Human SP-D Acts as an Innate Immune Surveillance Molecule Against Androgen-Responsive and Androgen-Resistant Prostate Cancer Cells

Surfactant Protein D (SP-D), a pattern recognition innate immune molecule, has been implicated in the immune surveillance against cancer. A recent report showed an association of decreased SP-D expression in human prostate adenocarcinoma with an increased Gleason score and severity. In the present study, the SP-D expression was evaluated in primary prostate epithelial cells (PrEC) and prostate cancer cell lines. LNCaP, an androgen dependent prostate cancer cell line, exhibited significantly lower mRNA and protein levels of SP-D than PrEC and the androgen independent cell lines (PC3 and DU145). A recombinant fragment of human SP-D, rfhSP-D, showed a dose and time dependent binding to prostate cancer cells via its carbohydrate recognition domain. This study, for the first time, provides evidence of significant and specific cell death of tumor cells in rfhSP-D treated explants as well as primary tumor cells isolated from tissue biopsies of metatstatic prostate cancer patients. Viability of PrEC was not altered by rfhSP-D. Treated LNCaP (p53+/+) and PC3 (p53 −/−) cells exhibited reduced cell viability in a dose and time dependent manner and were arrested in G2/M and G1/G0 phase of the cell cycle, respectively. rfhSP-D treated LNCaP cells showed a significant upregulation of p53 whereas a significant downregulation of pAkt was observed in both PC3 and LNCaP cell lines. The rfhSP-D-induced apoptosis signaling cascade involved upregulation of Bax:Bcl2 ratio, cytochrome c and cleaved products of caspase 7. The study concludes that rfhSP-D induces apoptosis in prostate tumor explants as well as in androgen dependent and independent prostate cancer cells via p53 and pAkt pathways.


INTRODUCTION
Prostate cancer, an adenocarcinoma of epithelial cell-origin, is the second most frequently diagnosed cancer among men (1). In its early stages, prostate cancer cells rapidly proliferate in an androgen dependent manner, and thus, are treated by androgen deprivation therapy (2). Conventional anti-cancer treatments such as chemotherapy and radiotherapy improve survival, but many patients encounter relapse and metastasis. Following the remittent stage, the cancer progresses to become androgenindependent where most therapeutic strategies fail.
Immunotherapy by stimulating pattern recognition receptors of the innate immune system such as toll-like receptors (TLRs) has shown promise as a preferred adjunct treatment against cancer (3). Imiquimod, a synthetic imidazoquinoline and an agonist that targets TLR-7 and induces the production of proinflammatory cytokines including IFN-α, IL-6, and TNF-α, inhibited cancer growth in the mouse prostate via apoptosis induction (4,5).
Collectins are pattern recognition proteins belonging to the C-type lectin family. They are composed of an N-terminal cysteine-rich region, a triple-helical collagen domain, an αhelical coiled-coil neck region, and C-terminal carbohydrate recognition domain (CRD) (6). Surfactant Protein D (SP-D) is one of the most studied collectins with a vital role in host defense against pathogens and allergens, and modulation of inflammatory response (6). Although SP-D was historically shown to be lung-resident being produced by type II alveolar and Clara cells (7), studies during the last decade have established its extrapulmonary existence in a range of tissues (8). SP-D has also been shown to be expressed in the male reproductive tracts of human and mice (9,10). Elevated levels of SP-D at inflamed sites in the prostate manifested protection against bacterial infection (11,12). Kankavi et al. (13) observed differential expression of SP-D in glandular structures of inflamed malignant and non-malignant human prostate tissues. There was a significant correlation between decreased levels of SP-D and increased Gleason score, a grading system based on the histologic pattern of arrangement of carcinoma cells, and tumor volume.
Previously, we reported a novel anti-cancer role of human SP-D and its recombinant fragment (rfhSP-D) comprising 8 Gly-X-Y repeats neck and CRD region, wherein they reduced the viability of a range of human cancer cell lines including eosinophilic leukemia cell line (AML14.3D10) (14). Importantly, survival of peripheral blood mononuclear cells (PBMCs) derived from healthy individuals was found to be unaltered (14). rfhSP-D treated AML14.3D10 cells showed a significant increase in apoptosis with reduced HMGA1 levels and increased levels of activated p53 and caspase 9 (14). SP-D has recently been shown to inhibit the proliferation, migration and invasion of A549 human lung adenocarcinoma cells by binding to N-glycans of epidermal growth factor receptor (EGFR) via its CRD region, and interfering with EGF signaling (15). In UV treated apoptotic Jurkat T cells, SP-D enhanced membrane and nuclear blebbing, suggesting involvement of SP-D in induction of apoptosis (16). Exogenous treatment of SKOV3 cells (an ovarian cancer cell line) with rfhSP-D led to increased caspase 3 cleavage and induction of pro-apoptotic genes Fas and TNF-α (17).
To take the next logical step from the reported anticancer role of SP-D in tumorigenic cell lines, we examined anti-prostate cancer role of rfhSP-D using tumor explants and primary cells derived from tissue biopsies of metastatic prostate cancer patients. rfhSP-D induced apoptosis selectively in various prostate cancer cells including the two prostate cancer cell lines (LNCaP and PC3) in a dose-and timedependent manner. Apoptotic signaling involved upregulation of p53 and downregulation of pAkt. Decreased levels of Bcl2, with a concomitant increase in Bax, cytochrome c and cleavage of caspase 7, confirmed rfhSP-D mediated induction of intrinsic apoptosis.

Ethics Statement
The study was approved by the Institutional Ethics Committee for Clinical Studies, ICMR-National Institute for Research in Reproductive Health; (Project No.: 260/2014) and Institutional Ethics Committee, Tata Memorial Hospital (Study no: 1467). Tissue biopsy samples were collected from 9 suspected metastatic prostate cancer patients (treatment naïve) undergoing Transrectal Ultrasound guided multiple core needle biopsy, with written informed consent. Chemotherapy was initiated in the patients after confirmation of metastasized prostate cancer. Information regarding the androgen dependency of the prostate cancer in these patients was not available. Average age of the study participants was 67.4 ± 3.97 years with Mean PSA (Prostate-specific antigen) level of 190.04 ± 85.12 ng/ml and Median Gleason score of 8 ± 0.83.

Explant and Cell Culture
Prostate tissue biopsies were collected in cold Phosphate Buffer saline containing 10% v/v fetal bovine serum (FBS; GIBCO) and 2% v/v Pen-Strep solution (GIBCO) and immediately processed in sterile conditions. Biopsies were cut into 5 mm explants, placed in the scratched areas of 35 mm tissue culture plate (Nunc) and cultured in RPMI 1640 (GIBCO) supplemented with 10% FBS, 1% Pen-Strep solution, Glucose (1 mg/ml) (HiMedia) and 1% Sodium Pyruvate (GIBCO) (complete RPMI) for a week. Primary Cancer Epithelial Cells (PrCEC) isolated from the explants were sub-cultured using 5 mM EDTA (Sigma) in PBS and passaged further. Isolated PrCEC (Passage 2) were examined for Cytokeratin (an epithelial cell marker) via immunofluorescence microscopy (Anti-cytokeratin antibody, Dako, 1:500) and by Real time RT-PCR analysis for CD164 expression, a marker for metastatic epithelial cell cancer (18). Isolated PrCEC were stained for the presence of Alpha-methylacyl-CoA racemase (AMACR) expression, which is upregulated in prostate cancer with highgrade prostatic intraepithelial neoplasia (HGPIN) than in the normal human prostate (19). PrEC, PrCEC and PC3 cells were stained using mouse monoclonal antibodies to AMACR (Novus Biologicals, 1:100) or matched-isotype control IgG for 1 h at 4 • C. Phycoerythrin-conjugated rat anti-mouse IgG (Molecular Probes, Eugene, USA) was used as secondary antibody. The gated cell population was determined using PE tagged anti-AMACR. Cells were analyzed via BD FACS Aria III (BD Biosciences, San Jose, California, USA) flow cytometer.

rfhSP-D Preparation
The recombinant fragment of human SP-D (rfhSP-D) was expressed in Escherichia coli BL21 (λDE3) pLysS (Invitrogen), purified and characterized, as described previously (15). Endotoxin level in the rfhSP-D preparation was determined using the QCL-1000 Limulus amebocyte lysate system (BioWhittaker Inc., USA). The assay was linear over a range of 0.1-1.0 EU/ml (10 EU = 1 ng of endotoxin) and the amount of endotoxin present in the preparations was found to be <4 pg/µg of rfhSP-D.
Interaction Between FITC Labeled rfhSP-D and Prostate Cells rfhSP-D was labeled with FITC dye (20) and incubated with prostate epithelial cells and prostate cancer cells at 5, 10, and 20 µg/ml concentration in staining buffer for 15, 30, 45, and 60 min at 4 • C in the presence of 2 mM CaCl 2 . Cells were washed to remove unbound rfhSP-D and fixed with 2% PFA for analysis via BD FACS Aria III (BD Biosciences, San Jose, California, USA). Data was analyzed using FCS Express 6 De Nova software. To assess the specificity of the interaction, PC3 cells were incubated with FITC labeled rfhSP-D in the presence of 5 mM CaCl 2 , or 5 mM EDTA, or 5 mM Glucose in PBS, pH 7.4. Staining buffer was used as control for these experiments.

Fluorescence Microscopy for Nuclear Morphology
PrEC, LNCaP, or PC3 cells (2 × 10 3 ) were grown on coverslips and incubated with rfhSP-D (20 µg/ml) for 48 h to analyze nuclear morphology following induction of apoptosis. Cells were fixed in 2% PFA and permeabilized using 1% v/v Triton X 100 (Sigma). Cells were incubated with Hoechst (1:10,000, Invitrogen) for 20 min in dark. Coverslips were mounted in vector shield (Vector laboratories, UK) and observed under confocal microscope (Zeiss, Germany).

TUNEL (Terminal Deoxynucleotidyl Transferase dUTP Nick end Labeling) Assay
Prostate tissue biopsies collected from metastatic prostate cancer patients were incubated with rfhSP-D (40 µg/ml) for 48 h in serum free RPMI medium at 37 • C under 5% v/v CO 2 . Five micrometer paraffin embedded sections of 10% NBF (neutralbuffered formalin) fixed prostate tissue biopsies were placed on poly L-lysine coated slides. The sections were fixed with chilled acetone followed by washing with PBS. Slides were incubated in TUNEL Mix (Roche Diagnostics), containing terminal deoxynucleotidyl transferase (TdT) and fluorescein labeled nucleotides for an hour in a moist chamber at 37 • C. The slides were washed and counterstained with DAPI (4 ′ ,6diamidino-2-phenylindole). TUNEL positive apoptotic cells (green-stained cells) were viewed under confocal microscope (Zeiss, Germany).

Annexin V Assay
For Annexin-V immunostaining, the manufacturer's protocol of Annexin V-FITC apoptosis detection kit (Calbiochem) was followed with some modifications. PrEC, LNCaP, or PC3 were treated with indicated concentrations of rfhSP-D and harvested at the end of 24 and 48 h. Cells were trypsinized and washed with ice cold PBS to remove culture supernatant, followed by incubation with FITC-tagged Annexin V for 20 min in dark. Subsequently, Annexin V was washed and 1 µl PI was added to stain the DNA. Cells were immediately analyzed via BD FACS Aria III.

Enzyme-Linked Immnosorbent Assay (ELISA) for SP-D and Cytochrome c
Cell lysates and culture supernatants were analyzed for SP-D levels (Duo Set Human SP-D, catlog no. DY1920, R & D Systems) and Cytochrome c (Human Cyt-C, catlog no. E1516Hu, BT Assay) using commercially available ELISA kits. Briefly, for quantification of cytochrome c, PrEC, LNCaP, PC3 cells, or Prostate tissue biopsies collected from metastatic prostate cancer patients were incubated with indicated concentrations of rfhSP-D for 48 h. Cell culture supernatant was analyzed for cytochrome c released by cells. Color development was stopped using 2N H 2 SO 4 and optical density was determined at 450 nm using a microplate reader (Beckman Coulter).

Real Time PCR
PrEC, PrCEC, PC3, and LNCaP cells (1 × 10 6 ) were plated in a six-well plate and total RNA was isolated using Trizol (Takara) that was further treated with DNase I (Thermo Scientific, Rockford, USA) at 37 • C for 30 min to eliminate genomic DNA contamination. One to two microgram of total RNA was reverse transcribed into cDNA using Superscript III first strand synthesis kit (Invitrogen, USA). One microliter of cDNA was used for real time PCR reactions using BioRad CFX96 TouchTM real-time PCR detection system and iQTM SYBR Green Supermix (Bio-Rad, Hercules, CA, USA). 18S was used as housekeeping control. Each qPCR experiment was performed in triplicates and each experiment was repeated 3 times. Primers were designed using NCBI Primer BLAST Software and their annealing temperatures and product sizes are mentioned in Table 1.

In silico Analysis of Human SP-D Promoter Region for Androgen Responsive Elements
In silico analysis was carried out to predict putative androgen responsive elements (ARE) in promoter regions of human SP-D gene using MatInspector Genomatix v3.4 Software, GmbH, Munchen (Germany). The transcription start site (TSS) for SP-D was determined from the Institute of Bioinformatics and Applied Biotechnology (IBAB) MGEx-Tdb database. Promoter regions −10,000 bp upstream and +1,000 bp downstream from the transcription start site (TSS sites) were submitted for analysis.

Statistical Analysis
GraphPad PRISM Software version 6.00 (GraphPad Software Inc., San Diego, CA) was used to plot the graphs and analyze the data using one-way ANOVA with Bonferroni corrections for comparison among prostate cells or unpaired two tailed Student's t-test for comparing the rfhSP-D treated groups with control. Data is represented as mean ± SD. Values of p < 0.05 were considered statistically significant.

Prostate Epithelial and Cancer Cells Expressed SP-D
Since it is known that prostate epithelial cells secrete SP-D (11,13), we evaluated SP-D expression and its regulation in PrEC, LNCaP, DU145, and PC3. Figure 1A shows that SP-D is expressed by prostate epithelial cells and prostate tumor cells; SP-D expression in LNCaP cells is significantly lower in comparison with DU145 (p < 0.05), PC3 (p < 0.05), and PrEC (p < 0.05) by western blot and ELISA (PrEC-1.48 ± 0.31 pg of SP-D/µg of total protein; LNCaP-1.26 ± 0.28 pg/µg; DU145-1.50 ± 0.35 pg/µg, data not shown). SP-D was detectable in the cell culture supernatants of various prostate cells (Range-150 ± 21 to 260 ± 35 pg/ml), suggesting that SP-D is secreted, though not sufficient enough to induce apoptosis (10-20 µg/ml) (14). It is reported that androgen induces proliferation of prostate epithelial cells and LNCaP cells via androgen receptors (22). Furthermore, SP-D expression in rodent prostate is altered on castration (11). Hence, we evaluated levels of SP-D transcripts in Dihydrotestosterone (DHT) treated PrEC, LNCaP, and PC3 cells. Treatment with DHT (10 nM) significantly upregulated SP-D transcripts in both PrEC and LNCaP cells by 1.9 fold but not in PC3 cells (androgen independent) ( Figure 1B). This suggests that SP-D expression is regulated by androgens (DHT) in androgen dependent cancer. Increased cell viability was observed in DHT treated PrEC (111.43 ± 4.69%) and LNCaP (119.23 ± 2.68%) cells as reported [ Figure 1B (ii)]. In silico analysis of promoter region of human SP-D gene elucidated 10 putative Androgen Responsive Elements (AREs) ( Table 2), suggesting that androgens may regulate SP-D expression. The functionality of identified ARE has not been evaluated in this study.

rfhSP-D Binds Differentially to Prostate Cancer Cells
FITC-labeled rfhSP-D showed a dose and time dependent binding to PrEC and prostate cancer cells (LNCaP, DU145, and PC3) (Figure 2). A significantly higher binding was observed with the androgen independent PC3 cells (MFI-743.86 ± 67.41) than the androgen dependent LNCaP cells (MFI-354.75 ± 54.11) (p < 0.05) (Figure 2A). PrEC (MFI-96.48 ± 21.07) showed comparatively less binding with the FITClabeled rfhSP-D than any of the cancer cells (PC3, LNCaP, p < 0.05). Binding of rfhSP-D to all cell types was calciumand carbohydrate-dependent that was inhibitable by EDTA and glucose (Figures 2B,C show representative binding to PC3 cells). DU145 cells showed results similar to that of PC3 cells (data not shown).

Anti-prostate Cancer Activity of rfhSP-D in PrCEC and Tumor Explants
We first carried out a pilot investigation on the effect of rfhSP-D on the primary cells derived from tissue biopsies of metastatic PCa patients (n = 9). Isolated PrCEC (Passage 2) showed positive staining for Cytokeratin ( Figure 3C). Expression of AMACR (which is upregulated in prostate cancer with high-grade prostatic intraepithelial neoplasia) in PrCEC was not significantly different than the PC3 cells whereas PrEC cells showed significantly decreased expression (Figure 3A). PrCEC and PC3 cells showed significantly upregulated expression of CD164 in comparison with PrEC ( Figure 3B). PrCEC cells isolated from the metastatic PCa patients showed significantly reduced viability in a dose-and time-dependent manner following treatment with rfhSP-D ( Figure 3D). Importantly, TUNEL assay confirmed increased apoptosis in rfhSP-D treated tissue biopsies of the same patients ( Figure 3E). Since these metastatic PCa patients were not evaluated for androgen-dependency, we could not assess if rfhSP-D induced cancer cell death would vary with androgen resistance. Therefore, we pursued further studies to delineate the molecular mechanisms for rfhSP-D mediated apoptosis in androgen dependent (LNCaP) and independent (PC3) cancer cell lines.

rfhSP-D Selectively Reduced the Viability of Prostate Cancer Cells
Reduction in the viability of PCa cells following rfhSP-D treatment was dose-and time-dependent,  irrespective of their androgen sensitivity whereas the viability of rfhSP-D treated PrEC was unaltered till 48 h ( Figure 4A). The half-maximal inhibitory concentration (IC 50 ) of rfhSP-D against the PCa cells is tabulated as Table 3.

rfhSP-D Induced Apoptosis in LNCaP and PC3 Cells
Annexin-V, a classical marker of the apoptosis, was evaluated by flow cytometry. A significant increase in the annexin-V positive cells was observed post 48 h in rfhSP-D treated LNCaP ( Figure 5A). rfhSP-D promoted karyorrhexis (fragmentation of nucleus) of LNCaP and PC3 cells at 48 h, as seen in Hoechst stained cells (Figure 5B).

rfhSP-D Triggered Intrinsic Mitochondrial Apoptosis in the Prostate Cancer Cells
p53 and PI3K/Akt are involved in several cellular physiological processes including apoptosis and cell proliferation. Our previous study had shown involvement of p53 in the rfhSP-D mediated apoptosis of AML14.3D10 cells (12). Here, rfhSP-D treated LNCaP cells with a wild type p53, showed significant upregulation of p53 ( Figure 6A). To understand the other likely mechanisms of apoptosis, p53 null PC3 cells were examined ( Figure 6B) for differential expression of several kinases (MAPK, JAK, Stat-1, ERK, and Akt). rfhSP-D significantly decreased the phospho-active forms of Akt in PC3 cells at 15 min ( Figure 6B iii and iv). Interestingly, treated LNCaP cells also showed significantly decreased pAkt at 30 min ( Figure 6B i and ii). Phosphorylated Akt leads to increased phosphorylation of Bad, an inactive form of Bad. The rfhSP-D induced decrease in pAkt sequentially led to a significant decrease in phosphorylated Bad in both the cell lines ( Figure 7A). rfhSP-D treatment further led to a significantly decreased level of Bcl2 and an increased level of Bax at 24 h ( Figure 7B). Both LNCaP and PC3 cells showed a significant increase in Bax to Bcl-2 ratio (p < 0.05) upon  (Figure 7D). Cleaved products of the executioner caspase 7 were significantly increased after 24 h following rfhSP-D treatment (Figure 7C). From these observations, we infer that rfhSP-D triggers intrinsic mitochondrial pathway of apoptosis in prostate cancer cells.

DISCUSSION
The present study established the anti-prostate cancer activity of rfhSP-D via induction of intrinsic apoptosis in explants and primary tumor cells isolated from tissue biopsies of metastatic prostate cancer patients and prostate cancer cell lines: LNCaP (androgen responsive) and PC3 (androgen resistant). Various attributes of apoptosis like phosphatidylserine externalization (Figure 5), mitochondrial dysfunction (Figure 7), and DNA fragmentation ( Figure 5) and various apoptotic markers (Figure 7) were observed in the rfhSP-D treated LNCaP and PC3 cells. Viability of the normal prostate epithelial cells (PrEC) was not altered in presence of rfhSP-D (Figure 1).  Human prostate cancer tissues frequently exhibit inactivation of the tumor suppressor gene p53 which is associated with therapeutic resistance (23). The p53 pathway plays a crucial role in the transmission of pro-apoptotic signals (24). Various therapeutic agents/candidates induce apoptosis of LNCaP (p53 +/+ ) cells by increasing the levels of phosphorylated p53 (25)(26)(27). Knockdown of p53 resulted in blockade of docetaxel induced apoptotic cell death in prostate cancer cells (28). Upregulation of activated p53 levels in a timedependent manner by rfhSP-D treatment suggested involvement of the p53 pathway in the induction of apoptosis of LNCaP cells. Previously, activation of p53 was also observed in the eosinophilic leukemic cells (AML14.3D10) undergoing apoptosis upon treatment with rfhSP-D (14,29). rfhSP-D treated AML14.3D10 cells showed significantly reduced levels of HMGA1, a survival protein (14). PC3 cells, a p53 null and highly metastatic prostate cancer cell line, also showed significant apoptosis following treatment with rfhSP-D, which suggested involvement of a p53 independent mechanism of apoptosis. Among 25% prostate cancer cases, diallelic deletion of the Phosphatase and tensin homolog (PTEN) gene and the associated increase in Akt phosphorylation correlates with hormone refractory prostate cancer (30). Decreased levels of activated Akt may lead to decreased levels of phosphorylated Bad (Bcl-2 associated death promoter). Dephosphorylated Bad interferes with interaction of activated Bcl2 with Bax. Thus, increased release of Bax triggers apoptosis (31,32). Decreased ratio of transcripts of Bcl-2 to Bax has been associated with cell death following an apoptotic stimulus (33,34). Both LNCaP and PC3 cells, upon treatment with rfhSP-D, showed a significant increase in Bax to Bcl-2 ratio (p < 0.05), suggesting that besides activation of p53 pathway, rfhSP-D also inhibited Akt-PI3K pathway leading to induction of Bax mediated apoptosis. The present study unravels PI3K/Akt, an anti-apoptotic pathway, as a novel target of rfhSP-D mediated anti-prostate cancer activity.
Mitochondria plays a central role in the initiation of intrinsic apoptosis pathway (35). rfhSP-D treatment mediated disruption of the mitochondria was elucidated by its reduced ability to oxidize methyl tetrazolium to form formazan crystals (MTT assay; Figure 4). Breakdown of mitochondria is followed by the release of cytochrome c (Figure 7), a key initial step in the irreversible apoptotic process (35,36). Culture supernatants from the rfhSP-D treated prostate cancer cells and cancer tissue biopsies showed significantly elevated levels of cytochrome c, confirming induction of intrinsic apoptosis. Once in the cytosol, the cytochrome c interacts with its adaptor molecule, Apaf-1, resulting in the processing and activation of pro-caspase-9 (37). Caspase-9, in turn, cleaves and activates pro-caspases−3 and−7, effector caspases responsible for the cleavage of various proteins leading to the biochemical and morphological features observed in apoptosis (38). Treatment of the prostate cancer cells with rfhSP-D resulted in the cleavage and activation of the effector caspase-7, thus leading to programmed cell death. Nuclear condensation, membrane and nuclear blebbing and flipping of phosphatidylserine (PS) on the cell membranes, the typical morphological characteristics of apoptotic cells, were observed in the rfhSP-D treated prostate cancer cells, confirming apoptotic cell death. The proposed mechanisms for the rfhSP-D mediated apoptosis in prostate cancer cells as revealed in this study along with some key molecules of these pathways interacting with/ induced by SP-D as reported previously have been depicted in Figure 7E.
Recent studies highlighted the suppressive role of SP-D in extrinsic apoptosis wherein native purified human SP-D interacted with Jurkat T cells and delayed the progression of Fas (CD95)-Fas ligand and TRAIL-TRAIL receptor induced, but not TNF-TNF receptor-mediated apoptosis (39). In a subsequent study from the same group in UV irradiated Jurkat T cells, SP-D reduced the activation of caspase-8, executioner caspase-3 and exposure of phosphatidylserine (PS) on the membranes of dying cells, with a concomitant increase in the formation of nuclear and membrane blebs (16). The involvement of rfhSP-D in extrinsic apoptotic pathway in prostate cancer cells needs to be explored further.
rfhSP-D induced a cell cycle arrest in G2/M and G0/G1 phases of LNCaP and PC3 cells, respectively. We had previously shown that the rfhSP-D induced elevated p21 expression and inhibited Cdc2 phosphorylation resulting in reduced activity of Cdc2cyclin B1 thus, leading to G2/M arrest in the AML14.3D10 cells (14). Several genes are common to the pathways involved in cell cycle regulation and apoptosis (40). The p53 protein plays a critical role both in the G1/S and G2/M checkpoint while Retinoblastoma (Rb) protein is a potent inhibitor of the G1 to S phase transition in the cell cycle (41,42). Following rfhSP-D treatment, PC3 cells, which are null p53 but express the wild type Rb gene, showed an arrest in the G1-phase with significant reduction in the S-phase cell population. This alluded to the involvement of a p53 independent mechanism in the rfhSP-D induced apoptosis in PC3 cells.
Increased interaction of rfhSP-D with the prostate cancer cells as compared to PrEC, could be due to differential expression of the receptors. Similar to our previous study with eosinophilic leukemic cells, rfhSP-D showed interaction with the prostate cancer cells via carbohydrate recognition domain (CRD) (14). CRD domain of SP-D is known to interact with CD14, TLR-2, TLR-4, EGFR, and SIRPα that are reported to be present on prostate cancer cells and normal prostate tissues (15,43,44). Binding of SP-D with pattern recognition receptors CD14, TLR-2, and TLR-4 may lead to blockade of their downstream proinflammatory and pro-survival signaling ( Figure 7E). Interaction with SIRPα, involved in the negative regulation of receptor tyrosine kinase-coupled signaling processes, may result in recruitment of PI3K, leading to a reduction in the activity of the downstream kinase Akt (45) (Figure 7E). A recent study showed that SP-D reduced EGF -EGFR binding through the interaction between CRD of SP-D and N-glycans of EGFR, thus, leading to downregulation of EGF signaling in the A549 human lung adenocarcinoma cells (15) (Figure 7E). Further studies are needed to elucidate the involvement of CRD and its interacting partners in downstream signaling of rfhSP-D in the prostate cancer cells.
Oncogenic mutations disturb the normal cellular functions, thus, allowing the tumor cells to undergo dysregulated proliferation, resist pro-apoptotic insults, invade normal tissues, and most importantly, escape apoptosis. Therefore, induction of apoptosis in the malignant tumors has evolved as a successful adjunct anti-cancer strategy. We report that rfhSP-D selectively triggered intrinsic apoptosis in androgen-dependent as well as androgen-independent prostate cancer cells, without affecting normal epithelial cells (PrEC). rfhSP-D also induced apoptotic cell death in the tissue biopsies from metastatic prostate cancer patients. Thus, SP-D is likely to act as an integral component of the human innate immune surveillance against cancer cells. A great advantage associated with the anti-cancer activity of rfhSP-D is induction of apoptosis by simultaneous targeting of multiple cellular signaling pathways including transcription factors, tumor cell survival factors, protein kinases resulting in the efficient and selective killing of prostate cancer cells. Murine models of prostate cancer including the patient-derived xenograft (PDX) mouse models that mimic human disease and 3D cell cultures derived from PDX are going to be valuable tools for the evaluation of therapeutic strategies using rfhSP-D (46).

DATA AVAILABILITY
The raw data supporting the conclusions of this manuscript will be made available by the authors, without undue reservation, to any qualified researcher.

ETHICS STATEMENT
The study was approved by the Institutional Ethics Committee for Clinical Studies, ICMR-National Institute for Research in Reproductive Health (Project No.: 260/2014) and Tata Memorial Hospital (Study no: 1467). Tissue biopsy samples were collected from 9 metastatic prostate cancer patients (treatment naïve) using Trans-rectal Ultrasound guided multiple core needle, with written informed consent. Chemotherapy was started in the case of patients following confirmation of metastasized prostate cancer. Information regarding the androgen dependency of the prostate cancer in these patients was not available. Average age of the study participants was 67.4 ± 3.97 years with Mean PSA (Prostate-specific antigen) level of 190.04 ± 85.12 ng/ml and Median Gleason score of 8 ± 0.83.