Mice
Male or female BALB/c nude mice were purchased from Shanghai Lingchang Biotechnology Co., Ltd (Shanghai, China). All animals were housed and maintained in pathogen-free conditions. All animal experiments were performed in compliance with the Guide for the Care and Use of Laboratory Animals and approved by the Institutional Biomedical Research Ethics Committee of the ShanghaiTech University or Shanghai Jiao Tong University School of Medicine (SJTU-SM).
Cell Culture
HCT116, RKO, COLO205, LOVO, SW480, SW620, HT29 cells were cultured according to the culture methods of ATCC.
Cell Cycle
Cells within 2–5×106 were collected, then wash the cells twice with 1 ml cold PBS, each time centrifuged at 400 g for 5 minutes at 4℃. Discard most of the supernatant and gently bounce the cell sediment evenly, so that the volume of cell suspension is about 50-100ul. Use a whirlpool oscillator to keep the cells in suspension, slowly drop by drop add 1 ml of ice-cold 75% ethanol (pre-cooled at -20℃). Fixed overnight at 4℃ or for 4 hours at -20℃. Cells were collected by centrifugation at 1000 g for 5 min at 4℃ before analysis. Wash cells once with 1 ml cold PBS, resuspend cells with 400 µl cold PBS. Add 400 µl PI Staining Solution/RNase solution, gently mix and incubate for 30–60 minutes in the dark at 4℃ then analyzed by flow cytometry. The maximum excitation wavelength is 488 nm.
Cell Apoptosis
HCT116 or RKO cells were stained with the FITC Annexin V apoptosis detection kit (556547, BD Biosciences) or PE Annexin V apoptosis detection Kit (559763, BD Biosciences) according to the manufacturer protocol and analyzed early- and late-stage apoptosis by FACS (FACS AriaTM IIII, BD Biosciences).
Cell Growth Curve and Colony Formation
For growth curve assay, a total of 1×104 HCT116 or RKO cells per well were seeded in 12-well plate, triplicate wells were seeded. Count cells every 24 or 48 hours for a total of 5–8 days. For colony formation assay, a total of 1000 HCT116 or RKO cells per well were seeded in 6-well plate, triplicate wells were seeded. After 10 days, cells were stained with Crystal Violet Staining Solution (C0121, Beyotime).
RNA Extraction and Real-time PCR
RNA extraction and real-time PCR analysis were performed according to previous protocol 56. The mRNA and miRNA levels were assessed by SYBR Green-based quantitative real-time PCR with gene-specific primers (Supplementary Table S1).
Western Blot
Western blot analysis was performed according to our previous method 57. The following primary antibodies were used: PUM1 (ab92545), PUM2 (ab92930), AGO1(ab5070), AGO2 (ab57113), Lamin A/C (ab108595) were purchased from Abcam. p21 (2947S), p27 (3686S), Tubulin (2125S), GAPDH (2118S) were purchased from Cell Signaling Technology.
RNA Immunoprecipitation (RIP)
After the cells grow to 90% density in a 10 cm dish, the medium is removed, wash twice with PBS, add 1 ml of cell lysate and lyse on ice for 30 min. Cells and lysate were collected with a cell scraper and centrifuged in 1.5 ml centrifuge tube at 120000 rpm for 15 min. Take the supernatant, reserve 50 µl as input, and divide the cell lysate into two parts. Add 2 µg PUMILIO antibody (Abcam) and 2 µg IgG, respectively, and incubate overnight at 4℃. Protein A magnetic beads were washed twice with cell lysate, and 50 µl of the washed magnetic beads were added to 1.5 ml centrifugal tube incubated overnight, and incubated at 4℃ for 4 hours. The incubated magnetic beads were washed five times with a magnetic frame and NT2 buffer for 5 min each time. Remove supernatant, collect magnetic beads, add TRIzol reagent (Invitrogen) and extract RNA.
Photoactivatable Ribonucleoside-Enhanced Crosslinking and Immunoprecipitation (PAR-CLIP)
PAR-CLIP was performed following the previous protocol 18. For each condition (namely, WT1-Pum1-PAR-CLIP, WT2-Pum1-PAR-CLIP, KO1-Pum1-PAR-CLIP, KO2-Pum1-PAR-CLIP), we grew 15 cm cell culture plates of HCT116 cells in medium supplemented with 100µM 4-thiouridine(4SU) for 14 hours. Cells were washed once with cold PBS and then cross-linked on ice using 0.15 J/cm2 of 365-nm ultraviolet light in a Stratalinker. Cells were then scraped from culture dishes, transferred into a 15-ml Falcon tube, washed once with PBS, pelleted by centrifugation at 500 g for 5 minutes and flash-frozen in liquid nitrogen for storage at -80℃.
On the day of the experiment, cell pellets were thawed on ice, suspended in 3 volumes (the volumes of cell pellets were about 2 ml, hence 6 ml of lysis buffer were added) of lysis buffer (20 mM Tris-HCl pH = 7.4, 150 mM NaCl, 1 mM EDTA, 1% (V/V) IGEPAL® CA-630, 0.5 mM DTT, cOmplete™ EDTA-free Protease Inhibitor Cocktail Tablets, and PhosStop Tablets) and incubated on ice for 30 minutes. Cell lysate was collected by centrifugation at 13,000 g at 4℃ for 30 minutes and further filtered by 0.45 µM pore size membrane filter (Millipore, Cat#: HAWP04700). Partial RNA digestion was performed before immunoprecipitation with 1 U/µl RNase T1 (Thermo Fisher Scientific, EN0541) at 22℃ for 15 minutes. In the meantime, the Protein A beads-antibody complexes were prepared, for each sample, 0.25 mg/ml Pum1 antibody (Abcam, Cat#: ab92545) was conjugated to 100 µl protein A beads at 4℃ for 2 hours, unbound antibody was removed and added the pre-cleared lysates to antibody-coupled beads and incubated overnight at 4℃.
The following day, wash the beads five times with wash buffer [20 mM Tris-HCl pH = 7.4, 200 mM NaCl, 0.05% (v/v) IGEPAL® CA-630, 0.5 mM DTT, cOmpleteTM EDTA-free Protease Inhibitor Cocktail Tablets, and PhosStop Tablets) and digest with 10U/µl RNase T1 at 22℃ for 15 minutes, then wash the beads three times in high salt wash buffer (20 mM Tris-HCl pH = 7.4, 300 mM NaCl, 0.05% (v/v) IGEPAL® CA-630, 0.5 mM DTT, cOmpleteTM EDTA-free Protease Inhibitor Cocktail Tablets, and PhosStop Tablets). Immunopurified protein–RNA complexes were dephosphorylated by resuspending beads in 200 µl dephosphorylation mix (1 × NEBuffer 3 (New England Biolabs, Cat#: B7003S), 0.5 U/µl calf intestinal alkaline phosphatase (Thermo Fisher Scientific, Catalog # M0290)) and incubated at 37℃ for 10 minutes. Then, wash the beads two times with 1 × NEBuffer 3 and two times in PNK buffer without DTT (50 mM Tris-HCl pH = 7.4, 50 mM NaCl, 10 mM MgCl2·H2O). Next, 3′ ligation was performed according to the NEBNext Multiplex Small RNA Library Prep Set for Illumina (Set 1) (New England Biolabs, Cat#: E7300S).
The next day, wash the beads two times with high salt wash buffer followed by two times with PNK buffer without DTT. Then the protein–RNA complexes were phosphorylated by resuspending beads in 100 µl phosphorylation mix (1 × T4 DNA ligase buffer (Thermo Fisher Scientific, Cat#: B69), 1 U/µl T4 Polynucleotide kinase (Thermo Fisher Scientific, Cat#: EK0031)) and incubated at 37℃ for 30 minutes. Then wash the beads three times with wash buffer. Subsequently, NuPAGE LDS Sample Buffer (Thermo Fisher Scientific, Cat#: NP0007) was added to each sample and incubated at 95℃ for 5 minutes. Protein–RNA complexes were resolved using NuPAGE 4–12% Bis-Tris-HCl Gels (Thermo Fisher Scientific, Cat#: NP0335BOX) and desired complexes were excised from gel using a clean scalpel. The gel pieces were transferred to a D-tube Dialyzer Midi tube (Millipore, Cat#: 71506) and 800 µl 1 × SDS buffer (Thermo Fisher Scientific, Cat#: NP0001) was added. After electroeluted the cross-linked RNA-RBP complex, transferred the solution to tubes, added 800 µl proteinase K buffer (100 mM Tris-HCl pH = 7.4, 150 mM NaCl, 12.5 mM EDTA, 2% (V/V) SDS (Bio-Rad, Cat#:1610418)) with 1.2mg/ml proteinase K (New England Biolabs, Cat#P8107S) and incubated at 55℃ for 30 minutes. Then recover the RNA with Trizol reagent according to the manufactory’s protocol. The following steps of library preparation were performed as described according to the NEBNext Multiplex Small RNA Library Prep Set for Illumina (Set 1).
Analysis of PAR-CLIP data
The forward read1 was taken from the paired-end sequenced PAR-CLIP data of each sample. This is because the RBP (RNA-binding protein)-bound fragments are short (< 70bp), the 150 bp long paired-end read1 and read2 will share the same fragment, thus one read (either read1 or read2) is enough. First, TrimGalore (version 0.4.4_dev, http://www.bioinformatics.babraham.ac.uk/projects/trim_galore/) was used to trim off the adapter sequences and only reads in the size range of 17–50 nt were retained for downstream analyses. Secondly, the size-selected reads were aligned to the human hg19 genome using 0, 1 and 2 mismatches sequentially using bowtie (version 1.2.1.1) 58 (sequential procedure: 1) map reads to the genome using 0 mismatch (-v 0), retain the mapped reads, and align the unmapped reads to the genome using 1 mismatch (-v 1); 2) retain the 1-mismatched reads and map the unmapped reads to the genome using 2 mismatches (-v 2)). A maximum of two mismatches were used as the RBP-binding sites will show up to T-to-C conversions during the treatment step of the PAR-CLIP protocol. To increase the signal-to-noise ratio, only reads resulting from up to 2 T-to-C conversions were selected, i.e., reads from other mismatches were discarded. Customized scripts and commands (available upon request) were developed to process aligned SAM files to identify reads that perfectly aligned to the genome, and those aligned to the genome with 1 or 2 mismatches resulting from T-to-C conversion(s) only. Subsequently, the three SAM files (0 mismatch, 1 T-C converted SAM, 2 T-C converted SAM) were merged and sorted using samtools (version 1.4.1) 59, and acted as the input for PARalyzer (v1.5) 60. The parameters in the “.ini” file used as the input for PARazyler are as follows:
BANDWIDTH = 3,
CONVERSION = T > C,
MINIMUM_READ_COUNT_PER_GROUP = 5, MINIMUM_READ_COUNT_PER_CLUSTER = 2, MINIMUM_READ_COUNT_FOR_KDE = 3,
MINIMUM_CLUSTER_SIZE = 11, MINIMUM_CONVERSION_LOCATIONS_FOR_CLUSTER = 2, MINIMUM_CONVERSION_COUNT_FOR_CLUSTER = 2, MINIMUM_READ_COUNT_FOR_CLUSTER_INCLUSION = 1, MINIMUM_READ_LENGTH = 20, MAXIMUM_NUMBER_OF_NON_CONVERSION_MISMATCHES = 1,
GENOME_2BIT_FILE = Homo_sapiens.GRCh19.dna.primary_assembly.2bit.
The bound regions identified by PARalyzer from the “resulting.cluster.file” were processed and intersected with the hg19 gene annotations using bedtools 61 for detailed annotation (for instance, 3’UTR, or CDS, 5’UTR, or intron of particular genes of the hg19 genome). Two replicates for each genotype (wild type and knockout) were analyzed separately. To identify the PUM1-bound genes, the bound genes from each replicates of the wild type were combined (take the union not the intersect), and those genes identified to be bound by PUM1 in the knockout were excluded, as they presumably contain non-specific binding sites (not resulting from PUM1 but other non-specific regions of the antibody). The bound regions were analyzed for enriched motif analysis and visualized in IGV 62 to infer the potential binding affinity (percentage of T-to-C conversions among all mapped reads covering the region of interest).
Mass Spectrometry and RNA Deep Sequencing
Mass spectrometry was performed with the TMT kit from Thermo Fisher (#90110) and followed the product instructions. The selection criteria for up-regulated proteins were that the number of unique peptides was more than 1 and the up-regulated fold was higher than 1.2. For RNA sequencing, wild type, Pum1−/− and Pum2−/− HCT116 cells were collected. Total RNA was isolated using TRIzol (Invitrogen, 15596026) according to the protocol of the manufacturer. mRNAs and small RNAs were sequenced with an Illumina HiSeq3000 platform (Lie Bing Co., Shanghai, China). The procedure on analyses of RNA-Seq data followed the same as described in Shi et al. 2020. Briefly, paired-end reads were mapped onto the hg38 reference using STAR, HTSeq-count software was used to quantify the reads mapped to each gene. The R package DESeq2 was used to identify significantly differentially expressed genes in Pum1−/−/ Pum2−/− compared with WILD TYPE HCT116 cells (adjusted p-value ≤ 0.05 or fold change ≥ 1.5).
Immunofluorescence Microscopy
Immunofluorescence microscopy assay was performed according to a previously described standard method 57.
Immunohistochemistry
Formalin-fixed, paraffin-embedded colorectal tumor tissue blocks from mice were used in our investigation. Immunohistochemical staining was performed using primary antibodies and HRP-conjugated secondary antibodies. The antibodies used include PUM1 (ab92545, Abcam), PUM2 (ab92930, Abcam), p21 (2947S, Cell Signaling Technology), p27 (3686S, Cell Signaling Technology) and anti-Ki67 (sc-7846, Santa Cruz Biotechnology Inc). TUNEL staining was performed using an in situ Cell Death Detection Kit (Roche) based on the manufacturer’s instructions.
Luciferase Reporter Assay
The 3′UTRs of p21, p27 were amplified and cloned into psiCHECK2 vector (C8021, Promega) using primers specified in Supplementary Table S1. In the 3′UTR reporter assay, HCT116 cells in 24-well plates were transfected with 100 ng of the psiCHECK2-p21-3′UTR-WT or psiCHECK2-p27-3′ UTR-WT or 3'UTR mutant plasmids with mutations in PUM1, miR-130a or miR-221/222 binding sites and 100 nM mimics or 100 nM antisense (RiboBio) or 500 ng pGL3-PUM1 plasmid along with 2 µl Lipofectamine 2000 (Invitrogen). Lysates were harvested after 48 hours of transfection. The reporter activity was measured with the Dual Luciferase Assay (E1910, Promega).
Tumorigenesis Assay
Four-week-old BALB/c nude mice (male) were purchased from Shanghai Lingchang Biotechnology Co., Ltd (Shanghai, China). All experiments were performed under the guidelines for the care and use of laboratory animals. BALB/c nude mice were randomly divided into two groups (n = 8 per group). Then, wild type, Pum1−/− and Pum2−/− HCT116 (2 × 106) or RKO (3 × 106) cells in 0.1 ml of medium mixture (medium: matrigel = 1: 1) were subcutaneously injected into the nude mice. The tumor size was measured at time intervals as indicated in Figs. 2M and 2P. The mice were euthanized by cervical dislocation and the tumors were dissected. The tumor weight was measured, and the tumor volume was calculated using the following formula: D × d2/2, with ‘D’ representing the longest diameter and ‘d’ representing the shortest diameter.
AOM/DSS Model
There are two types of mice used in the AOM/DSS model: Lgr5cre::Pum1flox/flox::Pum2flox/flox and Pum1flox/flox::Pum2flox/flox. Mice were first induced by intraperitoneal injection with tamoxifen (120 mg/kg) every other day for a total of 7 days. Sixteen days later, mice were injected intraperitoneally with 12.5 mg/kg azoxymethane (Sigma, St.Louis, MO). Dextran sodium sulfate (DSS) 2.5% (wt/vol) (molecular mass, 36–50 kilodaltons; Meilun biotechnology) was given for 5 days. Mice were then given regular drinking water for 14 days, followed by 1 additional cycle with DSS. In the third cycle, the DSS was adjusted from 2.5–2%. On day 65, mice were sacrificed for testing.
Preparation and Characterization of siRNA-loaded MSN
We previously developed a woven PEI (1.8 kDa)-coated, glutathione (GSH)-responsive, mesoporous silica nanoparticle (MSN) for nucleic acid (miRNA-145) delivery 44. The nanoparticles can quickly escape from the lysosome into the cytosol for miRNA-145 delivery, mediated by the proton sponge effect of PEI 58. The reducing environment (0.5–10 mM GSH) 45, 46 within tumor cells removed the PEI from the MSN surface through hydrolyzing the disulfide bond and thus allow nucleic acid release in the cytoplasm 44. Here, this smart nanocarrier was adopted for the delivery of siPum1 and 2, which were complexed with MSN to generate siPum1, 2@MSN. Transmission electron microscopy (TEM) images were acquired on a FEI Talos F200X system. The hydrodynamic size and zeta potential of nanoparticles were determined through dynamic light scattering (DLS) and measured using a ZetaSizer Nano ZS instrument (Malvern, Worcestershire, UK).
siRNA (FAM labeled) release test was performed in the presence of 5 mM GSH-contained PBS in the centrifuge tubes. The tubes were placed in gas bath at 37°C shaking at 100 rpm. At pre-determined time point (1, 2, 4, 8, 12, 24 h), empty DPPC liposomes (-30 mV, 120 nm, 25 mg/ml) was added to bind free PEI and PEI-siRNA complex. Then, the solution was transferred into Ultra Centrifugal Filters (molecular weight cut-off 50 kDa), and then centrifuged under 5000 rpm at 4°C for 30 min. FAM-labeled siRNA (Ex 488 nm, Em 520 nm) in the filtrate was collected for the quantification of nucleic acid release. The in vivo tumor targeting of the siRNA-loaded nanoparticles 24 h after i.v. injection was examined using the IVIS Spectrum/CT imaging system (PerkinElmer, USA) as previously described 44.
siRNA molecules specifically targeting the mRNA of Pum1 and Pum2 were purchased from RIBOBIO (China). siPum1: GGTCAGAGTTTCCATGTGA, siPum2: CTGAAGTAGTTGAGCGCTT 24. siRNA (FAM labeled) release test was performed in the presence of 5 mM GSH-contained PBS in the centrifuge tubes. The tubes were placed in gas bath at 37°C shaking at 100 rpm. At pre-determined time point (1, 2, 4, 8, 12, 24 h), empty DPPC liposomes (-30 nm, 120 nm, 25 mg/ml) was added to bind free PEI and PEI-siRNA complex. Then, the solution was transferred into Ultra Centrifugal Filters (molecular weight cut-off 50 kDa), and then centrifuged under 5000 rpm at 4°C for 30 min. FAM-labeled siRNA (Ex 488 nm, Em 520 nm) in the filtrate was collected for the quantification of nucleic acid release.
Antitumor Therapy in Orthotopic Colorectal Tumor Model
The orthotopic CRC model was developed in female BALB/c nude mice as described previously 44. 28 day after orthotopic tumor cell (HCT116-luc, COLO205-luc) inoculation, the tumor-bearing mice were randomly allocated into five groups (n = 5) according to the tumor bioluminescent intensity and treated with saline (control), MSN loaded with negative control siRNA with a scrambled sequence (siNC@MSN), and MSN loaded with siPum1 (siPum1@MSN), siPum2 (siPum1@MSN) or both (siPum1/2@MSN). The siRNA dose was 1 mg/kg on days 0, 3, 6, and 9, respectively. Each group included 10 mice. The tumor burden was monitored weekly using bioluminescence imaging (IVIS spectrum CT). The body weight was recorded throughout the study. At the end of the study (day 30), the mice were sacrificed, the tumors were removed and weighted. Other major organs included heart, liver, spleen, lung, and kidney were harvested for ex vivo bioluminescent imaging to examine the metastasis. In a separate study, on day 11 (2 d after the last injection of the nanoparticles), three mice from each group were sacrificed. The major organs, including heart, liver, spleen, lung, and kidney were collected for H&E histological assay for toxicity evaluation.
Accession Numbers
The accession number for the RNA-seq and PAR-CLIP data reported in this paper is GEO: PRJNA648706.