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
* This work was supported by National Institutes of Health Grant HL074324. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. ‡ Present address: Dept. of Biochemistry and Molecular Biology, University of Massachusetts, Amherst, MA 01003.
At low concentrations, hydrogen peroxide (H2O2) is a positive endogenous regulator of mammalian cell proliferation and survival; however, the signal transduction pathways involved in these processes are poorly understood. In primary human endothelial cells, low concentrations of H2O2 stimulated the rapid phosphorylation of the acidic C-terminal domain (ACD) of heterogeneous nuclear ribonucleoprotein C (hnRNP-C), a nuclear restricted pre-mRNA-binding protein, at Ser240 and at Ser225–Ser228. A kinase activity was identified in mouse liver that phosphorylates the ACD of hnRNP-C at Ser240 and at two sites at Ser225–Ser228. The kinase was purified and identified by tandem mass spectrometry as protein kinase CK1α (formerly casein kinase 1α). Protein kinase CK1α immunoprecipitated from primary human endothelial cell nuclei also phosphorylated the ACD of hnRNP-C at these positions. Pretreatment of endothelial cells with the protein kinase CK1-specific inhibitor IC261 prevented the H2O2-stimulated phosphorylation of hnRNP-C. Utilizing phosphoserine-mimicking Ser-to-Glu point mutations, the effects of phosphorylation on hnRNP-C function were investigated by quantitative equilibrium fluorescence RNA binding analyses. Wild-type hnRNP-C1 and hnRNP-C1 modified at the basal sites of phosphorylation (S247E and S286E) both avidly bound RNA with similar binding constants. In contrast, hnRNP-C1 that was also modified at the CK1α phosphorylation sites exhibited a 14–500-fold decrease in binding affinity, demonstrating that CK1α-mediated phosphorylation modulates the mRNA binding ability of hnRNP-C.
At low concentrations, hydrogen peroxide (H2O2) functions as an endogenous positive modulator of mammalian cell proliferation and cell survival (reviewed in Refs.
). Application of low concentrations (<10 μm)ofH2O2 to cultured mammalian cells stimulates cell growth and enhances survival, whereas the application of catalase inhibits proliferation and stimulates cell death. In mammalian systems, H2O2 is generated by plasma membrane NADPH oxidases in response to growth factor stimulation (
). Many of the downstream effects of growth factors are in fact inhibited by treatment of the cells with catalase. In contrast to the effects of low concentrations of H2O2, the application of higher concentrations (>20 μm) of H2O2 to cultured cells results in oxidative stress and generally inhibits cell proliferation and stimulates cell death. Although several stress-responsive pathways have been implicated in the response of mammalian cells to oxidative stress, the biochemical mechanisms by which low concentrations of H2O2 stimulate cell proliferation and enhance survival are poorly understood (
). Murine stem cells lacking hnRNP-C1/C2 are viable, but show impaired survival and decreased rates of proliferation and differentiation. Interestingly, heterologous expression of the gene for hnRNP-C1/C2 in yeast cells, which normally lack this gene, is also lethal (
). In this latter case, it appears that heterologously expressed hnRNP-C1/C2 translocates to the yeast nucleus and binds mRNA. However, yeast cells appear to lack the ability to stimulate the release of hnRNP-C1/C2 from the mRNA, resulting in inhibition of mRNA export from the nucleus.
Structurally, hnRNP-C1/C2 isolated from HeLa cells is a heterotetramer (C13C2) in which C1 and C2 are splice variants of the same gene, differing by the presence of an additional 13 amino acids in C2 (
). Each hnRNP-C subunit has four functional domains. There is an N-terminal RNA recognition motif (residues 8–87), followed by a basic high affinity RNA-binding domain (residues 140–179), which may be responsible for much of the affinity of the protein for RNA (
). Previous tandem mass spectrometry phosphate mapping studies utilizing endogenously phosphorylated protein from human endothelial cells revealed that, under basal conditions, hnRNP-C1/C2 is phosphorylated at Ser247 and Ser286 (
). However, the kinase responsible for the H2O2-stimulated phosphorylation has been unclear, as has the effect of this phosphorylation on the function of hnRNP-C. Here, we report that protein kinase CK1α (formerly casein kinase 1α), a kinase implicated in regulating cell cycle progression and pre-mRNA processing (
), mediates the H2O2-stimulated phosphorylation of hnRNP-C. Furthermore, we show that this CK1α-mediated phosphorylation modulates the RNA binding activity of hnRNP-C.
Expression and Purification of hnRNP-C ACDs—A DNA segment containing residues 217–293 of hnRNP-C1 was amplified from a nearly full-length clone of hnRNP-C1 (Invitrogen) using the PCR. The wild-type ACD was mutated using the QuikChange multisite-directed mutagenesis kit (Stratagene) to generate the double mutant S247E,S286E (ACD-2E) as well as the triple mutant S240E,S247E,S286E (ACD-3E). The amplified fragments corresponding to ACD, ACD-2E, and ACD-3E were each subcloned into the BamHI and EcoRI sites of pGEX-2T (Amersham Biosciences). Automated dideoxynucleotide sequencing of the constructs verified the authenticity of the inserts. The predicted protein products consisted of the ACD of hnRNP-C fused to glutathione S-transferase (GST). For the wild-type ACD, the amino acid sequence of the final protein product after cleavage from GST with thrombin consists of 79 residues, GSNDKSEEEQSSSSVKKDETNVKMESEGGADDSAEEGDLLDDDDNEDRGDDQLELIKDDEKEAEEGEDDRDSANGEDDS, corresponding to hnRNP-C1 residues Asn217–Ser293 preceded by “GS.” Ser residues mutated to Glu in the ACD-2E and/or ACD-3E construct are in boldface underlined type.
Escherichia coli BL21 cells (Novagen) transformed with the pGEX-hnRNP-C fusion construct were grown to A600 = 0.6 at 37 °C and then induced with 0.1 mm isopropyl β-d-thiogalactopyranoside for 1–2 h. The cells were pelleted at 4000 × g for 30 min and then suspended in phosphate-buffered saline (67 mm phosphate and 150 mm NaCl, pH 7.4) containing 5 mm dithiothreitol (DTT), 0.1% Nonidet P-40 (Roche Applied Science), and Complete protease inhibitor mixture (one tablet/50 ml; Roche Applied Science) at 180 ml/6 liters of culture. The cells were lysed by sonication, and the homogenate was centrifuged at 3000 × g for 30 min. Glutathione-Sepharose (Amersham Biosciences) was added to the resulting supernatant (0.75 ml of resin/1 liter of culture). The supernatant was then agitated on a rocker for 20 min and centrifuged at 500 × g for 5 min. The resulting supernatant was removed, and the resin was placed in a chromatography column and washed with 2 bed volumes of Tris-buffered saline (5 mm Tris and 125 mm NaCl, pH 7.5) containing 5 mm DTT. The resin was then incubated overnight at 25 °C as a 50% slurry with thrombin (80 units/0.75 ml of resin). All subsequent steps were performed at 4 °C. The resin was washed with 1 bed volume of Tris-buffered saline, and the eluate was concentrated to 0.5 ml with a Centriprep YM-3 concentrator (3000-Da molecular mass cutoff; Amicon, Inc.) at 3000 × g. The sample was then applied to a Superdex 200 HR 10/30 size exclusion chromatography column (10 × 300 mm; Amersham Biosciences) equilibrated in Tris-buffered saline containing 5 mm MgCl2 at a flow rate of 0.4 ml/min using a BioLogic DuoFlow chromatography system (Bio-Rad). The concentrations of the purified ACD constructs were determined by spectrophotometric analysis using ϵ = 31 (mg/ml)-1 cm-1 at 205 nm in 0.01% (v/v) Brij 35 (Sigma).
Purification of an hnRNP-C Kinase from Mouse Liver—Mouse liver (25 livers, ∼30 g; Pel-Freez Biologicals) was homogenized in 80 ml of cold 50 mm Tris, pH 7.4, containing 5 mm DTT and Complete protease inhibitor mixture (one tablet/50 ml) using a food processor. All subsequent steps were at 4 °C. The homogenate was centrifuged at 4000 × g for 30 min. To the supernatant was added 6 ml of red-agarose chromatography resin (Sigma). The resin was sedimented by centrifugation at 500 × g for 5 min and then washed three times each with 100 ml of Tris-buffered saline containing 5 mm DTT. hnRNP-C kinase activity was eluted from the resin using 12 ml of 50 mm Tris, 1.25 m NaCl, and 5 mm DTT, pH 7.4. The sample was then diluted 1:20 with water and applied to an UNO-Q anion exchange column (Bio-Rad) using a BioLogic DuoFlow chromatography system. The protein was eluted with a linear NaCl gradient (50 mm to 1.0 m) in 50 mm Tris and 5 mm DTT, pH 7.4. Fractions containing hnRNP-C kinase activity (see below) were pooled and diluted 1:3 with water (final concentration of 5 ml). The partially purified kinase was mixed with 0.7 ml of glutathione-Sepharose containing 2 mg of GST-ACD-2E fusion protein (see above). The slurry was brought to 5 mm MgCl2,10mm NaNO3, and 0.2 mm ADP and incubated for 20 min at 4 °C. The slurry was then centrifuged at 500 × g for 5 min, and the supernatant was removed. The resin was washed twice with 5 ml of 50 mm Tris and 5 mm DTT, pH 7.4. hnRNP-C kinase activity was then eluted by incubating the resin for 16 h at 4 °C with 5 ml of 50 mm Tris, 5 mm DTT, and 2 mm EDTA, pH 7.4. The sample was concentrated to 100 μl by centrifugation at 3000 × g with a Centricon YM-3 centrifugal concentrating device (Amicon, Inc.). The sample was then subjected to SDS-PAGE, and the entire sample lane was analyzed by tandem mass spectrometry following tryptic in-gel digestion (see below).
In Vitro Phosphorylation Assays—Samples were assayed for hnRNP-C kinase activity by incubating aliquots with 5 μg of purified recombinant ACD-2E or ACD-3E in 10 mm Tris, pH 7.5, containing 5 mm DTT, 5 mm MgCl2, 100 μm ATP, and 5 μCi of [γ-32P]ATP (PerkinElmer Life Sciences) at 37 °C in a reaction volume of 50 μl. Reactions were quenched after1hbythe addition of SDS-PAGE sample buffer (Bio-Rad) containing β-mercaptoethanol. For a positive control on the autoradiographs, the wild-type ACD was phosphorylated in vitro using these same conditions along with 0.5 units of protein kinase CK2 (Roche Applied Science). Samples were then subjected to reducing SDS-PAGE on a 4–20% polyacrylamide gel. After exposure, film was developed using an Eastman Kodak X-Omat processor.
Endothelial Cell Culture—Human umbilical vein endothelial cells (HUVECs) were purchased from Cambrex and cultured as described previously (
). HUVECs were grown to confluency in 10-cm dishes at passage 3 in M199 medium (Cambrex) containing 10% fetal calf serum, 100 μg/ml heparin (Sigma), and 50 μg/ml endothelial cell growth supplement (Sigma). 48 h after reaching confluence, the endothelial cells in M199 medium with fetal calf serum, heparin, and endothelial cell growth supplement were incubated in the absence or presence of 5 μm H2O2 for 20 min at 37 °C with or without pretreatment with 100 μm IC261 (Calbiochem) for 3 h prior to H2O2 stimulation. After treatment, the cells were washed twice with ice-cold phosphate-buffered saline; harvested by scraping in 10 mm Tris, 140 mm NaCl, and 1 mm EDTA, pH 8.0; and pelleted by spinning at 500 × g for 15 min. The cells were resuspended in 10 mm HEPES, 750 μm spermidine, 150 μm spermine, 20 mm NaF, 1 mm sodium orthovanadate, 2 mm EDTA, 5 mm DTT, and Complete protease inhibitor mixture (one tablet/50 ml), pH 7.9 (Buffer A), containing 0.1% Nonidet P-40; incubated on ice for 10 min; and then centrifuged at 16,000 × g for 5 min. The supernatant was removed, and the nuclear pellet was washed once with 0.8 ml of Buffer A.
Immunoprecipitation of Protein Kinase CK1α—HUVEC nuclei were lysed by application of 0.5 ml of 10 mm Tris and 1 m NaCl, pH 7.4, with incubation on ice for 20 min. The samples were centrifuged at 16,000 × g for 10 min. The supernatants were diluted with an equal volume of water and centrifuged again at 16,000 × g for 10 min. The nuclear extracts were incubated with 40 μl of agarose-conjugated anti-CK1α antibody (Santa Cruz Biotechnology Inc.) for 1 h at 4 °C. The resin was then isolated by centrifugation at 400 × g for 5 min. The resin was washed three times with 1 ml of 20 mm Tris and 500 mm NaCl, pH 7.5, followed by one wash with 1 ml of 10 mm Tris, pH 7.5. The final washed resin was analyzed for hnRNP-C kinase activity as described above.
Two-dimensional Immunoblots—HUVEC nuclei were resuspended in isoelectric focusing (IEF) sample buffer (9 m urea, 65 mm DTT, 1% CHAPS, and 0.1% Bio-Lyte 3/10 ampholyte (Bio-Rad)) using 100 μl/106 cells. The suspension was centrifuged at 16,000 × g for 15 min, and the resulting supernatant was then applied to a Bio-Spin 6 chromatography column (Bio-Rad) equilibrated with IEF sample buffer, followed by centrifugation at 1000 × g for 4 min. These desalted nuclear extracts were subjected to IEF using a PROTEAN IEF cell and 17-cm Ready-Strip IPG strips (Bio-Rad), which covered the pH range of 4–7. Upon completion of IEF, the ReadyStrip IPG strips were incubated first with equilibration buffer (375 mm Tris, 6 m urea, 2% SDS, and 20% glycerol, pH 8.8) containing 130 mm DTT for 10 min and then with equilibration buffer containing 135 mm iodoacetamide for 10 min. The strips were placed on 4–20% gradient polyacrylamide gels (Bio-Rad) and electrophoresed at 200 V. Gels were blotted onto polyvinylidene difluoride membranes, and the membranes were probed for hnRNP-C1/C2 using anti-hnRNP-C polyclonal primary antibody (1:100 dilution; Santa Cruz Biotechnology Inc.) and peroxidase-conjugated donkey anti-goat secondary antibody (1:5000 dilution; Jackson ImmunoResearch Laboratories, Inc.). Blots were developed using ECL Plus chemiluminescence detection kits (Amersham Biosciences).
Tandem Mass Spectrometry—Gel portions were subjected to tryptic in-gel digestion, followed by liquid chromatography-tandem mass spectrometry analysis of the extracted peptides as described previously (
). Specifically, excised gel portions were cut into ∼1-mm3 pieces. Gel pieces were washed and dehydrated with acetonitrile for 10 min, followed by removal of acetonitrile. Pieces were then completely dried in a SpeedVac. Rehydration of the gel pieces was with 50 mm ammonium bicarbonate solution containing 12.5 ng/μl modified sequencing grade trypsin (Promega) at 4 °C. After 45 min, the excess trypsin solution was removed and replaced with 50 mm ammonium bicarbonate solution to just cover the gel pieces. The samples were then placed overnight in a 37 °C room. Peptides were later extracted by removing the ammonium bicarbonate solution, followed by two washes, for 20 min each, with a solution containing 50% acetonitrile and 5% formic acid. The extracts were then dried in a SpeedVac (∼1 h). The samples were stored at 4 °C until analyzed.
On the day of analysis, the samples were reconstituted in 5 μl of HPLC solvent A (5% acetonitrile, 0.005% heptafluorobutyric acid, and 0.4% acetic acid). A nanoscale reverse-phase HPLC capillary column was created by packing 5-μm C18 spherical silica beads into a fused silica capillary (75-μm inner diameter × 12-cm length) with a flame-drawn tip. After equilibrating the column, each sample was pressure-loaded offline onto the column. The column was then reattached to the HPLC system. A gradient was formed, and peptides were eluted with increasing concentrations of HPLC solvent B (95% acetonitrile, 0.005% heptafluorobutyric acid, and 0.4% acetic acid).
As the peptides were eluted, they were subjected to electrospray ionization, at which time they entered into either an LCQ DECA ion trap mass spectrometer or an LTQ linear ion trap mass spectrometer (ThermoFinnigan, San Jose, CA). Eluting peptides were detected, isolated, and fragmented to produce a tandem mass spectrum of specific fragment ions for each peptide. Sites of phosphorylation were determined by matching the protein sequence with the acquired fragmentation pattern using the software program Sequest (ThermoFinnigan) by allowing for a modification comprising an additional 80 Da on Ser, Thr, or Tyr.
Expression and Purification of Full-length hnRNP-C1 Tetramers—A DNA segment containing full-length hnRNP-C1 (residues 1–293) (
) was amplified from a nearly full-length clone of hnRNP-C1 using the PCR. Wild-type hnRNP-C1 was mutated using the QuikChange multisite-directed mutagenesis kit to generate the S247E,S286E double mutant (2E-hnRNP-C1) as well as the S225E,S228E,S240E,S247E,S286E quintuple mutant (5E-hnRNP-C1). The amplified fragments corresponding to these three proteins were each subcloned into the NdeI and BamHI sites of pET-3b (Novagen). Automated dideoxynucleotide sequencing of the constructs verified the authenticity of the inserts.
E. coli Rosetta 2(DE3) bacteria (Novagen) transformed with the hnRNP-C1 constructs were grown to A600 = 0.6 at 25 °C and then induced with 0.4 mm isopropyl β-d-thiogalactopyranoside for 18 h at 25 °C. The cells were pelleted at 3000 × g for 30 min and stored at -80 °C. The cells were thawed and suspended in 3 volumes of 10 mm Tris, 8 m urea, 1 m NaCl, 5 mm EDTA, 5 mm DTT, and 1 mm phenylmethylsulfonyl fluoride, pH 7.4. The cells were lysed by sonication, and the homogenates were centrifuged at 10,000 × g for 20 min in a Sorvall RC-5 high speed centrifuge at 4 °C. The supernatant was collected and diluted 1:10 with 10 mm Tris, pH 7.4, containing 5 mm EDTA, 5 mm DTT, and 1 mm phenylmethylsulfonyl fluoride (Buffer B). All subsequent steps were performed at 4 °C. The diluted supernatant was centrifuged at 3000 × g for 20 min and filtered through cheesecloth. The sample was then applied to an Econo-Pac High Q anion exchange cartridge (Bio-Rad) at 2 ml/min using a BioLogic DuoFlow chromatography system. The proteins were eluted with a linear NaCl gradient (0.1–1 m) in Buffer B. Fractions containing hnRNP-C were pooled (4 ml) and diluted 1:10 with Buffer B. The sample was then applied to an S6 cation exchange column (Bio-Rad), and hnRNP-C was eluted with a linear NaCl gradient (0.1–1.0 m) in Buffer B. Fractions containing hnRNP-C were pooled (2 ml) and concentrated to 0.5 ml using a Centricon YM-3 centrifugal concentrating device. The concentrated sample was then applied at 0.4 ml/min to a Superdex 200 10/300 GL preparative gel filtration column (Amersham Biosciences) equilibrated with 10 mm Tris, pH 7.4, containing 1 m NaCl, 5 mm MgCl2, and 5 mm DTT. The purified protein (1.5 ml) was dialyzed overnight against 10 mm Tris, pH 7.4, containing 100 mm NaCl, 1 mm EDTA, and 5 mm DTT. The concentrations of the full-length hnRNP-C proteins were determined by analysis of colloidal Coomassie Blue-stained SDS-polyacrylamide gels using a GS-800 laser densitometer (Bio-Rad) with bovine serum albumin as the standard.
Quantitative Fluorescence RNA Binding Studies—Fluorescent ethenoadenosine-modified poly(A) was synthesized by reacting poly(A) with chloracetaldehyde as described previously (
). Equilibrium binding studies were performed at 25 °C in 10 mm Tris, pH 7.4, containing 0.1 m NaCl and 1 mm EDTA at a nucleotide (residue) concentration of 1 μm. As purified hnRNP-C was titrated into the binding mixture, the enhancement in ethenoadenosine-modified poly(A) fluorescence upon protein binding was monitored using a Cary Eclipse fluorescence spectrophotometer at an excitation wavelength of 310 nm and an emission wavelength of 410 nm. The raw fluorescence readings were corrected for dilution and for the background fluorescence of the protein, assessed in the absence of ethenoadenosine-modified poly(A). Data were fit by nonlinear regression using the program KaleidaGraph and the McGhee-von Hippel model for binding large ligands to a homogeneous lattice (
). According to this model, the binding of a large ligand such as a protein to a homogeneous lattice such as a segment of RNA can be described as follows:
and where ν is the moles of bound ligand (protein)/mol of lattice residues, L is the molar free ligand concentration, n is the number of lattice residues occupied by a single ligand or the binding site size, K is the intrinsic association constant in units of m-1, and ω is the cooperativity parameter such that ω > 1 indicates positive cooperativity in ligand binding. Here, the binding site size (n) was assumed to be constant at 230 nucleotides (lattice residues), as determined previously (
). Free ligand concentrations were calculated based on the total protein concentrations and the calculated bound protein concentrations obtained from preliminary nonlinear regression analyses.
Miscellaneous Methods—H2O2 concentrations of stock solutions were determined using ϵ = 81 m-1 at 230 nm. UV-visible absorption spectra were recorded on a Cary 50 Bio UV-visible spectrophotometer. Chemicals not otherwise specified were obtained from Sigma. Protein sequences were aligned using ClustalW.
Purification of an hnRNP-C Kinase from Mouse Liver—The purified recombinant hnRNP-C ACD was used as a substrate to identify hnRNP-C kinase activities in mouse liver extracts. Previously, the purified ACD was shown to be phosphorylated by protein kinase CK2 at the position corresponding to Ser247, but was not phosphorylated by multiple other kinases, including protein kinase A and multiple isoforms of protein kinase C. A double mutant containing Ser-to-Glu mutations at the two basal sites of phosphorylation (S247E,S286E, ACD-2E) was utilized to identify kinases capable of phosphorylating the H2O2-stimulated sites of phosphorylation. Such an activity was identified in mouse liver extracts, and the kinase was isolated using red-agarose affinity chromatography, ion exchange chromatography, and an affinity column consisting of the GST-ACD-2E fusion protein immobilized on glutathione-Sepharose (Fig. 1A). This kinase activity readily phosphorylated the ACD-2E protein and also the ACD-3E protein, although to a lesser extent. The ACD-3E protein contained a third mutation (S240E) at one of the H2O2-stimulated sites of phosphorylation.
Tandem mass spectral analyses demonstrated that the kinase phosphorylated ACD-2E at the position corresponding to Ser240 in full-length hnRNP-C1 and phosphorylated twice the stretch of 4 contiguous Ser residues corresponding to Ser225–Ser228 (Fig. 1, B and C). The high susceptibility of this second phosphopeptide to H3PO4 loss during acquisition of collision-induced dissociation spectra (
) prevented precise localization of these two phosphates. However, the two phosphates were observed at Ser225–Ser228. Thus, hnRNP-C kinase isolated from mouse liver phosphorylates hnRNP-C at precisely the same locations as shown to be phosphorylated endogenously in endothelial cells in response to low levels of H2O2 (
Protein Kinase CK1α Is an hnRNP-C Kinase—The partially purified protein kinase from mouse liver was subjected to liquid chromatography-tandem mass spectrometry to identify all of the proteins present. The only protein kinase identified was murine protein kinase CK1α (Fig. 2). The kinase was identified by the presence of two tryptic peptides, YASINAHLGIEQSR and ILQGGVGIPHIR. The analysis also revealed the presence of murine centaurin-α2 (data not shown), consistent with the previous observation that CK1α co-purifies with members of the centaurin-α family of proteins (
). Although a single Asp or Glu residue at this upstream position is often insufficient to direct CK1 phosphorylation, CK1 is also able to phosphorylate Ser residues 3 residues downstream of a stretch of multiple acidic residues without prior phosphorylation (
). In hnRNP-C1, the 3 consecutive Glu residues at positions 221–223 likely direct the initial CK1α-mediated phosphorylation at Ser225. The resulting phospho-Ser225 then likely directs a second CK1α phosphorylation at Ser228. It has also been recently reported that a cluster of negatively charged residues 6–13 residues downstream of a Ser residue may direct CK1 phosphorylation at that site (
). This downstream acidic cluster was reported to function along with a target SLS sequence in β-catenin and in NFAT (nuclear factor of activated T cells). In hnRNP-C1, Ser240 is located 5 residues upstream of the acidic cluster DDpSAEE, indicating that this stretch of acidic residues may direct the CK1α-mediated phosphorylation at Ser240. In hnRNP-C, this downstream acidic cluster appears to target CK1α phosphorylation in the absence of an SLS sequence. Thus, all of the identified CK1α-mediated sites of phosphorylation in hnRNP-C are in agreement with previous observations concerning substrate sequence requirements for this kinase.
Protein Kinase CK1α from HUVEC Nuclei Phosphorylates hnRNP-C—To show that protein kinase CK1α from endothelial cell nuclei can phosphorylate hnRNP-C, this kinase was immunoprecipitated from HUVEC nuclear extracts. The immunoprecipitated kinase was washed extensively with high salt buffer and then assessed for its ability to phosphorylate hnRNP-C ACD constructs. Protein kinase CK1α from HUVEC nuclei behaves identically to protein kinase CK1α isolated from mouse liver. The kinase phosphorylated both the ACD-2E and ACD-3E proteins, although the latter to a lesser extent (Fig. 3A). Furthermore, tandem mass spectral analysis showed that HUVEC nuclear CK1α phosphorylated ACD-2E at the position corresponding to Ser240 and twice at Ser225–Ser228 (Fig. 3, B and C). As before, the precise location of the two phosphates at Ser225–Ser228 could not be determined because of loss of H3PO4 during the acquisition of the collision-induced dissociation spectra. Thus, CK1α immunopurified from endothelial nuclei phosphorylates hnRNP-C at the precise locations where phosphorylation is stimulated by low levels of H2O2 in cultured endothelial cells (
). The immunopurified kinase activity from HUVEC nuclei was not altered by prior treatment of the cells with low concentrations of H2O2 (data not shown), suggesting that the kinase is purified in the fully active form.
Inhibition of Protein Kinase CK1α Prevents H2O2-stimulated Phosphorylation of hnRNP-C—To assess the role of protein kinase CK1α in phosphorylating hnRNP-C in cells, the effects of a CK1 family-specific inhibitor (IC261) (
), hnRNP-C was present in confluent endothelial cells predominantly in the biphosphorylated form (pI 5.05) and, to a lesser extent, in the triphosphorylated form (pI 5.00). Treatment of the cells with 5 μm H2O2 for 20 min enhanced the formation of hyperphosphorylated forms containing four to five phosphates/subunit at pI 4.90–4.95 (Fig. 4). However, in the presence of the CK1 inhibitor, the addition of H2O2 had no effect on the phosphorylation pattern of endogenous hnRNP-C1/C2. In addition, even in the absence of added H2O2, there was a clear reduction in the triphosphorylated form of the protein in the presence of the CK1 inhibitor. This effect was very similar to that observed previously by treating endothelial cells with catalase (
), phosphorylation of hnRNP-C by other CK1 family members in addition to CK1α cannot be ruled out. However, the purification of CK1α using an hnRNP-C ACD affinity column described here and the previously reported localization of CK1α at sites of pre-mRNA processing (
) strongly implicate CK1α as the kinase mediating the H2O2-stimulated phosphorylation of hnRNP-C. It is important to note that whereas CK1α appears to localize to nuclear speckles, hnRNP-C is present diffusely throughout the nucleoplasm without specific localization to nuclear speckles (
Protein Kinase CK1α Phosphorylation Sites in hnRNP-C Are Evolutionarily Conserved—To gain insight into the evolutionary conservation of the CK1α phosphorylation sites in hnRNP-C1/C2, the protein sequences from several vertebrate species were aligned using ClustalW (Fig. 5). All of the phosphorylation sites in hnRNP-C are highly conserved. The basal site of phosphorylation at Ser286 is invariant, being present in fish, amphibians, and mammalians. The CK2 basal phosphorylation site at Ser247 is nearly invariant, substituted with Tyr in fish. The CK1α phosphorylation site at Ser240 is also nearly invariant. Interestingly, this site contains a Glu residue in frogs, perhaps indicating constitutive phosphoserine effects at this position in this species. The CK1α phosphorylation sites at Ser225–Ser228 are also highly conserved. Whereas Ser225 is invariant, Ser226 and Ser228 are highly conserved, being present in all of the mammals and amphibians. In contrast, Ser227 is the least conserved Ser residue in this region, present only in humans and rabbits. Based on these variable degrees of conservation and on the known tendency of CK1α to phosphorylate Ser residues that have acidic or phosphorylated residues at the n-3 position, it can be concluded that, in humans, Ser225 and Ser228 are the most likely sites of phosphorylation in this contiguous stretch of 4 Ser residues. The high degree of conservation of the CK1α phosphorylation sites among the different vertebrate species suggests that CK1α-mediated phosphorylation of hnRNP-C1/C2 constitutes a signaling pathway widespread among vertebrate species.
Protein Kinase CK1α Modulates the RNA Binding Activity of hnRNP-C—To assess the effects of ACD phosphorylation on hnRNP-C function, full-length hnRNP-C1 tetramers were overexpressed in E. coli and purified to homogeneity using anion exchange, cation exchange, and size exclusion chromatography (Fig. 6A). The presence of phospho-Ser residues was mimicked by formation of Ser-to-Glu point mutations. Previously, mutating a Ser residue to Glu has been shown to mimic the effects of phosphorylation in several proteins, including MEK-1 (mitogen-activated protein kinase/extracellular signal-regulated kinase kinase-1) (
). The three proteins purified were wild-type hnRNP-C1, hnRNP-C1 with S247E and S286E mutations (2E-hnRNP-C1) to mimic basal phosphorylation, and hnRNP-C1 with S247E, S286E, S240E, S225E, and S228E point mutations (5E-hnRNP-C1) to mimic the effects of protein kinase CK1α phosphorylation. Upon size exclusion chromatography, all three proteins exhibited an aberrantly high apparent molecular mass of ∼360 kDa (data not shown), in good agreement with previous observations with both recombinant hnRNP-C1 tetramers and hnRNP-C tetramers isolated from cells (
). This model incorporates both an intrinsic affinity constant (K) and a cooperativity parameter (ω), with the overall affinity represented by the product (Kω) and thus an overall dissociation constant represented by 1/Kω. Wild-type hnRNP-C1 tetramers avidly bound the RNA with high affinity and marked positive cooperativity (Fig. 6B and Table I), with binding constants in good agreement with previously reported values (
). The 2E-hnRNP-C1 mutant showed RNA binding constants similar to those shown by the wild-type protein. Thus, the biphosphorylated form of hnRNP-C1/C2, which is the predominant form in mammalian cells, appears to bind mRNA with high affinity and substantial positive cooperativity. In contrast, the 5E-hnRNP-C1 mutant displayed markedly reduced RNA binding affinity. Because of the inability of this protein to saturate the RNA, the binding curve could not be fit with absolute certainty, but a range for the binding and cooperativity constants could be ascertained (Table I). The intrinsic affinity constant for the 5E-hnRNP-C1 mutant was decreased by 10–100-fold compared with those for the wild-type hnRNP-C1 and 2E-hnRNP-C1 proteins. In addition, the cooperativity constant was diminished by 1.4–5-fold. Thus, the overall binding affinity of the 5E-hnRNP-C1 mutant was decreased by 14–500-fold compared with those of the wild-type and 2E-hnRNP-C1 proteins. This result indicates that the CK1α-mediated phosphorylation of hnRNP-C1/C2 greatly reduces the mRNA binding affinity of the protein, thus modulating its function.
Table IRNA binding constants for hnRNP-C1 phosphorylation site mutants The plots in Fig. 6B were fit by nonlinear regression using the McGhee-von Hippel model for binding large ligands to a homogeneous lattice (
). A previous functional proteomic analysis revealed that primary human endothelial cells respond to low levels of H2O2 by the rapid and reversible phosphorylation of hnRNP-C1/C2, a nuclear restricted pre-mRNA-binding protein (
). In addition, cultured stem cells from these embryos show impaired survival and slowed rates of proliferation and differentiation. Although hnRNP-C1/C2 is clearly involved in pre-mRNA metabolism in the nucleus, the specific function of this protein has remained unclear. It has been proposed that this protein binds most (if not all) mRNA transcripts in the nucleus (
). Thus, this protein must dissociate from processed mRNA transcripts to allow for their export to the cytosol. In fact, heterologous expression of hnRNP-C in yeast is lethal, apparently because the protein enters the nucleus and prevents the export of mRNA transcripts (
). Whereas the protein is basally phosphorylated at Ser247 by protein kinase CK2 and at Ser286 by an as yet unidentified kinase, H2O2 stimulates phosphorylation at Ser240 and Ser225–Ser228 by protein kinase CK1α. The equilibrium RNA binding studies presented here indicate that the CK1α-mediated phosphorylation of hnRNP-C modulates the mRNA binding function of the protein, as would be required during mRNA processing and mRNA export. These studies are in general agreement with a previous study in which incubation of crude nuclear extracts with ATP and okadaic acid resulted in diminished binding of hnRNP-C to an RNA affinity resin (
). Because of this localization of CK1α, the kinase has been postulated to play an important role in pre-mRNA processing. In this regard, it has been demonstrated that this kinase is able to phosphorylate in vitro several SR protein splicing factors. From the data present here, it is clear that a major role for CK1α in the nucleus would be to regulate hnRNP-C1/C2 function during pre-mRNA processing and/or cell cycle progression.
The mechanism by which CK1α is regulated is not entirely clear. Although some members of the CK1 family, such as CK1δ and CK1ϵ, are activated by dephosphorylation of inhibitory autophosphorylation sites (
), this mechanism appears not to apply to CK1α, which is typically isolated in the fully active form. It is possible that CK1α-mediated phosphorylation may be regulated by sequestration of the kinase into specific complexes via protein-protein interactions (
). Such an activation mechanism may help to explain some of the paradoxical actions of this kinase. For example, protein kinase CK1α has been observed to be a positive regulator of cell cycle progression (
). Thus, it is likely that the nature of the specific complexes to which CK1α is targeted is a significant factor in determining the overall effect of CK1α-mediated phosphorylation on cellular homeostasis. Currently, the mechanisms by which CK1α is selectively targeted to specific protein complexes are largely unknown. Understanding the processes by which low concentrations of H2O2 stimulate CK1α-mediated phosphorylation in the nucleus is a focus of active investigation.
We gratefully acknowledge the Taplin Biological Mass Spectrometry Facility in the Department of Cell Biology at Harvard Medical School under the direction of Dr. Steven Gygi for acquisition of liquid chromatography-tandem mass spectrometry spectra.