlm.nih.gov/Blast.cgi) are given in Table 1.

Relative mRNA expression was determined by qPCR reactions performed in triplicates consisting of 10 µl final volume per well [5 µl Power SYBR Green MasterMix (Applied Biosystems), 75 nM of forward and reverse primers, and 500 ng cDNA], using the 7900HT Fast Real-Time PCR System (Applied Biosystems). Samples were initially incubated at 50°C (2 min) and 95°C (10 min), followed by 40 cycles (each cycle for 15 s at 95°C and 1 min at 60°C) for amplification of the template. Human 18S rRNA, an endogenous control, was used to normalize the gene expression. Relative quantification (RQ) value and formula: RQ = 2^−ΔΔCt was used to calculate the relative expression of each target.

Graphs were made and statistically analyzed using Graphpad Prism 6.0 by applying an unpaired two-way ANOVA test. Significance of values is based on *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001 between treated and untreated samples. Error bars represent the SD or SEM, as indicated in the figure legends.

The fluorescence microscopy analysis of rfhSP-D and FL-SP-D binding to Panc-1, MiaPaCa-2, and Capan-2 cells revealed its membrane localization following 1 h incubation at 4°C (Figure 1). The rfhSP-D probed with mouse anti-human SP-D-CY5 antibody and FL-SP-D probed with rabbit anti-human SP-D-FITC appeared evenly bound in clusters on the cell membrane, along with nucleus stained positively with Hoechst. All cell lines showed a similar rfhSP-D and FL-SP-D binding pattern. No CY5 or FITC fluorescence was detected in the untreated controls, probed with primary and secondary antibodies, for each cell line, suggesting the rfhSP-D and FL-SP-D binding observed in the treated cell lines was protein-specific.

Panc-1, MiaPaCa-2, and Capan-2 cell lines were individually treated with rfhSP-D for 24 h to assess whether rfhSP-D induced growth arrest. DNA binding dye, PI, was used to analyze the cell cycle for both treated and untreated cells via DNA quantitation. rfhSP-D induced inhibition of DNA synthesis in treated Panc-1 (68%) and MiaPaCa-2 (50%) in comparison to untreated Panc-1 (3%) and MiaPaCa-2 (2%) cells, respectively, as the cells were arrested in G1 phase (Figure 2). DNA synthesis was unaffected in the untreated cells for both cell lines since Panc-1 (43%) and MiaPaCa-2 (31%) were seen in S phase and Panc-1 (32%) and MiaPaCa-2 (33%) in the G2 phase of cell cycle. The growth arrest was, however, not seen in Capan-2 cell line following the rfhSP-D treatment (data not shown). Growth arrest at 24 h following rfhSP-D treatment prompted the determination of cell fate at a later time point; therefore, all cell lines were analyzed for likely apoptosis at 48 h.

The qualitative apoptosis analysis of Panc-1, MiaPaCa-2, and Capan-2 treated with FL-SP-D or rfhSP-D for 48 h using immunofluorescence microscopy (Figure 3A) showed that the cell membrane was disoriented and the PI bound to DNA in the treated cells as compared to untreated cells, where no florescence was detected, indicating that cells were undergoing apoptosis at 48 h.

The flow cytometry analysis to quantify apoptosis showed significant reduction in the viable cell percentage of Panc-1, MiaPaCa-2, and Capan-2. The rfhSP-D induced apoptosis in ~67% of Panc-1 cells at 48 h, out of which, ~57% Panc-1 cells were both FITC and PI positive and ~7% were FITC alone positive, suggesting annexin V/FITC binding to phosphatidylserine, a cell membrane phospholipid, which is externalized during early apoptotic stage and the passage of PI, a DNA stain, through the porous cell membrane into the nucleus in order to intercalate the DNA. Approximately, 10% cells were PI alone positive, which suggested that these cells were either dead or in late apoptotic stage. The percentage of viable cells, i.e., unstained, in the untreated sample was significantly higher (70%) as compared to treated (26%) (Figure 3B). The rfhSP-D induced apoptosis in MiaPaCa-2 was ~60%. However, rfhSP-D induced apoptosis in Capan-2 (~35%) cell line, which was not as much as in Panc-1 and MiaPaCa-2 cell lines (Figure 3B). The treatment with FL-SP-D (20 µg/ml) for 48 h induced apoptosis in approximately 25% of Panc-1 and MiaPaCa-2 cell lines, and less than 10% in Capan-2 cell line. No significant difference was seen with FL-SP-D (10 µg/ml) treatment for 48 h in all the cell lines investigated in this study (Figure 3C).

The cell viability analysis via MTT assay following rfhSP-D treatment showed ~60% decrease in the cell viability of Panc-1 and MiaPaCa-2 and 45% in Capan-2 as compared to untreated (Figure 3D) and BSA (10 and 20 µg/ml; data not shown) controls. The cell viability analysis via MTT assay following FL-SP-D treatment also showed consistent reduction as seen in flow cytometer analysis (Figure 3E). Apoptosis was further confirmed by analyzing the activation of caspase to determine the pathway involved.

Western blot analysis revealed that caspase 8 and 3 were cleaved in all the cell lines following treatment with rfhSP-D for 48 h (Figure 4). The cleavage of caspase 3, however, was not seen in the untreated cells and faint bands appeared for caspase 8 in the untreated cells (Figure 4), which further confirmed that cell death occurred via apoptosis. Interestingly, although Capan-2 cell line appeared unaffected in terms of cell cycle arrest at 24 h; yet, the cleaved bands for caspase 8 and 3 were seen in Capan-2 treated cells too, which suggested that rfhSP-D can affect the cancer cells via multiple pathways. Caspase 9 was tested as a marker for intrinsic apoptosis pathway; however, no difference was noted between treated and untreated cells (data not shown). Therefore, gene expressions were assessed for pro-apoptotic genes such as Bax, an intrinsic pathway marker, and Fas, an extrinsic pathway marker, to further determine the apoptotic pathway.

Human pancreatic cancer cells often escape apoptosis by downregulating apoptosis stimulators such as FasL/FasR (25), or pro-apoptotic proteins such as Bax (26). These pro-apoptotic genes, Bax and Fas, for time-points ranging from 2 to 24 h in all the cell lines, were analyzed. Bax was unaffected following the treatment with rfhSP-D in Panc-1 and MiaPaCa-2 cell lines at all time-points (data not shown), which, in addition to unaffected caspase 9, suggested that intrinsic pathway may not have been involved in causing the cell death in these cell lines. Fas expression was unaffected at earlier time-points up to 6 h (data not shown); however, it was upregulated at 12 and 24 h in Panc-1 (log₁₀ ~0.5), MiaPaCa-2 (log₁₀ ~1), and Capan-2 (log₁₀ ~1) cell lines (Figure 5A), which indicated that apoptosis induction by rfhSP-D is likely to take place via the extrinsic pathway. Western blot analysis also showed upregulation of Fas at the protein level in rfhSP-D treated cells as compared to untreated cells (Figure 5B). Since TNF-α and NF-κB are crucial factors in the apoptotic pathway and they can regulate Fas expression (27), the effect of rfhSP-D on the gene expression of TNF-α and NF-κB as well as translocation of NF-κB from the cytoplasm to nucleus was investigated.

The p53 transcript levels were measured by qPCR following the treatment with rfhSP-D at 2, 6, and 12 h in Capan-2 cells and compared with the p53 levels in untreated cells for each time-point. Interestingly, the levels of p53 were upregulated, most significantly at 12 h, which suggested that p53 may also have contributed to the apoptosis in Capan-2 cells (Figure 6).

Following treatment with rfhSP-D, the analysis of TNF-α mRNA expression levels showed a significant upregulation in Panc-1 (log₁₀ ~0.5), MiaPaCa-2 (log₁₀ ~1), and Capan-2 (log₁₀ ~1) at 12 and 24 h; however, no difference was observed at earlier time-points. Similar transcriptional upregulation was noted for NF-κB for Panc-1 (log₁₀ ~0.4), MiaPaCa-2 (log₁₀ ~0.8), and Capan-2 (log₁₀ ~0.6) at 12 and 24 h (Figure 7A). Immunofluorescence microscopy of Panc-1, MiaPaCa-2, and Capan-2 cell lines showed that NF-κB was translocated to the nucleus at 24 h, which was not seen in the untreated cells (Figure 7B). This further confirmed that NF-κB could play a key role in deciding the apoptotic fate of the pancreatic cancer cells following the rfhSP-D treatment.

The mTOR is often deregulated in the pancreatic cancer (28) and its activation is associated with poor prognosis (29). Upon treatment with rfhSP-D, mRNA expression of mTOR was downregulated in Panc-1 and MiaPaCa-2 cell line at 12 h (Figure 8A), however, no difference was seen in Capan-2 (data not shown). In addition, immunofluorescence analysis revealed that in comparison to the untreated cells, a significant decrease in the cytoplasmic levels and an increased accumulation of mTOR in the nucleus of MiaPaCa-2 cells was evident (Figure 8B), where it has been shown to be present in its inactive form (30).