If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
BioID with KPNB1 has limited efficiency for identifying bona fide importin cargos.
A new monoclonal antibody targets conserved residues 301-320 in KPBN1.
The antibody is specific for cytoplasmic Importin β1 complexes.
BAR with the novel antibody reveals cytoplasmic importin β1 interactomes.
Importin β1 (KPNB1) is a nucleocytoplasmic transport factor with critical roles in both cytoplasmic and nucleocytoplasmic transport, hence there is keen interest in the characterization of its subcellular interactomes. We found limited efficiency of BioID in the detection of importin complex cargos and therefore generated a highly specific and sensitive anti-KPNB1 monoclonal antibody to enable biotinylation by antibody recognition analysis of importin β1 interactomes. The monoclonal antibody recognizes an epitope comprising residues 301-320 of human KPBN1 and strikingly is highly specific for cytoplasmic KPNB1 in diverse applications, with little reaction with KPNB1 in the nucleus. Biotinylation by antibody recognition with this novel antibody revealed numerous new interactors of importin β1, expanding the KPNB1 interactome to cytoplasmic and signaling complexes that highlight potential new functions for the importins complex beyond nucleocytoplasmic transport. Data are available via ProteomeXchange with identifier PXD032728.
). The importin β protein family transports cargos with nuclear localization signals (NLS) into the nucleus. Most importin βs bind their cargos directly, while the canonical importin β1 (also termed karyopherin β1, KPNB1) uses one of the accessory importin α family members as cargo-binding subunits (
). Ran-regulated association of importin β1/α/NLS-cargo complexes in the cytoplasm and their dissociation in the nucleoplasm enables nucleocytoplasmic transport (Fig. 1A). In addition to its key role in nucleocytoplasmic transport, importin β1 forms cytoplasmic transport complexes with the microtubule motor dynein (Fig. 1B). Importin β1–dependent cytoplasmic transport has been implicated in long distance signaling in large cells, especially neurons, with roles in injury signaling, size regulation, and growth control (
Subcellular localization is an important determinant of biological function, hence there is keen interest in determining importin interactomes and importin-dependent protein partitioning within cells (
). Comprehensive characterization of importin interactomes is challenging due to the transient nature of importin–cargo interactions and the overlapping specificities of different importin complexes. Hence, most previous studies have focused on identifying importin carriers for specific cargo proteins, revealing a diverse range of both direct and indirect KPNB1 interactors (e.g. (
). We therefore sought methods for comprehensive subcellular studies of its interactomes. A diverse range of proximity-based biotinylation assays provide powerful tools to label and subsequently detect interacting proteins (
) suggested limited efficiency of the method for both N- and C-terminal fusions of KPNB1 with the modified biotin ligase BirA. We therefore generated a monoclonal antibody (mAb) to enable biotinylation by antibody recognition (BAR) (
). The antibody targets a 20 amino acid epitope that is highly conserved in mammalian KPNB1 and discriminates between nuclear and cytoplasmic protein complexes, allowing rigorous characterization of cytoplasmic importin β1 interactomes using the BAR method.
All animal procedures were performed in accordance with the guidelines approved by the Weizmann Institute of Science Institutional Animal Care and Use Committee. Animals were purchased from Envigo Ltd (Israel) or originate from in-house breeding. Female BALB/c mice (6–8 weeks, Envigo) were used for immunization and subsequent isolation of spleen cells. Follow-up antibody characterization experiments were performed on male C57BL6 mice (8–12 weeks, Envigo) and as well as male Wistar rats (8–12 weeks, Envigo). All animals were maintained at the Veterinary Resources department of the Weizmann Institute with free access to food and water.
Sciatic Nerve Crush
Animals were anesthetized with ketamine/xylazine (10 mg/kg of body weight intraperitoneal). Sciatic nerve (SN) crush was performed at the mid-thigh level using fine jeweler forceps in two adjacent positions for 30 s each. Only one side was subjected to crush injury, the contralateral side served as an uninjured control. SNs were excised 6 and 24 h after injury, fixed and processed for immunohistochemistry.
Reagents and Antibodies
Commercially available primary antibodies used in this study for immunoprecipitation (IP), immunofluorescence (IF), proximity ligation assay (PLA), Western blot (WB), and Simple Western capillary system (WES) were as follows: anti-GFP (abcam, ab6556, IF & WB 1:5000, PLA 1:2000), anti-NFH (abcam, ab72996, 1:2000), anti-Actin (MP, clone C4, 08691001, WB 1:5000), anti-GAPDH (Santa Cruz Biotechnology, sc-32233, WB 1:5000), anti-ERK1/2 (Sigma, M5670, WB 1:30000), anti-β-III tubulin TUJ1 (abcam, ab18207, IF 1:1000, WB: 1:6000), anti-Dynein heavy chain (DYNC1H1, Proteintech, #12345-1-AP, PLA 1:50), anti-importin beta 1 (MBS713065, rabbit, polyclonal, 1:1000), and rabbit anti-nucleolin (ProteinTech, 10556-1-AP, 1:100 for IF).
Plasmids and Constructs
Expression constructs for KPNB1 were based on pEGFP-N1 human full-length KPNB1 from Addgene (#106941). The KPNB1 coding DNA sequence was amplified from pEGFP-N1 human full-length KPNB1 (Addgene #106941) and mVenus coding DNA sequence was amplified from pTrix-mVenus-PA-Rac1 (Addgene #22007). For seamless protein engineering of importin β1 BioID constructs, all cloning steps were carried out using restriction-free cloning following published protocols (
). All primers were designed such that the protein domains (BirA∗, ImpB1, and mVenus) are separated by a small flexible GS-dipeptide. Briefly, the multiple cloning site of pcDNA3.1 mycBioID (Addgene #35700) was removed to generate the Myc-BirA∗ control. Subsequently, the KPNB1 coding sequence was inserted either downstream or upstream of BirA∗ to generate N- or C-terminally BirA∗-tagged importin β1 fusion constructs, respectively. The opposite terminus of importin β1 was fused to mVenus (amplified from pTrix-mVenus-PA-Rac1, Addgene #22007) which has a similar size to BirA∗ to ensure similar molecular behavior of both fusion constructs. Detailed cloning procedure is described in the Supplementary Data.
Generation of Monoclonal Antibodies
The antibody was raised against full-length recombinant mouse importin β1 protein (UniProt accession # P70168) with an N-terminal His-tag produced in yeast (MyBioSource: cat # MBS955236). Importin β1 protein was dissolved in 1xPBS and mixed with Complete Freund’s Adjuvant (CFA) in a double-glass syringe at a volume ratio of 1:1 (23 μg protein in 50 μl 1xPBS +50 μl CFA/mouse) for immunization. Five BALB/c female mice (6–8 weeks old) were used for immunization by footpad injection into both hindlimbs (50 μl/food pad). Two weeks later, mice were boosted with the same dose (boost 1: 23 μg protein in 100 μl/mouse) intradermally in several spots in the belly area and limbs (close to lymph nodes). Retroorbital bleeding was performed 1 week later and serum was tested by ELISA to determine the mouse with the highest immune response. Four weeks after boost 1, the mouse with the highest immune response received a prefusion boost (IP injection of 20 μg protein in 200 μl 1xPBS, without CFA) and was sacrificed 4 days later to collect immune cells from the spleen. Approximately, 100 million spleen cells were harvested and collected in Dulbecco’s modified Eagle’s medium (DMEM) for the production of hybridoma cell lines. The timeline of the immunization procedure is shown in supplemental Fig. S2A.
Hybridomas were generated as previously described (
). BALB/c female mouse splenocytes were fused with NS0 nonsecreting murine myeloma cells using 41% PEG in DMEM (Roth). Hybrid cells were selected in growth medium supplemented with HAT (hypoxanthine, aminopterin, and thymidine) (100× HAT media supplement, Sigma-Aldrich). Culture supernatants from wells with viable clones were screened by an indirect ELISA using recombinant importin β1 protein. Stable hybridoma clones secreting importin β1 specific antibodies were obtained after two cloning cycles by a limiting dilution assay and were weaned from HAT supplements. Hybridoma cells were grown in complete DMEM (Biochrom) supplemented with 15% fetal bovine serum (FBS, Biochrom), 2 mM L-glutamine, 10,000 units penicillin, 10,000 units streptomycin, 1 mM Na pyruvate, with or without HAT supplements (1:100 from stock). The final positive hybridoma cells secreting antibodies against importin β1 were stored in liquid nitrogen. All procedures involving experimental mice were performed under controlled laboratory conditions in strict accordance with Institutional Animal Care and Use Committee guidelines. A schematic of the selection process and hybridoma propagation is shown in supplemental Fig. S2B. For more detailed protocols, see Supplementary Material.
Enzyme-Linked Immunosorbent Assay
Indirect ELISA was performed in 96-well plates (NUNC-442404), coated with full-length importin β1 protein (1 μg/ml in 1xPBS) overnight at 4 °C, and then blocked by incubation with 10 mg/ml bovine serum albumin (BSA) in 1xPBS for 1 h at RT. Hybridoma supernatant or purified and diluted Abs were applied and incubated for 1 h at RT, followed by HRP-conjugated anti-mouse Fab fragments (1 h RT, 1:60,000) as the secondary antibody (Sigma A9917-1 ml). HRP substrate (TMB Sigma T0565-100 ml) was added subsequently and the absorbance was recorded at 630 nm using a microplate reader (Thermo Fisher Scientific). In between steps, plates were washed in PBST (PBS containing 0.05% Tween-20) three times. For linear ELISA, coating was done with a serial dilution of the full-length importin β1 protein (0.001, 0.01, 0.1, 1 μg/ml), to explore the linear range of antibody-antigen recognition.
Antibody Isotyping & Purification
Immunoglobulin isotyping test was performed using a mouse mAb isotyping kit, according to the manufacturer’s instructions (SBA Clonotyping System-HRP, SouthernBiotech, 5300-05). Monoclonal antibodies were purified by Protein A affinity purification, using an AKTA PURE system. The purified fraction was desalted in 1XPBS, using a HiPrep desalting column and stored at 2 to 8 °C.
Adult mouse dorsal root ganglia (DRG) neuron culture was performed as previously described (
). Briefly, the ganglia were dissociated with 100 U of papain (P4762, Sigma) followed by 1 mg/ml collagenase-II (11179179001, Roche) and 1.2 mg/ml dispase-II (04942078001, Roche). The DRGs were triturated in 1xHBSS, 10 mM glucose, and 5 mM Hepes (pH 7.35) using a fire-coated Pasteur pipette and then layered on 20% percoll in L15 media and recovered by centrifugation at 1000g for 8 min. Cells were washed briefly in growth media (F12, 10% FBS, Primocin (100 μg/ml, InvivoGen #ant-pm-1)) and plated on poly-L-lysine (P4832, Sigma) and laminin (23017-015, Invitrogen)-coated glass coverslips. Culture media and serum were purchased from Thermo Fisher Scientific. DRG neurons were grown for 24 to 48 h in culture at 37 °C and 5% CO2.
Cell Line Cultures
HeLa cells (human, female, RRID: CVCL-0030), HEK-293 (human, female, RRID: CVCL_0045), NIH/3T3 (mouse, male, RRID: CVCL_0594), and Neuro-2A (mouse, male, RRID: CVCL_0470) cells were purchased from ATCC (Cat# CCL-2, CRL-1573, CRL-1658, CCL-131, respectively). All cell lines were cultured in DMEM (Gibco), supplemented with 10% FBS (Gibco), 100 U/ml penicillin, and 100 μg/ml streptomycin. All cells were incubated at 37 °C and 5% CO2.
Protein Extraction for SDS Page, Western Blot, and WES
For total protein extraction, tissues were collected in RIPA buffer (150 mM NaCl, 1.0% NP-40, 0.5% sodium deoxycholate, 0.1% SDS, 50 mM Tris, pH 8.0.), supplemented with protease/phosphatase/RNase inhibitors (complete protease inhibitor EDTA-free (Roche 1187358000), phosphatase inhibitor cocktail 2 (1/1000, Sigma 5726), phosphatase inhibitor cocktail 3 (1/1000, Sigma P0044), and RNAse inhibitor (200 U/ml, RNAseIn, Promega N251B)). Tissue samples were homogenized in Dounce Tissue Grinders (WHEATON 33, 1 ml #357538, 7 ml #357542) or, using plastic pestles, in an Eppendorf tube.
The axoplasm for biochemical analysis was extracted from mouse, rat, or bovine SN as previously described (
). To minimize glia contamination, transport buffer (TB) was used in this extraction protocol (20 mM Hepes, 110 mM KAc, 5 mM MgAc, pH 7.4 supplemented with protease/phosphatase/RNase inhibitors). All protein extracts were incubated on ice for 20 min, followed by a spin at 10,000g for 10 min at 4 °C.
For WB, proteins were blotted on nitrocellulose membranes using a Trans-Blot Turbo Transfer System (Bio-Rad). Membranes were blocked with 5% nonfat dry milk in 1xTBST buffer for 1 h. Primary antibodies were incubated for 1 h at room temperature or overnight at 4 °C with shaking. Secondary HRP-conjugated antibodies were diluted 1:10,000 in 1xTBST and incubated for 1 h at room temperature. Blots were developed using Radiance ECL substrate (Azure biosystems) or SuperSignal West Femto (Thermo Fisher Scientific) substrates on an Amersham Imager 680. Automated capillary electrophoresis immuno-quantification runs were conducted on a WES instrument (ProteinSimple) as described (
). The indicated molecular weights are based on an internal standard spiked into every capillary.
Proximity Ligation Assay
PLA was performed on cultured DRG neurons using the Duolink system (Sigma) according to the manufacturer's instructions with minor modifications. DRG neurons (isolated from C57BL6 mice) were fixed in 4% paraformaldehyde (PFA) after 48 h in culture. Blocking was performed in 5% donkey serum with 1% BSA in 1xPBS for 30 min, followed by primary antibody incubation for 1 h at room temperature. Combinations: rabbit anti-DYNC1H1 (Proteintech, #12345-1-AP, 1:50) or rabbit anti-importin β1 (MBS713065, 1:1000), together with the antibody generated in the present study mouse anti-importin β1 (mAbKPNB1-301-320, 1 μg/ml). As controls, each antibody was used by itself, and furthermore, we used mAbKPNB1-301-320 in combination with another rabbit antibody (GFP, abcam, ab6556, 1:2000). Interactions (PLA signals) were detected using Sigma PLA probe anti-mouse minus DUO92004, anti-rabbit plus DUO92002, and detection kits Red DUO92008 or FarRed DUO92013 according to the manufacturer's instructions and incubation times. The PLA protocol was followed by IF staining for neurofilament heavy chain (abcam, ab72996, 1:2000, 45 min at RT) with Alexa 488 donkey anti-chicken secondary antibody (45 min at RT). The quantification of the PLA signal was performed on neurofilament heavy chain–positive axons using Fiji (
DRG neurons were grown on 3-mm carbon-coated Sapphire disks (Wohlwend GmbH) and fixed 48 h after plating by high-pressure freezing using a Leica EM ICE (Leica Microsystems GmbH), followed by freeze substitution in a temperature-controlled device, AFS2 (Leica Microsystems GmbH), washing, embedding in Lowicryl HM20, and ultrathin sectioning (70–90 nm). Sections were collected on nickel grids. For immunostaining, grids were washed, incubated for 30 min in blocking solution (0.5% gelatin, 0.5% BSA, 0.2% glycine, 3% NGS in PBS), and then incubated for 2 h in mAbKPNB1-301-320 (1 mg/ml stock, dilutions: 1:10, 1:100) with or without blocking peptide KPNB1301–320, with a ratio of 4.6 nM mAbKPNB1-301-320 to 9.5 nM peptide. After washing with PBS containing 0.2% glycine, the sections were incubated for 15 min in blocking solution and then for 30 min in 10-nm colloidal-gold–conjugated goat anti mouse antibodies (Electron Microscopy Sciences) (1:20), followed by sequential washing in PBS containing 0.2% glycine and in bi-distillated water. The grids were then stained with Reynolds lead citrate and analyzed under 120 kV on a Thermo Fisher Scientific Tecnai T12 transmission electron microscope equipped with a bottom mounted TVIPS TemCam-XF416 4k × 4k CMOS camera using TVIPS EMplified software (https://www.tvips.com/imaging-software/em-menu/).
Pull-Down From Rat Brain
Whole brain was isolated from adult Wistar rats (male, 8–12 weeks, Envigo) and homogenized in 7 ml Dounce Tissue Grinders (WHEATON 33, 7 ml #357542) in transport buffer, supplemented with NP-40 (20 mM Hepes, 110 mM KAc, 5 mM MgAc, pH 7.4 supplemented with complete protease inhibitor, phosphatase inhibitor cocktail 2 & 3, as well as 0.5% NP-40 for the initial lysis, then for the IP diluted to 0.035%). For each IP sample, 1 mg of total brain extract was used and adjusted with TB to a total volume of 300 μl and a final concentration of 0.035% NP-40. 10% (30 μl) was taken as input. The remaining lysate for the IP was incubated for 3 h at 4 °C with overhead rotation with 10 μg of mAbKPNB1-301-320 or control (preblocked mAbKPNB1-301-320: 10 μg Ab with 50 μg blocking peptide (KPNB1301–320) for 1 h). Subsequently, 100 μl Protein G magnetic beads (Dynabeads, Thermo Fisher Scientific 10004D), preblocked with salmon sperm DNA (10 μl DNA per 100 μl beads for 1 h 4 °C), were added to the brain lysate-Ab mixture and incubated for additional 2 h at 4 °C. Beads were washed in several steps with different buffers as follows: TB-0.1% NP-40 for 3 min, TB-0.5% NP-40 3 min, TB-1% NP-40 1 min, TB- no detergent 3 min. For simple Western blotting, elution from beads was conducted by denaturing the proteins from the beads with WES sample buffer supplemented with DTT to a final concentration of 40 mM for 5 min at 95 °C.
BioID in HeLa Cells
HeLa cells were transfected using jetPEI (Polyplus-transfectionSA) according to the manufacturer’s protocol and supplemented with 50 μM Biotin (Molecular Probes) for 24 h. Cells were grown on 10 cm plates for mass spectrometry analyses. For streptavidin-IP from whole cell lysates, cells were lysed with 1.5 ml/plate lysis buffer (50 mM Tris HCl, pH 7.5, 150 mM NaCl, 0.25% (w/v) NP-40, 1 mM MgCl2, 5 mM KCl, 2.5 mM EDTA, 1% (w/v) Triton X-100, 0.1% (w/v) SDS, 1 mM DTT, 1× Complete protease inhibitor, EDTA-free (Roche)), sonicated (30% amplitude, ten cycles of 5.5 s pulse on, 9.9 s pulse off), and lysates from six plates were incubated with 600 μl streptavidin beads for 4 h. Beads (Dynabeads MyOne streptavidin C1, Thermo Fisher Scientific) were then washed twice with wash buffer I (2% (w/v) SDS), once with wash buffer II (0.1% (w/v) deoxycholic acid, 1% (w/v) Triton X-100, 1 mM EDTA, 500 mM NaCl, 50 mM Tris HCl, pH 7.5), once with wash buffer III (0.5% (w/v) deoxycholic acid, 0.5% (w/v) NP-40, 1 mM EDTA, 250 mM LiCl, 10 mM Tris HCl, pH 7.5), twice with 50 mM Tris HCl, pH 7.5, and once with (20 mM Tris HCl pH 8.0, 2 mM CaCl2) (
HEK 293T cells were grown in DMEM high glucose 5 g/l (Sigma-Aldrich) supplemented with 10% heat inactivated FBS. Cells were seeded at a density of 1.2e4 cells/cm2 and monitored over 24 h until reaching 70% confluency. Cells were then fixed in 4% PFA and permeabilized with Triton X-100 for 10 min at RT, subsequently, incubated for 3 h at 4 °C with blocking solution. 2.7 μg/ml mAbKPNB1-301-320 was added in blocking buffer and incubated overnight at 4 °C. The next day, cells were washed three times with 1xPBST and incubated with secondary HRP-conjugated antibody for 3 h at 4 °C prior to tyramide signal amplification according to the manufacturer’s protocol. Cells were harvested and boiled for 1 h at 99 °C for reverse crosslinking followed by streptavidin beads affinity purification and processing for proteolytic digest and mass spectrometry.
Adult DRG neuronal cultures were prepared as previously described (
). The culture was maintained for 72 h with daily supplement of glia inhibitor AraC (10 μM). The cells were then fixed in 4% PFA for 20 min at room temperature followed by three washes with 1xPBST, permeabilization with 0.1% Triton X-100, and incubation in blocking solution for 3 h at 4 °C. 2.7 μg/ml mAbKPNB1-301-320 was added in blocking buffer and incubated overnight at 4 °C. The next day, cells were washed three times with 1xPBST and incubated with secondary HRP-conjugated antibody for 3 h at 4 °C. The tyramide signal amplification was performed according to the manufacturer's protocol (Tyramide SuperBoost Kits with Alexa Fluor Tyramides, Invitrogen). The cells were then harvested and boiled for 1 h at 99 °C for reverse crosslinking followed by purification of the streptavidin beads affinity before processing for mass spectrometry.
Mass Spectrometry Analysis for BioID Assay Samples
Sample-incubated streptavidin magnetic beads were resuspended in 20 μl 5 mM DTT, 100 mM NH4HCO3 and incubated for 30 min at room temperature. Subsequently, iodoacetamide was added to a final concentration of 7.5 mM and the samples were incubated for 30 additional minutes. 0.5 μg of sequencing grade trypsin (Promega) was added to each sample and incubated at 37 °C overnight. Supernatants of the beads were recovered, and beads were digested again using 0.5 μg trypsin in 100 mM NH4HCO3 for 2 h. Peptides from both consecutive digestions were combined and recovered by solid phase extraction using C18 ZipTips (Millipore), eluted in 15 μl 50% acetonitrile, 0.1% formic acid, and resuspended in 5 μl 0.1% formic acid for analysis by LC-MS/MS.
Peptides resulting from trypsinization were analyzed on a QExactive Plus (Thermo Scientific), connected to a NanoAcquity Ultra Performance UPLC system (Waters). A 15-cm EasySpray C18 column (Thermo Scientific) was used to resolve peptides (90-min 2–30% gradient with 0.1% formic acid in water as mobile phase A and 0.1% formic acid in acetonitrile as mobile phase B). Mass spectrometer was operated in positive ion mode and in data-dependent mode to automatically switch between MS and MS/MS. MS spectra were acquired between 350 and 1500 m/z with a resolution of 70,000. The top 10 precursor ions with a charge state of 2+ or higher over the selected threshold (1.7E4) were fragmented by higher-energy dcollisional dissociation using a normalized collision energy of 25 with an isolation window of 4 m/z. MS/MS spectra were acquired in centroid mode with resolution of 17,500 from m/z = 100. A dynamic exclusion window was applied which prevented the same m/z from being selected for 10 s after its acquisition.
For peptide and protein identification, peak lists were generated using PAVA in-house software (
) with the following parameters: enzyme specificity was set as Trypsin, and up to two missed cleavages per peptide were allowed. Carbamidomethylation of cysteine residues was allowed as fixed modification. N-acetylation of the N-terminus of the protein, loss of protein N-terminal methionine, pyroglutamate formation from peptide N-terminal glutamines, and oxidation of methionine were allowed as variable modifications. Mass tolerance was 10 ppm in MS and 30 ppm in MS/MS. The false positive rate was estimated by searching the data using a concatenated database which contains the original SwissProt database, as well as a version of each original entry where the sequence has been randomized. A 1% false discovery rate was permitted at the protein and peptide level. All spectra identified as matches to peptides of a given protein (Peptide Spectral Matches, PSMs) were reported, and this number is used for label-free quantitation of relative protein abundances in the samples as ratios of each protein PSMs to the average PSMs for that protein on the appropriate control samples (cells transfected with Myc-BirA). Absent values were given a PSM value of 0.5.
Mass Spectrometry Analysis of BAR Assay Samples
On-bead digestion was performed as indicated above. Peptides resulting from trypsinization were analyzed on an Orbitrap Lumos Fusion (Thermo Scientific), connected to a NanoAcquity Ultra Performance UPLC system (Waters). A 15-cm EasySpray C18 column (Thermo Scientific) was used to separate peptides in a 60-min 2 to 30% gradient with 0.1% formic acid in water as mobile phase A and 0.1% formic acid in acetonitrile as mobile phase B. MS was operated in positive ion mode and in data-dependent mode to automatically switch between MS and MS/MS. MS spectra were acquired between 375 and 1500 m/z with a resolution of 120,000. For each MS spectrum, multiple charged ions over the selected threshold (2e4) were selected for MS/MS in cycles of 3 s, with an isolation window of 1.6 Th. Precursor ions were fragmented by higher-energy dcollisional dissociation using a normalized collision energy of 30. MS/MS spectra were acquired in centroid mode with resolution 30,000 from m/z = 110. A dynamic exclusion window was applied which prevented the same m/z from being selected for 30 s after its acquisition. MS data analysis was performed as indicated for BioID samples above, but in the case of data from DRG mouse neurons, the mouse subset (17,009 entries) of the SwissProt.2019.07.31 database was used to search the data.
Experimental Design and Statistical Rationale
BioID and DRG BAR experiments were performed in three biological replicates, while BAR on HEK cells was in duplicates. Control samples used in BioID were Myc-BirA∗ and samples omitting mAbKPNB1-301-320 for BAR experiments. Proteins whose corresponding peptides count was less than two in all replicas were excluded from the analysis in BioID experiments. Proteins whose corresponding peptides were not found in all replicas as well as keratin contaminants were excluded from the analysis in BAR experiments. To assess the statistical significance of protein fold changes between treatment conditions, log2 fold change ratios were averaged across replicates and p-values were calculated using Prism software (GraphPad) applying multiple unpaired t test (two-stage step-up Benjamini, Krieger, and Yekutieli (
Immunofluorescence and Immunohistochemistry on Cryosections
Sensory neurons were grown on glass coverslips coated with Poly-L-lysin and Laminin for 2 days and then fixed using 4% PFA, permeabilized with 0.2% Triton X for 10 min at 25 °C, and blocked for 1 h at room temperature with 5% BSA in 1XPBST. Coverslips with neurons were incubated in mAbKPNB1-301-320 with or without blocking peptide KPNB1301–320, with a ratio of 4.6 nM mAbKPNB1-301-320 to 9.5 nM peptide, overnight at 4 °C. After three five-minute washes in 1xPBS, cells were incubated in donkey anti-mouse FITC secondary antibodies (1:200 each; Jackson Immuno-Res) for 1 h at room temperature. After three 5 min washes in 1xPBS, coverslips were rinsed in ddH2O, then mounted with Prolong Gold Antifade with DAPI.
SNs of adult WT mice were dissected at the designated times after crush injury and noninjured controls, fixed in 4% PFA for 12 h at 4 °C. Then nerves were washed three times in 1XPBST (10 min each), incubated overnight in 30% sucrose and subsequently embedded in OCT freezing medium. The 12 μm thick sections were cut with a cryostat and mounted on glass slides. Freshly cut sections were incubated in blocking solution (5% BSA in 1XPBS) for 2 h at 4 °C and incubated overnight with primary antibody (in 1XPBST with 2% BSA). Primary antibody solution was carefully removed and slides were washed three times (5 min each) at room temperature with 1XPBST. In the next step, the slides were incubated in donkey anti-mouse FITC secondary antibodies (1:1000 each; Jackson Immuno-Res) for 2 h at room temperature. After three 5-min washes in 1XPBST, slides were mounted with Prolong Gold Antifade.
Acquisition and Image Analysis
All fluorescent images including PLA were collected using an Olympus FV1000 Confocal laser-scanning microscope at 60× magnification with oil-immersion objective (Olympus UPLSAPO, NA 1.35) or using Fluoview (FV10i), a fully automated confocal laser-scanning microscope, with water immersion objective (Olympus UPLANSAPO 60X, NA 1.2). Image analysis was performed using Fiji software (
N2A cells were seeded in six-well plates with 120,000 cells per well, one day before siRNA transfection. The cells were transfected with custom siRNA against the 3′UTR of mouse importin β1 (siKPNB1) or siControl #5, using DharmaFECT 4 transfection reagent (Dharmacon) according to the manufacturer’s protocol. Forty eight hours after transfection, the cell growth medium was replaced by serum-free medium to slow down proliferation and to activate the differentiation process. Seventy two hours after transfection, the cells were harvested and total protein was extracted in RIPA buffer containing 0.1% SDS and protease inhibitors. Twenty micrograms of total protein per each sample were run on gel for WB analysis.
Data represent mean ± SEM, unless otherwise noted. Groupwise analyses were conducted by one- or two-way ANOVA with post hoc tests as described in the figure legends. Pairwise analyses were conducted by two-tailed Student’s t tests (unpaired, unless otherwise noted; see Figure Legends). Statistical analyses were conducted using GraphPad Prism, Synergy Kaleidagraph, or Microsoft Excel. Statistically significant p values are shown as ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001 and ∗∗∗∗p < 0.0001. Physical subnetwork analysis was performed using String database version 11.5 (
) to identify all statistically enriched terms, with a p value cut-off of 0.01 and min enrichment of 1.5. The significant terms were hierarchically clustered into a tree diagram, based on Kappa-statistical similarities among their gene memberships. A 0.3 kappa score was applied for term clustering. A selected subset of representative terms from this cluster was converted into a network layout, with each term represented by a circle node and its cluster identity shown in color. Terms with a similarity score >0.3 are linked by an edge (the thickness of the edge represents the similarity score). The network was visualized with Cytoscape (v3.1.2) with a ‘force-directed’ layout and with edge bundled for clarity.
BioID Proximity Labeling of Importin β1 Interactors in HeLa Cells
We first tested BioID for proximity labeling of importin β1 interactomes. The effective labeling range of this method has been estimated as ∼10 nm (
), hence fusions of BirA∗ at the N-terminus of importin β1 might not label NLS-containing cargos that bind the complex via importin α (Fig. 2, A and B). We therefore generated constructs with BirA∗ fused to either N- or C-terminus of the human importin β1 ORF, with a Myc epitope tag and Venus fluorescent protein at the opposite termini (Fig. 2B). These and a control construct comprising only Myc-tagged BirA∗ were expressed in HeLa cells and expression was validated with either Myc or anti-importin β1 antibodies. Cells were pulsed with biotin for 24 h, and cell lysates were subjected to streptavidin pull-downs for mass spectrometric analyses. A total of 223 specific hits were identified (supplemental Table S1), of which 54 hits, primarily nuclear pore complex proteins, were shared by both constructs (Fig. 2, C–E and supplemental Fig. S1). Additional candidates include proteins involved in cell division/spindle assembly, RNA splicing, mRNA metabolism and export from the nucleus, and proteins that participate in nuclear import (supplemental Fig. S1). However, importin α′s were identified only with the C-terminal BirA∗ construct, and the dataset lacked many classical NLS-dependent importin cargos that are known to be expressed in HeLa cells.
In order to determine if NLS-dependent cargos can be efficiently biotin tagged by regular BioID, we designed NLS test constructs featuring either the bipartite NLS of Nucleoplasmin (NLS1) or the monopartite SV40 NLS (NLS2) fused to either the C- or N-terminus of mCherry carrying a FLAG epitope (supplemental Table S2). We cotransfected HeLa cells with these constructs with the Venus-Impβ1-BirA construct and incubated with biotin for 24 h. The importin-β–binding domain (IBB) of Importin α fused to mCherry (IBB-mCherry-FLAG) construct was used as a positive control and a FLAG-mCherry construct as a negative control. Robust expression and nuclear localization of IBB-mCherry and all NLS-mCherry fusions were validated by imaging. To test whether these constructs are biotinylated, cells were lysed, biotinylated proteins were captured by streptavidin pull-down, and eluates were probed with anti-FLAG antibody. The IBB-Cherry-FLAG was biotinylated as expected, but neither of the NLS-Cherry test constructs were labeled (data not shown). Since this might be due to the limited biotinylation range of BirA∗ (
), we generated a series of constructs incorporating linkers of different lengths and flexibility between BirA∗ and importin β1 to bring the BirA∗ enzyme closer to the cargo (supplemental Table S2). These new BioID constructs were cotransfected with NLS1-Cherry-FLAG, and biotinylation of the NLS construct was tested by Streptavidin pull-down and probing of eluates with anti-FLAG antibody. Strikingly, none of the constructs were able to biotinylate the NLS-Cherry-FLAG test construct, whereas the positive control (IBB-Cherry-FLAG) cotransfected with the mVenus-Impβ1-BirA∗ construct underwent robust biotinylation (data not shown). This strongly suggests that steric hindrance and limited range may restrict the efficiency of importin β1–targeted BioID for tagging of endogenous importin cargos.
Tagging an Importin α for BioID instead of importin β1 is not straightforward, since both the N- and C-termini of Importin α are essential for its functions (
). We therefore set out to generate a high-quality mAb for importin β1.
Antibody Generation and Epitope Mapping
Full-length recombinant mouse importin β1 protein was used to immunize mice for mAb production. A conventional immunization protocol was conducted on five animals, and the highest responding mouse was selected after verification of serum titer and specificity (supplemental Figs. S2 and S3). The selected mouse underwent a prefusion boost four days prior to spleen cells collection and hybridoma fusion using standard procedures (
). Cell culture supernatants from ∼960 hybridoma clones were screened for secretion of anti-importin β1 antibodies, yielding 20 positive clones. Further screening of these 20 clones included WB for endogenous KPNB1 in bovine axoplasm (supplemental Fig. S3C), IF in different cell types (HeLa, 3T3, mouse DRG, supplemental Fig. S3D), and ELISA, using linear dilution series (supplemental Fig. S3E). Clones were scored on a scale from 0 (no signal) to 3 (highest signal) in all applications and the top eight clones underwent subcloning, antibody purification, and isotyping (supplemental Fig. S3F). Purified antibody from these eight clones were validated again by ELISA (supplemental Fig. S3G). All these eight antibodies were able to detect importin β1, with some mainly showing nuclear signal, while others revealed a clear preference for cytoplasmic importin β1. The antibody with the highest overall score from those with preferential immunoreactivity for cytoplasmic importin β1 was chosen for further characterization.
We used a custom peptide library screen to identify the epitope recognized by the selected mAb. Fifty eight overlapping peptides with a length of 20 amino acids (aa) each were synthesized to tile the entire amino acid (aa) sequence of murine importin β1, conjugated to N-terminal biotin, and screened in ELISA format on streptavidin-coated 96-well plates (supplemental Fig. S4A). This procedure led to the identification of a 20 aa epitope sequence corresponding to residues 301-320 of human KPNB1, within HEAT repeats 7 and 8. The alignment of the mouse, rat, human, and bovine importin β1 protein sequence shows complete homology in the epitope region (Fig. 3A). We used PyMOL to visualize the identified epitope in the importin β1 structure (
), as shown in Figure 3B. To validate the affinity of the mAb for the identified sequence, we conducted reciprocal linear ELISAs, using either varying peptide dilutions (Fig. 3C) or varying mAb dilutions (Fig. 3D) and observed a linear signal increase until saturation. The mAb effectively recognizes Importin β1 from different species in a number of applications, including ELISA, WB, IP (supplemental Fig. S4), and IF (supplemental Fig. S5). A direct comparison of the mAb performance in IF staining on several cell lines derived from different species is shown in supplemental Fig. S5.
To test the possibility of poor nuclear penetration of the antibody, 3T3 cells were treated with 0.1%, 1%, and 5% Triton. The cells were then simultaneously incubated with the mAb and rabbit anti-Nucleolin, which was used as a positive control for antibody recognition of nuclear proteins (
). The results showed a clear nuclear localization of nucleolin immunoreactivity, but importin β1 signal remained prominently cytoplasmic (supplemental Fig. S6). We also tested the effect of various fixatives and fixation times on mAbKPNB1-301-320 epitope recognition in HEK cells. The results showed that increasing the fixation time and adding 0.1% glutaraldehyde to 4% PFA reduced signal intensity but did not affect cytoplasmic preference in recognition of importin β1 by the mAb (supplemental Fig. S7). We then verified whether the identified peptide epitope can serve as a blocking agent in IP assays. Preincubation of the mAb with the peptide prior to IP from rat brain lysates showed that KPNB1301-320 competes with binding of endogenous full-length importin β1, as demonstrated by reduced signal in the IP and no evident depletion in the supernatants (supplemental Fig. S8). Based on the validated epitope identity, the antibody is henceforth termed mAbKPNB1-301-320.
mAbKPNB1-301-320 Efficiently Recognizes Importin β1 in Sensory Neurons
Locally translated importin β1 is retrogradely transported from axon to soma in injured sensory neurons and is also associated with dynein in growing axons (
). We set out to test mAbKPNB1-301-320 in recognition of importin β1 in sensory axons. Indeed, immunostaining of cultured neurons revealed strong cytoplasmic staining, which was absent in cells incubated in parallel with a blocking peptide comprising the epitope sequence (Fig. 4, A and B). Specific recognition of cytoplasmic and axonal importin-β1 by mAbKPNB1-301-320 was also confirmed by immunogold labeling in electron microscopy sections of cultured sensory neurons (Fig. 4C) and by in situ PLA (supplemental Fig. S9, A and C). The latter method allows the detection of two epitopes, either on the same protein or on two interacting proteins, within a radius of 40 nm (
). Here, we used mAbKPNB1-301-320 together with a commercially available polyclonal anti-KPNB1 to detect importin-β1 in the axons of cultured DRG neurons. Furthermore, we used PLA to validate the ability of mAbKPNB1-301-320 to recognize importin β1 in association with dynein, robustly detecting colocalization of importin β1 and dynein heavy chain (Dync1h1) in the axons of growing DRG neurons (Fig. 4, D and E).
In order to test the ability of mAbKPNB1-301-320 to detect axonal importin β1 upregulation in vivo, we used the mouse model of SN crush injury. Nerves were isolated six hours after crush, fixed, and processed for immunohistochemistry for frozen sections. mAbKPNB1-301-320 detected significant upregulation of importin β1 protein in injured SN axons (Fig. 5, A and B). Local upregulation of importin β1 in axons after SN crush was also detected when using mAbKPNB1-301-320 for immunohistochemistry on paraffin sections (supplemental Fig. S10, A and B). Furthermore, we used PLA to validate the ability of mAbKPNB1-301-320 to detect importin β1 in association with dynein in the SN. PLA signal significantly increased after injury, confirming local translation of importin β1 protein upon nerve injury and its binding to dynein for retrograde transport (Fig. 5, C–E). Little or no background was detected in both injured and noninjured PLA probe and antibody controls (supplemental Fig. S11).
mAbKPNB1-301-320 for BAR Analyses of Importin β1 Cytoplasmic Interactomes
After thorough validation of mAbKPNB1-301-320, we tested its utility in capturing importin β1 interactors using the BAR method on HEK cells (supplemental Fig. S12). A total of 276 importin β1 proximal proteins were identified in this analysis (supplemental Fig. S12). Candidate proteins include nucleocytoplasmic transport proteins and strictly cytoplasmic interactors, such as the translation initiation factor EIF4B and the stress granule protein G3BP1 (supplemental Fig. S12, C and D and supplemental Table S3). Comparison of this dataset to a previously published BioID dataset for KPNB1 in HEK cells (
) revealed approximately 80% novel candidates, with a bias for cytoplasmic components, as compared to more nucleoplasmic representation in the BioID dataset (supplemental Fig. S12B). We then proceeded to use mAbKPNB1-301-320 for BAR analyses on DRG neurons grown for 3 days in culture. Streptavidin staining confirmed that biotinylation occurred in cytoplasm with negligible background signal from the nucleus (Fig. 6A). A total of 432 specific candidates were identified in this experiment (Fig. 6B and supplemental Table S4). String database analysis revealed that ∼28% of these proteins are annotated as of neuron projection origin and ∼18% as axonal proteins. Overall, 93% of the candidates were annotated as cytoplasmic proteins, some of these in overlap with hits annotated as occurring also in the nucleus. This overlap is not surprising for a nucleocytoplasmic transport protein. A number of functionally distinct complexes were identified, including components of the axonal growth cone, the dynein retrograde motor complex, axonal-dendritic transport proteins, Rho GTPase signaling systems, and markers for the cellular leading edge (Fig. 6, C and D, and supplemental Fig. S13). We also identified known importin β1 cargos such as ribosomal proteins, STAT3, and vimentin, supplemental Table S4 (
). These results confirm mAbKPNB1-301-320 as a high-quality probe for importin β1 interactomes.
In this study, we generated and validated a new mAb for importin β1 and demonstrated its utility in proximity biotinylation analyses of importin β1 interactomes using the BAR method. We turned to the BAR approach after initial attempts to use importin β1/BirA∗ fusions for BioID, which identified various nuclear and nucleocytoplasmic transport interactors but revealed poor efficiency for identifying cytoplasmic interactors of the importins complex. The chemistry on which the BAR method is based has a significantly larger labeling range than BioID (
), and the method does not require the introduction of exogenous tags to the protein of interest. These characteristics are advantageous for comprehensive coverage of interactomes of interest, but conversely, they may decrease the specificity of the analysis. Hence, the effective use of BAR is critically dependent on the antibody quality. mAbKPNB1-301-320 was shown to be a highly specific and sensitive importin β1 probe in a broad range of applications, cell types, and species. Mapping localized the epitope to amino acids 301 to 320 in importin β1, and importantly, the linear 20-mer peptide epitope is effective as a blocking reagent in different assays. Visualization of the epitope in the importin β1 crystal structure localizes it to a concave feature within the stalk motif that does not overlap with Ran-binding sites (Fig. 3B) (
). The epitope protrudes by ∼20 Å from the surface of the protein and is clearly accessible to the antibody in cytoplasmic and axonal compartments, but less available within nuclei. Specific functions of the stalk motif are unknown, and the subcellular specificity of our new mAb might suggest a difference in occupancy of this region between nucleus and cytoplasm. The significance of this subcellular specificity for importin β1 functions will be an intriguing topic for future exploration, and the antibody provides an excellent tool for this purpose.
Previous characterizations of importin β1 interactomes used a diversity of approaches, including BioID (
). Different cell types and biological paradigms were used in these studies, complicating direct comparison of those datasets with ours. However, gene ontology analyses allow more comprehensive comparisons, and these are shown for subcellular localization and canonical pathway analyses in Figure 7. Unsurprisingly, both BAR datasets from the present study showed a higher representation of cytoplasmic proteins than the previous datasets, reflecting the subcellular preference of the antibody. This subcellular specificity is also reflected in the signaling pathways identified by BAR versus previous BioID and pull-down approaches (Fig. 7A), highlighting the need for diverse approaches in multiple cell types for truly comprehensive profiling of such interactomes. It should however be noted that the BioID studies conducted with KPNB1 to date have all used first or second generation BirA∗ enzymes, and it is possible that enhanced activity or faster acting variants developed recently (
In summary, we have generated an antibody that discriminates between subcellular conformations of importin β1 and have shown that it provides a highly specific and sensitive probe for cytoplasmic importin complexes. BAR analyses utilizing the new antibody as a targeting agent reveal significant new categories of the importin β1 interactome, with focus on cytoplasmic and signaling proteins that highlight new functional roles of the importins complex that go beyond nucleocytoplasmic transport.
The mass spectrometry proteomics data have been deposited to the ProteomeXchange Consortium via the PRIDE (
) partner repository with the dataset identifier PXD032728. The mAbKPNB1-301-320 antibody will be available through the Ximbio repository (https://ximbio.com/). Data are available via ProteomeXchange with identifier PXD032728.
The authors declare that they have no conflicts of interest with the contents of this article.
We sincerely thank Prof. Zelig Eshhar and Tova Waks for their advice on monoclonal antibody generation; Vladimir Kiss for assistance with microscopy; Dr. Reinat Nevo for software expertise on image analysis; Dr. Vera Shinder and Dr. Eyal Shimoni for assistance with electron microscopy; and Dr. Dalia Gordon for helpful discussions. Electron microscopy studies were conducted at the Irving and Cherna Moskowitz Center for Nano and Bio-Nano Imaging at the Weizmann Institute of Science.
Funding and additional information
We gratefully acknowledge funding from the Israel Science Foundation (ISF 1337/18 to M. F. & I. R.), and the Dr. Miriam and Sheldon G. Adelson Medical Research Foundation (to A. L. B. & M. F.). C. A. A. was supported by a long-term fellowship from the Human Frontier Science Program Organization. M. F. is the incumbent of the Chaya Professorial Chair in Molecular Neuroscience at the Weizmann Institute of Science .
D.-A. S., S. A., E. D.-M., C. A. A., M. F., and I. R. methodology; D.-A. S., S. A., E. D.-M., V. S., C. A. A., O. L., H. H., J. A. O.-P., N. D., and I. R. investigation; D.-A. S., S. A., E. D.-M., V. S., C. A. A., O. L., H. H., N. D., and I. R. formal analysis; A. L. B., M. F., and I. R. supervision; D.-A. S., S. A., E. D.-M., M. F., and I. R. writing–original draft; D.-A. S., S. A., E. D.-M., V. S., C. A. A., O. L., H. H., J. A. O.-P., N. D., A. L. B., M. F., and I. R. writing–review and editing.