Background
Colorectal cancer is one of the deadliest cancers, accounting for approximately 10% of cancer-related deaths [
1]. While the survival rates of colorectal cancer patients with localized and regional tumors are almost 89 and 79%, respectively, those with distant metastatic tumors are lower than 15%, which leads to the high mortality of this cancer type [
2]. Treatment options for metastatic colorectal cancer are limited. While local therapies such as surgery of metastatic sites or radiofrequency ablation have mitigated the morbidity of patients, in many cases including metastasis with multiple sites in the body, the patients are not eligible for these options and are thus treated with systemic therapy [
1]. For the last two decades, extensive efforts to develop better therapeutic strategies have increased the overall survival of metastatic cancer patients from 12 to nearly 30 months [
3]. However, the short duration of overall survival, 30 months, still demands a better understanding of metastatic colorectal cancer to develop more efficacious diagnostic/prognostic factors and thus further improve patient prognosis.
LAD1 (ladinin-1) is a protein originally known as a collagenous anchoring filament protein of basement membranes in mammalian epidermal cells [
4,
5]. However, a more recent study found cytosolic localization of LAD1 in mammary epithelial cells [
6]. Moreover, LAD1 has been implicated in the progression of different cancers. Comparative proteomic analyses found that LAD1 is abundant in lung adenocarcinoma but not in normal lung tissue and benign lung nodules [
7]. Proteomic profiling of laryngeal cancer tissues showed that LAD1 protein is enriched specifically in metastatic tissues but not in paired adjacent normal tissues and primary tumor tissues [
8]. Moreover, analysis in METABRIC, a large clinical dataset of approximately 2000 breast cancer patients [
9], showed that a high abundance of LAD1 transcripts is associated with poor prognosis in breast cancer patients [
6]. A study of mouse thyroid cancer models revealed a molecular link between LAD1 expression and oncogenic signaling pathways in cancer by showing that the BRAFV600E mutation induced the expression of LAD1 transcripts [
10]. Similarly, compared with normal tissues, the expression of the LAD1 transcript was elevated in human thyroid cancers carrying oncogenic mutations in genes such as BRAF, RET and RAS [
10]. This finding suggests that different oncogenic signaling pathways upregulate LAD1 expression. Collectively, these studies demonstrate that aggressive progression of various cancers is accompanied by upregulation of LAD1 expression. Nevertheless, the molecular function of LAD1 in the progression of these cancers remains unclear. In addition, studies assessing the involvement of LAD1 in colorectal cancer progression are lacking.
In this study, we aimed to explore the expression of LAD1 and its functional involvement in colorectal cancer progression by using cell-based systems and a xenograft mouse model. Furthermore, we assessed the clinical relevance of LAD1 expression by examining public databases and patient tissue arrays.
Methods
Survival analyses
Survival analyses of data extracted from GSE14333 and GSE24549 were performed by survival analysis in the R package. Patients were stratified into two groups: high LAD1 expression (high) and low LAD1 expression (low) based on the median expression value. The log-rank test was used to compare the survival rates, and Cox proportional hazard regression analysis was carried out to determine the hazard ratios.
Gene ontology analyses
To identify biological processes and pathways related to LAD1 expression in colorectal cancer, we performed Gene Ontology analyses with a list of genes that positively correlated with LAD1 expression. Positively correlated genes to LAD1 (R > 0.4, p-value< 0.001; R: correlation coefficient) were extracted from GSE14333 and GSE24549 using R packages. Finally, Gene Ontology analyses were performed with DAVID tools (DAVID Bioinformatics Resources 6.8).
Cell cultures
The human colorectal cancer cell lines SW480, SW620, Caco-2, HT-29 and DLD-1 were purchased from ATCC (American Type Culture Collection) and maintained in DMEM (Welgene, LM011–05) and RPMI (Welgene, LM011–01) supplemented with 10% FBS (Gibco, 16000–044) and 1% antibiotics (Gibco, 15240–062). All of the indicated cell lines were cultured at 37 °C in a humidified atmosphere with 5% CO
2 (Heraeus BB15 CO2 Incubator, Thermo Fisher). All the cell lines used in this study were authenticated and validated using STR genotyping performed by HPBIO, Inc. (Seoul, South Korea). Results of STR genotyping were provided in supplemental Table
3.
Antibodies
The following antibodies were purchased from commercial suppliers: anti-LAD1 (Novus Biologicals, NBP1–86126), anti-E-cadherin (Santa Cruz, sc-7870), anti-N-cadherin (Santa Cruz, sc-7939), anti-vimentin (Santa Cruz, sc-32,322), β-actin (Santa Cruz, sc-47,778), α-tubulin (Santa Cruz, sc-23,948), anti-mouse and rabbit secondary peroxidase-conjugated antibodies (Millipore, AP124P and AP132P), DAPI (Invitrogen, 62,248), Alexa594-conjugated phalloidin (Invitrogen, A12381), anti-rabbit secondary antibody Alexa488 conjugate (Invitrogen, A-11034) and anti-Ki67 (Abcam, ab16667).
RNA interference
siRNAs against LAD1 (si-LAD1#1-sense: 5′-CAGUGAAGUUGGGAGAGAA-3′; si-LAD1#2-sense: 5′-CAGACAACACAGUGAAGUU-3′) and negative control siRNA (sense: 5′-CCUACGCCACCAAUUUCGU-3′, Bioneer, SN-1003) were purchased from Bioneer. Human colorectal cancer cell lines were seeded in 6-well plates at 3X105 cells per well, and then, 25 nM siRNAs were transfected for 48 h using RNAiMAX reagents (Invitrogen,13,778–150).
Plasmid transfection
For transient plasmid transfection, SW480 cells were seeded at 2.5X105 cells in 6-well plates. Next day, pCMV6-XL5-LAD1 (ORIGEN, LAD1 Human Untagged Clone, SC116651) was transfected with lipofectamine 3000 (Invitrogen, L3000–015) for 48 h. After transfection, cells that overexpressed LAD1 were used for Transwell assay.
RNA extraction and reverse transcription (RT)-PCR analysis
Total RNA was extracted using TRIzol (Ambion, TRI Reagent Solution). cDNA was synthesized using the extracted total RNA, oligo dT (Bioneer, N-7503), dNTP mixture (Enzynomics, N002) and reverse transcriptase (Invitrogen, RevertAid reverse transcriptase, EP0442). RT-PCR and quantitative Real-Time PCR (qRT-PCR) were performed to analyze gene expression using synthesized cDNA and gene-specific primers (Bioneer). For RT-PCR, cDNA was amplified with gene-specific primers and PCR master mixture (Bioneer, AccuPower PCR Master Mix) by PCR machinery (MiniAmp Plus Thermal Cycler, Thermo fisher). The PCR products were analyzed via electrophoresis on agarose gel. qRT-PCR was carried out using a CFX96 Real-Time PCR system (Bio-Rad). qPCR reactions were prepared in mixtures with cDNA template, 300 nM forward and reverse primers and ready-to-use reaction master mix (Bio-Rad, iQ SYBR Green Supermix). The conditions of thermal cycling were as follows: initial incubation at 95 °C for 5 min; 45 cycles of 95 °C for 10s, annealing at 60 °C for 30s and extension at 72 °C for 30s. mRNA gene expression was normalized to β-Actin (ACTB) mRNA. Supplemental Table
1 provides the primer sequences used for RT-PCR.
Protein extraction and western blot
Whole-cell lysates were prepared by cold RIPA buffer (50 mM Tris-HCl, pH 8.0, 150 mM NaCl, 1% Triton X-100, 0.5% sodium deoxycholate, and 0.1% SDS). Proteins were quantified with the Bradford assay, and equal amounts of proteins were used for each sample. To resolve by SDS-PAGE, 5X SDS sample buffer (250 mM Tris-HCl, pH 6.8, 10% SDS, 50% glycerol, 0.5 M dithiothreitol, and 0.25% bromophenol) was added to cell lysates and boiled at 100 °C for 10 min. After SDS-PAGE, proteins were transferred to nitrocellulose membranes (PALL, Biotrace™NT), and membranes were blocked using 5% skim milk in TBST buffer. Primary antibodies incubated with membranes and HRP-conjugated secondary antibody were used to detect chemiluminescent signals from specific target proteins. The immunoblotting signals were visualized by using LAS3000 (Luminescent Image analyzer, FUGIFILM).
Cell viability assay
For the cell viability assay, LAD1 knockdown cells (2X103 cells per well) were seeded into black 96-well plates (Corning, 3603). After 72 h, 20 μl of CellTiter-Blue Reagent (Promega, CellTiter-Blue® Cell viability assay) was added to each well and incubated for 2 h. The fluorescence signal at Ex/Em = 530/590 nm was detected by a fluorometer (SYNERGY/HTS, BioTek).
Transwell assay
Transwell inserts (Corning, 3422) were precoated with human collagen-type IV for 1 h. Specifically, for the invasion assay, the upper chamber was coated with Matrigel (Corning, 354,234) for 1 h. Cells were suspended in serum-free DMEM, and 5X10
4–1X10
5 cells were seeded in each well containing DMEM with 10% FBS as previously described [
11]. After incubation for 24 h, the migrated or invaded cells were stained with 0.5% crystal violet solution. Cells that had migrated or invaded were counted manually (200x magnification field, 5 field per replicate).
A QCM™ gelatin invadopodia assay (EMD Millipore Corp, ECM670) was used to evaluate invadopodia formation in SW620 cells transiently knocked down with siRNA. In detail, poly-L-lysine was diluted in distilled water to coat 8-well glass chamber slides (Thermo Scientific, Lab-Tek chamber slide, 177,402). After a PBS wash, glutaraldehyde was added to each well for 15 min to activate the poly-L-lysine surface. Fluorescently labeled gelatin was coated onto each well for 10 min and disinfected with 70% ethanol. Cells at approximately 50% density (2X104 cells per the condition) were seeded in each well and incubated in previously indicated cell culture conditions for 48 h before taking fluorescent images with a fluorescence microscope (Inverted Microscope Eclipse Ti-S, Nikon).
Immunofluorescence
Immunofluorescence staining was performed as described previously [
12]. Briefly, cells were seeded at 1X10
4 cells on precoated glass plates (Ibidi, 89,626) and fixed in 4% PFA and permeabilized using 0.3% Triton X-100 in PBS for 10 min. After blocking with 1% bovine serum albumin in PBST, the primary antibody specific for LAD1 was incubated, and then Alexa 488-conjugated goat anti-rabbit antibody was incubated for 1 h. Finally, Alexa594-conjugated phalloidin and DAPI were used to stain actin and the nucleus. Fluorescent images were analyzed by a fluorescence microscope (Inverted Microscope Eclipse Ti-S, Nikon).
Lentivirus production and constructing stable cell lines
For in vivo experiment, SW620 cells stably expressing shSCR and shLAD1 RNAs were constructed as described below. Lentivirus expressing shSCR and shLAD1 RNAs were prepared from 293 T cells by co-transfection with shRNA plasmid (shSCR: SHC002, MISSION pLKO.1-puro, CCGGCAACAAGATGAAGAGCACCAACTCGAGTTGGTGCTCTTCATCTTGTTGTTTTT; shLAD1–1: TRCN0000077998, CCGGGTGTGCAACCACTTACCCTTTCTCGAGAAAGGGTAAGTGGTTGCACACTTTTTG), psPAX2 and pMD2.G using lipofectamine reagents (Invitrogen, 18,324–012). After 48 h, lentiviral-supernatant was harvested and filtered with 0.45uM syringe filters (ADVANTEC, 25CS045AS). SW620 cells were infected with each lentivirus for 12 h and stabilized for another 1 day, and then puromycin selection was conducted for 72 h as described previously [
11].
Female CAnN.Cg-Foxn1nu/CrlOri (BALB/C-nu) mice (4 weeks old, 12-14 g) were purchased from OrientBio (Seongnam-si, Gyeonggi-do, South Korea). Mice were housed (specific pathogen free, individually ventilated cages with ad libitum feeding, wood bedding, 3 mice/cage) for 4 weeks prior to intrasplenic injection of SW620 cells expressing shSCR or shLAD1 RNAs. Cages with bedding, rodent diet (Harlan Teklad, 2018, sterilized) and autoclaved water were replaced twice a week during a whole period of animal experiment. Daily light and dark cycles (12-h light and 12-h darkness) were maintained with constant physiological conditions (temperature, 20 and 26 °C; humidity, 40–60%). Mice were randomly assigned into two groups before intrasplenic injection. Inhaled anesthesia was performed with 2% isoflurane with oxygen. A total of 2X106 cells were resuspended in 50 μl of HBSS media and injected into the spleen via a 30-gauge needle. Two minutes after the injection, the splenic vessels were tied off with a suture. After surgery, mice were recovered under heating lamp for a while. Health status (body weight; recorded once a week, body temperature, activity, and behavior during handling) of mice were daily monitored until termination of the experiment. Mice were sacrificed by cervical dislocation 7 weeks after intrasplenic injection, and the liver was excised (n = 3 per group). The liver was examined by the naked eye to count metastatic nodules, and liver lobes were fixed in 10% formalin. This animal study was approved by Korea Research Institute of Bioscience & Biotechnology-IACUC (Daejeon, South Korea).
Immunohistochemistry
Paraffin-embedded tumor-bearing liver sections were stained with hematoxylin and eosin (H&E) and an anti-Ki67 antibody [
11‐
13]. Histology of liver-containing metastatic nodules was analyzed by H&E staining, and proliferation of tumors in each sample was analyzed by labeling with the anti-Ki67 antibody. Slide sections were deparaffinized by xylene and rehydrated with ethanol. Antigen retrieval was performed with sodium citrate buffer, pH 6.0, as recommended in the datasheet of the primary antibody. After the blocking step with goat serum in PBS, the primary antibody was incubated overnight at 4 °C. Endogenous peroxidase was blocked by 3% H
2O
2 incubation, and then slides were incubated with biotinylated goat anti-rabbit antibody. For the visualization of staining, DAB (Vector Laboratories, Inc., SK-4100) was used as a chromogen. Finally, slides were counterstained with hematoxylin.
Human colon cancer tissue array staining
A human colon cancer tissue array slide (CDA3-G) was purchased from SuperBioChips (Seoul, South Korea). The tissue array slide contained 55 samples of colon cancer and normal tissue specimens. In addition, 9 samples were matched primary and metastasis tissues from different organs. Supplemental Table
2 included details of metastasis tissues. Immunohistochemistry of the tissue array slide was performed as previously indicated. The staining results of the tissue array were empirically graded by a pathologist.
Statistical analysis
All data are presented as means of three independent experiments ± standard deviation. GraphPad Prism 7 software (GraphPad Software, Inc.) were used for statistical analyses. Student’s t-tests (two-tailed, unpaired) for two group comparisons and one-way ANOVA followed by Tukey’s post hoc test for more than two group comparisons were used to determine significant differences (P < 0.05).
Discussion
The low survival rate of patients with regional and distant metastatic tumors compared with those with localized diseases raises the need for more extensive efforts to improve the outcomes of patients with metastasis [
2]. Biomarkers related to metastatic progression facilitate efficacious clinical approaches to overcome the poor prognosis of metastatic cancer. Metastasis is a complex process involving multiple steps, such as invasion of tumor cells into the tissue surrounding the primary tumor, entry and extravasation through the bloodstream and colonization at secondary sites [
2]. Tumorigenic factors in tumor cells whose functions are involved in metastatic pathways have the potential to be metastatic biomarkers. In this study, we suggest LAD1 as a positive regulator of the metastatic progression of colorectal cancer. Our findings show that (i) the expression of LAD1 transcript correlates with poor prognosis of colorectal cancer patients, (ii) depletion of LAD1 impairs the metastatic potential of colorectal cancer cells in vitro and in vivo, and (iii) elevated expression of LAD1 protein in tissues is associated with the metastatic stage of colorectal cancer.
LAD1 was originally proposed as a component of the basement membrane in mice [
4]. More recently, an extensive characterization of LAD1 in human mammary epithelial cells showed that most LAD1 protein was localized in the cytosol [
6]. Furthermore, the latter revealed colocalization of LAD1 with actin filament, filamins, and actin-cross-linking cytoskeletal proteins. The same study also reported that LAD1 physically interacted with stratifin (14–3-3σ), which promoted breast tumor invasion by regulating actin filament turnover [
20]. We also observed intracellular localization of LAD1 partly together with filamentous actin. However, how LAD1 promotes the migration and invasion of cancer cells remains obscure. LAD1 KD failed to change the expression of genes involved in metastatic pathways such as EMT (E-cadherin and N-cadherin) [
17] and matrix degradation (MMPs) [
18]. In addition, LAD1-depleted cells displayed no prominent alteration in the expression of integrins that modulate signaling pathways in cancer cells [
19]. Nonetheless, LAD1 loss mediated defects in invadopodia activity to degrade the extracellular matrix [
21] and consequently impaired the invasion and migration of cancer cells. Considering these results, we speculate that LAD1 is unlikely to be involved in gene expression for cancer cell motility. Alternatively, based on its cellular localization, we hypothesize that LAD1 may modulate the reorganization of the cytoskeleton possibly through interaction with actin filaments and thus promote the invasive movement of cancer cells. Extensive interactome studies could help further understand the molecular mechanism of LAD1 in metastatic cancer. Although understanding the exact function of LAD1 requires additional studies in future, our findings in this study suggest that LAD1 has a potential being a prognostic biomarker for metastatic colorectal cancer.
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.