Cell culture

The MDA-MB-231 cell line was a generous gift from Dr. Zaver Bhujwalla (Johns Hopkins School of Medicine, Baltimore, MD). MDA-MB-231 cells were authenticated by short-tandem repeat profiling, performed by ATCC. MDA-MB-231 KRT19 KO cells generated using the CRISPR/Cas9 system and cells stably expressing GFP or GFP-KRT19 generated using the lentiviral system were described previously [26]. Stable transductants were selected using hygromycin (100 µg/ml). Cells were grown in Dulbecco’s modified essential medium (DMEM; Gibco, Grand Island, NY) supplemented with 10% Fetal Bovine Serum (GE Healthcare, Logan, UT) and 1% penicillin/streptomycin (Gibco) in a humidified incubator at 5% CO2 and 37 °C. To measure cell proliferation, 50,000 cells were plated on each well of six-well plates. Cells were counted using a hemocytometer 24, 48, 72, and 96 h after plating.

Plasmids and siRNAs

To overexpress HNRNPK, cDNA was cloned out of pCMV6-AC HNRNPK (OriGene, Rockville, MD) using the primers described previously [4] and cloned into pmCherry-C1 (Takara Bio Inc., San Jose, CA). To silence HNRNPK expression, Accell Human SMARTpool HNRNPK siRNAs (Dharmacon, Lafayette, CO) was used. The siRNA sequences targeting HNRNPK are described in Table 1. AllStars negative control SI03650318 was purchased from Qiagen (Germantown, MD).

Table 1 Sequences for Accell Human SMARTpool HNRNPK siRNAs from Dharmacon Transient transfection

For siRNA, RNAimax lipofectamine (Invitrogen, Waltham, MA) was used to transfect HNRNPK smartpool siRNA (Dharmacon) or non-targeting siRNA (Qiagen) according to the manufacturer’s protocol. For overexpression plasmids, jetOptimus DNA transfection reagent (Polyplus, Illkirch, France) was used according to the manufacturer’s protocol.

PAR-CLIP analysis

Both parental and KRT19 KO cells were treated with 100 µM 4thiouridine (4SU) for 16 h, washed with PBS and irradiated with 0.15 mJ cm2, 365 nm ultraviolet light in a Spectrolinker XL-1500 UV crosslinker to crosslink RNA to HNRNPK. Cytoplasmic pools of cells were collected by lysing them with a buffer containing 20 mM 4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid (HEPES), pH 8, 1 mM ethylenediaminetetraacetic acid (EDTA), 1.5 mM magnesium chloride, 10 mM potassium chloride, 1 mM dithiothreitol, 1 mM sodium orthovanadate, 1 mM sodium fluoride, 1 mM phenylmethylsulfonyl fluoride, 0.5 mg/mL benzamidine, 0.1 mg/ml leupeptin, and 1.2 mg/mL aprotinin on ice for 15 min. Then, 7.5 µL of 10% NP-40 detergent was added to the cells and lysates were centrifuged for 1 min at 14000 g and 4°C, and supernatants were collected as cytosolic fractions. 10 U/ul of RNAse T1 was then added to the lysates followed by the immunoprecipitation using anti-HNRNPK antibody and Protein G magnetic beads. The beads were resuspended in one bead volume of dephosphorylation buffer after washing them with lysis buffer. On beads 3’ adapter ligation was performed, followed by on beads phosphorylation step. Then, the protein-RNA complexes were resolved using SDS-PAGE and fluorescent protein-RNA bands were recovered at the corresponding size ~ 60 kDa. Proteinase K digestion was performed followed by RNA recovery by acid phenol/chloroform extraction and ethanol precipitation. After RNA isolation, cDNA library preparation was carried out [30]. Library preparation was done using cDNA and sequences by Illumina platform. Obtained reads were processed and referenced to the hg19 genome assembly (hg19). Data analysis was performed by using PARalyzer settings where T-to-C mutation sequences were filtered. Data from the sequencer were converted to fastq files by being demultiplexed using Bcl2fastq (v2.20.0). Adapter barcoded samples were demultiplexed by using cutadapt (1.15 with python 3.6.4) with retaining the adapters. Also, libraries were demultiplexed by using 5’ adapter barcode by removing the 5’ adapter barcode anchored to 5’ end of the read. Moreover, data were further processed by PARpipe (https://github.com/ohlerlab/PARpipe) through implementing Paralyzer as well as annotation of groups and clusters. Potential PCR duplicates were further removed by allowing the use of additional UMIs in the new protocol and preserve the true number of reads in the final bam file, during preprocessing fastq sequences were collapsed prior to adapter and UMI removal [33, 44].

RNA-seq analysis

For RNA-seq analysis, total RNAs from three biological replicates each of untransfected parental and KRT19 KO cells, transfected with siHNRNPK, control siRNA, HNRNPK ∆NLS or vector were extracted. With siHNRNPK, control siRNA was extracted, while with HNRNPK ∆NLS, vector was extracted. Ribosomal RNA was depleted of using the NEBNext® rRNA Depletion Kit and cDNA libraries were prepared using the NEBNext® Ultra™ Directional RNA Library Prep Kit for Illumina® (NEB, Ipswich MA). RNA was barcoded using the NEBNext Multiplex Oligos for Illumina (NEB). All samples were multiplexed and sequenced on the Illumina HiSeq 3000 platform using 50 cycles single-end sequencing. Reads were aligned to hg19 using TopHat2 [45]. Cufflinks and Cuffdiff were used to quantify transcripts and determine differential expression. The top 500 upregulated genes with minimum 4 RPKM and p-value 0.05 were subjected to Enrichr tool, which uses curated gene sets to give biological meaning that can be used for further validation [2].

Antibodies and other reagents

The following antibodies were used in this study anti-GAPDH (FL-335), anti-K19 (A-3), anti-K8 (C51), anti-K18 (C-04), anti-PARP (F-2), anti-HNRNPK (3C2) and anti-MDM2 (SMP14) were from Santa Cruz Biotechnology (Santa Cruz, CA); anti-p53 (DO-2) was from (MilliporeSigma, St. Louis, MO, USA); anti-GFP (12A6) was from the Developmental Studies Hybridoma Bank (Iowa City, IA); and anti-RFP (5F8) from ChromoTek (Munich, Germany).

Western blotting

Cells were washed with 1X PBS, and cell lysates were prepared in cold Triton lysis buffer (1% Triton X-100; 40 mm HEPES, pH 7.5; 120 mm sodium chloride; 1 mm EDTA; 1 mm phenylmethylsulfonyl fluoride; 10 mm sodium pyrophosphate; 1 µg/ml each of chymostatin, leupeptin, and pepstatin; 10 µg/ml each of aprotinin and benzamidine; 2 µg/ml antipain; 1 mm sodium orthovanadate; and 50 mm sodium fluoride). Cell lysates were centrifuged to remove cell debris. Protein concentration was determined using the Bio-Rad Protein Assay (Bio-Rad) with bovine serum albumin as standard then was prepared in Laemmli SDS-PAGE sample buffer. Aliquots of protein lysate were resolved by SDS-PAGE, transferred to nitrocellulose membranes (Bio-Rad), cut into smaller pieces when necessary, and immunoblotted with the indicated antibodies, followed by horseradish peroxidase-conjugated goat anti-mouse or goat anti-rabbit IgG (MilliporeSigma, St. Louis, MO, USA) and Amersham ECL Select Western Blotting Detection Reagent or Pierce ECL Western Blotting Substrate (Thermo Scientific, Hudson, NH). Signals were detected using ChemiDoc Touch Imager (Bio-Rad) or CL1500 Imaging System (Thermo Fisher Scientific). For Western blot signal quantitation, the Image Lab software (Bio-Rad) was used.

Biochemical subcellular fractionation

Subcellular fractionation was performed as described previously [4]. After rinsing with cold 1X PBS, cold lysis buffer (20 mM HEPES pH 8,1 mM EDTA, 1.5 mM Magnesium chloride, 10 mM Potassium chloride, 1 mM Dithiothreitol, 1 mM sodium orthovanadate, 1 mM Sodium fluoride, 1 mM phenylmethylsulfonyl fluoride, 0.5 mg/mL benzamidine, 0.1 mg/ml leupeptin, and 1.2 mg/mL aprotinin) was used to lyse the cells. Then 7.5 µL of 10% NP-40 detergent was added to the cells after incubating them on ice for 15 min. After adding NP-40 detergent to the cells, lysates were centrifuged for 1 min at 14,000 g and 4 °C, and supernatants were collected as cytosolic fractions. Pellets were washed 4 times with 1X PBS and incubated for 40 min at 4 °C (1% Triton X-100, 40 mM HEPES (pH 7.5), 120 mM sodium chloride, 1 mM EDTA, 1 mM phenylmethylsulfonyl fluoride, 10 mM sodium pyrophosphate, 1 µg/ml each of chymostatin, leupeptin and pepstatin, 10 µg/ml each of aprotinin and benzamidine, 2 µg/ml antipain, 1 mM sodium orthovanadate, 50 mM sodium fluoride). Supernatants were collected as nuclear fractions after centrifugation for 10 min at 13,800 g and 4 °C.

Co-immunoprecipitation

Cells were washed with 1X PBS and cell lysates prepared in cold triton lysis buffer (1% Triton X-100; 40 mm HEPES (pH 7.5); 120 mm sodium chloride; 1 mm EDTA; 1 mm phenylmethylsulfonyl fluoride; 10 mm sodium pyrophosphate; 1 µg/ml each of chymostatin, leupeptin, and pepstatin; 10 µg/ml each of aprotinin and benzamidine; 2 µg/ml antipain; 1 mm sodium orthovanadate; 50 mm sodium fluoride) supplemented with 2% empigen for anti-K19 IP, or cold NP-40 lysis buffer (0.25% NP-40; 50 mM Tris (pH 8.0); 100 mM sodium chloride; 1 mm phenylmethylsulfonyl fluoride; 10 mm sodium pyrophosphate; 1 µg/ml each of chymostatin, leupeptin, and pepstatin; 10 µg/ml each of aprotinin and benzamidine; 2 µg/ml antipain; 1 mm sodium orthovanadate; 50 mm sodium fluoride) for anti-HNRNPK IP. Cell lysates were centrifuged to remove cell debris, and protein concentration was determined using the Bio-Rad Protein Assay with BSA as standard. Aliquots of cell lysate were then incubated with the indicated antibody or IgG control, and immune complexes were captured using Protein G Sepharose (GE Healthcare).

Immunofluorescence (IF) staining

IF staining of cells was performed as described previously [26]. Cells grown on glass coverslips (VWR, Radnor, PA) were washed with 1X PBS, fixed in 4% paraformaldehyde in 1X PBS for 35 min, and permeabilized in 0.1% Triton X-100 for 20 min or 0.01% digitonin for 5 min. Samples were blocked in 5% normal goat serum (NGS; RMBIO, Missoula, MT) in 1X PBS before staining with primary antibodies diluted at 1:400 ratio in 5% NGS blocking buffer and a mixture of 1:1000 of Alexa Fluor 488-conjugated goat anti-mouse secondary antibody (Invitrogen) and 1:5000 DAPI (MilliporeSigma, St. Louis, MO, USA) in 1X PBS was added for 1 h incubation at RT. After 1X PBS washes, coverslips were mounted on microscope slides with a mounting medium containing 1,4-diaza-bicyclo[2.2.2]octane (Electron Microscopy Sciences, Hatfield, PA). Fluorescence images were taken using the Olympus optical elements fluorescence microscope (Olympus Optical Co., Japan).

Proximity ligation assay

The Duolink in situ proximity ligation assay (PLA) was performed according to the manufacturer’s protocol (MilliporeSigma, St. Louis, MO, USA). In brief, cells were plated on glass coverslips, rinsed three times with PBS and fixed in 3.7% formaldehyde in 1X PBS for 20 min. The cells were permeabilized in 0.01% digitonin for 5 min and blocked with 5% NGS in 1X PBS for overnight at 4 °C. After blocking, cells were then incubated with antibodies against HNRNPK, K19 and IgG in 1X PBS containing 5% NGS overnight at 4 °C, followed by incubation with corresponding secondary antibodies conjugated with PLA probes for 60 min at 37 °C in the dark. Cells were washed three times in 1X PBS. Duolink and DAPI signals were detected using Olympus optical elements fluorescence microscope (Olympus). Images are analyzed using ImageJ [46].

Colony formation assay

Colony formation assay was performed as described previously [25]. On each well of six-well plates, 1,000 cells were seeded and grown in 2 ml DMEM media for 14 days or 10,000 cells were seeded and grown for 10 days. The colonies were fixed with 4% formaldehyde and then stained with 0.5% crystal violet. Images were taken using ChemiDoc Touch Imager (Bio-Rad, Hercules, CA) and area of colonies were determined by ImageJ software (National Institutes of Health). Three biological replicates were analyzed.

Graphs and statistics

All graphs in the manuscript are shown as mean ± standard error of mean. For comparisons between two datasets, a Student’s t test (tails = 2, type = 1) was used, and statistically significant p-values ≤ 0.05 are indicated in the figures and figure legends. GraphPad prism and excel were used to generate graphs in this manuscript.