MCB
Home Help [Feedback] [For Subscribers] [Archive] [Search] [Contents]
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Other Versions of this Article:
MCB.02014-06v1
27/5/1758    most recent
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrowReprints and Permissions
Right arrow Copyright Information
Right arrow Books from ASM Press
Right arrow MicrobeWorld
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Adolph, D.
Right arrow Articles by Ostareck-Lederer, A.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Adolph, D.
Right arrow Articles by Ostareck-Lederer, A.
Molecular and Cellular Biology, March 2007, p. 1758-1770, Vol. 27, No. 5
0270-7306/07/$08.00+0     doi:10.1128/MCB.02014-06
Copyright © 2007, American Society for Microbiology. All Rights Reserved.

Deciphering the Cross Talk between hnRNP K and c-Src: the c-Src Activation Domain in hnRNP K Is Distinct from a Second Interaction Site{triangledown}

Dörte Adolph, Nadine Flach, Katharina Mueller, Dirk H. Ostareck,* and Antje Ostareck-Lederer*

Institute of Biochemistry, Martin Luther University Halle-Wittenberg, Halle (Saale), Germany

Received 27 October 2006/ Accepted 4 December 2006


   ABSTRACT
Top
 Abstract
Introduction
 Materials and Methods
 Results
 Discussion
 References
 
The protein tyrosine kinase c-Src is regulated by two intramolecular interactions. The repressed state is achieved through the interaction of the Src homology 2 (SH2) domain with the phosphorylated C-terminal tail and the association of the SH3 domain with a polyproline type II helix formed by the linker region between SH2 and the kinase domain. hnRNP K, the founding member of the KH domain protein family, is involved in chromatin remodeling, regulation of transcription, and translation of specific mRNAs and is a target in different signal transduction pathways. In particular, it functions as a specific activator and a substrate of the tyrosine kinase c-Src. Here we address the question how hnRNP K interacts with and activates c-Src. We define the proline residues in hnRNP K in the proline-rich motifs P2 (amino acids [aa] 285 to 297) and P3 (aa 303 to 318), which are necessary and sufficient for the specific activation of c-Src, and we dissect the amino acid sequence (aa 216 to 226) of hnRNP K that mediates a second interaction with c-Src. Our findings indicate that the interaction with c-Src and the activation of the kinase are separable functions of hnRNP K. hnRNP K acts as a scaffold protein that integrates signaling cascades by facilitating the cross talk between kinases and factors that mediate nucleic acid-directed processes.


   INTRODUCTION
Top
 Abstract
 Introduction
Materials and Methods
 Results
 Discussion
 References
 
The Src family of nonreceptor tyrosine kinases comprises nine members, including Src, Lck, Hck, Fyn, Blk, Lyn, Fgr, Yes, and Yrk. These signaling proteins catalyze the transfer of the ATP {gamma}-phosphate to specific tyrosine residues in substrate proteins, leading to the modulation of protein-protein and protein-RNA interactions or enzymatic activities. Therefore, their catalytic activity is subject to tight regulation (4, 50). All Src family members are myristoylated at the N terminus and characterized by a specific domain structure, including the Src homology 3 (SH3) domain, the SH2 domain, a kinase domain, and a short C-terminal tail. The catalytic activity is down-regulated by a series of intramolecular interactions that impose conformational constraints on the catalytic domain, making it inaccessible for the substrate. This inactive conformation is established in c-Src by the phosphorylation and subsequent binding of tyrosine 527 (Y527) to the SH2 domain and by the interaction between the SH3 domain and a linker region between SH2 and the kinase domain, which bears a proline motif (4, 47, 55, 57, 58). Dephosphorylation of Y527, or engagement of the SH2 or SH3 domains with other proteins, leads to the stimulation of Src kinase activity by disrupting the intramolecular constraints imposed on the kinase domain (1, 2, 5, 33, 34, 61). Once the kinase is released from the repressed state, intermolecular autophosphorylation of Tyr 416 (Y416) in the activation loop of the kinase domain rapidly occurs, resulting in the reconfiguration of the activation loop and a fully active kinase (46, 57, 58).

Activated Src phosphorylates a large number of proteins, leading to the activation of the mitogen-activated protein kinase pathways, and plays a key role in the stimulation of DNA synthesis, cell cycle regulation, proliferation, and differentiation (4, 50). Constitutive activation of Src results in cell transformation (24) and is associated with numerous human cancers (26), making Src an important proto-oncogene and potential drug target (43). Therefore the elucidation of the regulatory structure-function relationships of Src and its activators is important for the understanding of physiological and oncogenic signal transduction processes.

Heterogeneous nuclear ribonucleoprotein K (hnRNP K) belongs to the family of heterogeneous ribonucleoproteins, which participate in the processing of pre-mRNAs and mRNA export from the nucleus (9, 10). hnRNP K is implicated in chromatin remodeling (42), in transcription control of the c-myc gene (32, 48) and the translation initiation factor eIF4 gene (28), and in RNA splicing (13). An N-terminal bipartite nuclear localization signal and an hnRNP K-specific nuclear shuttling signal confer the capacity for bidirectional transport across the nuclear envelope (30, 31). Extracellular signal-regulated kinase (ERK)-dependent serine phosphorylation mediates the cytoplasmic accumulation of hnRNP K (19). In the cytoplasm, hnRNP K functions as a translational regulator of specific mRNAs, such as c-myc mRNA (12), renin mRNA (44, 45), human papillomavirus type 16 L2 capsid protein mRNA (6), and reticulocyte-15-lipoxygenase (r15-LOX) mRNA (36-38). The mechanism of hnRNP K action has been extensively studied in r15-LOX mRNA translational regulation. r15-LOX, a key enzyme in erythroid cell differentiation, participates in the breakdown of mitochondria in mature reticulocytes, which is a prerequisite for erythrocyte formation. Its premature expression in erythroid precursor cells is restricted by translational silencing, mediated by hnRNP K and hnRNP E1 binding, individually or together, to the differentiation control element (DICE) in the r15-LOX mRNA 3' untranslated region. The silencing complex blocks 80S ribosome assembly (36, 37). In mature reticulocytes, r15-LOX mRNA translation becomes activated (38). Interestingly, hnRNP K, but not hnRNP E1, activates c-Src specifically and is phosphorylated by the kinase (39). The function of hnRNP K as an inhibitor of r15-LOX mRNA translation is abolished by c-Src-dependent phosphorylation of tyrosine 458 (Y458) in KH domain 3 that confers binding to the DICE (29). hnRNP K contains three proline-rich motifs which have been shown to interact with the isolated SH3 domain of the Src kinase family members c-Src, Fyn, and Lyn in vitro (49, 53, 54). The arginine residues (R256, 258, 268, 296, 299) of hnRNP K, which are located between the proline-rich motifs, are asymmetrically dimethylated by the protein arginine methyltransferase 1 (PRMT1) (40). The methylation of those arginine residues inhibits the activation of c-Src by hnRNP K, suggesting that the proline-rich motifs are important for the activation of c-Src in vitro and in vivo (40). Thus, hnRNP K acts as a scaffold protein that integrates signaling cascades by facilitating cross talk between ERK, c-Src, and PRMT1 with factors that mediate mRNA-directed processes.

Here we address the question how the multifunctional protein hnRNP K interacts with and specifically activates c-Src. We have identified the proline residues in hnRNP K that are necessary and sufficient for the activation of c-Src and determined a second sequence motif in hnRNP K that mediates the stable interaction with c-Src in vitro and in vivo. Interestingly, our findings indicate that the interaction with c-Src and the specific activation of the kinase are separable functions of hnRNP K.


   MATERIALS AND METHODS
Top
 Abstract
 Introduction
 Materials and Methods
Results
 Discussion
 References
 
Plasmids. The cDNA clones pSGT-Src, pSGT-Src(K249E/P250G) [pSGT-Src(KP)], and pSGT-Src(Y416F) were described before (16, 17, 39). The plasmid pCA-Lyn was a kind gift of Peter Klinken (51). The plasmid pSG5-Fyn was a kind gift of Guilio Superti-Furga (11). The plasmid pcDNA-Lck (60) was kindly provided by Bart Sefton. The plasmids pSG5-Luc-DICE and pSG5-His-hnRNP K are described in reference 39. pSG5-His-hnRNP K({Delta}P1-3) was constructed using pSG5-hnRNP K from which amino acids (aa) 240 to 337 had been deleted (a kind gift of Maria Shnyreva), and an N-terminal His tag was generated by the insertion of an oligonucleotide coding for 10 histidine residues between SmaI/XhoI. The plasmid pET16b-hnRNP K has been described in reference 37. The proline-to-alanine substitution variants pET16b-hnRNP K or pSG5-His-hnRNP K P1, P2(1), P2(2), P2(3), P3(1), P3(2), and (P2(2,3)P3(1,2)) were generated using a site-directed mutagenesis kit (Stratagene) according to the manufacturer's protocol. The plasmid pET16b-P2-3 (Pro285-Gly318) was generated by PCR using forward (5'GGGAATTCCATATGCCTCGTCGAGGACCACCTCCC3') and reverse (5'CGCGGATCCTTATCCCCCTCTAGGTGGTGGTGG3') primers and inserted into NdeI/BamHI of pET16b (Novagen). The plasmids GST-hnRNP K(127-237) and GST-hnRNP E1 were kind gifts of Henrik Leffers (8). GST-hnRNP K(318-463) and GST-hnRNP K(1-337) were kind gifts of Karol Bomsztyk (53). The hnRNP K deletion variants GST-hnRNP K(1-145), GST-hnRNP K(1-218), GST-hnRNP K(1-242), GST-hnRNP K(1-250), GST-hnRNP K({Delta}P1-3), and GST-hnRNP K(P2[2,3]P3[1,2]) were PCR amplified using primers GST-hnRNP K({Delta}P1-3) and GST-hnRNP K(P2[2,3]P3[1,2]) (forward, 5'CCGGAATTCGTGGAAACTGAACAGCCAGAAGAAACC3'; reverse, 5'TTTTCCTTTTGCGGCCGCTTAGAAAAACTTTCCAGAATACTGCTTCACA3'), GST-hnRNP K(1-145) (forward, 5'CCGGAATTCGTGGAAACTGAACAGCCAGAAGAAACC3'; reverse, 5'TTTTCCTT TTGCGGCCGCTTACGCAGTCAAAGTCACTTCCTTTATAGTG3'), GST-hnRNP K(1-218) (forward, 5'CCGGAATTCGTGGAAACTGAACAGCCAGAAGAAACC3'; reverse, 5'AAGGAAAAAAGCGGCCGCTTAGATGGGAGACTCAGATATAAGATCAAATGTCTTTATGC3'), GST-hnRNP K(1-242) (forward, 5'CCGGAATTCGTGGAAACTGAACAGCCAGAAGAAACC3'; reverse, 5'TTTTCCTTTTGCGGCCGCTTACATCATTGTAAAACCACCATAATC ATAGG3'), and GST-hnRNP K(1-250) (forward, 5'CCGGAATTCGTGGAAACTGAACAGCCAGAAGAAACC3'; reverse, 5'AAGGAAAAAAGCGGCCGCTTATGGGCGTCCGCGACGGTCATCAAACA3') and inserted into EcoRI/NotI of pGEX-4T-1 (Pharmacia). The deletion variants Venus2 hnRNP K(1-218) [V2-hnRNP K(1-218)] and V2-hnRNP K(1-242) were generated by NotI digestion of the respective pGEX variants followed by Klenow treatment, BamHI digestion, and cloning into pVenus2C BamHI/KspAI. V2-hnRNP K, V2-hnRNP K({Delta}P1-3), and V2-hnRNP K(P2[2,3]P3[1,2]) were generated by PCR (forward primer, 5'CCATAATCATAGGTTTCATCGTAAAAATTGGGCTCAGATATAAGATCAAGGATGATCTTTATCGACTC3'; reverse primer, 5'TTTTCCTTTTGCGGCCGCTTAGAAAAACTTTCCAGAATACTG CTTCACAC3'). V2-hnRNP K({Delta}216-226) and V2-hnRNP K{({Delta}216-226),(P2[2,3]P3[1,2])} were generated by three PCRs. For the first PCR, the forward primer was 5'GCAGTCGACGGTACCGTGGAAACT GAACAGC3' and the reverse primer was 5'CCATAATCATAGGTTTCATCGTAAAAATTGGGCTCAGATATAAGATCAAGGATGATCTTTATCGACTC3'. For the second PCR, the forward primer was 5'GAGTCGATAAAGATCATCCTTGATCTTATATCTGAGCCCAATTTTTACGATGAAACCTATGATTATGG3' and the reverse primer was 5'TTTTCCTTTTGCGGCCGCTTAGAAAAACTTTCCAGAATACTGCTTCACAC3'. For the third PCR, the forward primer was 5'CCATAATCATAGGTTTCATCGTAAAAATTGGGCTCAGATATAAGATCAAGGATGATCTTTATCGACTC3' and the reverse primer was 5'TTTTCCTTTTGCGGCCGCTTAGAAAAACTTTCCAGAATACTGCTTCACAC3'. All primers were based on an hnRNP K construct with or without the P2(2,3)P3(1,2) mutation. V2-hnRNP (K-E1-K) and V2-hnRNP {(K-E1-K)(P2[2,3]P3[1,2])}, containing aa 169 to 179 of hnRNP E1 instead of hnRNP K aa 216 to 226, were generated by a combination of three PCRs. For the first PCR, the forward primer was 5'GCAGTCGACGGTACCGTGGAAACTGAACAGC3' and the reverse primer was 5'CATGACTCTCCCTTGCGGAGACTGGGAGAGCGTCTCAGATATAAGATCAAGGATGATCTT3'. For the second PCR, the forward primer was 5'ACGCTCTCCCAGTCTCCGCAAGGGAGAGTCATGCCCAATTTTTACGATGAAACCTATGAT3' and the reverse primer was 5'TTTTCCTTTTGCGGCCGCTTAGAAAAACTTTCCAGAATACTGCTTCACAC3'. For the third PCR, the forward primer was 5'GCAGTCGACGGTACCGTGGA AACTGAACAGC3' and the reverse primer was 5'TTTTCCTTTTGCGGCCGCTTAGAAAAACTTTCCAGAATACTGCTTCACA3'. All primers were based on an hnRNP K construct with or without the P2(2,3)P3(1,2) mutation. The products of the PCRs were digested with NotI, a procedure followed by Klenow treatment for blunt-end formation and KpnI digestion. The vector pVenus2C, in which the PCR products were ligated, was digested with KpnI and SmaI. V1-cSrc and the inactive variant V1-Src(Y416F) were PCR amplified using forward (5'GCAGTCGACGGTACCGTGGGGAGCAGC3') and reverse (5'CGCGGATCCTAGGTTCTCTCCAGGCTG3') primers and inserted into KpnI/BamHI of pVenus1C. Plasmids Venus1C and -2C were kind gifts of S. Hüttelmaier.

Recombinant proteins. His-hnRNP K and its proline-to-alanine substitution variants were prepared as described previously (37) with the exception that the lysis and elution buffers contained 1 M KCl. Proteins were dialyzed against 100 mM KCl, 20 mM HEPES (pH 7.4), 5% glycerol, 1 mM EDTA, 1 mM dithiothreitol (DTT). Glutathione S-transferase (GST) fusion proteins were expressed and purified as described for the pGEX system (Pharmacia). GST-tagged hnRNP K and deletion variants used in the in vitro protein-protein interaction assay were dialyzed against 100 mM KCl, 20 mM HEPES (pH 7.4), 5% glycerol, 0.05% Nonidet P-40, 1 mM EDTA, 1 mM DTT. Protein concentrations were determined by the Bradford assay and by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) and Coomassie staining using bovine serum albumin as a standard.

In vitro Src activation assay. c-Src, Src(KP), or Src(Y416F) expressed in HeLa cells was immunopurified with a Src-specific polyclonal antibody (Abcam) from 20 µg of total protein. Equal amounts of the immunopurified kinases were incubated in the presence of 1 µg (18 pmol) of purified recombinant His-hnRNP K or its respective substitution variants and deletion variants in a kinase buffer in the presence of 300 nM ATP at 30°C for 10 min (39). As shown below (see Fig. 5), 25-ng (5.35 pmol), 50-ng (10.7 pmol), and 100-ng (21.4 pmol) portions of the peptide representing P2 and P3 (Pro285-Gly318) were used. After SDS-PAGE separation, proteins were analyzed in Western blot assays using antibodies against the His tag, Src, and p-Tyr.


Figure 5
View larger version (24K):
[in this window]
[in a new window]

  		FIG. 5. The proline motifs P2 and P3 are sufficient to activate c-Src in vitro. Src(wt) (lanes 1 to 6) was immunopurified from transfected HeLa cells. Equal amounts, as shown in a Western blot assay with a Src-ab, were incubated in the presence of ATP with equal amounts of recombinant His-hnRNP K (lane 2) or its variant His-hnRNP K(P2[2,3]P3[1,2]) (lanes 3 to 6), as shown in a Western blot assay with an anti-His antibody (His-ab) and increasing amounts of the peptide representing P2 and P3 (aa 285 to 318) (lanes 4 to 6). Phosphorylation of His-hnRNP K or its variant was analyzed in a Western blot assay with a pY-ab.

 
Analysis of protein-protein interactions. Coupled transcription-translation assays (7) contained, in a 12-µl reaction volume, 1 µg of unlinearized plasmid DNA, T7 RNA polymerase, 0.7 mM ribonucleoside triphosphates, and 0.5 mCi/ml [35S]methionine (Perkin Elmer) in addition to the nuclease-treated rabbit reticulocyte lysate mix (23).

For studying in vitro protein-protein interactions, glutathione-Sepharose 4B beads (Amersham Bioscience) were blocked overnight in reaction buffer (20 mM HEPES, pH 7.4, 100 mM KCl, 0.05% Nonidet P-40, 1 mM DTT) supplemented with 2 mg bovine serum albumin/ml, 200 ng tRNA/µl, and 200 ng of glycogen/µl. For each reaction, 130 pmol of GST-tagged hnRNP K or its deletion variants or no protein, GST alone, or GST-hnRNP E1 (as negative controls) was immobilized on 20 µl of preblocked beads and incubated with in vitro transcribed/translated 35S-labeled c-Src protein (10 µl of transcription-translation reaction mixture) in 600 µl of reaction buffer. After 45 min of incubation with head-over-tail rotation at 4°C, the beads were washed three times with reaction buffer. The bound proteins were eluted by being heated in protein sample buffer, fractionated by 10% SDS-PAGE, and visualized by autoradiography.

HeLa cell transfection and analysis. HeLa cells were maintained in Dulbecco's modified Eagle's medium with L-glutamine and pyruvate (Invitrogen) supplemented with 10% fetal bovine serum (Biochrom) and penicillin-streptomycin (Invitrogen). Cells were transiently transfected by use of the calcium phosphate method (18). For an experiment shown below (see Fig. 2), 5 µg of pSG5-LUC-DICE and, for normalization, 5 µg of simian virus 40-ß-galactosidase control plasmid, as well as 5 µg pSG5-U1A, pSG5-His-hnRNP K, or pSG5-His-hnRNP K({Delta}P1-3), were cotransfected with 5 µg pSGT vector, pSGT-c-Src, or pSGT-Src(Y416F), respectively. For another experiment shown below (see Fig. 4), 5 µg pSG5-His-hnRNP K or the proline-to-alanine substitution variants were cotransfected with 5 µg of pSGT, pSGT-Src(Y416F), pSGT-Src(KP), or pSGT-Src as indicated. For an additional experiment shown below (see Fig. 6), 5 µg pSGT-Src, pCA-Lyn, pSG5-Fyn, or pcDNA-Lck was cotransfected with 5 µg pSG5, pSG5-His-hnRNP K, or pSG5-hnRNP K(P2[2,3]P3[1,2]) as indicated. HeLa cell lysate preparation and immunoprecipitation were performed as described previously (39) with a monoclonal penta-His antibody (QIAGEN). Western blot assays were performed using antibodies against His (Santa Cruz), v-Src (Oncogene), p-Tyr (Santa Cruz), Src(Y416) (Cell Signaling), Lyn (Santa Cruz), Fyn (Santa Cruz), Lck (Santa Cruz), and GAPDH (Abcam). The luciferase activity assay (see Fig. 2) was carried out as described in reference 39. As shown below (see Fig. 8), 5 µg V2-hnRNP K or its proline-to-alanine substitution variants and deletion or insertion variants were cotransfected with 5 µg of V1-c-Src or 5 µg V1-Src(KP) or V1-Src(Y416F) as indicated.


Figure 2
View larger version (35K):
[in this window]
[in a new window]

  		FIG. 2. hnRNP K({Delta}P1-3) fails to activate, but still interacts with, c-Src. (A) HeLa cells were transiently cotransfected with LUC-DICE cDNA (lanes 1 to 7), with cDNAs coding for U1A as a specificity control (lane 1), His-hnRNP K (lanes 2 to 4), or His-hnRNP K({Delta}P1-3) (lanes 5 to 7), with Src(wt) (lanes 3 and 6) or the inactive variant Src(Y416F) (lanes 4 and 7), or without cDNAs (lane 8). HeLa cell lysate was resolved by SDS-PAGE and analyzed in Western blot assays with antibodies against the His tag (His-ab), Src (Src-ab), and GAPDH (GAPDH-ab) as a loading control. His-tagged hnRNP K and His-hnRNP K({Delta}P1-3) were immunoprecipitated with an anti-His antibody, resolved by SDS-PAGE, and analyzed by Western blot assays with antibodies against the His tag (His-ab), Src (Src-ab), or p-Tyr (pY-ab). Peptides recognized by the pY-ab nonspecifically are marked as nonspecific (n.s.). (B) LUC activity assay showing the relative (rel.) enzymatic activity levels of the expressed LUC protein.

 

Figure 4
View larger version (53K):
[in this window]
[in a new window]

  		FIG. 4. Proline motifs P2 and P3 of hnRNP K are responsible for c-Src activation in vivo. (Left) HeLa cells were cotransfected with cDNAs coding for His-hnRNP K (lanes 1 to 4) or its Pro-Ala variants (lanes 5 to 11), the inactive form Src(Y416F) (lane 2), the constitutively active form Src(KP) (lane 3), or Src(wt) (lanes 4 to 11). (A) HeLa cell lysate was resolved by SDS-PAGE and analyzed in Western blot assays with antibodies against the His tag [His-ab], Src [Src-ab], Src phosphorylated at Y416 [Src(pY416)-ab], and GAPDH [GAPDH-ab] as a loading control. (B) His-tagged hnRNP K or its variants were immunoprecipitated with an anti-His antibody, resolved by SDS-PAGE, and analyzed by Western blot assays with antibodies against the His tag (His-ab), p-Tyr (pY-ab), and Src (Src-ab). (Right) HeLa cells were cotransfected with His-hnRNP K (lane 13) or its Pro-Ala variants (lanes 14 to 20) and with the constitutively active variant Src(KP) (lanes 13 to 20). (A) HeLa cell lysate was resolved by SDS-PAGE and analyzed in Western blot assays with antibodies against the His tag [His-ab], Src [Src-ab], Src phosphorylated at Y416 [Src(pY416)-ab], and GAPDH [GAPDH-ab] as a loading control. (B) His-tagged hnRNP K or its variants were immunoprecipitated with an anti-His antibody, resolved by SDS-PAGE, and analyzed by Western blot assays with antibodies against the His tag (His-ab), p-Tyr (pY-ab), and c-Src (Src-ab). wt, wild type.

 

Figure 6
View larger version (45K):
[in this window]
[in a new window]

  		FIG. 6. hnRNP K activates c-Src but not wild-type Lyn, Fyn, or Lck. HeLa cells were transiently transfected with c-Src (lanes 1 to 3), Lyn (lanes 5 to 7), Fyn (lanes 9 to 11), or Lck (lanes 13 to 15) and with His-hnRNP K (lanes 2, 6, 10, and 14) or His-hnRNP K(P2[2,3]P3[1,2]) (lanes 3, 7, 11, and 15) or no cDNAs (lanes 4, 8, 12, and 16). (A) HeLa cell lysate was resolved by SDS-PAGE and analyzed in Western blot assays using antibodies against the His tag (His-ab), against Src, Lyn, Fyn, or Lck (kinase-ab), against p-Tyr (pY-ab), and against GAPDH (GAPDH-ab) as a loading control. (B) His-tagged hnRNP K or its variant was immunoprecipitated with an anti-His antibody, resolved by SDS-PAGE, and analyzed by Western blot assays using antibodies against the His tag (His-ab), against Src, Lyn, Fyn, or Lck (kinase-ab), and against p-Tyr (pY-ab). The Western blots for Lyn and Fyn and p-Tyr had to be overexposed to detect a signal.

 

Figure 8
Figure 8
View larger version (50K):
[in this window]
[in a new window]

  		FIG. 8. Visualization of the interaction between hnRNP K and c-Src in living cells using BiFC analysis. (A) HeLa cells were cotransfected with cDNAs coding for V1-Src, V2-hnRNP K singly or together, V1-Src(Y416F), and V2-hnRNP K, V1-Src, and V2-hnRNP K(P2[2,3]P3[1,2]) as indicated. (B) HeLa cells were cotransfected with V1-Src together with V2-hnRNP K(1-218), V2-hnRNP K(1-242), V2-hnRNP K({Delta}216-226), V2-hnRNP K{({Delta}216-226),(P2[2,3]P3[1,2])}, V2-hnRNP (K-E1-K), and V2-hnRNP {(K-E1-K)(P2[2,3]P3[1,2])} as indicated. For panels A and B, the expression of c-Src or Src(Y416F) and hnRNP K or its variants was determined by specific antibodies and Cy3 (c-Src)- or Cy5 (hnRNP K)-labeled secondary antibodies. The Venus BiFC analysis visualized direct protein-protein interaction. DAPI, 4',6'-diamidino-2-phenylindole.

 
Immunofluorescence microscopy. For single or double immunofluorescence, HeLa cells were processed essentially as described previously (21). An E600 (Nikon) microscope equipped with a digital camera (Hamamatsu) was used for conventional fluorescence microscopy. Images were acquired by Lucia software (Nikon) and compiled with Adobe Photoshop. The transfection efficiency and the number of cells in which a Venus signal was detected were determined in three independent experiments by counting 50 cells per experiment during microscopy. For immunofluorescence, antibodies against hnRNP K were supplied by Matritech, and those against v-Src were supplied by Oncogene.


   RESULTS
Top
 Abstract
 Introduction
 Materials and Methods
 Results
Discussion
 References
 
The domain of hnRNP K that confers interaction with c-Src is distinct from that activating the kinase. hnRNP K shares its function as a DICE-binding protein and a translational regulator of r15-LOX mRNA with hnRNP E1 (37). In contrast to hnRNP K, hnRNP E1 does not activate c-Src specifically and is not phosphorylated by c-Src (39). To identify the structural elements important for this functional difference, we compared the primary structures of hnRNP K and hnRNP E1. Both proteins show similar organizations of the KH domains but striking differences as well: hnRNP K contains three proline-rich motifs (P1, aa 265 to 278; P2, aa 285 to 297; P3, aa 303 to 318) (Fig. 1) which interact with the isolated SH3 of c-Src in vitro (49, 53, 54). These proline-rich motifs are not present in hnRNP E1. A deletion variant of hnRNP K lacking the proline-rich motifs [hnRNP K({Delta}P1-3)] was used to analyze their function in the activation of c-Src required for the tyrosine phosphorylation of hnRNP K (39). HeLa cells were cotransfected (Fig. 2A and B) with plasmids coding for U1A, a nonrelated RNA-binding protein used as a negative control (lane 1), His-hnRNP K (lanes 2 to 4) or hnRNP K({Delta}P1-3) (lanes 5 to 7), and c-Src (lanes 3 and 6) or the inactive variant of the kinase [Src(Y416F)] (lanes 4 and 7), as well as luciferase (LUC) reporter plasmids bearing a DICE in the LUC mRNA 3' untranslated region (Fig. 2A and B, lanes 1 to 7). Protein expression was monitored by Western blot assays of total lysate by use of antibodies against the His tag and c-Src (Fig. 2A). To test whether hnRNP K({Delta}P1-3) is able to activate and/or interact with c-Src, His-hnRNP K or His-hnRNP K({Delta}P1-3) was immunoprecipitated with an anti-His antibody (His-ipp.), followed by Western blot assays (Fig. 2A). The specificity of the interaction between His-hnRNP K and c-Src was shown in previous immunoprecipitation experiments (39). When His-hnRNP K and c-Src were cotransfected, hnRNP K was coprecipitated with c-Src and became phosphorylated, as detected by an anti-phosphotyrosine antibody (pY-ab) (Fig. 2A, lane 3). Furthermore, we have shown earlier (39) that c-Src is activated sixfold in the in vitro Src kinase assay. Tyrosine phosphorylation was below the detection level when hnRNP K({Delta}P1-3) was cotransfected with c-Src (Fig. 2A, lane 6), indicating that hnRNP K({Delta}P1-3) fails to activate the tyrosine kinase in vivo. Interestingly, hnRNP K({Delta}P1-3) still interacts with c-Src (Fig. 2A, lane 6 and 7). Furthermore hnRNP K({Delta}P1-3) is still able to inhibit the expression of a LUC-DICE-bearing mRNA (Fig. 2B, lane 5), consistent with the finding that KH domain 3 of hnRNP K mediates binding to the DICE (29). Translation inhibition by hnRNP K({Delta}P1-3) was not relieved when c-Src was cotransfected in addition (Fig. 2B, lane 6), consistent with the result that hnRNP K({Delta}P1-3) is not able to activate the kinase and hence is not phosphorylated (Fig. 2A, lane 6). However, hnRNP K({Delta}P1-3) is phosphorylated by the constitutive active form Src(KP) (data not shown). From these results, we conclude that the domain of hnRNP K that confers interaction with c-Src is distinct from that activating the kinase.


Figure 1
View larger version (18K):
[in this window]
[in a new window]

  		FIG. 1. Sequence alignment of human hnRNP K and hnRNP E1. The positions of KH domains 1, 2, and 3 are indicated with red bars. The proline (pro)-rich motifs are underlined in black. The c-Src phosphorylation site of hnRNP K that affects DICE binding is indicated by a red arrow (29), and the tyrosine residues identified as targets of c-Src by mass spectrometry are marked with blue arrows (39). The classification of the individual proline-rich motifs as class I or class II according to their orientation in the context of the bound SH3 domain is indicated in red or green, respectively.

 
Analysis of the proline-rich motifs of hnRNP K in vitro and in vivo. Proline motifs can be classified as class I (RxxPxxP) or class II (PxxPxR) according to the orientation of the bound amino acids in the context of the SH3 domain (14, 15, 27) (Fig. 1). The proline motif P1 represents a class II motif [aa 272 to 277], P2 bears two class I motifs [P2(1), aa 286 to 292, and P2(2), aa 287 to 293] and one class II motif [P2(3), aa 291 to 296], and P3 bears one class I [P3(1), aa 305 to 311] and one class II [P3(2), aa 311 to 316] motif (Fig. 1, lower panel). We have substituted the class I- and class II-determining proline residues by alanines and tested the respective hnRNP K variants for their abilities to activate c-Src in vitro (Fig. 3). c-Src was immunopurified from transfected HeLa cells through the use of an anti-Src antibody (Src-ab) and incubated with recombinant His-hnRNP K or its respective Pro-Ala variants (His-ab) under phosphorylation conditions. The phosphorylation of hnRNP K and its variants as a substrate of c-Src was detected by use of the pY-ab. hnRNP K P2(2), P2(3), P3(1), and P3(2) showed significantly reduced phosphorylation (Fig. 3, lanes 7 to 10) compared to hnRNP K (Fig. 3, lane 4). No phosphorylation of the substitution variant combining the four proline-to-alanine exchanges {hnRNP K(P2[2,3]P3[1,2])} (Fig. 3, lane 11) was detectable. This indicates that P2 and P3 bear the proline residues that are structurally required for c-Src activation in vitro.


Figure 3
View larger version (35K):
[in this window]
[in a new window]

  		FIG. 3. The proline motifs P2 and P3 of hnRNP K are responsible for c-Src activation in vitro. The inactive c-Src variant Src(Y416F) (lane 2), the constitutively active form Src(KP) (lane 3), and Src(wt) (lanes 4 to 12) were immunopurified from transfected HeLa cells. Equal amounts, as shown in a Western blot assay with a Src-ab, were incubated in the presence of ATP with equal amounts of recombinant His-hnRNP K (lanes 1 to 4) or its Pro-Ala variants (lanes 5 to 11), as shown in a Western blot assay with an anti His-antibody (His-ab). Phosphorylation of His-hnRNP K or its variants was analyzed in a Western blot assay with a pY-ab. wt, wild type.

 
Next we examined the same hnRNP K variants in vivo. HeLa cells were cotransfected (Fig. 4A and B) with cDNAs coding for His-hnRNP K (lanes 1 to 4) or its Pro-Ala variants (lanes 5 to 11) as indicated, the inactive variant of c-Src, Src(Y416F) (lane 2), the constitutively active form Src(KP) (lane 3), or c-Src (lanes 4 to 11). Protein expression was monitored by Western blot assays of total lysate using antibodies against the His tag, c-Src, c-Src phosphorylated at Y416, which indicates Src kinase activity, and GAPDH as a loading control (Fig. 4A, lanes 1 to 12). To test the respective hnRNP K Pro-Ala substitution variants for their abilities to activate c-Src, His-hnRNP K or its Pro-Ala substitution variants were immunoprecipitated with an anti-His antibody, as indicated, followed by Western blot assays using antibodies against the His tag (Fig. 4B, lanes 1 to 12). The phosphorylation of hnRNP K and its variants by c-Src was detected by a pY-ab (Fig. 4B, lanes 1 to 12). The coprecipitation of His-hnRNP K and its Pro-Ala variants with c-Src was determined by a Src antibody (Fig. 4B, lanes 1 to 12). hnRNP K P2(2), P2(3), P3(1), and P3(2) showed significantly reduced phosphorylation (Fig. 4B, lanes 7 to 10) compared to hnRNP K (Fig. 4B, lane 4). No phosphorylation of the substitution variant combining the four Pro-Ala exchanges {hnRNP K(P2[2,3]P3[1,2])} (Fig. 4B, lane 11) was detectable, consistent with a barely detectable phosphorylation of Y416 in c-Src (Fig. 4A, lane 11). Furthermore, the experiment showed that the Pro-Ala substitution interferes only with the activation of c-Src [Src(pY416)-ab] (Fig. 4A, lanes 4 to 11) and not with the interaction between hnRNP K and c-Src [Src-ab] (Fig. 4B, lanes 4 to 11). Importantly, when hnRNP K or its Pro-Ala variants were cotransfected with the constitutively active variant of c-Src [Src(KP)] (Fig. 4A and B, lanes 13 to 21), no reduction of hnRNP K phosphorylation was observed in the absence of the proline residues implicated in c-Src activation (Fig. 4B, lanes 13 to 21). From these experiments, we conclude that P2 and P3 in hnRNP K are required for c-Src activation in vitro and in vivo but not for the phosphorylation of hnRNP K by active Src.

The proline motifs P2 and P3 are sufficient to activate c-Src. In order to confirm that the proline motifs P2 and P3 of hnRNP K mediate the activation of c-Src, we used a peptide representing P2 and P3 (aa 285 to 318) in an in vitro Src activation assay (Fig. 5). c-Src immunopurified from transfected HeLa cells was incubated with recombinant His-hnRNP K, which both activates c-Src and functions as a substrate (Fig. 5, lane 2), or with His-hnRNP K(P2[2,3]P3[1,2]), which fails to activate c-Src (Fig. 5, lane 3). Increasing amounts of the peptide (the highest point represents equimolar amounts of peptide compared to full-length hnRNP K [Fig. 5, compare lanes 2 and 6]) were added, and this resulted in a phosphorylation of hnRNP K(P2[2,3]P3[1,2]), which still functions as a substrate (Fig. 5, lanes 4 to 6).

In summary, these experiments show that the proline motifs P2 and P3 of hnRNP K are necessary and sufficient for the activation of c-Src by hnRNP K (Fig. 4A and B, lanes 1 to 12, and Fig. 5) but are required neither for the stable interaction of both proteins (Fig. 4B, lanes 1 to 21) nor for the phosphorylation of hnRNP K by active Src (Fig. 4A and B, lanes 13 to 21).

hnRNP K activates c-Src specifically. It has been shown earlier that the isolated proline-rich motifs of hnRNP K interact with the isolated SH3 domain of the Src kinase family members c-Src, Lyn, and Fyn in vitro (49, 53, 54) and that the constitutively active form of Lck phosphorylates hnRNP K in vitro (41).

In order to investigate whether another member of the Src family is also a target of hnRNP K, we analyzed the interaction of hnRNP K with Lyn, Fyn, and Lck and its potential to activate these kinases. HeLa cells were cotransfected (Fig. 6A and B) with plasmids coding for c-Src (lanes 1 to 3), Lyn (lanes 5 to 7), Fyn (lanes 9 to 11), or Lck (lanes 13 to 15) and with His-hnRNP K (lanes 2, 6, 10, and 14) or the Pro-Ala substitution variant His-hnRNP K(P2[2,3]P3[1,2]), which lacks the potential to activate c-Src in vitro and in vivo, as a control (lanes 3, 7, 11, and 15). Protein expression was analyzed in total lysates by Western blot assays using antibodies against the His tag, antibodies against the respective kinases as indicated, and GAPDH as a loading control. Protein phosphorylation was monitored with an anti-phosphotyrosine antibody (Fig. 6A). The cotransfection of hnRNP K and c-Src served as a control (Fig. 6A, lanes 1 to 3). c-Src was activated by hnRNP K but not by hnRNP K(P2[2,3]P3[1,2]) (Fig. 6A, lanes 2 and 3). No significant increase of protein phosphorylation could be detected when hnRNP K was cotransfected with Lyn, Fyn, or Lck (Fig. 6A, compare lanes 2, 6, 10, and 14). This indicates that hnRNP K, in contrast to c-Src, does not activate wild-type Lyn, Fyn, or Lck. To determine the interaction of Lyn, Fyn, and Lck with hnRNP K and the phosphorylation of hnRNP K by these kinases, His-hnRNP K or His-hnRNP K(P2[2,3]P3[1,2]) was immunoprecipitated with a His antibody (His-ipp.), followed by Western blot assays (Fig. 6B). A coprecipitation of hnRNP K and hnRNP K(P2[2,3]P3[1,2]) was clearly detectable with c-Src and Lck (Fig. 6B, lanes 2 and 3 and 14 and 15, respectively). In contrast, an interaction with Lyn and Fyn could be detected only with very long exposures (Fig. 6B, lanes 6 and 7 and 10 and 11, respectively). Tyrosine phosphorylation of hnRNP K was detectable only when c-Src was cotransfected and not when Lyn, Fyn, and Lck were cotransfected (Fig. 6B, lanes 2, 6, 10, and 14). Phosphorylated hnRNP K coprecipitated phosphorylated c-Src, as detected with a long exposure (Fig. 6B, lane 2). These data show that hnRNP K neither binds nor activates wild-type Lyn and Fyn in a manner comparable to that seen for c-Src. Interestingly, although hnRNP K interacts with Lck, it does not activate the kinase. This further reflects the specificity of the biochemical and functional interplay between hnRNP K and c-Src.

Amino acids 218 to 242 of hnRNP K are important for interaction with c-Src in vitro and in vivo. Besides the interaction between proline-rich motifs P2 and P3, which is required for the activation of c-Src, hnRNP K must contain a further interaction site, because hnRNP K({Delta}P1-3), although lacking the potential to activate, still binds to c-Src (Fig. 2A). To obtain information concerning the sequence element(s) in hnRNP K that mediate c-Src binding, we generated GST-tagged deletion variants of hnRNP K. These GST fusion proteins, bound to glutathione-Sepharose, were incubated with in vitro-translated [35S]Met-labeled c-Src (Fig. 7). As negative controls, no protein, GST or GST-hnRNP E1 was coupled (data not shown). An interaction of hnRNP K and the variant hnRNP K({Delta}P1-3) with c-Src could be detected, while peptides bearing the individual KH domains K(1-121), K(127-237), and K(318-463) did not bind c-Src. Since hnRNP E1 neither interacts with nor activates c-Src (39), we analyzed the amino acid sequences of hnRNP K and hnRNP E1 for motifs that are present in hnRNP K but not in hnRNP E1 or that differ between the two proteins in addition to the proline-rich domain (Fig. 1). One of these motifs is located between amino acids 121 and 140 of hnRNP K. Two deletion variants were prepared, hnRNP K(1-145) and an extended version hnRNP K(1-218), additionally carrying KH domain 2 to ensure that the embedded sequence is properly folded. Both proteins did not interact with c-Src.


Figure 7
View larger version (9K):
[in this window]
[in a new window]

  		FIG. 7. Summary of the results of the in vitro protein-protein interaction assays. GST-pull down experiments with recombinant GST-hnRNP K or deletion variants representing the peptides indicated on the left. c-Src was synthesized in the presence of [35S]Met in a coupled transcription-translation assay in rabbit reticulocyte lysate. Equal amounts of 35S-labeled c-Src were incubated with GST-tagged hnRNP K or its deletion variants coupled to glutathione-Sepharose. The bound c-Src was released from the GST-tagged proteins under denaturing conditions and analyzed by SDS-PAGE and autoradiography. The binding (bdg.) of 35S-labeled c-Src to the GST-tagged hnRNP K or its deletion variants is indicated on the left.

 
The second motif can be divided in two parts, aa 218 to 226, which differs between the two proteins, and aa 227 to 242, which is not present in hnRNP E1. The deletion variant hnRNP K(1-242) showed binding to c-Src, as did further-extended versions, hnRNP K(1-250) and hnRNP K(1-337) (Fig. 7). According to this analysis, amino acids 218 to 242 of hnRNP K contain a potential c-Src-binding motif.

To study the interaction between hnRNP K and c-Src in vivo, we used the bimolecular fluorescence complementation (BiFC) assay (20) to visualize the direct interaction of these proteins in living cells. This approach is based on the complementation between nonfluorescent fragments of the yellow fluorescent protein, which shows enhanced fluorescence due to the amino acid exchange F46L (Venus) (35). Venus protein was divided in two fragments (Venus 1 and Venus 2) at nonconserved amino acid residues and fused to the N-terminal ends of hnRNP K or its deletion variants (Venus 2) and c-Src or its respective variants (Venus 1). When the proteins hnRNP K and c-Src interact, the two Venus domains are expected to come in close contact, leading to the reconstitution of a Venus yellow fluorescent protein (Fig. 8). To confirm the results of the in vitro protein-protein binding studies, HeLa cells were cotransfected with cDNAs coding for Venus 1 fused to c-Src (V1-Src) and Venus 2 fused to hnRNP K (V2-hnRNP K), as well as variants of both proteins, as indicated (Fig. 8). When expressed individually, V1-Src, detected as Cy3 fluorescence, was localized mainly to the cytoplasm, whereas V2-hnRNP K, visualized by the Cy5 dye, showed nuclear localization with nucleolar exclusion. The transfection efficiency was above 80%. Expression of either V1-Src or V2-hnRNP K did not result in a detectable Venus fluorescence signal. When V1-Src and V2-hnRNP K were cotransfected, a heteromeric complex was formed, as indicated by the Venus fluorescence, and V1-Src became localized to the nucleus. That is supported by V1-Src exclusion from the nucleoli, as it was detected for V2-hnRNP K in 90% of the transfected cells (Fig. 8A). The inactive version of Src, V1-Src(Y416F) (Fig. 2 and 4), interacted with V2-hnRNP K to a level comparable to that seen for V1-Src in 90% of the transfected cells. Therefore, kinase activity of Src is not a prerequisite for the interaction with hnRNP K, consistent with the results of the immunoprecipitation experiments shown in Fig. 2 and 4. The Pro-Ala variant of hnRNP K, V2-hnRNP K(P2[2,3]P3[1,2]), which did not activate c-Src in vitro and in vivo (Fig. 3, 4, and 6) but still interacted with c-Src (Fig. 4 and 6), was also bound to V1-Src, as indicated by Venus fluorescence in the BiFC assay, in more than 90% of the cells (Fig. 8A). To analyze the hnRNP K variants used in the GST pull-down experiment (Fig. 7) in vivo, we cotransfected V1-Src with V2-hnRNP K(1-218) or V2-hnRNP K(1-242). As seen in the in vitro assay, V2-hnRNP K(1-242) did interact with V1-Src (above 90%) but V1-Src did not interact with V2-hnRNP K(1-218) (below 5%) (Fig. 8B). Consistent with the results of the in vitro protein-protein interaction studies (Fig. 7), this indicates that amino acids 218 to 242, located between KH domain 2 and the proline-rich domain, are likely to be involved in the interaction between hnRNP K and c-Src. Interestingly, this region contains amino acids 216 to 226, which differ between hnRNP K and hnRNP E1 (Fig. 1). In order to analyze this sequence motif, we generated variants of hnRNP K lacking amino acids 216 to 226 {V2-hnRNP K({Delta}216-226) and V2-hnRNP K[({Delta}216-226),(P2{2,3}P3{1,2})]}. V2-hnRNP K({Delta}216-226) and V2-hnRNP K{({Delta}216-226),(P2[2,3]P3[1,2])} did not bind to V1-Src (below 10%) (Fig. 8B). When we replaced aa 216 to 226 by the respective amino acids of hnRNP E1 (Fig. 1) in the hnRNP K sequence, binding to V1-Src was also abolished (below 20%) (Fig. 8B). This result indicates that amino acids 216 to 226 represent a motif important for the stable interaction of hnRNP K with c-Src.


   DISCUSSION
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
References
 
Src tyrosine kinase family members are involved in signaling pathways that regulate cell growth, differentiation, and transformation (4, 50). c-Src-dependent phosphorylation of ZBP1 modulates its binding to the beta-actin mRNA and thereby controls the temporal and spatial expression of beta-actin, which is important for cell migration and growth cone guidance (22, 59).

hnRNP K is a multifunctional protein that acts in different signaling pathways. The ERK-dependent serine phosphorylation of hnRNP K leads to its cytoplasmic accumulation (19). In the cytoplasm, hnRNP K functions as a specific activator of c-Src and is a substrate of that tyrosine kinase. c-Src-dependent phosphorylation modulates the r15-LOX mRNA-binding activity of hnRNP K and its function in the control of mRNA translation during erythroid cell maturation (29, 38, 39). The activation of c-Src is regulated by PRMT1-dependent arginine methylation of hnRNP K (40). Therefore, hnRNP K functions as a scaffold protein that integrates cross talk between ERK, c-Src, and PRMT1 to modulate mRNA translation control specifically.

Here we describe the identification and functional characterization of the structural determinants in hnRNP K that confer the interaction with c-Src and its specific activation. To our surprise, we found that the sequence motif of hnRNP K that is required for the specific activation of c-Src is structurally separated from a second site of stable interaction.

Specific activation of c-Src by hnRNP K. hnRNP K, but not hnRNP E1, contains three proline-rich motifs, P1 (aa 265 to 278), P2 (aa 285 to 297), and P3 (aa 303 to 318) located between KH domains 2 and 3 (Fig. 1). From in vitro binding studies it was known that peptides containing these proline-rich motifs interact with the isolated SH3 domain of c-Src (49, 53, 54). Furthermore, we showed that binding of hnRNP K to c-Src and phosphorylation of hnRNP K by the tyrosine kinase involve the SH3 domain of c-Src in vivo (39). In this study, we found that an hnRNP K deletion variant lacking all three proline-rich motifs, hnRNP K({Delta}P1-3), fails to activate the kinase but still interacts with c-Src (Fig. 2). Using a detailed mutational analysis, we could identify the proline residues in the proline-rich motifs P2 and P3 as necessary and sufficient for the activation of c-Src in vitro and in vivo. These proline residues are arranged in class I (RxxPxxP) or class II (PxxPxR) motifs according to their orientation in SH3 domain binding (14, 15, 27). In motif P2, a class I (P290, 293) and class II (P291, 294) pattern can be mapped, and in motif P3, a class I (P308, 311) and class II (P311, 314) pattern likewise can be mapped (Fig. 3 to 5). It is likely that these proline motifs displace the intramolecular SH3-linker interaction in c-Src, leading to its activation. The importance of the structural integrity of the proline-rich motifs is supported by our finding that the PRMT1-dependent asymmetric dimethylation of the arginine residues (R256, 258, 268, 296, 299), which are located within these proline-rich motifs, inhibits the activation of c-Src (40).

For c-Src, Fyn, Lyn, and Lck, either an interaction with hnRNP K (Src, Lyn, Fyn) or the phosphorylation of hnRNP K by the constitutively active form of Lck has been shown in vitro (41, 49, 53, 54). We have studied Src, Lyn, Fyn, and Lck and found that in contrast to c-Src, neither Lyn nor Fyn nor Lck is activated by hnRNP K (Fig. 6). The Nef protein of human immunodeficiency virus type 1 has been shown to activate Hck, Lyn, and c-Src but not Fgr, Fyn, Lck, or Yes (33, 52). The structural determinant for the activation of Hck has been mapped to the variable loop (RT loop), which is positioned close to the conserved SH3 residues implicated in the binding of proline-rich motifs. A single amino acid substitution in the RT loop of Fyn converts this kinase into a substrate of activation by Nef (25). The importance of this loop for intra- and intermolecular interactions has also been shown for c-Src (11). Therefore, the different amino acid sequences in the RT loops of c-Src, Lyn, Fyn, and Lck might explain why c-Src but not Lyn, Fyn, or Lck is activated by hnRNP K. In addition, sequence specificity in the proline-rich motifs of hnRNP K may contribute to the selective activation of c-Src. For example, a lysine residue in the peptide PPPALPPKKR from the C3G protein makes this ligand highly specific for the c-Crk SH3 domain, whereas the same peptide does not bind the SH3 domains of the kinases Nck, c-Src, and Abl (3, 56). Interestingly, although hnRNP K does not activate Lck, is does interact with this kinase (Fig. 6). Because Lck is a key regulator of T-cell activation, its interaction with hnRNP K as a docking platform might make it accessible for other activators. Thereby, hnRNP K could become a substrate of Lck in such a complex, which might explain the phosphorylation of hnRNP K by the constitutively active form of Lck (41).

Interaction of hnRNP K with c-Src. Besides the proline-rich motifs, hnRNP K bears three additional sequence motifs that are missing from or differ substantially in hnRNP E1 (Fig. 1). Initially, we could identify amino acids 218 to 242 as the binding site for c-Src both in vitro and in vivo (Fig. 7 and 8B). Interestingly, four tyrosine residues identified as targets of c-Src are located in that region (39). Because the inactive variant of c-Src, Src(Y416F), does bind to hnRNP K to a level comparable to those seen for c-Src and the constitutively active variant Src(KP), phosphorylation of these four tyrosine residues is not a prerequisite for the interaction with either variant of Src (Fig. 4 and 8A), and binding to the SH2 domain of c-Src can be excluded. This is consistent with our earlier finding that c-Src and a Src variant lacking the SH2 domain show identical levels of binding to hnRNP K in vivo (39). Since hnRNP K(1-218) does not bind to c-Src in vitro or in vivo and hnRNP K(1-242) shows c-Src interaction comparable to that of full-length hnRNP K in vitro and in vivo, we further analyzed this sequence motif, which is located between KH domain 2 and the proline-rich motifs of hnRNP K. Interestingly, the sequences of aa 216 to 226 differ between hnRNP K and hnRNP E1 (Fig. 1). Clearly the deletion variants hnRNP K({Delta}216-226) and hnRNP K{({Delta}216-226),(P2[2,3]P3[1,2])} do not bind to c-Src (Fig. 8B). When aa 216 to 226 in hnRNP K are replaced by the respective sequence of hnRNP E1, the binding to c-Src is reduced to the same extent as for the deletion variant in vivo (Fig. 8B). This is consistent with our finding that hnRNP E1 does not interact with c-Src (39). A class I proline-rich motif (RxxPxxP) is located between aa 221 and 227. However, an hnRNP K Pro-Ala substitution variant of that motif does not differ in c-Src binding from hnRNP K in vivo (data not shown). From this we conclude that the SH3 domain of c-Src is not involved in the stable interaction with hnRNP K.

Taken together, these data demonstrate the likelihood that either the linker region between the SH2 and the kinase domain of c-Src or the kinase domain itself is involved in the interaction with the sequence motif. Our data suggest that the sequence aa 216 to 226 functions as a docking site for c-Src that brings the kinase into a complex with its activator hnRNP K and other potential substrates binding to it.

Our experiments show that the c-Src interaction and activation domains in hnRNP K can be separated in vitro and in transfected cells. In the cellular context, the respective domains are both present in hnRNP K. Thereby, the protein not only interacts with c-Src, as it does with Lck, but also is capable of activating c-Src in a specific manner. This supports the function of hnRNP K as a multifunctional scaffold protein that mediates the cross talk between signaling pathways that control cell differentiation and maturation.


   ACKNOWLEDGMENTS
 
Plasmids were kindly provided by K. Bomsztyk, S. Hüttelmaier, P. Klinken, H. Leffers, B. Sefton, M. Shnyreva, and G. Superti-Furga. We are grateful to M. W. Hentze, in whose lab we (A.O.-L. and D.H.O.) initiated the work on the activation of c-Src by hnRNP K. We thank N. Stöhr and S. Hüttelmaier for support with fluorescence microscopy and E. Wahle for critical reading of the manuscript.

This work was supported by a Heisenberg Fellowship of the Deutsche Forschungsgemeinschaft (DFG) to A.O.-L. (Os 290/1-1), as well as by grants from the DFG to D.H.O. (Os 135/2-1, 2) and A.O.-L. (Os 290/2-1).


   FOOTNOTES
 
* Corresponding author. Mailing address: Institute of Biochemistry, Martin Luther University Halle-Wittenberg, Kurt-Mothes-Str. 3, 06120 Halle (Saale), Germany. Phone for Dirk H. Ostareck: 49-(0)345-5524816. Fax: 49-(0)345-5527014. E-mail: dostareck{at}biochemtech.uni-halle.de. Phone for Antje Ostareck-Lederer: 49-(0)345-5524949. Fax: 49-(0)345-5527014. E-mail: aostareck{at}biochemtech.uni-halle.de. Back

{triangledown} Published ahead of print on 18 December 2006. Back


   REFERENCES
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 References
 
1. Alexandropoulos, K., and D. Baltimore. 1996. Coordinate activation of c-Src by SH3- and SH2-binding sites on a novel, p130 Cas-related protein, Sin. Genes Dev. 10:1341-1355.[Abstract/Free Full Text]

2. Alonso, G., M. Koegl, N. Mazurenko, and S. A. Courtneidge. 1995. Sequence requirements for binding of Src family tyrosine kinases to activated growth factor receptors. J. Biol. Chem. 270:9840-9848.[Abstract/Free Full Text]

3. Ball, L. J., R. Kuhne, J. Schneider-Mergener, and H. Oschkinat. 2005. Recognition of proline-rich motifs by protein-protein-interaction domains. Angew. Chem. Int. Ed. Engl. 44:2852-2869.[CrossRef][Medline]

4. Brown, M. T., and J. A. Cooper. 1996. Regulation, substrates and functions of Src. Biochim. Biophys. Acta 1287:121-149.[Medline]

5. Burnham, M. R., P. J. Bruce-Staskal, M. T. Harte, C. L. Weidow, A. Ma, S. A. Weed, and A. H. Bouton. 2000. Regulation of c-SRC activity and function by the adapter protein CAS. Mol. Cell. Biol. 20:5865-5878.[Abstract/Free Full Text]

6. Collier, B., L. Goobar-Larsson, M. Sokolowski, and S. Schwartz. 1998. Translational inhibition in vitro of human papillomavirus type 16 L2 mRNA mediated through interaction with heterogenous ribonucleoprotein K and poly(rC)-binding proteins 1 and 2. J. Biol. Chem. 273:22648-22656.[Abstract/Free Full Text]

7. Craig, D., M. T. Howell, C. L. Gibbs, T. Hunt, and R. J. Jackson. 1992. Plasmid cDNA-directed protein synthesis in a coupled eukaryotic in vitro transcription-translation system. Nucleic Acids Res. 20:4987-4995.[Abstract/Free Full Text]

8. Dejgaard, K., and H. Leffers. 1996. Characterization of the nucleic-acid binding activity of KH domains. Different properties of different domains. Eur. J. Biochem. 241:425-431.[Medline]

9. Dreyfuss, G., V. N. Kim, and N. Kataoka. 2002. Messenger-RNA-binding proteins and the messages they carry. Nat. Rev. Mol. Cell Biol. 3:195-205.[CrossRef][Medline]

10. Dreyfuss, G., M. J. Matunis, S. Pinol-Roma, and C. G. Burd. 1993. hnRNP proteins and the biogenesis of mRNA. Annu. Rev. Biochem. 62:289-321.[CrossRef][Medline]

11. Erpel, T., G. Superti-Furga, and S. A. Courtneidge. 1995. Mutational analysis of the Src SH3 domain: the same residues of the ligand binding surface are important for intra- and intermolecular interactions. EMBO J. 14:963-975.[Medline]

12. Evans, J. R., S. A. Mitchell, K. A. Spriggs, J. Ostrowski, K. Bomsztyk, D. H. Ostareck, and A. E. Willis. 2003. Members of the poly (rC) binding protein family stimulate the activity of the c-myc internal ribosome entry segment in vitro and in vivo. Oncogene 22:8012-8020.[CrossRef][Medline]

13. Expert-Bezancon, A., L. P. Le Caer, and J. Marie. 2002. hnRNP K is a component of an intronic splicing enhancer complex that activates the splicing of the alternative exon 6A from chicken-beta-tropomyosin pre-mRNA. J. Biol. Chem. 277:16614-16623.[Abstract/Free Full Text]

14. Feng, S., J. K. Chen, H. Yu, J. A. Simon, and S. L. Schreiber. 1994. Two binding orientations for peptides to the Src SH3 domain: development of a general model for SH3-ligand interactions. Science 266:1241-1247.[Abstract/Free Full Text]

15. Feng, S., C. Kasahara, R. J. Rickles, and S. L. Schreiber. 1995. Specific interactions outside the proline-rich core of two classes of Src homology 3 ligands. Proc. Natl. Acad. Sci. USA 92:12408-12415.[Abstract/Free Full Text]

16. Gonfloni, S., A. Weijland, J. Kretzschmar, and G. Superti-Furga. 2000. Crosstalk between the catalytic and the regulatory domains allows bidirectional regulation of Src. Nat. Struct. Biol. 7:281-286.[CrossRef][Medline]

17. Gonfloni, S., J. C. Williams, K. Hattula, A. Weijland, R. K. Wierenga, and G. Superti-Furga. 1997. The role of the linker between the SH2 domain and the catalytic domain in the regulation and function of Src. EMBO J. 16:7261-7271.[CrossRef][Medline]

18. Graham, F. L., and A. J. Van der Eb. 1973. A new technique for the assay of infectivity of human adenovirus 5 DNA. Virology 52:456-464.[CrossRef][Medline]

19. Habelhah, H., K. Shah, L. Huang, A. Ostareck-Lederer, A. L. Burlingame, K. M. Shokat, M. W. Hentze, and Z. Ronai. 2001. Erk phosphorylation drives cytoplasmic accumulation of hnRNP-K and inhibition of mRNA translation. Nat. Cell Biol. 3:325-330.[CrossRef][Medline]

20. Hu, C. D., Y. Chinenov, and T. K. Kerppola. 2002. Visualization of interactions among bZIP and Rel family proteins in living cells using bimolecular fluorescence complementation. Mol. Cell 9:789-798.[CrossRef][Medline]

21. Hüttelmaier, S., S. Illenberger, I. Grosheva, M. Rüdiger, R. H. Singer, and B. M. Jockusch. 2001. Raver1, a dual compartment protein, is a ligand for PTB/hnRNPI and microfilament attachment proteins. J. Cell Biol. 155:775-786.[Abstract/Free Full Text]

22. Hüttelmaier, S., D. Zenklusen, M. Lederer, J. Dictenberg, M. Lorenz, X. Meng, G. J. Bassell, J. Condeelis, and R. H. Singer. 2005. Spatial regulation of ß-actin translation by Src-dependent phosphorylation of ZBP1. Nature 438:512-515.[CrossRef][Medline]

23. Jackson., R. J., and T. Hunt. 1983. Preparation and use of nuclease-treated rabbit reticulocyte lysates for the translation of eukaryotic messenger RNA. Methods Enzymol. 96:50-74.[Medline]

24. Kato, J. Y., T. Takeya, C. Grandori, H. Iba, J. B. Levy, and H. Hanafusa. 1986. Amino acid substitutions sufficient to convert the nontransforming p60c-src protein to a transforming protein. Mol. Cell. Biol. 6:4155-4160.[Abstract/Free Full Text]

25. Lee, C. H., B. Leung, M. A. Lemmon, J. Zheng, D. Cowburn, J. Kuriyan, and K. Saksela. 1995. A single amino acid in the SH3 domain of Hck determines its high affinity and specificity in binding to HIV-1 Nef protein. EMBO J. 14:5006-5015.[Medline]

26. Levin, V. A. 2004. Basis and importance of Src as a target in cancer. Cancer Treat. Res. 119:89-119.[Medline]

27. Lim, W. A., F. M. Richards, and R. O. Fox. 1994. Structural determinants of peptide-binding orientation and of sequence specificity in SH3 domains. Nature 372:375-379.[CrossRef][Medline]

28. Lynch, M., L. Chen, M. J. Ravitz, S. Mehtani, K. Korenblat, M. J. Pazin, and E. V. Schmidt. 2005. hnRNP K binds a core polypyrimidine element in the eukaryotic translation initiation factor 4E (eIF4E) promoter, and its regulation of eIF4E contributes to neoplastic transformation. Mol. Cell. Biol. 25:6436-6453.[Abstract/Free Full Text]

29. Messias, A. C., C. Harnisch, A. Ostareck-Lederer, M. Sattler, and D. H. Ostareck. 2006. The DICE-binding activity of KH domain 3 of hnRNP K is affected by c-Src mediated tyrosine phosphorylation. J. Mol. Biol. 361:470-481.[CrossRef][Medline]

30. Michael, W. M., M. Choi, and G. Dreyfuss. 1995. A nuclear export signal in hnRNP A1: a signal-mediated, temperature-dependent nuclear protein export pathway. Cell 83:415-422.[CrossRef][Medline]

31. Michael, W. M., P. S. Eder, and G. Dreyfuss. 1997. The K nuclear shuttling domain: a novel signal for nuclear import and nuclear export in the hnRNP K protein. EMBO J. 16:3587-3598.[CrossRef][Medline]

32. Michelotti, E. F., G. A. Michelotti, A. I. Aronsohn, and D. Levens. 1996. Heterogeneous ribonucleoprotein K is a transcription factor. Mol. Cell. Biol. 6:2350-2360.

33. Moarefi, I., F. Lafevre-Bernt, F. Sicheri, M. Huse, C.-H. Lee, J. Kuriyan, and W. T. Miller. 1997. Activation of the Src-family tyrosine kinase Hck by SH3 domain displacement. Nature 385:650-653.[CrossRef][Medline]

34. Mori, S., L. Rönnstrand, K. Yokote, A. Engström, S. A. Courtneidge, L. Claesson-Welsh, and C. H. Heldin. 1993. Identification of two juxtamembrane autophosphorylation sites in the PDGF beta-receptor; involvement in the interaction with Src family tyrosine kinases. EMBO J. 12:2257-2264.[Medline]

35. Nagai, T., K. Ibata, E. S. Park, M. Kubota, K. Mikoshiba, and A. Miyawaki. 2002. A variant of yellow fluorescent protein with fast and efficient maturation for cell-biological applications. Nat. Biotechnol. 20:87-90.[CrossRef][Medline]

36. Ostareck, D. H., A. Ostareck-Lederer, I. N. Shatsky, and M. W. Hentze. 2001. Lipoxygenase mRNA silencing in erythroid differentiation: the 3'UTR regulatory complex controls 60S ribosomal subunit joining. Cell 104:281-290.[CrossRef][Medline]

37. Ostareck, D. H., A. Ostareck-Lederer, M. Wilm, B. J. Thiele, M. Mann, and M. W. Hentze. 1997. mRNA silencing in erythroid maturation: hnRNP K and hnRNP E1 regulate 15-lipoxygenase translation from the 3' end. Cell 89:597-606.[CrossRef][Medline]

38. Ostareck-Lederer, A., and D. H. Ostareck. 2004. Control of mRNA translation and stability in haematopoietic cells: the function of hnRNPs K and E1/E2. Biol. Cell 96:407-411.[CrossRef][Medline]

39. Ostareck-Lederer, A., D. H. Ostareck, C. Cans, G. Neubauer, K. Bomsztyk, G. Superti-Furga, and M. W. Hentze. 2002. c-Src mediated phosphorylation of hnRNP K drives translation activation of specifically silenced mRNAs. Mol. Cell. Biol. 22:4535-4543.[Abstract/Free Full Text]

40. Ostareck-Lederer, A., D. H. Ostareck, K. P. Rücknagel, A. Schierhorn, B. Moritz, S. Hüttelmaier, N. Flach, L. Handoko, and E. Wahle. 2006. Asymmetric arginine dimethylation of hnRNP K by PRMT1 inhibits its interaction with c-Src. J. Biol. Chem. 281:11115-11125.[Abstract/Free Full Text]

41. Ostrowski, J., D. S. Schullery, O. N. Denisenko, Y. Higaki, J. Watts, R. Aebersold, M. Stempka, M. Geschwendt, and K. Bomsztyk. 2000. Role of tyrosine phosphorylation in the regulation of the interaction of heterogenous nuclear ribonucleoprotein K protein with its protein and RNA partners. J. Biol. Chem. 275:3619-3628.[Abstract/Free Full Text]

42. Ostrowski, J., Y. Kawata, D. S. Schullery, O. N. Denisenko, and K. Bomsztyk. 2003. Transient recruitment of the hnRNP K protein to inducibly transcribed gene loci. Nucleic Acids Res. 14:3954-3962.

43. Parang, K., and G. Sun. 2004. Design strategies for protein kinase inhibitors. Curr. Opin. Drug Discov. Dev. 7:617-629.[Medline]

44. Persson, P. B., A. Skalweit, R. Mrowka, and B. J. Thiele. 2003. Control of renin synthesis. Am. J. Physiol. Regul. Integr. Comp. Physiol. 285:R491-R497.[Abstract/Free Full Text]

45. Skalweit, A., A. Doller, A. Huth, T. Kahne, P. B. Persson, and B. J. Thiele. 2003. Posttranscriptional control of renin synthesis: identification of proteins interacting with renin mRNA 3'-untranslated region. Circ. Res. 92:419-427.[Abstract/Free Full Text]

46. Smart, J. E., H. Oppermann, A. P. Czernilofsky, A. F. Purchio, R. L. Erikson, and J. M. Bishop. 1981. Characterization of sites for tyrosine phosphorylation in the transforming protein of Rous sarcoma virus (pp60v-src) and its normal cellular homologue (pp60c-src). Proc. Natl. Acad. Sci. USA 78:6013-6017.[Abstract/Free Full Text]

47. Superti-Furga, G., and S. Courtneidge. 1995. Structure-function relationships in Src family and related protein tyrosine kinases. Bioessays 17:321-330.[CrossRef][Medline]

48. Takimoto, M., T. Tomonaga, M. Matunis, M. Avigan, and H. Krutzsch. 1993. Specific binding of heterogeneous ribonucleoprotein particle protein K to the human c-myc promoter, in vitro. J. Biol. Chem. 268:18249-18258.[Abstract/Free Full Text]

49. Taylor, S. J., and D. Shalloway. 1994. An RNA-binding protein associated with Src through its SH2 and SH3 domains in mitosis. Nature 368:867-871.[CrossRef][Medline]

50. Thomas, S. M., and J. S. Brugge. 1997. Cellular functions regulated by Src family kinases. Annu. Rev. Cell Dev. Biol. 13:513-609.[CrossRef][Medline]

51. Tilbrook, P. A., E. Ingley, J. H. Williams, M. L. Hibbs, and S. P. Klinken. 1997. Lyn tyrosine kinase is essential for erythropoietin-induced differentiation of J2E erythroid cells. EMBO J. 16:1610-1619.[CrossRef][Medline]

52. Trible, R. P., L. Emert-Sedlak, and T. Smithgall. 2006. HIV-1 NEF selectively activates Src family kinases Hck, Lyn, and c-Src through direct SH3 domain interaction. J. Biol. Chem. 281:27029-27038.[Abstract/Free Full Text]

53. Van Seuningen, I., J. Ostrowski, X. R. Bustelo, P. R. Sleath, and K. Bomsztyk. 1995. The K protein domain that recruits the interleukin 1-responsive K protein kinase lies at adjacent to a cluster of c-Src and Vav SH3-binding sites. Implications that K protein acts as a docking platform. J. Biol. Chem. 270:26976-26985.[Abstract/Free Full Text]

54. Weng, Z., S. M. Thomas, J. J. Rickles, J. A. Taylor, A. W. Brauer, C. Seidel-Dugan, M. W. Michael, G. Dreyfuss, and J. S. Brugge. 1994. Identification of Src, Fyn, and Lyn SH3-binding proteins: implications for a function of SH3 domains. Mol. Cell. Biol. 14:4509-4521.[Abstract/Free Full Text]

55. Williams, J. C., A. Weijland, S. Gonfloni, A. Thompson, S. A. Courtneidge, G. Superti-Furga, and R. K. Wierenga. 1997. The 2.35 Å crystal structure of the inactivated form of chicken Src: a dynamic molecule with multiple regulatory interactions. J. Mol. Biol. 274:757-775.[CrossRef][Medline]

56. Wu, X., B. B. Knudsen, S. M. Feller, J. Zheng, A. Sali, D. Cowburn, H. Hanafusa, and J. Kuriyan. 1995. Structural basis for the specific interaction of lysine-containing proline-rich peptides with the N-terminal SH3 domain of c-Crk. Structure 3:215-226.[Medline]

57. Xu, W., A. Doshi, M. Lei, M. J. Eck, and S. C. Harrison. 1999. Crystal structures of c-Src reveal features of its autoinhibitory mechanism. Mol. Cell 3:629-638.[CrossRef][Medline]

58. Xu, W., S. C. Harrison, and M. J. Eck. 1997. Three-dimensional structure of the tyrosine kinase c-Src. Nature 385:595-602.[CrossRef][Medline]

59. Yao, J., Y. Sasaki, Z. Wen, G. J. Bassell, and J. Q. Zheng. 2006. An essential role for ß-actin mRNA localization and translation in Ca2+-dependent growth cone guidance. Nat. Neurosci. 10:1265-1273.[CrossRef]

60. Yurchak, L. K., and B. Sefton. 1995. Palmitoylation of either Cys-3 or Cys-5 is required for the biological activity of the Lck tyrosine protein kinase. Mol. Cell. Biol. 15:6914-6922.[Abstract]

61. Zheng, X. M., Y. Wang, and C. J. Pallen. 1992. Cell transformation and activation of pp60c-src by overexpression of a protein tyrosine phosphatase. Nature 359:336-339.[CrossRef][Medline]

Molecular and Cellular Biology, March 2007, p. 1758-1770, Vol. 27, No. 5
0270-7306/07/$08.00+0     doi:10.1128/MCB.02014-06
Copyright © 2007, American Society for Microbiology. All Rights Reserved.



This article has been cited by other articles:


This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)