Importin 13 (IPO13) is a key member of the importin β superfamily, which can transport cargoes both into and out of the nucleus to contribute to a variety of important cellular processes. IPO13 is known to undergo phosphorylation, but the impact of this on function has not been investigated. Here, we show for the first time that IPO13 is phosphorylated by cAMP-dependent protein kinase A specifically at serine 193. Results from fluorescence recovery after photobleaching and fluorescence loss in photobleaching approaches establish that negative charge at serine 193 through phosphorylation or point mutation both reduces IPO13 nuclear import and increases its nuclear export. Importantly, phosphorylation also appears to enhance cargo interaction on the part of IPO13, with significant impact on localization, as shown for the Pax6 homeobox-containing transcription partner. This is the first report that IPO13 can be phosphorylated at Ser193 and that this modification regulates IPO13 subcellular localization and nucleocytoplasmic transport function, with important implications for IPO13's role in development and other processes.
Nucleocytoplasmic transport of macromolecules is integral to eukaryotic cell function, with a central role played by members of the importin/karyopherin superfamily of transporters that, in concert with the monomeric guanine nucleotide-binding protein Ran in activated GTP-bound form, carry cargoes into and out of the nucleus through the nuclear envelope-embedded nuclear pore complexes [1,2]. A key member of this family, Importin 13 (IPO13, imp13, lgl2 or Kap13), was originally identified as a hormonally and developmentally regulated transporter in fetal rat lung [3,4]; importantly, it has since been shown to mediate both nuclear import and export , with many specific import and export cargoes thus far identified; these include factors with important roles in organ development (Arx, NF-YB/C heterodimer, Pax6, ubiquitin-conjugating enzyme 9) and hormone regulation (e.g. the glucocorticoid receptor) [5–14]. IPO13 adopts an open conformation in the cytoplasm that enables it to interact with import cargo and transport it into the nucleus, where cargo release is effected by RanGTP binding to IPO13. The nuclear IPO13–RanGTP complex has a more closed conformation, which enables the recognition of export cargoes such as eIF1A, followed by transport to the cytoplasm; hydrolysis of GTP by Ran is facilitated by RanGTPase-activating protein 1 (RanGAP1), effecting Ran release from IPO13, with association of new import cargoes completing release of the export cargo from IPO13 [5,15].
Although phosphorylation of nuclear transport cargoes has been known for many years to play an important role in regulating nucleocytoplasmic transport in response to signaling [16–26] and cell cycle-dependent phosphorylation of Crm1 (exportin1) is known to promote Ran-dependent mitotic spindle assembly , little is known regarding phosphorylation of importin family members themselves, and the functional impact thereof on their nucleocytoplasmic transport function. One exception is ERK-dependent phosphorylation of exportin-5 (XPO5) that suppresses pre-miRNA export from the nucleus by reducing pre-miRNA loading .
IPO13 is known to be phosphorylated, but the functional impact thereof has never been assessed. Here, we address this directly for the first time, showing that Serine 193 (S193) is a target of specific phosphorylation by the cAMP-dependent protein kinase A (PKA). PKA phosphorylation increases IPO13 cytoplasmic localization through increasing its export and slowing its nuclear import activities. It also enhances IPO13 interaction with nuclear transport cargoes such as Pax6, with resultant impact on cargo localization. The results have important implications for IPO13's role in development and other processes.
Materials and methods
Plasmids encoding IPO13 with enhanced yellow fluorescence protein (EYFP) tag and glucocorticoid receptor (GR) with GST tag have been described [12,29]. To generate a plasmid encoding FLAG-tagged IPO13, the rat IPO13 coding sequence was PCR-amplified and inserted into the EcoRI/Xho restriction sites of plasmid vector pCMV-3×FLAG. To generate a plasmid encoding EGFP-tagged N-terminal IPO13 (amino acids 1–307, N-IPO13), the appropriate PCR-amplified sequence was inserted into the XhoⅠ/BamHI restriction sites of plasmid vector pEGFP-C1. Plasmid encoding mCherry-tagged Pax6 was constructed by inserting the sequence encoding full-length human Pax6 into the HindIII/BamHI restriction sites of plasmid vector mCherry-C1. Plasmids expressing GST-tagged rat full-length and N-terminal IPO13, human eIF1A, truncated human Pax6 (amino acids 208–288, M-Pax6) and C-terminal human Arx (amino acids 288–564, C-Arx) were generated by inserting PCR-amplified sequences into the BamHⅠ/XhoⅠ restriction sites of plasmid vector pGEX-4T-2.
Site-directed mutagenesis for the Ser193 site in IPO13 was performed by QuikChange PCR as described previously , with plasmid IPO13-EYFP  as a template and using mutagenic primers: A193-F 5′- GGTCTGGTACGGGCCGCCCTGGCGGTG-3′ and A193-R 5′- GGCGGCCCGTACCAGACCTTTGCGGTA-3′; and D193-F 5′- GGTCTGGTACGGGCCGACCTGGCGGTG-3′ and D193-R 5′- GTCGGCCCGTACCAGACCTTTGCGGTA-3′. Mutations were confirmed by DNA sequencing.
Cell culture and transfection
Cells of the human HeLa cervical cancer, 293T embryonic kidney lines and B104 rat neuroblastoma were cultured and maintained in Dulbecco's modified Eagle medium containing 10% fetal bovine serum and 1% penicillin–streptomycin in a 5% CO2 humidified incubator at 37°C, respectively. They were transfected using polyethyleneimine or Lipofectamine® 3000 (#L3000015, Invitrogen, U.S.A.) according to the manufacturer's specifications. Preparation of cell lysates was as described previously .
Protein expression and purification
Expression in Escherichia coli and purification of GST fusion proteins were as described previously . Cells were grown at 37°C to an optical density of 0.8 at 600 nm and then induced by the addition of 1 mM isopropyl-β-d-thiogalactoside for 4 h at 25°C (or at 37°C for C-Arx-GST). Cells were subsequently collected and lysed by sonication. Expressed proteins were purified using glutathione–Sepharose 4B beads (#17-0756-05 GE Healthcare, U.S.A.) according to the manufacturer's instructions. In the case of protein for in vitro phosphorylation, eluted protein was desalted by dialysis and the concentration was determined using a BCA kit (#23225, Thermo, U.S.A.) according to the manufacturer's instructions.
Generation of S193 phospho-specific antibody (anti-pS193)
The peptides RKGLVRAS(p)LAVEC and RKGLVRASLAVEC were synthesized by ABclonal Biotech (Wuhan, China) and injected into rabbits. Serum was harvested after 40 days and added to a sepharose column with the RKGLVRAS(p)LAVEC peptide precoupled to it using the SulfoLink Couping Gel (#20401, Thermo, U.S.A.). The column was washed with wash buffer [320 mM NaCl in 1× PBS (phosphate-buffered saline)] and eluted with elution buffer (0.1 M glycine–HCl, pH 2.8). The eluate was pH-neutralized and added to a sepharose column precoupled with the RKGLVRASLAVEC peptide, and the flow-through was collected and established to contain the purified anti-pS193 antibody.
Dot blots were performed as previously described , with nitrocellulose membranes blocked in 5% bovine serum albumin (BSA) for 1 h at room temperature prior to incubation with the pS193 antibody (1 : 2000), followed by goat anti-rabbit horseradish peroxidase (HRP) secondary antibody, and detection of binding using the enhanced ECL agent (#NCI5080, Thermo, U.S.A.).
SDS–PAGE and Western analysis
Samples were separated by SDS–polyacrylamide gel electrophoresis (PAGE) and stained with Coomassie brilliant blue R-250 or transferred to a PVDF membrane (#IPVH00010, Millipore) for Western analysis . The membrane was blocked with 5% milk or BSA for 1 h at room temperature before incubation with either anti-FLAG (#F1804, Sigma, 1 : 5000), anti-GST (#sc-138, Santa Cruz, 1 : 5000), anti-pS193 (1 : 2000), anti-GFP (#sc9998, Santa Cruz, 1 : 2000) or anti-pSer (#sc81514, Santa Cruz, 1 : 1000) antibody followed by the corresponding HRP-coupled secondary antibodies, respectively.
In vitro phosphorylation
In vitro phosphorylation was performed as previously described [31–33]. Recombinant IPO13- or IPO13:A193-GST or peptide substrates were phosphorylated using 90 mU/reaction of purified PKA-C subunit (#14-440, Millipore, U.S.A.) or lysate from HeLa cells expressing PKA-C-HA for 30 min at 30°C in assay buffer [8 mM MOPS (pH 7), 10 mM MgCl2 and 0.2 mM EDTA] containing 0.1 mM ATP and 1 µCi 32P ATP. Reactions were either stopped for SDS–PAGE by boiling and followed by phosphoimaging on the Typhoon Trio, or spotting onto Whatman P-81 phosphocellulose paper and washed in 0.5% phosphoric acid for quantitative analysis by scintillation counting as previously described .
Co-immunoprecipitation was performed as previously described . Lysates from HEK293T cells expressing FLAG-IPO13 with PKA-C-HA were incubated with anti-HA antibody, and mixtures were rotated slowly for 3 h at 4°C. Protein A/G agarose (#2042, Thermo, U.S.A.) was added to mixtures and then mixtures were rotated for another 2 h at 4°C. Beads were collected by centrifugation at 250 g for 2 min and then washed in lysis buffer three times, prior to boiling in SDS loading buffer for SDS–PAGE/Western analysis.
GST pull-down assay
GST pull-down was performed as described previously . Bacterial cells expressing GST-tagged proteins were sonicated in PBS, and supernatants were incubated with Glutathione–Sepharose 4B beads. Lysates from HEK293T cells expressing FLAG-IPO13 with or without PKA-C-HA were incubated with beads for 3 h at 4°C. Washed beads were boiled in SDS loading buffer for SDS–PAGE/Western analysis.
Immunostaining was performed as described previously . Coverslips were incubated with anti-FLAG (#F1804, Sigma, 1 : 1000), anti-HA (#sc-805, Santa Cruz, 1 : 500), anti-IPO13 (#NBP2-14994, Novus, 1 : 100) or anti-pS193 (1 : 100) primary antibody for 3 h at room temperature, followed by an Alexa 488-labeled goat anti-mouse (#ab150076, abcam, 1 : 1000) or Alexa 568-labeled goat anti-rabbit (#ab175471, abcam, 1 : 1000) secondary antibody for 1 h at room temperature, and subsequent staining with 4′,6-diamidino-2-phenylindole (DAPI). Where indicated, B104 cells were preincubated with 10 µM epinephrine .
Confocal laser scanning microscopy and image analysis
Live cell imaging was performed as described previously  at 100× using a FluoView™ FV1000 Confocal Microscope or a Nikon C1 invert Confocal Microscope equipped with an FCS2 live cell chamber maintained at 37°C (Bioptechs, Butler, PA, U.S.A.). Multiple cell images for each sample/condition were subjected to image analysis [14,35] to determine the nuclear to cytoplasmic fluorescence ratio (Fn/c), calculated according to the following equation Fn/c = (Fn − Fb)/(Fc − Fb), where Fn is the nuclear fluorescence, Fc is the cytoplasmic fluorescence and Fb is the background on digitized images using the ImageJ FIJI software. Where indicated, HeLa cells were preincubated with 50 µM forskolin and 0.5 mM IBMX before imaging [36,37].
Fluorescence recovery after photobleaching and fluorescence loss in photobleaching
Fluorescence recovery after photobleaching (FRAP) experiments were performed as previously described [14,29,35,38,39]; briefly, HeLa cells expressing IPO13-EYFP or IPO13:A193-EYFP were imaged live using an Olympus Fluoview 1000 confocal laser scanning microscopy (CLSM) (100× oil immersion objective, 37°C). The area of the nucleus was photobleached (12 scans, 12.5 µs/pixel, 100% laser power) and the cells were immediately scanned at 5% laser power (10 µs/pixel) at 20 s intervals for up to 460 s to monitor changes of both the bleached area of the nucleus and the adjacent cytoplasm. The Fn/c was determined and the data were subjected to exponential curve-fitting to determine the fractional recovery of Fn/c (FrecFn/c), t1/2 and the initial rate, which is expressed as the change in Frec (Fn/c)/s for the linear portion of the recovery curve [14,40].
Fluorescence loss in photobleaching (FLIP) experiments were similar, except that bleaching was repeated after acquiring each image at 15 s intervals for up to 300 s [29,41]. The relative level of nuclear fluorescence above background was analyzed, followed by exponential curve-fitting of the data to determine the fractional loss of Fn − b (FrecFn − b) and t1/2 with the initial rate, which is expressed as a change in Frec (Fn − b)/s for the linear portion of the loss curve [14,29,41].
Statistical analysis was performed using Student's t-test using the GraphPad Prism 7 software.
PKA can phosphorylate IPO13 at Ser193
Although the impact on function has not been addressed, mass spectrometric analysis indicates that IPO13 is phosphorylated at four distinct serine/threonine (S/T) sites (http://www.phosphosite.org — see Table 1). Significantly, two of these are in the N-terminal region of IPO13, which is the binding site for several cargoes as well as RanGTP [6,42], in addition to being responsible for IPO13 nuclear localization (Figure 1A) . Since the S193 phosphorylation site is absolutely conserved in human, mouse and rat IPO13 (see Table 1), it appeared to be a prime candidate for the observed phosphorylation. Based on Phosphomotif Finder (http://www.hprd.org/PhosphoMotif finder), S193 is predicted to be a target for specific phosphorylation by cAMP-dependent PKA, and hence, we tested if purified PKA catalytic subunit (PKA-C) could phosphorylate the N-terminal region of IPO13 expressed in bacteria as a GST fusion protein. Phosphorimaging revealed that 32Pi was incorporated into N-terminal IPO13-GST, as well as a Histone H1-GFP fusion protein positive control, but not the GFP alone negative control, implying that PKA is able to phosphorylate N-terminal IPO13 specifically (Figure 1B).
N-terminal IPO13 can be phosphorylated by PKA at S193 both in vivo and in vitro.
|Residues of IPO131||Human||Mouse||Rat|
|Residues of IPO131||Human||Mouse||Rat|
Sites of Ser/Thr phosphorylation detected in importin13 (IPO13) by mass spectrometry. Results are from phosphosite (www.phosphosite.org); the single letter amino acid code is used.
To examine the Ser193 site in more detail, we used site-directed mutagenesis to substitute Ala for Ser193 in full-length IPO13-GST (IPO13-GST), and the resultant protein, IPO13:A193-GST, was expressed in bacteria along with IPO13-GST. After purification, the proteins were subjected to phosphorylation using the purified PKA-C subunit, as well as lysate from HeLa cells transfected to ectopically express HA-tagged PKA-C (PKA-C-HA). IPO13-GST protein was phosphorylated by both the purified PKA-C subunit and the cell lysates to a 50% higher extent than IPO13:A193-GST. No 32Pi incorporation was observed for GFP control (Figure 1C). The level of 32Pi incorporation of IPO13:A193-GST was comparable to that for N-terminal IPO13-EGFP, implying that phosphorylation site(s) additional to Ser193 may be present in IPO13 that are outside the IPO13 N-terminus (i.e. within residues 308–963). HEK293T cells transfected to co-express FLAG-IPO13 and PKA-C-HA were subjected to immunoprecipitation using an antibody specific to HA (Figure 1D); Western analysis revealed the co-immunoprecipitation of FLAG-IPO13 with PKA-C-HA, as well as confirming IPO13 phosphorylation using an antibody to phosphoserine.
To further validate S193 as a phosphorylation site, an antibody specifically detected at the phosphorylated S193 site (anti-pS193) was generated (see Supplementary Figure S1). IPO13-GST or IPO13:A193-GST was phosphorylated in vitro by pure PKA-C subunit, with phosphorylated IPO13-GST, but not IPO13:A193-GST able to be detected using the anti-pS193 antibody (Figure 2A). To confirm that Ser193 of IPO13 can be specifically phosphorylated by PKA in a cell context, IPO13-EYFP or IPO13:A193-EYFP was co-expressed with PKA-C-HA in transfected HEK293T cells, with the anti-pS193 antibody able to detect a band in cell lysates for IPO13-EYFP but not for IPO13:A193-EYFP (Figure 2B). Finally, to confirm that S193 phosphorylation was PKA-specific, FLAG-IPO13 was co-expressed with either PKA-C-HA or FLAG-tagged Ca2+ phospholipid-dependent kinase (PKC-FLAG) in HeLa cells, and cell lysates were analyzed for phosphorylation using the anti-pS193 antibody. As shown in Figure 2C, pS193 antibody detected FLAG-IPO13 in the presence of co-expressed PKA but not PKC, consistent with IPO13 S193 being a specific target of PKA. This result is consistent with the idea that PKA can phosphorylate IPO13 Ser193 in a cellular context. That S193 phosphorylation could be reduced by 1 h treatment with the PKA-specific inhibitor H89 (Figure 2C) implies that phosphorylation at the site is dynamic, and strongly dependent on the relative phosphorylation/dephosphorylation activities within the cell.
Confirmation of PKA-dependent phosphorylation of IPO13 phosphorylation at Ser193 using a phosphosite-specific antibody.
PKA activation regulates nucleocytoplasmic distribution of IPO13
To test for effects of S193 phosphorylation on IPO13 subcellular localization, FLAG-IPO13 and PKA-C-HA were co-expressed in HeLa cells, and the cells were fixed and immunostained 24 h later for microscopic visualization. As evident in Figure 3A, wild-type IPO13 is predominantly nuclear in the absence, but cytoplasmic in the presence, of PKA-C-HA, quantitative analysis indicating that whereas >90% of cells have predominantly nuclear IPO13 in the absence of co-transfected PKA, this falls to <40% in its presence (Figure 3B). Immunostaining using the anti-Ser193 antibody revealed that phosphorylation in the presence of PKA correlated strongly with cytoplasmic IPO13 (Figure 3A—bottom merge panels); >85% of cells treated with FSK/IBMX showed predominantly cytoplasmic staining for phospho-Ser193-IPO13 (not shown), consistent with the idea that phosphorylation of Ser193 may induce relocation of IPO13 from the nucleus to the cytoplasm. To begin to confirm this in a dynamic context, transfected HeLa cells expressing IPO13-EYFP were analyzed live in the absence or presence of treatment with a combination of the adenylate cyclase activator forskolin (FSK) and phosphodiesterase inhibitor 3-isobutyl-1-methylxanthine (IBMX) to raise intracellular levels of cAMP [36,37,43]. As shown in Figure 3C, IPO13-EYFP nuclear localization was reduced within 15 min of treatment, quantitative analysis to determine the nuclear to cytoplasmic fluorescence ratio (Fn/c) confirming the significantly (P < 0.05) reduced nuclear accumulation (Fn/c of ∼1.2 at 30 min post-treatment, compared with an initial Fn/c of 2; see Figure 3D). In contrast, no significant change in IPO13 localization was observed in the presence of the PKC activator phorbol-12-myristate-13-acetate (PMA)  (Figure 3E,F). These results imply that specific activation of PKA can induce rapid cytoplasmic relocalization of IPO13 from the nucleus to the cytoplasm, and that this is maintained for ∼40 min.
Active PKA drives dynamic relocalization of IPO13 to the cytoplasm.
Negative charge at Ser193 reduces IPO13 nuclear localization
To confirm the role of Ser193 in IPO13 relocation in response to PKA activation, IPO13-, IPO13:A193- or IPO13:D193-EYFP were expressed in HeLa cells followed by CLSM imaging 24 h later, revealing reduced nuclear accumulation of IPO13:D193-EYFP compared with IPO13-EYFP or IPO13:A193-EYFP (Figure 4). Quantitative analysis confirmed significantly (P < 0.05) almost 30% reduced nuclear accumulation (Figure 4B), consistent with the idea that negative charge at Ser193, normally provided by PKA phosphorylation, reduces IPO13 nuclear accumulation; that the effects for the D193 phosphomimetic IPO13 construct are similar but not identical with those for IPO13 in the presence of ectopically expressed PKA implies that phosphorylation of Ser193 probably impacts IPO13 localization not only through charge effects, but potentially also through induced conformational change.
Negative charge at Ser193 reduces nuclear accumulation of IPO13.
Phosphorylation regulates IPO13 nuclear import and export
To investigate the role of Ser193 phosphorylation in regulating the dynamics of IPO13 trafficking, we first applied the FRAP technique as previously described [14,35,38,39] to measure the kinetics of nuclear import of IPO13-EYFP in living cells. IPO13-EYFP was expressed in HeLa cells with or without FSK/IBMX treatment, and then nuclear fluorescence bleached using a high-power laser as described in Materials and Methods. The recovery of nuclear fluorescence was then monitored over time by CLSM to assess effects on the nuclear import rate (Figure 5A), and quantitative analysis/curve-fitting performed as per Figure 5; pooled data for multiple samples are presented in Figure 5C. In the presence of FSK/IBMX, nuclear import of IPO13-EYFP was significantly reduced (Figure 5A,B), the pooled data revealing significantly (P < 0.05) reduced maximal fractional recovery and the rate of nuclear import in terms of both t1/2 (almost 50% increase and the initial rate upon FSK/IBMX treatment; Figure 5C). Clearly, PKA activation results in reduced/slowed nuclear import of IPO13.
PKA activation reduces the kinetics of IPO13 nuclear import.
IPO13 nuclear export activity was analyzed using the FLIP approach essentially as previously described  to assess IPO13 nuclear export. HeLa cells expressing IPO13-EYFP or IPO13:A193-EYFP were treated with or without FSK/IBMX, and then cells were imaged by CLSM as the cytoplasm is bleached continuously using a high-power laser (Figure 6A). FSK/IBMX treatment increased IPO13 nuclear export of IPO13 (Figure 6A,B), with quantitative analysis revealing a significantly (P < 0.01) faster nuclear export rate (30% lower t1/2 — Figure 6C). The clear implication is that PKA activation facilitates IPO13 nuclear export. In contrast with IPO13-EYFP, IPO13:A193-EYFP showed a negligible change in the absence or presence of FSK/IBMX, implying that Ser193 is the key PKA site within IPO13 regulating nuclear export.
Ser193 phosphorylation by PKA increases nuclear export of IPO13.
Phosphorylation enhances IPO13–cargo interaction: impact on nucleocytoplasmic distribution
To investigate the impact of PKA phosphorylation of IPO13 on cargo recognition, M-Pax6- and eIF1A-GST, together with GST alone as a control, were expressed in bacteria and subsequently purified as described in Materials and Methods and then immobilized on Glutathione–Sepharose 4B beads. Lysates from HEK293T cells co-expressing FLAG-IPO13 with PKA-C-HA were then incubated with the beads, followed by Western analysis of the bound eluates. Figure 7A shows that the amount of IPO13 bound to M-Pax6 or eIF1A [5,11] was markedly increased in the presence of PKA-C-HA. Similar results were observed for other IPO13 cargoes such as the GR and C-ARX [8,12] (Supplementary Figure S2A). Interaction between IPO13 and RanGTP, in the form of the RanQ69L derivative , however, was not altered (Supplementary Figure S2B). Thus, PKA phosphorylation of IPO13 appears to enhance the interaction of IPO13 with its cargoes, but not RanGTP.
PKA phosphorylation enhances IPO13 interaction with cargoes, resulting in reduced nuclear localization.
The impact of PKA phosphorylation/increased cytoplaspmic localization of IPO13 on nucleocytoplasmic distribution of Pax6 was assessed in HeLa cells co-expressing mCherry-tagged full-length human Pax6 with IPO13-EYFP or IPO13:A193-EYFP without or with treatment with FSK/IBMX. Strikingly, cytoplasmic accumulation of Pax6 was increased in the presence of IPO13-EYFP and FSK/IBMX treatment (Figure 7B), quantitative analysis indicating significantly (P < 0.05) ∼25% lower nuclear accumulation (Figure 7C). In contrast, FSK/IBMX treatment had only a modest effect on Pax6 in the presence of co-expressed IPO13:A193-EYFP. In cells not expressing IPO13-EYFP, there was no effect of FSK/IBMX treatment on Pax6 nucleocytoplasmic distribution (Supplementary Figure S3), consistent with the idea that the observed effects in the presence of ectopically expressed IPO13 relate to phosphorylation of IPO13 rather than of Pax6. The results strongly imply that Ser193 phosphorylation of IPO13 reduces its availability for Pax6 nuclear import, by increasing IPO13 nuclear export activity and hence cytoplasmic localization. To begin to assess whether PKA phosphorylation might affect IPO13-mediated nuclear export, the subcellular localization of elongation initiation factor eIF4G, a known cargo of IPO13 , was assessed as a GFP fusion protein in the presence or absence of ectopically expressed FLAG-IPO13 without or with PKA-C-HA (Supplementary Figure S4). All combinations showed identical strongly cytoplasmic localization of eIF4G2, quantitative analysis (not shown) indicating the lack of a significant difference in nucleocytoplasmic trafficking in the presence or absence of PKA. The implication is that phosphorylation of IPO13 may predominantly affect nuclear import but not export activity of the transporter, although this will obviously require detailed analysis in the future.
Hormonal activation of PKA can drive relocalization of endogenous IPO13 to the cytoplasm in B104 cells
To confirm the physiological relevance of the above observations, we decided to assess the effect on localization and phosphorylation of IPO13 in B104 rat neuroblastoma cells in response to treatment with the hormone epinephrine, which triggers activation of adenylate cyclase through Gs [34,45–47]; cells were then fixed and immunostained for total IPO13. As shown in Figure 8A, endogenous IPO13 is predominantly nuclear in B104 cells, but is markedly more cytoplasmic after epinephrine treatment; this was confirmed by quantitative analysis, indicating significantly (P < 0.01) reduced nuclear localization of endogenous IPO13 (Fn/c of ∼1.2 in the presence of, compared with a value of 2 in the absence of, epinephrine treatment; see Figure 8B). Importantly, Ser193 phosphorylation in this experimental system could be confirmed by Western analysis using the pS193 antibody on immunoprecipitates from cells treated with or without epinephrine (Supplementary Figure S5). These results indicate that the observations for ectopically expressed IPO13 above also relate to endogenous IPO13, the localization and presumably function of which can be regulated by PKA activation in response to hormonal stimulation.
Hormonal PKA activation can drive relocalization to the cytoplasm of endogenous IPO13.
This is the first study to demonstrate specific phosphorylation of IPO13 by PKA, resulting in altered nucleocytoplasmic shuttling, increased cytoplasmic localization and the impact thereof on nuclear import cargoes. In brief, phosphorylation of IPO13 by PKA at serine 193 both reduces the nuclear import kinetics and enhances the nuclear export of IPO13. Ser193 phosphorylation also increases interaction with nuclear import and export cargoes of IPO13, the end result being increased cytoplasmic distribution of cargoes, such as the homeobox protein Pax6. The physiological relevance of our study is indicated by the important finding that endogenous IPO13 can be induced to relocalize to the cytoplasm upon hormonal stimulation of B104 neuroblastoma cells.
It has proved difficult to examine the role of phosphorylation/post-translation modification in regulating the activity of karyopherin β superfamily members, since recombinant karyopherins expressed in bacteria generally appear to be fully active in in vitro transport [5,8,48,49]. Thus far, SUMOylation of yeast Kap114 has been shown to promote dissociation of Kap114/cargo import complexes , ERK phosphorylation of exportin 5 by ERK has been reported to impair pre-miRNA loading, resulting in global miRNA down-regulation , and Crm1 phosphorylation is known to enhance Ran-dependent interaction between Crm1 and RanGAP1–RanBP2, thereby promoting mitotic spindle assembly in mitosis . The present results for regulation of IPO13 transport function by PKA represent a new finding with important implications. Although it is not possible at this stage to discount the possibility that phosphorylation sites other than Ser193 in IPO13 (see Table 1) could play a role, the previous observations of regulated IPIO13 nucleocytoplasmic trafficking in lung development  may well relate directly to our results here, whereby the increased nuclear import rate of IPO13 observed in rat embryonic lung from D18 to 21  correlates with decreased PKA activity.
Here, we show for the first time that PKA phosphorylation of Ser193 increased IPO13 association with cargoes such as Pax6 and eIF1A. Ser193 is located just upstream of HEAT repeat 5 of IPO13, which is known to play an important role in eIF1A recognition . It thus seems likely that Ser193 phosphorylation may modulate conformation in this region of IPO13 to facilitate eIF1A interaction, and potentially interaction with the other cargoes examined here, possibly by the same mechanism. Interestingly, Pax6 nucleocytoplasmic transport has been suggested to be regulated by glucose concentration ; a possibility in this context is that this may relate to the cAMP/PKA pathway [53,54], with resultant altered IPO13 transport activity largely responsible. Detailed examination of this intriguing possibility is a focus of future work in this laboratory.
enhanced yellow fluorescence protein
fluorescence loss in photobleaching
fluorescence recovery after photobleaching
polyacrylamide gel electrophoresis
protein kinase A
PKA catalytic subunit
RanGTPase-activating protein 1
X.L. performed the experimental studies shown in the manuscript under the supervision of D.A.J. and T.T. W.L. and X.S. developed the pSer193 phospho-specific antibody, and R.G.D. and K.M.W. generated some of the plasmid constructs used. X.L., T.T. and D.A.J. designed the studies, analyzed the experiments and wrote the paper. All the authors reviewed the results and approved the manuscript.
The authors acknowledge funding support from the National Natural Science Foundation of China [#31270810] and the Ministry of Science and Technology of China [#2013CB910803] to T.T. and from the National Health and Medical Research Council (Australia; project grant APP1022206 and Senior Principal Research Fellowship APP1103050 to D.A.J.) and the Chinese Scholarship Council (scholarship to X.L.).
The authors acknowledge the assistance of the Monash Micro Imaging Facility and the Monash Institute of Medical Research, Monash University, Clayton.
The Authors declare that there are no competing interests associated with the manuscript.