Advertisement
JBC

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Originally published In Press as doi:10.1074/jbc.M310794200 on November 17, 2003

J. Biol. Chem., Vol. 279, Issue 6, 4612-4624, February 6, 2004
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
279/6/4612    most recent
M310794200v1
Right arrow Submit a Letter to Editor
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
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 arrowRequest Permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by de Graaf, K.
Right arrow Articles by Becker, W.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by de Graaf, K.
Right arrow Articles by Becker, W.
Social Bookmarking
Add to CiteULike   Add to Complore   Add to Connotea   Add to Del.icio.us   Add to Digg   Add to Reddit   Add to Technorati  
What's this?

Characterization of Cyclin L2, a Novel Cyclin with an Arginine/Serine-rich Domain

PHOSPHORYLATION BY DYRK1A AND COLOCALIZATION WITH SPLICING FACTORS*

Katrin de Graaf{ddagger}, Paul Hekerman{ddagger}, Oliver Spelten{ddagger}, Andreas Herrmann§, Len C. Packman¶, Konrad Büssow||, Gerhard Müller-Newen§, and Walter Becker{ddagger}**

From the {ddagger}Institut für Pharmakologie und Toxikologie, Medizinische Fakultät der RWTH Aachen, Wendlingweg 2, 52074 Aachen, Germany, §Institut für Biochemie, Medizinische Fakultät der RWTH Aachen, Pauwelstrasse 30, 52074 Aachen, Germany, the Department of Biochemistry, University of Cambridge, 80 Tennis Court Road, Cambridge CB2 1GA, United Kingdom, and ||Max-Planck-Institut für Molekulare Genetik, Heubnerweg 6, 14059 Berlin, Germany

Received for publication, October 1, 2003 , and in revised form, November 14, 2003.


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 EXPERIMENTAL PROCEDURES
 RESULTS
 DISCUSSION
 REFERENCES
 
A novel method employing filter arrays of a cDNA expression library for the identification of substrates for protein kinases was developed. With this technique, we identified a new member of the cyclin family, cyclin L2, as a substrate of the nuclear protein kinase DYRK1A. Cyclin L2 contains an N-terminal cyclin domain and a C-terminal arginine/serine-rich domain (RS domain), which is a hallmark of many proteins involved in pre-mRNA processing. The gene for cyclin L2 encodes the full-length cyclin L2, which is predominantly expressed in testis, as well as a truncated splicing variant (cyclin L2S) that lacks the RS domain and is ubiquitously expressed in human tissues. Full-length cyclin L2, but not cyclin L2S, was associated with the cyclin-dependent kinase PITSLRE. Cyclin L2 interacted with splicing factor 2 in vitro and was co-localized with the splicing factor SC35 in the nuclear speckle compartment. Photobleaching experiments showed that a fusion protein of green fluorescent protein and cyclin L2 in nuclear speckles rapidly exchanged with unbleached molecules in the nucleus, similar to other RS domain-containing proteins. In striking contrast, the closely related green fluorescent protein-cyclin L1 was immobile in the speckle compartment. DYRK1A interacted with cyclin L2 in pull-down assays, and overexpression of DYRK1A stimulated phosphorylation of cyclin L2 in COS-7 cells. These data characterize cyclin L2 as a highly mobile component of nuclear speckles and suggest that DYRK1A may regulate splicing by phosphorylation of cyclin L2.


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
EXPERIMENTAL PROCEDURES
 RESULTS
 DISCUSSION
 REFERENCES
 
DYRK1A (dual specificity tyrosine phosphorylation-regulated kinase 1A) is a nuclear protein kinase that appears to play an important role in brain development (for a review, see Ref. 1). The first evidence for a role of DYRK1A in neurogenesis came from its similarity with the Drosophila orthologue, the minibrain kinase (2). Mutant flies with a reduced expression of minibrain show a significant reduction of neurons in distinct areas of the brain, correlated with specific behavioral defects (2). In addition, the human DYRK1A gene is located on chromosome 21 in the "Down syndrome critical region" (3–5). Consistent with a potential involvement of the DYRK1A gene in the pathogenesis of mental retardation in Down syndrome, overexpression of DYRK1A in transgenic mice causes learning deficits and neurodevelopmental alterations (5, 6). Furthermore, mice heterozygous for a targeted deletion of the Dyrk1a gene (Dyrk1a+/- mice) develop abnormal brain morphology (7). In early embryonal development, DYRK1A is expressed in neural progenitor cells during the transition from proliferating to neurogenic divisions (8). Altogether, these findings suggest a relevant function of DYRK1A during neuronal development, related to proliferation and/or differentiation.

In recent years, several transcription factors have been identified as substrates of DYRK1A (for a review, see Ref. 9). Particularly, overexpression of DYRK1A has been shown to modulate the transactivation potential of FKHR (forkhead in rhabdomyosarcoma) (10, 11), cAMP-response element-binding protein (12), and GLI1 (13). Cytoplasmic proteins like eukaryotic initiation factor 2B{epsilon}, tau (14), and dynamin (15) have also been reported to be phosphorylated by DYRK1A, but these proteins might equally well be targets of the cytoplasmic kinases DYRK2 or DYRK3. However, although DYRK1A contains a canonical nuclear localization signal and is indeed predominantly targeted to the nucleus (16, 17), a significant fraction of the kinase is also present in the cytoplasm (18). Within the nucleus, DYRK1A is localized in "nuclear speckles," which represent the splicing factor compartment (17, 19). The precise function of DYRK1A in the nucleus is still largely unknown, and it is an open question whether any of the known substrates of DYRK1A is a key mediator of its functions on cell differentiation and proliferation. We have therefore performed a more extensive search for interacting proteins and substrates to address this problem.

Several methods have been employed to identify novel substrates of protein kinases. Routine interaction screens like the yeast two-hybrid system, phage display, and phage expression library screening, yield in many cases substrates of kinases (20–23). Other approaches are based on the specificity of substrate recognition by kinases and aim to identify substrate proteins by searching data bases with experimentally determined or predicted consensus phosphorylation sequences (24, 25). More direct strategies include the phosphorylation screening of phage expression libraries (26) or of cell lysates (27), and the identification of phosphoproteins by functional proteomics and mass spectrometry (28).

Here we have employed a new method which allowed us to screen in parallel for interaction partners and substrates of the kinase DYRK1A. In this approach, we used an arrayed protein expression library (29, 30) both for interaction screening by overlaying with the radiolabeled recombinant kinase (31), and for a substrate screen by solid phase phosphorylation (26). The use of the filter array allowed us to directly identify positive clones without the need of further rounds of purification. Several DYRK1A-interacting and/or phosphorylated proteins were identified by this approach, and one protein was found in both screens. This protein turned out to be a novel member of the cyclin family and was designated cyclin L2.

Here we characterize cyclin L2 as a novel substrate of DYRK1A. Furthermore, we describe the alternative transcripts generated from the cyclin L2 gene and show that full-length cyclin L2, but not the truncated splicing variant cyclin L2S, is co-localized with the splicing factor SC35 in nuclear speckles. Using photobleaching techniques, we demonstrate that cyclin L2 and the closely related isoform, cyclin L1, differ strikingly with regard to their intranuclear mobility.


   EXPERIMENTAL PROCEDURES
TOP
 ABSTRACT
 INTRODUCTION
 EXPERIMENTAL PROCEDURES
RESULTS
 DISCUSSION
 REFERENCES
 
Production of the Filter Array—The construction of the cDNA expression library and the production of high density filter arrays have been described previously (29). Human fetal brain cDNA was directionally cloned into the pQE-30NST vector (GenBankTM accession number AF074376 [GenBank] ) for isopropyl-1-thio-{beta}-D-galactopyranoside-inducible expression of His6-tagged fusion proteins. The filter array (22 x 22 cm) used in the present study contained a library of 37,830 double-spotted plasmid clones (library number 800; RZPD, Berlin, Germany) that were enriched for putative expression clones as described earlier (29). The colonies were lysed by alkali treatment (0.5 M NaOH, 1.5 M NaCl, 10 min), and the filter was neutralized for 2 x 5 min in 1 M Tris-HCl, pH 7.5, 1.5 M NaCl. Bacterial debris was carefully wiped off in TBS1 supplemented with 0.5% (v/v) Triton X-100, 0.1% (v/v) Tween 20 (TBS contains 20 mM Tris-HCl, pH 7.5, 0.5 M NaCl), and then the filter was washed twice for 10 min in TBS with 0.5% (v/v) Triton X-100, 0.1% (v/v) Tween 20 and once for 10 min in TBS.

Probe Labeling for Overlay Assays—The probe for the interaction screen, GST-DYRK1A-{Delta}C-K188R, was expressed from the pGEX-2TK vector (GenBankTM accession number U13851 [GenBank] ), which allows labeling of the fusion protein on a vector-encoded phosphorylation site by cAMP-dependent protein kinase. The C-terminally truncated mutant of DYRK1A was used as a probe in order to avoid potential nonspecific binding of the repeat structures in the C-terminal domain of DYRK1A (32). The kinase-negative point mutant (K188R) of DYRK1A should exhibit enhanced enzyme-substrate interactions. For labeling, 20–25 µg of GST-DYRK1A-{Delta}C-K188R bound to glutathione S-Sepharose beads (Amersham Biosciences) were incubated in a volume of 110 µl with 10 units of cAMP-dependent protein kinase (New England Biolabs) in 50 mM Tris-HCl, pH 7.5, 10 mM MgCl2, 6 mg/ml dithiothreitol for 30 min at 4 °C in the presence of 45 µCi of [{gamma}-32P]ATP (3000 Ci/mmol). The reaction was stopped by adding 10 mM EDTA to a final concentration of 2 mM. The Sepharose beads were then washed seven times with 500 µl of ice-cold PBST (137 mM NaCl, 2.7 mM KCl, 10 mM Na2HPO4, 1.8 mM KH2PO4, pH 7.4, 0.5% (w/v) Triton X-100) in order to remove free [{gamma}-32P]ATP. The labeled protein was then eluted for 10 min on ice in 500 µl of glutathione elution buffer (10 mM reduced glutathione in 50 mM Tris-HCl, pH 8.0). The typical yield of this reaction was about 15 µCi of incorporated 32P in the eluted protein.

Interaction Screening by DYRK1A Overlay—In order to allow refolding of denatured proteins (33), the filter array or nitrocellulose membranes with electroblotted proteins were first incubated for 1 h at room temperature in 6 M guanidinium HCl, 50 mM Tris-HCl, pH 8.3, 50 mM dithiothreitol, 2 mM EDTA and then rinsed with TBS, followed by an overnight incubation at 4 °C in 150 mM NaCl, 10 mM Tris-HCl, pH 7.4, 2 mM EDTA, 1% BSA, 0.1% Nonidet P-40. To reduce false positive signals due to binding of GST, the membrane was blocked by incubation in interaction buffer (0.5% BSA, 75 mM KCl, 20 mM HEPES, pH 7.7, 2.5 mM MgCl2, 1 mM dithiothreitol, 0.1 mM EDTA, 0.05% Nonidet P-40) containing 1 µg/µl GST for 1 h at room temperature. The membrane was then incubated with 6 x 104 cpm/ml of 32P-labeled GST-DYRK1A-{Delta}C-K188R in interaction buffer for 18 h at 4 °C. After washing three times for 10 min each at room temperature with interaction buffer, the membrane was dried and subjected to autoradiography.

Substrate Screening by Solid Phase Phosphorylation—The filter with the expression array or polyvinylidene fluoride membranes with electroblotted proteins were preincubated in kinase buffer (25 mM HEPES, pH 7.4, 0.5 mM dithiothreitol, 5 mM MgCl2, 5 mM MnCl2), 100 mM NaCl, 3% BSA containing 3 µM cold ATP for 1 h at room temperature to mask proteins with autophosphorylating and/or ATP-binding activities (26). The solid phase phosphorylation was done by incubation with 0.2 units/ml GST-DYRK1A-{Delta}C in kinase buffer containing 3 µM [{gamma}-32P]ATP (5 µCi/ml) for 45 min at room temperature. This construct has been shown to have the same substrate specificity but higher activity than the full-length kinase when produced in Escherichia coli (34). One unit of protein kinase activity was that amount that catalyzed the phosphorylation of 1 nmol of DYRKtide, the standard synthetic peptide substrate (34), in 1 min in kinase buffer containing 100 µM ATP and 200 µM DYRKtide. After phosphorylation, the membrane was washed twice with 30 mM Tris-HCl, pH 7.4, once with 30 mM Tris-HCl, pH 7.4, 0.05% Nonidet P-40, twice with 30 mM Tris-HCl, pH 7.4, without Nonidet P-40, once with 1 M HCl, and again twice with 30 mM Tris-HCl, pH 7.4, each washing step for 10 min. The membrane was air-dried and exposed to x-ray film.

Bacterial Expression—The plasmids for bacterial expression of GSTDYRK1A-{Delta}C (34), GST-DYRK1A-{Delta}C-K188R, GST-DYRK1A-{Delta}C{Delta}N (35), and GST-DYRK2 (17) have been described previously. GSTDYRK1A-NT contains amino acids 1–177 of the N terminus of DYRK1A; GST-DYRK1Acat contains amino acids 133–499 of DYRK1A. Plasmids from the human fetal brain expression library that encode His6-tagged fusion proteins (Table I) were obtained from the RZPD Deutsches Ressourcenzentrum für Genomforschung (Berlin, Germany) (for cloning and vector information see, on the World Wide Web, www.rzpd.de, library number 800). The expression plasmids for GST-SF2/ASF and GST-hTRA2{beta}1 were kindly provided by Stefan Stamm (Erlangen, Germany) (36). The bacterial expression plasmid coding for GST-CycL2 was constructed by cloning the cDNA from clone N16596 (corresponding to amino acids 270–520 of cyclin L2) into pGEX-2TK.


View this table:
[in this window]
[in a new window]
  		TABLE I
Identification of DYRK1A-interacting proteins and substrates of DYRK1A

 
GST- and His-tagged fusion proteins were expressed in E. coli and affinity-purified using glutathione-Sepharose 4B (Amersham Biosciences) or nickel-charged nitrilotriacetic acid-agarose beads (Qiagen, Hilden, Germany). Proteins were eluted from the beads under native conditions (reduced glutathione or imidazole). His6-tagged cyclin L2 expressed from the pQE-30NST vector (RZPD clone N16596) was purified from inclusion bodies as described previously (37). The purified protein was refolded by gel filtration through a Sephadex G25 column (NAPTM-5 column, Amersham Biosciences) with 10 mM Tris-HCl, pH 7.4, 100 mM NaCl as equilibration and elution buffer. For kinase assays, imidazole was removed from His6-tagged fusion protein by ultrafiltration using Centricon concentrators (Amicon, Beverly, MA).

Mammalian Expression—The expression clone for GFP-DYRK1A has been described (17). GFP-CLK3 was constructed by cloning the insert of pGEX-CLK3 (38) into pEGFP-C1 (Clontech) and expresses C-terminal to the GFP tag amino acids 33–490 of rat CLK3. GFP constructs of cyclin L2 were cloned in pEGFP-C1 and comprise the indicated amino acids C-terminal of GFP (numbered as in Fig. 3A): GFP-CycL2, amino acids 1–520; GFP-CycL2-{Delta}N, amino acids 223–520; GFP-CycL2S, amino acids 1–220 of cyclin L2 plus 6 specific amino acids as indicated in Fig. 3A. GFP-CycL1 contains the full open reading frame of human cyclin L1 (amino acids 1–526; GenBankTM accession number AF180920 [GenBank] ). A plasmid for expression of untagged cyclin L2 was generated by deletion of GFP from GFP-CycL2.



View larger version (37K):
[in this window]
[in a new window]
  		FIG. 3.
Structure of the CCNL2 gene and gene products. A, amino acid sequence of cyclin L2 and cyclin L2S. The sequence of cyclin L2 was deduced from an uncharacterized data base entry (AY037150 [GenBank] ). Cyclin L2S consists of amino acids 1–220 of cyclin L2 and 6 additional amino acids (italics). The cyclin domain is boxed, and the RS domain is underlined. The amino acids of the antigenic peptide used for antibody production are highlighted by boldface type. B, structural comparison of cyclin L2, cyclin L2S, and cyclin L1. The sequence similarity of the cyclin domain (Cyc) and the RS domain (RS) is indicated as the percentage of identical amino acids. C, structural organization of the human CCNL2 gene and derived transcripts. Exons are shown as filled boxes, and the alternatively spliced exon 6a is shaded. Three differentially spliced transcripts (T1, T2, and T3) identified in this work and by searching of the public expressed sequence tag data bases are shown. Stop codons that terminate the open reading frames are indicated by asterisks. Putative polyadenylation signals of T1/T2 (at nucleotides 2073 and 2284 of AY037150 [GenBank] ) and T3 (nucleotide 1173 of BC016333 [GenBank] ) are indicated (AAUAAA and AAUACA). The localization of the hybridization probes used for Northern blot analysis (Fig. 4A) is indicated below the gene (E6a-5', E6a-3', and E11).

 
In Vitro Kinase Assays—For in vitro-kinase assays (Fig. 2), amounts of substrates were adjusted to 750 ng of myelin basic protein (Sigma) according to the intensities of the Coomassie Blue-stained bands on SDS-PAGE. Substrates were incubated for 15 min at 30 °C with 0.5 units/ml GST-DYRK1A-{Delta}C in kinase buffer supplemented with 100 µM [{gamma}-32P]ATP (66.6 µCi/ml). The samples were separated by SDS-PAGE and stained with Coomassie Blue, and phosphorylated proteins were detected by autoradiography of dried gels.



View larger version (104K):
[in this window]
[in a new window]
  		FIG. 2.
In vitro phosphorylation of RS domain-containing proteins by DYRK1A. The indicated RS-rich proteins were expressed in E. coli as GST fusion proteins (hTRA2{beta}1, SF2/ASF, and CYCL2) or His6-tagged fusion proteins (SFRS4, SFRS5, and SWAP2). The affinity-purified proteins were phosphorylated in vitro by GST-DYRK1A-{Delta}C, separated by SDS-PAGE, and stained with Coomassie Blue to assess the amounts of the different substrate proteins (lower panel). Note that the preparations of the recombinant proteins contain many truncated products. Incorporation of 32P was visualized by autoradiography of the dried gel (upper panel). For comparison, myelin basic protein (MBP) was analyzed as a well characterized in vitro substrate of DYRK1A. GST phosphorylation is shown for background control of the GST fusion proteins.

 
Mass Spectrometry—For identification of DYRK1A phosphorylation sites, His6-tagged cyclin L2 (produced from the RZPD clone N16596) was phosphorylated by GST-DYRK1Acat. In order to maximize the yield of the phosphorylated protein, the reaction was carried out for 1 h and with high amounts of the kinase (1:1 molar ratio). Coomassie-stained bands were cut and processed for analysis by mass spectrometry as described previously (35).

Northern Blot Analysis—Northern blots containing RNA from different human tissues were purchased from Clontech. For differential detection of the splicing variants of CCNL2 and CCNL1, templates for specific hybridization probes were generated by PCR from a human testis cDNA library (catalog number HL1161x; Clontech). Probe CCNL2-E11 matches the 3'-untranslated region of CCNL2 (nucleotides 1477–2070 in AY037150 [GenBank] ); probe CCNL2-E6a-5' matches the 5'-end of exon 6a (nucleotides 663–787 in AK074112 [GenBank] ); and probe CCNL2-E6a-3' matches the 3'-end of exon 6a (nucleotides 1913–2600 in AK074112 [GenBank] ). Probe CCNL1-E14 matches the 3'-end of the coding region of CCNL1 (nucleotides 1237–1660 of AF180920 [GenBank] ). CCNL1-E7+8 contains exon 7, exon 8, and the intron in between (exon definition by Dickinson et al. (39); nucleotides 39525–39874 of human bac clone AC104411 [GenBank] ). Probes were labeled with 32P by random oligonucleotide priming, and Northern blots were hybridized at probe concentrations of 1–5 x 106 cpm/ml.

PCR Analysis of Cyclin L2 Transcripts—Mouse cDNAs were produced using the First Strand cDNA synthesis kit (Amersham Biosciences). A panel of human cDNAs was purchased from Clontech (Human MTCTM Panel I). For specific detection of the transcripts T1–T3, we combined a common forward primer matching exon 2 (E2for, 5'-ATTCCGGAATGGCTACCGGGCAGGTGTTG-3') with primers that were specific for either exon 6a (E6arev, 5'-AGTCTAGAGCCAGCCTTCG-3') or exon 6b (E6brev, 5'-AAGACGTCGGTGCGAAGG-3'). These oligonucleotides were designed to match both the murine and the human mRNA sequences. PCRs were performed with the JumpStart REDAccuTaqTM DNA polymerase (Sigma) at an annealing temperature of 52 °C, each step (denaturing, annealing, polymerization) for 30 s. Products were separated by electrophoresis through 1.5% agarose gels and visualized by ethidium bromide staining. The PCR products amplified from the human cDNAs were hardly detectable by ethidium bromide staining and were therefore evaluated by Southern blot analysis with a probe matching exons 3–5 (nucleotides 301–664 of AY037150 [GenBank] ). The identity of the PCR products was verified by sequencing.

Antibodies—The following antibodies were commercially obtained: monoclonal antibodies for GFP (MBL, Nagoya, Japan), SC35 (Sigma), and phosphothreonine-proline (p-Thr-Pro-101; Cell Signaling Technology) and rabbit polyclonal antibodies for GFP (Molecular Probes, Inc., Eugene, OR) and PITSLRE (Santa Cruz Biotechnology, Inc., Santa Cruz, CA). Horseradish peroxidase-coupled secondary antibodies were purchased from Perbio Science, Bonn, Germany (anti-rabbit IgG) and Amersham Biosciences (anti-mouse IgG). Alexa Red-conjugated secondary antibody for immunofluorescence was obtained from Molecular Probes. An antiserum specific for the long form (see "Results") of cyclin L2 was raised by immunization of rabbits with a peptide (PYKGSEIRGSRK) corresponding to amino acids 437–448 within the C-terminal serine/arginine-rich (RS) domain (see Fig. 3A).

Cell Culture, Immunoprecipitation, and Immunoblotting—COS-7 cells were grown in Dulbecco's modified Eagle's medium high glucose, supplemented with 10% fetal calf serum. Cells were transfected using FuGENE 6 (Roche Applied Science) as suggested by the manufacturer. For detection of phosphorylated proteins, cells were washed twice in 50 mM HEPES, pH 7.5, 150 mM NaCl, lysed in 500 µl of boiling SDS lysis buffer (20 mM Tris-HCl, pH 7.5, 1% (w/v) SDS), and incubated in a boiling water bath at 100 °C for 5 min. Insoluble material was removed by centrifugation, and the lysates were diluted with 2 ml of water and 2.5 ml of 2x immunoprecipitation buffer (40 mM Tris-HCl, pH 7.4, 300 mM NaCl, 0.4 mM phenylmethylsulfonyl fluoride (PMSF), 1% (v/v) Triton X-100). The proteins were immunoprecipitated by incubating with 2 µl of polyclonal GFP-specific antiserum and 40 µl of protein A-Sepharose (Amersham Biosciences) at 4 °C overnight. The Sepharose was washed twice in wash buffer (20 mM Tris-HCl, pH 7.4, 150 mM NaCl, 0.2 mM phenylmethylsulfonyl fluoride) with 0.1% Triton X-100 and twice in wash buffer without Triton X-100, and bound proteins were eluted by boiling in Laemmli sample buffer. Samples were then subjected to Western blot analysis with the indicated primary antibodies.

For co-immunoprecipitations, the cells were washed twice in PBS and lysed under native conditions in lysis buffer (50 mM Tris-HCl, pH 7.5, 100 mM NaCl, 5 mM EDTA, 10 mM NaF) supplemented with 0.1% Nonidet P-40 and inhibitors (1 mM NaVO4, 1 mM phenylmethylsulfonyl fluoride, 5 µg/ml aprotinin, 5 µg/ml leupeptin, 3 µg/ml pepstatin). The lysate was sonicated and centrifuged to remove insoluble material. After immunoprecipitation at 4 °C overnight as described above, the Sepharose beads were washed with lysis buffer containing 0.1% Nonidet P-40 and boiled in Laemmli sample buffer.

In Vivo Phosphorylation of Cyclin L2—Two days after transfection with GFP expression plasmids (1.5 µg of pEGFP-CycL2 and 1.5 µg of either pEGFP, pEGFP-DYRK1A, or pEGFP-CLK3), COS-7 cells (seeded in 6-cm plates) were washed once and then preincubated in phosphate-free medium for 20 min. Phosphate-free medium was obtained from Sigma (catalog no. D3656) and supplemented with 3.7 g/liter NaHCO3, 0.11 g/liter sodium pyruvate, and 10% dialyzed phosphate-free fetal calf serum (catalog no. F0392; Sigma). The medium was replaced with 2 ml of the same medium containing 30 µCi of radioactive carrier-free H332PO4 (Hartmann Analytic GmbH, Braunschweig, Germany). The cells were incubated for 2.5 h in the presence of 32P before SDS lysates were prepared for immunoprecipitation. Immunoprecipitated proteins were separated by SDS-PAGE and blotted onto nitrocellulose membrane, and incorporation of 32P was detected by exposure of the dried blots to phosphor storage screens.

Detection of GFP Fusion Proteins and Immunofluorescence—COS-7 cells were seeded in 24-well plates onto coverslips and transiently transfected with 0.3 µg of expression plasmids for GFP fusion proteins. Two days after transfection, the cells were washed with PBS and fixed with 3% paraformaldehyde in PBS for 20 min at room temperature. After washing with PBS, the cells were permeabilized with 0.2% Triton X-100 in PBS for 10 min. For detection of endogenous SC35, the cells were blocked with 3% (w/v) BSA in PBS for 20 min and then incubated for 1 h with the primary antibody (1:100 in blocking solution) at room temperature. After washing and incubation with secondary antibody (Alexa Red-conjugated anti-mouse antibody 1:100 in 3% BSA/PBS), the cells were washed extensively in PBS, mounted with Fluoromount-G (Southern Biotechnology Associates), and revealed with a confocal laser-scanning microscope, Zeiss LSM510.

Photobleaching Analysis—Photobleaching experiments were essentially performed as described (40). Briefly, living COS-7 cells were analyzed 48 h after transfection with pEGFP-CycL1 or pEGFP-CycL2. GFP was excited by the 488-nm line of the argon laser, and emission was detected using a 505–550-nm band path filter. For fluorescence recovery after photobleaching (FRAP) analysis, the region of interest (nuclear speckle) was bleached using the 488-nm laser at 100% intensity, and pictures were taken over a 10-min period of observation. For quantitative FRAP analysis, the optical slice was set to 1 µm, and bleaching was performed with 35 iterations. The fluorescence intensity was measured every second within a region of 1 µm in diameter. For fluorescence loss in photobleaching (FLIP) analysis, a region of 1 µm in diameter of the nucleoplasm, but distinct from nuclear speckles, was bleached consecutively using the 488-nm laser at 100% intensity. Pictures were taken over a 5-min period.

Pull-down Assay—To generate baits for pull-down experiments, bacterially expressed GST fusion proteins were purified by affinity purification with glutathione-Sepharose as described above, but without elution from the beads. These immobilized proteins were incubated with lysates of transfected COS-7 cells for 3 h at 4 °C under continuous agitation in 20 mM Tris-HCl, pH 7.5, 150 mM NaCl, 0.2 mM phenylmethylsulfonyl fluoride, and 0.5% BRIJ97 (Sigma). After washing five times with the same buffer, the bound proteins were eluted by incubation in Laemmli sample buffer at 96 °C, and the samples were subjected to Western blot analysis.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 EXPERIMENTAL PROCEDURES
 RESULTS
DISCUSSION
 REFERENCES
 
Screening for DYRK1A-interacting Proteins—In order to detect DYRK1A-binding proteins or domains, we screened a high density colony filter array containing 37,830 clones of a cDNA expression library from human fetal brain by overlaying with radiolabeled GST-DYRK1A-{Delta}C-K188R. The coordinates of positive clones were determined on the autoradiogram (Fig. 1A, left panel; only a section of the array is shown), and 23 clones were selected for further characterization. To verify the screening results, the recombinant proteins were expressed in E. coli and subjected to blot overlay assays with GST-DYRK1A-{Delta}CK188R. The probe bound to all of the proteins (Fig. 1B, left panel; only part of the clones is shown). cDNA sequencing showed six cDNA clones encoded fusion proteins derived from artificial reading frames (untranslated region or out-of-frame fusion). The clones with correctly fused reading frames are listed in Table I. Eight of the 17 clones contained cDNAs for known or putative nuclear proteins, including a splicing factor (SRp75), a transcription factor (RP58), and a nucleolar protein.



View larger version (83K):
[in this window]
[in a new window]
  		FIG. 1.
Screening of a human fetal brain expression library by overlay assay and solid phase-phosphorylation. A, screening of the expression array. Two replicas of the expression array were subjected to either interaction screening by overlaying with 32P-labeled DYRK1A-{Delta}C-K188R (left panel) or to solid phase phosphorylation by DYRK1A-{Delta}C (right panel). Positive colonies were identified by autoradiography and are marked with their RZPD clone numbers (see Table I). Clone P23533 [GenBank] encodes cyclin L2. Only a section of the array is shown. B, confirmation of the array results. Total cellular lysates from bacteria transformed with the indicated clones were separated by SDS-PAGE and blotted onto nitrocellulose membranes. Proteins were stained by Ponceau Red (lower panels) for evaluation of expression levels before the blots were subjected to overlay assay or solid phase phosphorylation. A randomly chosen clone (F10592 [GenBank] ) that was not phosphorylated in the array experiment served as a negative control. Clone P23533 [GenBank] encodes cyclin L2. Representative data for selected clones are shown. Migration of molecular mass markers is indicated in kDa.

 
Substrate Screening of a Protein Expression Library—An identical filter as the one used for interaction screening was subjected to solid phase phosphorylation by GST-DYRK1A-{Delta}C in the presence of [{gamma}-32P]ATP, and phosphorylated fusion proteins were detected by autoradiography (Fig. 1A, right panel; only a section of the array is shown). 34 strongly reacting cDNA clones were further characterized, including three clones that were also identified in the interaction screen. By solid phase phosphorylation of bacterially expressed fusion proteins, all of the selected proteins were confirmed to be in vitro substrates of DYRK1A (Fig. 1B, right panel; only part of the clones is shown). A randomly chosen clone that gave no detectable signal on the array was not phosphorylated by DYRK1A, demonstrating the reproducibility of the array results (control in Fig. 1B). Sequencing the cDNA inserts resulted in six known (one of them represented by two clones) and two hypothetical proteins being identified (Table I). The remaining 26 fusion proteins represented cloning artifacts (ribosomal RNA, untranslated regions of mRNA, out of frame fusions). All of the identified proteins either contain nuclear localization signals or are known nuclear proteins.

Phosphorylation of Splicing Factors by DYRK1A—Three clones encoding proteins with RS domains were identified both as DYRK1A-interacting proteins and as substrates of the kinase (Table I); clone M24584 codes for the splicing factor SRp75, and N16596 and P23533 [GenBank] both contain partial cDNA coding for an RS domain-containing protein that we designated cyclin L2 (see below). A total of four proteins with RS domains were identified (SFRS4, cyclin L2, SFRS5, and SWAP2) as in vitro substrates of DYRK1A, suggesting that DYRK1A may be involved in splicing catalysis. We therefore performed in vitro kinase assays to compare the phosphorylation of RS domain-containing proteins by DYRK1A. In addition to the proteins identified in the present screen, two additional RS domain-containing proteins (hTRA2{beta}1 and SF2/ASF) and a well characterized in vitro substrate of DYRK1A (myelin basic protein) (34) were included in this experiment. As shown in Fig. 2, cyclin L2 and hTRA2{beta}1 were phosphorylated with markedly higher intensities than the other RS proteins. Thus, DYRK1A does not indiscriminately phosphorylate serine residues in the context of RS domains. In the present study, we concentrated on the characterization of cyclin L2 and its interaction with DYRK1A, because this protein was independently identified in both screens and because cyclin L2 was an excellent substrate of DYRK1A (see below; Fig. 2).

Structure of Cyclin L2—The amino acid sequence derived from the cDNA clones N16596 and P23533 [GenBank] definitely encodes an RS domain (21 RS dipeptide motifs within 141 amino acids; Fig. 3A) but shows no extended sequence similarity with any other RS domain-containing protein. By a data base search, we identified an uncharacterized full-length mRNA sequence corresponding to these partial cDNAs (GenBankTM accession number AY037150 [GenBank] ). The deduced protein encoded by this cDNA exhibits high sequence similarity with cyclin L1 (HUGO gene symbol CCNL1), a recently characterized protein with an N-terminal cyclin domain and a C-terminal RS domain (Fig. 3B) (39, 41). Therefore, we designated the protein encoded by AY037150 [GenBank] cyclin L2 (HUGO gene symbol CCNL2). The mouse homologues of cyclin L1 and cyclin L2 have previously been named Ania-6a and Ania-6b, respectively (41). The cyclin domain of cyclin L2 (amino acids 69–296, as indicated in Fig. 3A) was defined by comparison with cyclin H, which has been modeled by Andersen et al. (42). The cyclin domains of cyclin L1 and L2 are most closely related with cyclin K (HUGO gene symbol CCNK; 33% identical amino acids) and cyclin T1/T2 (HUGO gene symbols CCNT1 and CCNT2; 30 and 32% identical amino acids). The RS domains of cyclin L1 and L2 contain 45% identical amino acids, a range that was also found for RS domains of known splicing factors.

Genomic Organization of CCNL2—The human CCNL2 gene is located on chromosome 1 (1p36.33). The gene spans about 13 kb and comprises 11 exons (Fig. 3C). The comparison of the genomic sequence with public expressed sequence tag sequences allowed us to deduce the existence of three different transcripts, T1–T3 (Fig. 3C). The different transcripts are generated by alternative splicing of exon 6a and by the use of alternative polyadenylation sites within exon 6a. Inclusion of exon 6a introduces a stop codon, and therefore transcripts T1 and T3 encode a truncated version of cyclin L2 that lacks the RS domain (cyclin L2S, "s" represents "short," Fig. 3, A and B). Because exon 6a is skipped in transcript T2, the resulting mRNA (AY037150 [GenBank] ) encodes the long version of the protein, which is named cyclin L2 and includes the RS domain.

Alternatively Spliced Transcripts of CCNL2 and CCNL1— We performed Northern blot analyses with exon-specific hybridization probes (Fig. 3C) and PCR analysis in order to determine the expression pattern of CCNL2 in different tissues and to verify the existence of the hypothesized transcripts of the CCNL2 gene. On Northern blots of RNA from different human tissues, two major bands of about 4.5 and 2.4 kb were detected by a probe corresponding to the 3'-untranslated region in exon 11 of CCNL2 (Fig. 4A, upper left panel). These sizes correspond very well to the calculated lengths of the transcripts T1 and T2 (see Fig. 3C). Transcript T1, which encodes cyclin L2S, was ubiquitously found in all tissues tested, although signal intensities varied. In ovary, testis, prostate, pancreas, skeletal muscle, and liver, T1 appeared to be more abundant than in blood, colon, lung, and brain. In contrast, transcript T2 was weakly detected in a few tissues (skeletal muscle, and spleen), and only testis RNA contained comparable levels of both transcripts (T1 and T2). The 2.4-kb transcript (T2) appears as a double band on the autoradiogram, because there are two polyadenylation sites in exon 11 that are separated by 211 bp (see Fig. 3C). To confirm the identification of T1 and T2, we hybridized identical blots with a probe derived from the 5'-end of exon 6a (Fig. 4A, lower left panel). As expected, this probe detected the same 4.5-kb band (T1) as the 3'-untranslated region probe but did not hybridize with the 2.4-kb transcript (T2). In addition, we identified a band with an approximate size of 1.2 kb. We hypothesize that this band corresponds to a transcript comprising exons 1–5 (a total of 677 bp) and about 500 bp of exon 6a (transcript T3 in Fig. 3C). Indeed, exon 6a harbors a potential polyadenylation signal (AAUACA) (43) at nucleotide 1165 of BC016333 [GenBank] (GenBankTM accession number). We identified this clone in the human data bases as a full-length mRNA, which represents transcript T3. Consistent with this identification of the 1.2-kb band, a probe specific for the 3'-end of exon 6a (E6a-3') detected only the 4.5-kb band and reacted neither with the 1.2-kb transcript nor the 2.4-kb transcript (data not shown).



View larger version (78K):
[in this window]
[in a new window]
  		FIG. 4.
Differential expression of cyclin L1 and cyclin L2 splicing variants. A, Northern blot analysis of cyclin L1 and cyclin L2 in human tissues. Northern blots containing total RNA from human tissues were hybridized with gene-specific probes that were derived from exon 11 of CCNL2 (E11) and the last exon of CCNL1 (L1–E14; upper panels) or from the 5'-end of exon 6a in CCNL2 (E6a-5') and the corresponding region of CCNL1 (L1-E7+8; lower panels). The localization of the probes is shown in Fig. 3C. The identification of the transcripts (T1, T2, and T3) according to the scheme in Fig. 3C is indicated. pa, pancreas; ki, kidney; sm, skeletal muscle; li, liver; lu, lung; pl, placenta; br, brain; he, heart; bl, peripheral blood lymphocytes; co, colon; si, small intestine; ov, ovary; te, testis; pr, prostate; th, thymus; sp, spleen. B, design of primers for the detection of cyclin L2 splicing variants by PCR. The common forward primer (E2for) and two reverse primers matching either exon 6a (E6arev) or exon 6b (E6brev) were designed to specifically amplify transcripts T1/T3 or T2. The short polymerization time of 30 s did not allow the amplification of T1 by primer E6brev. Control PCRs of T1/T3 (primers E2for and E6arev) and T2 (primers E2for and E6brev) with a human testis cDNA library as template are shown in the right panel. C, PCR analysis of cyclin L2 splicing variants. First-strand cDNAs from the indicated human and mouse tissues were subjected to PCR analysis with primers E2for, E6arev, and E6brev in a single reaction. Reaction products from human tissues were detected by Southern blot hybridization with a probe recognizing both products. PCR products from mouse tissues were visualized by ethidium bromide staining.

 
The structure of the CCNL1 gene resembles that of CCNL2, and there are alternative splicing variants of cyclin L1 that result in a long and a short form of the protein (41, 39), similar to cyclin L2 and cyclin L2S. To directly compare the expression patterns of CCNL1 and CCNL2, we hybridized the same Northern blots with CCNL1-specific probes that corresponded to the probes used for detection of CCNL2 mRNA. Similar to CCNL2, a probe specific for the last exon of CCNL1 detected two bands of 4.3 and 2.3 kb (Fig. 4A, upper right panel). This result suggests that alternative transcripts of CCNL1 are generated in the same way as T1 and T2 of CCNL2. Consistent with this identification of the bands in the Northern blot, a probe equivalent to E6a-5' detected only the longer transcript of CCNL1 (Fig. 4A, lower right panel). No transcript of CCNL1 was detected that corresponded to T3 of CCNL2. As in CCNL2, a stop codon in the longer transcript results in a truncated version of the protein. These data suggest that alternative splicing occurs in the same way in both isoforms of cyclin L; inclusion of exon 6a of CCNL2 or the corresponding region of CCNL1 leads to a larger transcript, which encodes a truncated protein lacking the RS-domain and part of the cyclin domain.

To confirm the identification of the bands detected by Northern blot analysis, we characterized the ratio of the cyclin L2 transcripts in different tissues by PCR analysis. We designed exon-specific primers to amplify transcripts T1/T3, coding for the short protein (cyclin L2S), and transcript T2, coding for the long form (cyclin L2) (Fig. 4B). Combination of the three primers in a single reaction (multiplex PCR) allowed us to compare the expression ratios of T1/T3 with T2 in different tissues (Fig. 4C). It should be noted that the ratio of the PCR products is not necessarily equal to the ratio of the respective cDNA sequences, because the different products may have been amplified with different efficiencies. However, the data indicate whether this ratio differs between different samples (e.g. brain and testis cDNA). As shown in Fig. 4C, transcript T2 was detected in all tissues analyzed except for human brain. Nevertheless, transcript T2 was readily amplified from brain cDNA when only T2-specific primers were used for the PCR (data not shown). Consistent with the Northern blot data, stronger signals were obtained in most tissues for transcripts T1/T3 than for T2, except for testis, which contained relatively higher levels of T2. In mouse tissues, the ratio of T2 and T1/T3 was also higher in testis than in brain (Fig. 4C). Moreover, an antibody specific for the C-terminal domain of cyclin L2 detected a band of about 65 kDa in Western blots of mouse testis, matching the size of full-length cyclin L2 as deduced from transcript T2 (data not shown).

Localization of Cyclin L1 and Cyclin L2 to the Nuclear Speckle Compartment—Many splicing factors with RS domains are localized in subnuclear structures called nuclear speckles (44). Therefore, we generated GFP fusion constructs of cyclin L2, cyclin L2S, and cyclin L1 in order to study the subcellular localization of these proteins by fluorescence microscopy. As shown in Fig. 5, the fluorescent signal of both GFP-CycL1 and GFP-CycL2 appeared as discrete, punctate staining within the nucleus. Moreover, these structures were co-stained by an antibody directed against the splicing factor SC-35, a well characterized marker protein for the splicing factor compartment (45). Interestingly, cells overexpressing GFP-CycL1 consistently exhibited an altered appearance of the SC-35 containing speckles compared with untransfected cells. Nuclei of transfected cells contained a higher number of speckles, which were of smaller size (Fig. 5, lower panels). Overexpression of GFPCycL2 did not significantly affect the pattern of speckles (Fig. 5, upper panels). In contrast to GFP-CycL2, GFP-CycL2S, which lacks the RS domain, was evenly distributed to both cytoplasm and nucleus, as previously observed for the short splice variant of cyclin L1 (41). No speckled appearance of GFP-CycL2S was observed in the nucleus.



View larger version (42K):
[in this window]
[in a new window]
  		FIG. 5.
Differential subnuclear localization of cyclin L1, cyclin L2 and cyclin L2S. COS-7 cells transfected with the indicated GFP fusion proteins were fixed and after permeabilization were labeled using an anti-SC35 antibody. GFP (green) and Alexa Red-coupled secondary antibody (red) were visualized by confocal laser-scanning microscopy. Scale bars, 10 µm.

 
Cyclin L2 in Nuclear Speckles Is a Highly Mobile Protein— Nuclear speckles are considered to be storage compartments from which proteins involved in RNA transcription and processing are constantly recruited to sites of active RNA synthesis (44, 46). We used FRAP to analyze the mobility of GFP-CycL1 and GFP-CycL2 in living COS-7 cells. After bleaching of GFP fluorescence in a single nuclear speckle, recovery of fluorescence over time was monitored by live cell imaging. For GFPCycL2, recovery of fluorescence was observed within 20–60 s after bleaching (Fig. 6A), indicating that the protein in the speckles is highly mobile and rapidly exchanges with unbleached GFP-CycL2 molecules in the nucleus. Quantitative analysis showed that the recovery of fluorescence was fast (initial t of 3–5 s) but incomplete (Fig. 6C). This result shows that GFP-CycL2 in nuclear speckles consists of a highly mobile and an immobile fraction. In striking contrast, no recovery of fluorescence was observed after bleaching of GFP-CycL1 (Fig. 6A). Even 10 min after bleaching, no recovery was detected in 14 of 15 cells monitored (data not shown). This result indicates that GFP-CycL1 is an immobile component of the nuclear speckles.



View larger version (72K):
[in this window]
[in a new window]
  		FIG. 6.
FRAP and FLIP analysis of GFP-CycL1 and GFP-CycL2. A, FRAP analysis. COS-7 cells transfected with GFP-CycL1 or GFP-CycL2 were imaged before and during fluorescence recovery after photobleaching of a nuclear speckle with high intensity laser light at the time points indicated (A). The area of the bleach spot is indicated with a circle. B, FLIP analysis. The circled area within the nucleus of COS-7 cells transfected with GFP-CycL1 or GFP-CycL2 was continuously photobleached. Fluorescence intensity is shown in a false color code prior to bleaching (0 s) and after three consecutive bleach periods (10, 20, and 60 s). C, quantitative FRAP analysis. Regions of 1 µm in diameter within nuclear speckles of GFP-CycL2-transfected COS-7 cells were bleached, and subsequently the recovery of fluorescence was measured every second. The diagram depicts data from five speckles of a single cell and is a representative result from the analysis of three different cells.

 
To confirm that GFP-CycL1 and GFP-CycL2 indeed differed in their intranuclear mobility, we performed FLIP analysis of both proteins in living COS-7 cells. Repetitive bleaching of a small area of the nucleus (1 µm) free of nuclear speckles significantly reduced the total nuclear GFP fluorescence of GFPCycL2 expressing cells, indicating that a substantial fraction of the protein has moved through the area subjected to bleaching within the period of 60 s (Fig. 6B). In contrast, GFP-CycL1 fluorescence was affected exclusively in the bleaching area itself but remained unchanged in the remaining volume of the nucleus. Consistent with the FRAP data, this result indicates that cyclin L1 does not freely diffuse within the nucleus.

Mapping the Binding Domain for Cyclin L2—To identify the domain responsible for the interaction of DYRK1A and cyclin L2, we performed pull-down experiments with deletion mutants of DYRK1A (Fig. 7). GST-DYRK1A-{Delta}C, the construct used for interaction screening, lacks the noncatalytic C-terminal domain. GST-DYRK1A-{Delta}C{Delta}N expresses only the catalytic domain and GST-DYRK1A-NT contains only the N terminus and a small part of the catalytic domain (Fig. 7). The serine/arginine-rich splicing factor, SF2 (also called ASF or SRp30a), was used for comparison, because it is known that several splicing factors bind to each other by heterodimerization of their RS domains (47).



View larger version (29K):
[in this window]
[in a new window]
  		FIG. 7.
Mapping of the cyclin L2 binding domain in DYRK1A. Total cellular lysate from GFP-CycL2-{Delta}N-transfected COS-7 cells was incubated with GST or the indicated GST fusion constructs of DYRK1A, DYRK2, and SF2 immobilized to glutathione-Sepharose beads. After washing five times, proteins retained on the Sepharose and 5% of the input (GFP-CycL2-{Delta}N-containing lysate) were resolved by SDS-PAGE followed by immunoblotting (WB) using a GFP-specific antibody (upper panel). Ponceau staining shows the amounts of the bait proteins (lower panel). The bands corresponding the full-length protein are marked by asterisks. DYRK1A-NT has a calculated size of 44 kDa, indicating that most of the protein is truncated. The baits used are schematically shown in the right panel. The percentage of identical amino acids between DYRK1A and DYRK2 in the DH box and the catalytic domain (cat) is indicated.

 
As shown in Fig. 7, similar amounts of GFP-CycL2{Delta}N were pulled down by SF2 and by DYRK1A-{Delta}C. This result indicates that the C-terminal part of cyclin L2 can interact comparably well with other RS domain-containing proteins and with the protein kinase DYRK1A. Cyclin L2 bound also to the related kinase, DYRK2. Notably, sequence similarity between DYRK2 and DYRK1A is limited to the catalytic domain and a short segment immediately N-terminal of the catalytic domain that we earlier designated the "DYRK homology box" (DH box) (48). The DH box contains a large number of negative charges (35) and may thus bind the RS domain of cyclin L2 by ionic interactions. Both DYRK1A-{Delta}N{Delta}C, which lacks the DH box, and the N-terminal domain alone bound cyclin L2 only very weakly. Taken together, these results suggest that optimal binding of cyclin L2 to DYRK1A (or DYRK2) requires the presence of both the catalytic domain and the DH box.

Interaction of Cyclin L2 with PITSLREp110—Cyclin L1 has been reported to associate with the p110 isoform of PITSLRE (PITSLREp110), an orphan cyclin-dependent kinase (41, 39) (official gene symbol CDC2L1). Because of the high sequence similarity of the cyclin domains of cyclin L1 and cyclin L2, we tested whether cyclin L2 also interacted with PITSLREp110 (Fig. 8). In co-immunoprecipitation experiments with GFP fusion proteins of cyclin L1, cyclin L2, and cyclin L2S, we observed that PITSLREp110 co-precipitated with cyclin L2, although less efficiently than with cyclin L1. Neither cyclin L2S nor an unrelated control protein (the non-RS-splicing factor SF3B1) bound detectably to PITSLREp110.



View larger version (70K):
[in this window]
[in a new window]
  		FIG. 8.
Co-immunoprecipitation of cyclin L1, L2, and L2S with PITSLREp110. COS-7 cells were transfected with GFP constructs of cyclin L1, cyclin L2, cyclin L2S, or an unrelated control protein (SF3B1). Fusion proteins were immunoprecipitated with a GFP-specific antiserum, and immunoprecipitates (IP) were analyzed by Western blotting (WB) with anti PITSLRE antibody (upper panel) and anti-GFP antibody (lower panel). The p110 band of PITSLRE is marked by an arrow. Lysate, one-fiftieth of input was loaded on the gel.

 
Identification of in Vitro Phosphorylation Sites—Our finding that different arginine/serine-rich splicing factors were not equally well phosphorylated by DYRK1A (Fig. 2) raises the question of which sites in cyclin L2 are targeted by DYRK1A. Bacterially expressed His6-cyclin L2 was phosphorylated by DYRK1A in vitro and subjected to SDS-PAGE. As shown in Fig. 9, two slower migrating bands appeared after phosphorylation of cyclin L2 by DYRK1A. The three bands were individually analyzed for phosphorylated amino acids by mass spectroscopy in comparison with unphosphorylated cyclin L2. Three phosphoserines were identified in the slower migrating bands (Fig. 9; Ser-330, Ser-338, and Ser-369). All of these serine residues are located N-terminal of proline residues, consistent with our previous classification of DYRK1A as a "proline-directed" kinase. It should be noted that other phosphorylation sites may have escaped detection, since the predicted size of the tryptic fragments from the RS domain would have been too small to successfully purify and analyze by the present methodology.



View larger version (26K):
[in this window]
[in a new window]
  		FIG. 9.
Phosphorylation of cyclin L2 by DYRK1A in vitro. Bacterially expressed His6-CycL2 was phosphorylated with GSTDYRK1Acat (1:1 molar ratio), and the resulting bands were resolved by SDS-PAGE (right lane) in parallel with unphosphorylated His6-CycL2. The three bands were cut, and the indicated phosphorylation sites in each single band were determined by mass spectrometry of tryptic peptides. The amino acid sequences surrounding the three phosphorylated sites are shown.

 
In Vivo Phosphorylation of Cyclin L2—To analyze whether cyclin L2 is a phosphoprotein in vivo, COS-7 cells transfected with GFP constructs of cyclin L2 alone or with co-transfected DYRK1A, respectively, were incubated with 32P-labeled phosphate. CLK3 was compared with DYRK1A, because kinases of the CLK family are known to phosphorylate RS domains of arginine/serine-rich splicing factors (49). As shown in Fig. 10A, immunoprecipitated cyclin L2 was in fact a phosphoprotein when expressed in mammalian cells. Coexpression of DYRK1A or CLK3 did not detectably alter the level of phosphate incorporation, possibly because cyclin L2 was already strongly phosphorylated by endogenous protein kinases. However, a more slowly migrating form of GFP-CycL2 was observed in the immunoprecipitates from DYRK1A-overexpressing cells. This band was also detected when total cellular lysates of DYRK1A-expressing COS-7 cells were directly analyzed by Western blotting (Fig. 10B) but not after transfection of a kinase-negative point mutant of DYRK1A (DYRK1A-K188R). This result strongly suggests that cyclin L2 is directly phosphorylated by DYRK1A in living cells.



View larger version (34K):
[in this window]
[in a new window]
  		FIG. 10.
Phosphorylation of cyclin L2 by DYRK1A in vivo. COS-7 cells were co-transfected with 1.5 µg of plasmids encoding GFP-CycL2 and either GFP, GFP-DYRK1A, GFP-DYRK1A-K188R, or GFP-CLK3 as indicated above the lanes. A, cells were incubated with radioactive H3PO4 for 2.5 h, and GFP fusion proteins were immunoprecipitated under denaturing conditions with GFP antiserum. Phosphorylation was detected by autoradiography. B, total cellular lysates were subjected to Western blot (WB) analysis with a GFP-specific antibody. Two differently exposed films of the same blot are shown to visualize the shifted band of GFP-CycL2 (upper panel) or the detection of DYRK1A and DYRK1A-K188R (lower panel).

 

   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 EXPERIMENTAL PROCEDURES
 RESULTS
 DISCUSSION
REFERENCES
 
In this study, we have made use of an arrayed cDNA expression library to identify interaction partners as well as substrates of the nuclear protein kinase DYRK1A. Screening of phage libraries by overlay assay with radiolabeled proteins as well as substrate screening by solid phase phosphorylation have previously been employed to identify kinase substrates (21, 26). The main advantage of using the filter array for screening is the possibility to directly isolate positive clones without further rounds of purification. The second advantage is that the expression plasmids isolated from the library can be directly used for functional assays. These advantages made it feasible to select 58 positive clones to verify the original screening results with recombinant proteins (Fig. 1B). All tested clones were able to bind to DYRK1A and/or to be phosphorylated by DYRK1A, demonstrating the reliability of the screening results. Disregarding the cloning artifacts, we identified a high proportion of nuclear proteins that are potential downstream targets of DYRK1A, in particular the splicing proteins and transcription factors listed in Table I. The present work provides a more detailed characterization of a novel member of the cyclin family, cyclin L2, as an interaction partner and substrate of DYRK1A.

Cyclin L2 and the closely related isoform, cyclin L1 (originally named cyclin L (39)) differ from all other members of the cyclin family by the presence of a C-terminal RS domain. This unusual domain structure is conserved in the homologous proteins from invertebrates and plants (Drosophila, Caenorhabditis, and Arabidopsis (41)), providing evidence for a conserved function of the RS domain in these proteins. The RS domain is a hallmark of many proteins involved in pre-mRNA processing, including the serine/arginine-rich splicing factors (SR proteins), which contain at least one RNA-binding motif, and the SR protein-related polypeptides (50, 51, 47, 52). SR protein-related polypeptides constitute an inhomogeneous family of proteins that contain the RS domain in combination with additional domains (e.g. the protein kinase domains in CLK1-CLK4 and CrkRS/CRK7 or the DEAD/H box in RNA helicases) (47, 52). The modular domain architecture of cyclin L1 and L2 is thus a typical feature of SR protein-related proteins, although unusual for members of the cyclin family.

RS domains mediate protein-protein interactions with other proteins that contain RS repeats, and they target proteins to the nuclear speckle compartment (50). Here we show that cyclin L2 interacts with the alternative splicing factor SF2/ASF (Fig. 7) and is co-localized with the SR protein SC35 in nuclear speckles (Fig. 5). This subcellular distribution is very likely mediated by the RS domain, since a construct of cyclin L2 that lacked the major part of the cyclin domain (cyclin L2{Delta}1–222) exhibited a subnuclear distribution indistinguishable from that of the full-length construct (data not shown). In contrast, cyclin L2S, which lacks the RS domain, was distributed both in the nucleus and in the cytoplasm (Fig. 5). Cyclin L1 was also found to be co-localized with SC35, as previously shown by Berke et al. (41). However, cyclin L1 and cyclin L2 differed strikingly in regard to their intranuclear mobility.

Photobleaching experiments showed that a significant fraction of GFP-CycL2 in nuclear speckles was rapidly exchanged by unbleached molecules within the nucleus (FRAP analysis; Fig. 6, A and C). Moreover, FLIP analysis showed that a considerable portion of nuclear GFP-CycL2 transits a given small region of the nucleus within a time period of 60 s (Fig. 6B), indicating that the protein diffuses freely in the nucleoplasm when not associated with speckles. In contrast, GFP-CycL1 was found immobile in both FRAP and FLIP analysis. To our knowledge, the observation that GFP-CycL1 is an immobile component of nuclear speckles is unprecedented in the analysis of proteins with RS domains (for a review, see Ref. 44). The different mobilities of cyclin L1 and cyclin L2 indicate that the proteins have different functions despite their highly similar structures. Presently, we can only speculate on the reason for the different mobilities of cyclin L1 and cyclin L2. The fact that cyclin L1 was co-immunoprecipitated with PITSLREp110 argues against the possibility that the GFP fusion protein is grossly misfolded in the cell. Instead, the association of cyclin L1 with its cyclin dependent kinase, PITSLREp110, may constrain the mobility of cyclin L1, whereas the association of cyclin L2 and PITSLREp110 appears weaker (Fig. 8).

Altogether, the presence of the RS domain, the interaction with SF2/ASF, and the localization in the nuclear speckle compartment suggest a role of cyclin L2 in the regulation of RNA splicing. Cyclin L1 has previously been reported to stimulate an in vitro splicing reaction (39). Notably, however, speckles have also been shown to contain proteins other than splicing factors, including 3'-end mRNA-processing factors, the cyclin T1-CDK9 complex, and the hyperphosphorylated form of RNA polymerase II (53–55). Thus, a function of cyclin L2 in the control of transcription or another step of pre-mRNA processing cannot be excluded.

Based on sequence comparisons of the cyclin box, cyclin L2 belongs to a group of cyclins involved in transcriptional regulation, comprising cyclins H, K, T1, and T2 (Fig. 11).



View larger version (10K):
[in this window]
[in a new window]
  		FIG. 11.
Relationship of cyclin L2 with other members of the cyclin family. The dendrogram was constructed from an alignment of the cyclin (Cyc) domains with the help of the Clustal program (66). The branch of transcriptional cyclins is highlighted by boldface type. Data base (Swiss-Prot) accession numbers are as follows: cyclin K, O75909 [GenBank] ; cyclin T1, O60563 [GenBank] ; cyclin T2, O60583 [GenBank] ; cyclin H, P51946 [GenBank] ; cyclin A1, P78396 [GenBank] ; cyclin B1, P14635 [GenBank] ; cyclin D1, P24385 [GenBank] ; cyclin E1, P24864 [GenBank] .

 
These cyclins, in association with the corresponding cyclin-dependent kinases, are components of the basal transcription factor TFIIH (cyclin H-CDK7) or the transcriptional elongation factor P-TEFb that phosphorylates the C-terminal domain of the large subunit of RNA polymerase II (cyclin T-CDK9 and cyclin K-CDK9) (56). The present data show that cyclin L2 binds in vivo to the cyclin-dependent kinase PITSLREp110, as previously also demonstrated for cyclin L1 (41, 39). PITSLRE (also named CDK11 (57) or Cdc2-like 1 (official gene symbol CDC2L1)) exists in more than 20 splicing variants (58). PITSLREp110 isoforms are localized in the nucleus and are co-localized in nuclear speckles with the splicing protein RNPS1 (59). Here we show that cyclin L2 is co-localized with SC35 in the splicing factor compartment, supporting the hypothesis that cyclin L2 and PITSLREp110 are associated in vivo. Recent reports provide evidence that PITSLRE participates in a signaling pathway that coordinates transcription and RNA processing events (57, 60). It should be noted that in our coimmunoprecipitation experiments (Fig. 8), cyclin L2 did not interact as strongly with PITSLREp110 as it did with cyclin L1, and we cannot exclude the possibility that related kinases such as CDK10/PISSLRE or CDK7 may act as the catalytically active partner of cyclin L2.

Interestingly, we have identified a short variant of cyclin L2 (cyclin L2S) that lacks the RS domain and part of the cyclin box. Sequence comparisons of the human gene for cyclin L2 (CCNL2) and its transcripts allowed us to deduce the alternatively spliced transcripts for the long and the short variant (Fig. 3C). The exon/intron structure of the gene for cyclin L1 (CCNL1) is nearly identical to that of CCNL2, and CCNL1 results in the same alternative splicing variants coding for full-length cyclin L1 and a short isoform of cyclin L1 (cyclin L1S, designated Ania-6a25 by Berke et al. (41)). By Northern blot analysis with exon-specific probes, we could unambiguously identify the larger transcripts of about 4.5 kb (T1 in Fig. 4A) as the mRNA for the short polypeptides (cyclin L1S and L2S). This result is in agreement with the analysis of the murine cyclin L1 transcripts by Berke et al. (41) and improves the characterization of the human CCNL1 gene and its transcripts performed by Dickinson et al. (39).

The alternative splicing variants of cyclin L1 and L2 are differentially expressed in human tissues (Fig. 4A). In most tissues, the mRNA levels for full-length cyclin L1 (T2) were comparable with or higher than those for cyclin L1S. In contrast, the mRNA for full-length cyclin L2 was detected mainly in testis, whereas the short variant, cyclin L2S, was nearly ubiquitously expressed. Furthermore, we found that the transcripts of CCNL1 and CCNL2 were differentially regulated by extracellular stimuli; phorbol ester and cycloheximide specifically induced the mRNA for full-length cyclin L1 in human RT112 keratinocytes, whereas levels of the cyclin L2 transcripts were only weakly affected by cycloheximide and not at all by phorbol ester.2 These results are in accordance with a more extensive analysis of cyclin L1 expression in rat neurons by Berke et al. (41), who identified CCNL1 as an immediate early gene induced by cocaine and dopamine. Taken together, the present characterization of the cyclin L2 gene suggests that CCNL1 and CCNL2 originated by gene duplication and have acquired different patterns of expression by divergent evolution of their promoters.

The fact that both splicing variants are conserved in the paralogous genes CCNL1 and CCNL2, together with their differential patterns of expression, argues against the possibility that cyclin L1S and L2S are products of irregular splicing. However, the cyclin domain and the RS domain are truncated or completely absent, respectively, in cyclin L1S and L2S. Cyclin L2S neither interacts with PITSLREp110, most likely because of its incomplete cyclin box (as defined by Andersen et al. (42)), nor is localized to nuclear speckles. Identical results have been reported for cyclin L1S (39, 41). The existence of these short variants with no apparent function is reminiscent of the alternative splicing in the CLK kinases (61). CLK1 is regulated through alternative splicing of the CLK1 pre-mRNA, yielding mRNAs encoding the catalytically active kinase and truncated polypeptides lacking the kinase domain. CLK1 regulates splicing by phosphorylation of SR proteins and controls its own expression, since the presence of the active kinase favors the production of the truncated, inactive polypeptide (61). There are other known cases of splicing factors that are subject to autoregulation by alternative splicing (62, 63), and we hypothesize that the existence of the short splicing variants of cyclins L1 and L2 provides a means to control the levels of the full-length proteins at the level of splicing.

We have characterized cyclin L2 both as a substrate and an interaction partner of the protein kinase DYRK1A. It appears plausible that these proteins also interact in vivo, since DYRK1A has previously been found to be localized in nuclear speckles (17, 19). The original cyclin L2 clone isolated from the expression library lacked the cyclin domain, thus allowing the conclusion that DYRK1A interacts with the RS domain of cyclin L2. The results of the pull-down assays indicate that the catalytic domain of DYRK1A alone is able to bind cyclin L2 but that this interaction is greatly enhanced by the presence of a small conserved region preceding the catalytic domain, called the DH box. This region contains a large number of acidic residues (35) and is conserved in all kinases of the DYRK family and in several related kinases that are involved in the regulation of splicing (CLK family, PRP4 (48)). It appears reasonable to assume that the negatively charged DH box is engaged in ionic interactions with the positively charged RS domain.

DYRK1A phosphorylated cyclin L2 in vitro within the C-terminal domain, but the three in vitro phosphorylation sites identified in the present study were not located within RS repeat motifs (Fig. 9). This finding confirms our previous classification of DYRK1A as a proline-directed kinase (35). Notably, none of the target sites in cyclin L2 contains an arginine residue in position P-2 or P-3 that we and others have previously found to be critical for substrate recognition (14, 34, 64). We could further show that cyclin L2 is indeed a phosphoprotein when expressed in COS-7 cells. The phosphate content of cyclin L2 was not significantly altered by co-expressed DYRK1A (Fig. 10A), probably because RS domains are generally highly phosphorylated by endogenous kinases (47). However, overexpression in COS-7 cells of catalytically active DYRK1A, but not of a kinase-inactive mutant, induced a mobility of shift of cyclin L2, resembling the result of the in vitro phosphorylation of His6-CycL2 by DYRK1A (Fig. 9). This result strongly suggests that DYRK1A directly phosphorylates cyclin L2 in living cells. The phosphorylation of SR proteins is necessary for their recruitment from nuclear speckles to sites of transcription/pre-mRNA processing (44, 65). Interestingly, Álvarez et al. (19) have recently reported that kinase-active DYRK1A, but not an inactive point mutant, induced speckle disassembly. However, it remains to be determined whether phosphorylation of cyclin L2 is involved in this action of DYRK1A.


   FOOTNOTES
 
* This work was supported by Deutsche Forschungsgemeinschaft Grants Be 1967/1-3 and SFB542. 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. Back

** To whom correspondence should be addressed: Institut für Pharmakologie und Toxikologie, Medizinische Fakultät der RWTH Aachen, Wendlingweg 2, 52074 Aachen, Germany. Tel.: 49-241-8089136; Fax: 49-241-8082433; E-mail: walter.becker{at}post.rwth-aachen.de.

1 The abbreviations used are: TBS, Tris-buffered saline; CDK, cyclin-dependent kinase; FLIP, fluorescence loss in photobleaching; FRAP, fluorescence recovery after photobleaching; GFP, green fluorescent protein; GST, glutathione S-transferase; PBS, phosphate-buffered saline; RS domain, arginine- and serine-rich domain; BSA, bovine serum albumin; DH box, DYRK homology box. Back

2 K. de Graaf and W. Becker, unpublished data. Back


   ACKNOWLEDGMENTS
 
We thank Stefan Stamm (Erlangen, Germany) for the gift of expression plasmids (GST-TRA2{beta}1 and GST-SF2). The excellent technical assistance of Hanna Czajkowska and Birgit Feulner is gratefully acknowledged. We thank Hans-Georg Joost and Peter C. Heinrich for critical discussion and continuous support of this work. We are grateful to Bob Kosier for critical reading of the manuscript.



   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 EXPERIMENTAL PROCEDURES
 RESULTS
 DISCUSSION
 REFERENCES
 

  1. Hämmerle, B., Elizalde, C., Galceran, J., Becker, W., and Tejedor, F. J. (2003) J. Neural Transm. Suppl., in press
  2. Tejedor, F., Zhu, X. R., Kaltenbach, E., Ackermann, A., Baumann, A., Canal, I., Heisenberg, M., Fischbach, K. F., and Pongs, O. (1995) Neuron 14, 287-301[CrossRef][Medline] [Order article via Infotrieve]
  3. Guimera, J., Casas, C., Pucharcos, C., Solans, A., Domenech, A., Planas, A. M., Ashley, J., Lovett, M., Estivill, X., and Pritchard, M. A. (1996) Hum. Mol. Genet. 5, 1305-1310[Abstract/Free Full Text]
  4. Ohira, M., Seki, N., Nagase, T., Suzuki, E., Nomura, N., Ohara, O., Hattori, M., Sakaki, Y., Eki, T., Murakami, Y., Saito, T., Ichikawa, H., Ohki, M. (1997) Genome Res. 7, 47-58[Abstract/Free Full Text]
  5. Smith, D. J., Stevens, M. E., Sudanagunta, S. P., Bronson, R. T., Makhinson, M., Watabe, A. M., O'Dell, T. J., Fung, J., Weier, H. U., Cheng, J. F., and Rubin, E. M. (1997) Nat. Genet. 16, 28-36[Medline] [Order article via Infotrieve]
  6. Altafaj, X., Dierssen, M., Baamonde, C., Marti, E., Visa, J., Guimera, J., Oset, M., Gonzalez, J. R., Florez, J., Fillat, C., and Estivill, X. (2001) Hum. Mol. Genet. 10, 1915-1923[Abstract/Free Full Text]
  7. Fotaki, V., Dierssen, M., Alcantara, S., Martinez, S., Marti, E., Casas, C., Visa, J., Soriano, E., Estivill, X., and Arbones, M. L. (2002) Mol. Cell Biol. 22, 6636-6647[Abstract/Free Full Text]
  8. Hämmerle, B., Vera-Samper, E., Speicher, S., Arencibia, R., Martinez, S., and Tejedor F. J. (2002) Dev. Biol. 246, 259-273[CrossRef][Medline] [Order article via Infotrieve]
  9. Galceran, J., de Graaf, K., Tejedor, F. J., and Becker, W. (2003) J. Neural Transm. Suppl., in press
  10. Woods, Y. L., Rena, G., Morrice, N., Barthel, A., Becker, W., Guo, S., Unterman, T. G., and Cohen, P. (2001) Biochem. J. 355, 597-607[CrossRef][Medline] [Order article via Infotrieve]
  11. von Groote-Bidlingmaier, F., Schmoll, D., Orth, H. M., Joost, H. G., Becker, W., and Barthel, A. (2003) Biochem. Biophys. Res. Commun. 300, 764-769[CrossRef][Medline] [Order article via Infotrieve]
  12. Yang, E. J., Ahn, Y. S., and Chung, K. C. (2001) J. Biol. Chem. 276, 39819-39824[Abstract/Free Full Text]
  13. Mao, J., Maye, P., Kogerman, P., Tejedor, F. J., Toftgard, R., Xie, W., Wu, G., and Wu, D. (2002) J. Biol. Chem. 277, 35156-35161[Abstract/Free Full Text]
  14. Woods, Y. L., Cohen, P., Becker, W., Jakes, R., Goedert, M., Wang, X., and Proud, C. G. (2001) Biochem. J. 355, 609-615[CrossRef][Medline] [Order article via Infotrieve]
  15. Chen-Hwang, M. C., Chen, H. R., Elzinga, M., and Hwang, Y. W. (2002) J. Biol. Chem. 277, 17597-17604[Abstract/Free Full Text]
  16. Song, W. J., Chung, S. H., and Kurnit, D. M. (1997) Biochem. Biophys. Res. Commun. 231, 640-644[CrossRef][Medline] [Order article via Infotrieve]
  17. Becker, W., Weber, Y., Wetzel, K., Eirmbter, K., Tejedor, F. J., and Joost, H. G. (1998) J. Biol. Chem. 273, 25893-25902[Abstract/Free Full Text]
  18. Marti, E., Altafaj, X., Dierssen, M., de la Luna, S., Fotaki, V., Alvarez, M., Perez-Riba, M., Ferrer, I., and Estivill, X. (2003) Brain Res. 964, 250-263[CrossRef][Medline] [Order article via Infotrieve]
  19. Álvarez, M., Estivill, X., and De La Luna, S. (2003) J. Cell Sci. 116, 3099-3107[Abstract/Free Full Text]
  20. Yang, X., Hubbard, E. J., and Carlson, M. (1992) Science 25, 680-682
  21. Chen, Z., Indjeian, V. B., McManus, M., Wang, L., and Dynlacht, B. D. (2002) Dev. Cell 3, 339-350[CrossRef][Medline] [Order article via Infotrieve]
  22. Airhart, N., Yang, Y. F., Roberts, C. T., Jr., and Silberbach, M. (2003) J. Biol. Chem. 278, 38693-38698[Abstract/Free Full Text]
  23. Zhou, T., Aumais, J. P., Liu, X., Yu-Lee, L. Y., and Erikson, R. L. (2003) Dev. Cell 5, 127-138[CrossRef][Medline] [Order article via Infotrieve]
  24. Yaffe, M. B., Leparc, G. G., Lai, J., Obata, T., Volinia, S., and Cantley, L. C. (2001) Nat. Biotechnol. 19, 348-353[CrossRef][Medline] [Order article via Infotrieve]
  25. Brinkworth, R. I., Breinl, R. A., and Kobe, B. (2003) Proc. Natl. Acad. Sci. U. S. A. 100, 74-79[Abstract/Free Full Text]
  26. Fukunaga, R., and Hunter, T. (1997) EMBO J. 16, 1921-1933[CrossRef][Medline] [Order article via Infotrieve]
  27. Knebel, A., Morrice, N., and Cohen, P. (2001) EMBO J. 20, 4360-4369[CrossRef][Medline] [Order article via Infotrieve]
  28. Lewis, T. S., Hunt, J. B., Aveline, L. D., Jonscher, K. R., Louie, D. F., Yeh, J. M., Nahreini, T. S., Resing, K. A., and Ahn, N. G. (2000) Mol. Cell 6, 1343-1454[CrossRef][Medline] [Order article via Infotrieve]
  29. Büssow, K., Cahill, D., Nietfeld, W., Bancroft, D., Scherzinger, E., Lehrach, H., and Walter, G. (1998) Nucleic Acids Res. 26, 5007-5008[Abstract/Free Full Text]
  30. Büssow, K., Nordhoff, E., Lübbert, C., Lehrach, H., and Walter, G. (2000) Genomics 65, 1-8[CrossRef][Medline] [Order article via Infotrieve]
  31. Zhao, Z., Lim, L., and Manser, E. (2001) Methods 24, 194-200[CrossRef][Medline] [Order article via Infotrieve]
  32. Kentrup, H., Becker, W., Heukelbach, J., Wilmes, A., Schürmann, A., Huppertz, C., Kainulainen, H., and Joost, H. G. (1996) J. Biol. Chem. 271, 3488-3495[Abstract/Free Full Text]
  33. Takayama, S., and Reed, J. C. (1997) Methods Mol. Biol. 69, 171-184[Medline] [Order article via Infotrieve]
  34. Himpel, S., Tegge, W., Frank, R., Leder, S., Joost, H. G., and Becker, W. (2000) J. Biol. Chem. 275, 2431-2438[Abstract/Free Full Text]
  35. Himpel, S., Panzer, P., Eirmbter, K., Czajkowska, H., Sayed, M., Packman, L. C., Blundell, T., Kentrup, H., Grötzinger, J., Joost, H. G., and Becker, W. (2001) Biochem. J. 359, 497-505[CrossRef][Medline] [Order article via Infotrieve]
  36. Beil, B., Screaton, G., Stamm, S. (1997) DNA Cell Biol. 16, 679-690[Medline] [Order article via Infotrieve]
  37. Himpel, S., Joost, H. G., and Becker, W. (1999) Anal. Biochem. 274, 138-141[CrossRef][Medline] [Order article via Infotrieve]
  38. Becker, W., Kentrup, H., Heukelbach, J., and Joost, H. G. (1996) Biochim. Biophys. Acta 1312, 63-67[Medline] [Order article via Infotrieve]
  39. Dickinson, L. A., Edgar, A. J., Ehley, J., and Gottesfeld, J. M. (2002) J. Biol. Chem. 277, 25465-25473[Abstract/Free Full Text]
  40. Giese, B., Au-Yeung, C. K., Herrmann, A., Diefenbach, S., Haan, C., Küster, A., Wortmann, S., Roderburg, C., Heinrich, P. C., Behrmann, I., and Müller-Newen, G. (2003) J. Biol. Chem. 278, 39205-39213[Abstract/Free Full Text]
  41. Berke, J. D., Sgambato, V., Zhu, P. P., Lavoie, B., Vincent, M., Krause, M., and Hyman, S. E. (2001) Neuron 32, 277-287[CrossRef][Medline] [Order article via Infotrieve]
  42. Andersen, G., Busso, D., Poterszman, A., Hwang, J. R., Wurtz, J. M., Ripp, R., Thierry, J. C., Egly, J. M., and Moras, D. (1997) EMBO J. 16, 958-967[CrossRef][Medline] [Order article via Infotrieve]
  43. Beaudoing, E., Freier, S., Wyatt, J. R., Claverie, J. M., and Gautheret, D. (2000) Genome Res. 10, 1001-1010[Abstract/Free Full Text]
  44. Lamond, A. I., and Spector, D. L. (2003) Nat. Rev. Mol. Cell Biol. 4, 605-612[CrossRef][Medline] [Order article via Infotrieve]
  45. Fu, X. D., and Maniatis, T. (1990) Nature 343, 437-441[CrossRef][Medline] [Order article via Infotrieve]
  46. Misteli, T. (2000) Science 291, 843-847
  47. Graveley, B. R. (2000) RNA 6, 1197-1211[CrossRef][Medline] [Order article via Infotrieve]
  48. Becker, W., and Joost, H. G. (1999) Prog. Nucleic Acids Res. Mol. Biol. 62, 1-17[Medline] [Order article via Infotrieve]
  49. Colwill, K., Pawson, T., Andrews, B., Prasad, J., Manley, J. L., Bell, J. C., and Duncan, P. I. (1996) EMBO J. 15, 265-275[Medline] [Order article via Infotrieve]
  50. Fu, X. D. (1995) RNA 1, 663-680[Medline] [Order article via Infotrieve]
  51. Blencowe, B. J., Bowman, J. A., McCracken, S., and Rosonina, E. (1999) Biochem. Cell Biol. 77, 277-291[CrossRef][Medline] [Order article via Infotrieve]
  52. Boucher, L., Ouzounis, C. A., Enright, A. J., and Blencowe, B. J. (2001) RNA 7, 1693-1701[Abstract]
  53. Mortillaro, M. J., Blencowe, B. J., Wei, X., Nakayasu, H., Du, L., Warren, S. L., Sharp, P. A., and Berezney, R. (1996) Proc. Natl. Acad. Sci. U. S. A. 93, 8253-8257[Abstract/Free Full Text]
  54. Schul, W., van Driel, R., and de Jong, L. (1998) Exp. Cell Res. 238, 1-12[CrossRef][Medline] [Order article via Infotrieve]
  55. Herrmann, C. H., and Mancini, M. A. (2001) J. Cell Sci. 114, 1491-1503[Abstract]
  56. Oelgeschläger, T. (2002) J. Cell Physiol. 190, 160-169[CrossRef][Medline] [Order article via Infotrieve]
  57. Hu, D., Mayeda, A., Trembley, J. H., Lahti, J. M., and Kidd, V. J. (2003) J. Biol. Chem. 278, 8623-8629[Abstract/Free Full Text]
  58. Gururajan, R., Lahti, J. M., Grenet, J., Easton, J., Gruber, I., Ambros, P. F., and Kidd, V. J. (1998) Genome Res. 8, 929-939[Abstract/Free Full Text]
  59. Loyer, P., Trembley, J. H., Lahti, J. M., and Kidd, V. J. (1998) J. Cell Sci. 111, 1495-1506[Abstract]
  60. Trembley, J. H., Hu, D., Hsu, L. C., Yeung, C. Y., Slaughter, C., Lahti, J. M., and Kidd, V. J. (2002) J. Biol. Chem. 277, 2589-2596[Abstract/Free Full Text]
  61. Duncan, P. I., Stojdl, D. F., Marius, R. M., and Bell, J. C. (1997) Mol. Cell Biol. 17, 5996-6001[Abstract]
  62. Jumaa, H., and Nielsen, P. J. (1997) EMBO J. 16, 5077-5085[CrossRef][Medline] [Order article via Infotrieve]
  63. Mattox, W., and Baker, B. S. (1991) Genes Dev. 5, 786-796[Abstract/Free Full Text]
  64. Campbell, L. E., and Proud, C. G. (2002) FEBS Lett. 510, 31-36[CrossRef][Medline] [Order article via Infotrieve]
  65. Misteli, T., Cáceres, J. F., Clement, J. Q., Krainer, A. R., Wilkinson, M. F., and Spector, D. L. (1998) J. Cell Biol. 143, 297-307[Abstract/Free Full Text]
  66. Higgins, D. G., Thompson, J. D., and Gibson, T. J. (1996) Methods Enzymol. 266, 383-402[Medline] [Order article via Infotrieve]

Add to CiteULike CiteULike   Add to Complore Complore   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us   Add to Digg Digg   Add to Reddit Reddit   Add to Technorati Technorati    What's this?


This article has been cited by other articles:


Home page
 J. Cell Sci.Home page
J. Fernandez-Martinez, E. M. Vela, M. Tora-Ponsioen, O. H. Ocana, M. A. Nieto, and J. Galceran
Attenuation of Notch signalling by the Down-syndrome-associated kinase DYRK1A
J. Cell Sci., May 15, 2009; 122(10): 1574 - 1583.
[Abstract] [Full Text] [PDF]

Home page
 Hum Mol GenetHome page
F. K. Wiseman, K. A. Alford, V. L.J. Tybulewicz, and E. M.C. Fisher
Down syndrome--recent progress and future prospects
Hum. Mol. Genet., April 15, 2009; 18(R1): R75 - R83.
[Abstract] [Full Text] [PDF]

Home page
 J. Biol. Chem.Home page
O. Bogacheva, O. Bogachev, M. Menon, A. Dev, E. Houde, E. I. Valoret, H. M. Prosser, C. L. Creasy, S. J. Pickering, E. Grau, et al.
DYRK3 Dual-specificity Kinase Attenuates Erythropoiesis during Anemia
J. Biol. Chem., December 26, 2008; 283(52): 36665 - 36675.
[Abstract] [Full Text] [PDF]

Home page
 J. Biol. Chem.Home page
P. Loyer, J. H. Trembley, J. A. Grenet, A. Busson, A. Corlu, W. Zhao, M. Kocak, V. J. Kidd, and J. M. Lahti
Characterization of Cyclin L1 and L2 Interactions with CDK11 and Splicing Factors: INFLUENCE OF CYCLIN L ISOFORMS ON SPLICE SITE SELECTION
J. Biol. Chem., March 21, 2008; 283(12): 7721 - 7732.
[Abstract] [Full Text] [PDF]

Home page
 GENES CELLSHome page
J.-i. Yomoda, M. Muraki, N. Kataoka, T. Hosoya, M. Suzuki, M. Hagiwara, and H. Kimura
Combination of Clk family kinase and SRp75 modulates alternative splicing of Adenovirus E1A.
Genes Cells, March 1, 2008; 13(3): 233 - 244.
[Abstract] [Full Text] [PDF]

Home page
 FASEB J.Home page
A. Herrmann, K. Fleischer, H. Czajkowska, G. Muller-Newen, and W. Becker
Characterization of cyclin L1 as an immobile component of the splicing factor compartment
FASEB J, October 1, 2007; 21(12): 3142 - 3152.
[Abstract] [Full Text] [PDF]

Home page
 Mol. Biol. CellHome page
M. Alvarez, X. Altafaj, S. Aranda, and S. de la Luna
DYRK1A Autophosphorylation on Serine Residue 520 Modulates Its Kinase Activity via 14-3-3 Binding
Mol. Biol. Cell, April 1, 2007; 18(4): 1167 - 1178.
[Abstract] [Full Text] [PDF]

Home page
 Mol. Cell. Biol.Home page
H.-H. Chen, Y.-C. Wang, and M.-J. Fann
Identification and Characterization of the CDK12/Cyclin L1 Complex Involved in Alternative Splicing Regulation.
Mol. Cell. Biol., April 1, 2006; 26(7): 2736 - 2745.
[Abstract] [Full Text] [PDF]

Home page
 Hum Mol GenetHome page
J. P. Venables, C. F. Bourgeois, C. Dalgliesh, L. Kister, J. Stevenin, and D. J. Elliott
Up-regulation of the ubiquitous alternative splicing factor Tra2{beta} causes inclusion of a germ cell-specific exon
Hum. Mol. Genet., August 15, 2005; 14(16): 2289 - 2303.
[Abstract] [Full Text] [PDF]

Home page
 Mol. Biol. CellHome page
P. A. Kelly and Z. Rahmani
DYRK1A Enhances the Mitogen-activated Protein Kinase Cascade in PC12 Cells by Forming a Complex with Ras, B-Raf, and MEK1
Mol. Biol. Cell, August 1, 2005; 16(8): 3562 - 3573.
[Abstract] [Full Text] [PDF]

Home page
 J. Biol. Chem.Home page
L. Yang, N. Li, C. Wang, Y. Yu, L. Yuan, M. Zhang, and X. Cao
Cyclin L2, a Novel RNA Polymerase II-associated Cyclin, Is Involved in Pre-mRNA Splicing and Induces Apoptosis of Human Hepatocellular Carcinoma Cells
J. Biol. Chem., March 19, 2004; 279(12): 11639 - 11648.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
279/6/4612    most recent
M310794200v1
Right arrow Submit a Letter to Editor
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
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 arrowRequest Permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by de Graaf, K.
Right arrow Articles by Becker, W.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by de Graaf, K.
Right arrow Articles by Becker, W.
Social Bookmarking
Add to CiteULike   Add to Complore   Add to Connotea   Add to Del.icio.us   Add to Digg   Add to Reddit   Add to Technorati  
What's this?

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
 All ASBMB Journals   Molecular and Cellular Proteomics 
 Journal of Lipid Research   ASBMB Today 
Copyright © 2004 by the American Society for Biochemistry and Molecular Biology.
Advertisement
spacer
Advertisement
Advertisement