JBC Papers in Press.

Published on April 22, 2004 as Manuscript M402008200

CD2BP2-GYF interacts with the spliceosomal protein SmB/B' 1

Recognition sequences for the GYF domain reveal a possible spliceosomal

function of CD2BP2

Michael Kofler, Katja Heuer, Tobias Zech, and Christian Freund*

Protein Engineering Group, Forschungsinstitut für Molekulare Pharmakologie,

Robert-Rössle-Str. 10, 13125 Berlin, Germany

Phone: ++49-30-94793-181

Fax: ++49-30-94793-189

freund@fmp-berlin.de

Downloaded from http://www.jbc.org/ by guest on April 14, 2019

*Corresponding Author

Running Title: CD2BP2-GYF interacts with the spliceosomal protein SmB/B'

Summary

Protein-protein interactions are often mediated by small domains that recognize

solvent exposed peptide sequences. Deciphering the recognition code for these

adapter domains is an important step in the understanding of multi-protein assemblies.

Here, we investigate the sequence requirements for the CD2BP2-GYF domain, a

proline-rich sequence binding module previously shown to be involved in T cell

signalling. We show that the signature R/K/GxxPPGxR/K defines a preferred peptide-

binding motif that is present in several proteins of the splicing machinery.

Specifically, the core snRNP protein SmB/B’ contains several PPPPGMR motifs that

interact with the CD2BP2-GYF domain in vitro and in vivo. The colocalization of

CD2BP2 and SmB proteins in the nucleus of Jurkat T cells and HeLa cells suggests a

Copyright 2004 by The American Society for Biochemistry and Molecular Biology, Inc.
CD2BP2-GYF interacts with the spliceosomal protein SmB/B' 2

function of the GYF domain of CD2BP2 in mediating protein-protein interactions

within the spliceosome.

Introduction

Proline-rich sequence recognition plays an important role in the assembly of multi-

protein complexes during the course of eukaryotic signal transduction and is mediated

by a set of protein folds that share characteristic features. To date, the super-family of

proline-rich sequence recognition domains comprises profilin (1), the SH3 (2, 3), the

Downloaded from http://www.jbc.org/ by guest on April 14, 2019

WW (4), the EVH1 (5), the GYF (6, 7), the UEV (8, 9) and probably the ligand

binding domain of prolyl-4-hydroxylase (10). Each domain uses a set of aromatic

amino acid residues for peptide recognition that is conserved within the respective

fold family. Within the GYF domain the glycine-tyrosine-phenylalanine tripeptide is

part of a bulge-helix-bulge motif that contains several aromatic amino acid side-

chains that are essential for the binding of the CD2 cytoplasmic domain. The recently

solved structure of the CD2BP2-GYF domain in complex with the CD2 peptide

SHRPPPPGHRV showed that the peptide adopts an extended conformation and forms

a polyproline type II helix involving residues Pro4 – Pro7 (Fig. 1) (11, 12). The major

binding surface of the GYF domain accommodates Pro6 and Pro7 of the ligand and is

defined primarily by the aromatic residues Tyr6, Trp8, Tyr17, Tyr20, Trp28, Tyr33

and Phe34 of the GYF domain. Charge complementarity outside the central

hydrophobic interaction site is suggested to confer additional specificity (Fig. 1). The

CD2 target sequence for the CD2BP2-GYF domain comprises two PPPPGHR motifs

that probably can each bind a CD2BP2-GYF domain and thereby enhance the overall

affinity of the interaction (11). The PPPPGHR motifs of CD2 are conserved among
CD2BP2-GYF interacts with the spliceosomal protein SmB/B' 3

species and have been shown to mediate CD2 dependent signal transduction (13). In

vitro, the CD2BP2-GYF domain and the SH3 domain of the Fyn tyrosine kinase bind

to the CD2 tail motifs with similar affinity and show that a single proline-rich

sequence can interact with members of different fold families (11). Inversely, a single

recognition domain might well interact with proline-rich sequences of several target

proteins, but novel protein interactions for the CD2BP2-GYF domain have not been

reported so far.

We now show that the signature +/GxxPPGx+ defines a preferred motif for CD2BP2-

Downloaded from http://www.jbc.org/ by guest on April 14, 2019

GYF domain interactions and that several sequences from spliceosomal proteins that

resemble this motif are bound by the CD2BP2-GYF domain in vitro. In particular,

sequences from the core snRNP SmB/B’ proteins interact strongly and define

potential new target sites for the CD2BP2-GYF domain. The binding epitopes of

these peptides and the CD2-derived peptides are very similar as shown by NMR shift

mapping experiments. Furthermore, pulldown experiments with GST-GYF (CD2BP2)

show the precipitation of SmB/B’ from Jurkat cell lysates. An EGFP-CD2BP2 fusion

protein is primarily compartmentalized to the nucleus of Jurkat and HeLa cells and

colocalizes with the SmB/B’ protein. The latter findings reveal a role of CD2BP2

aside from mediating CD2 triggered signal transduction and suggest a function of

GYF domain containing proteins in splicing or processes that are associated with

splicing.
CD2BP2-GYF interacts with the spliceosomal protein SmB/B' 4

Materials and Methods

Expression constructs

Full-length human CD2BP2 was amplified from IMAGE clone IRALp962D081Q2 by

PCR using Vent Polymerase (NEB) and cloned into the Xho 1 and BamH 1 restriction

sites of the pEGFP-C3 vector (Clontech). The amplified sequence of full length

human SmB (IMAGE clone IMAGp998D118415Q3 from the Deutsches

Ressourcenzentrum) was cloned into the Hind III and Sal 1 restriction sites of the

vector pEYFP-C1 (Clontech). Amino acids 280–342 of the CD2BP2 protein,

Downloaded from http://www.jbc.org/ by guest on April 14, 2019

comprising the GYF domain were cloned into pGEX4T-1 (Amersham Biosciences)

and pTFT74 (14) for the expression of GST-GYF and untagged GYF domain,

respectively. The sequences of all constructs were verified by DNA sequencing.

Protein preparation

Proteins were expressed in E. coli BL21(DE3) and purified from the soluble fraction
15
after sonication. The untagged N-labelled GYF domain was obtained from E. coli
15
grown in defined media supplemented with NH4Cl and purified by ion exchange

chromatography (Mono Q® HR 10/10, Pharmacia) and subsequent gel filtration

(Superdex® 75, Pharmacia) in 50 mM sodium-phosphate buffer, pH 6.3. GST–GYF

(CD2BP2) purification included affinity chromatography (glutathione sepharose)

according to the manufacturers manual (Amersham Biosciences) and dialysis against

PBS.
CD2BP2-GYF interacts with the spliceosomal protein SmB/B' 5

Binding studies of membrane bound peptides

Single substitutions of SHRPPPPGHRV and a set of different proline-rich peptides

were generated by semiautomated spot synthesis (15, 16) (Abimed; Software LISA,

Jerini AG) on Whatman 50 cellulose membranes as described (17). Membranes were

probed with GST fusion protein as described elsewhere (18). Briefly, the membranes

were incubated with GST-GYF (CD2BP2; 40 µg/ml) over night. After washing,

bound GST fusion protein was detected with rabbit polyclonal anti-GST antibody (Z-

5, Santa Cruz) and horseradish peroxidase coupled anti-rabbit IgG antibodies

(Rockland). An enhanced chemiluminescence substrate (SuperSignal West Pico,

Downloaded from http://www.jbc.org/ by guest on April 14, 2019

Pierce Illinois) on a LumiImagerTM (Boehringer Mannheim GmbH) was used for

detection.

NMR Titration

All NMR experiments were performed at 298 K using a Bruker DRX600 instrument

equipped with a standard triple-resonance probe. Data processing and analysis were

carried out with the XWINNMR (Bruker), Prosa/XEASY (19) and the Sparky (20)

software packages. In the NMR titration experiments, increasing amounts of synthetic

peptides of the sequence NH2-PMGRGAPPPGMMGPPPGMRPPM–COOH (SmB-

1), NH2-GTPMGMPPPGMRPPPPGMRGLL–COOH (SmB-2) or NH2-SHRPPPPG-

15
HRV-COOH (CD2 peptide) were added to a 0.4 mM sample of the N-labelled GYF

domain. The gradual change of chemical shifts in the HSQC spectra allowed the

resonances of the ligand-bound GYF domain to be unambiguously assigned. The sum

15
of the chemical shift changes for N and 1H atoms in a sample with peptide was

determined as: [(∆1H_cs)2 + (∆15N_cs)2]1/2, where ∆1H_cs is in units of 0.1 p.p.m.,

and ∆15N_cs is in units of 0.5 p.p.m.

CD2BP2-GYF interacts with the spliceosomal protein SmB/B' 6

Cell culture and lysis

Jurkat cells were maintained in RPMI 1640 medium supplemented with 10 % (v/v)

fetal bovine serum (FBS) in a 37 °C humidified incubator. Cells were washed 3 times

with PBS and lysed for 30 min by gentle rocking in lysis buffer (25 mM Tris pH 7.4,

137 mM NaCl, 1 % (v/v) NP-40, 0.5 mM PMSF, aprotinin and leupeptin (both

5 µg/ml), 5 mM EDTA, 2 mM sodium orthovanadate). Cell lysates were centrifuged

in order to remove the insoluble cell debris and were either used directly or snap

frozen and stored at -80 °C.

Human HeLa S3 cells were maintained in DMEM containing 10 % FBS (Biochrom)

Downloaded from http://www.jbc.org/ by guest on April 14, 2019

supplemented with streptomycin and penicillin.

GST Pulldown assay

100 µg GST or GST fusion proteins were incubated with 50 µl of GSH Sepharose

beads (Amersham Biosciences) for 2 h at 4 °C. Beads were then washed three times

with PBS and a small aliquot was analysed by SDS PAGE to ascertain that the

loading of each fusion construct was similar. The pulldown experiment was carried

out by first preclearing the cell lysates with 50 µl of GST-loaded glutathione beads.

Following removal of the beads, pre-cleared lysates (~300 x 106 cells per experiment)

were incubated with GST, GST-GYF (CD2BP2) or GST-hSH3 (18) loaded beads for

2–4 h at 4°C. Where required, a specific (SmB-2) or non-specific peptide (Pep-1:

NH2-KETWWETWWTEWSQPKKKRKV-COOH) was added to the lysates to a final

concentration of 1 mM. Beads were washed 5 times with 10 ml of PBS. Bound

proteins were eluted in gel loading buffer by boiling, and separated by SDS-PAGE.

Proteins were then transferred onto a polyvinylidene difluoride membrane (PVDF).

Blots were probed with an anti-SmB/B’ antibody (sc-5485; Santa Cruz) and a
CD2BP2-GYF interacts with the spliceosomal protein SmB/B' 7

horseradish peroxidase conjugated secondary antibody. The immunoblots were then

developed with an enhanced chemiluminescence substrate (SuperSignal West Pico,

Pierce Illinois) and detected by a LumiImagerTM (Boehringer Mannheim GmbH).

Transfections and Fluorescence Microscopy

3 x 106 Jurkat J77 cells were transfected by nucleoporation using 2.0 µg of DNA, the

Amaxa Nucleofection Kit V (Amaxa Biosystems, Cologne, Germany) and a

Nucleofector device (program T-14). HeLa S3 cells were grown on glass coverslips in

12 well plates and transiently transfected with Lipofectamine plus (GIBCO) and

Downloaded from http://www.jbc.org/ by guest on April 14, 2019

0.4 µg DNA / well according to the manufacturers protocol. 24 h after transfection

Jurkat and HeLa cells were fixed on poly-L-lysine coated slides (Menzel-Glaeser)

with 4 % paraformaldehyde / PBS for 10 min at room temperature and the nuclei were

stained with Hoechst 33258 (5 µg/ml in H2O) for 2 min. After being washed twice

with PBS all cells were mounted in Flouromount G mounting medium for

fluorescence microscopy (Southern Biotechnology). Live transfected HeLa cells were

also examined by confocal laser scanning microscopy in standard medium 24 h post

transfection. Conventional fluorescence analysis was performed using the Leica

DMLB (Deerfield, IL) microscope. Images were recorded with a cooled Sensiocam

CCD camera. For confocal laser scanning microscopy a LSM 510 (Carl Zeiss, Jena)

with a 100 x / 1.3 objective and an argon laser (488 nm) was used. Images were

processed by using the AXIOVISION (Carl Zeiss, Jena) and Adobe Photoshop

(Adobe Systems, Mountain View, CA) programs.

CD2BP2-GYF interacts with the spliceosomal protein SmB/B' 8

Results

Peptide substitution analysis of the CD2BP2 target sequence SHRPPPPGHRV

In order to delineate the contribution of individual residues of the peptide

SHRPPPPGHRV to the interaction with the CD2BP2-GYF domain, a substitution

analysis of this peptide was performed (Fig. 2). Individual peptides were synthesised

on a cellulose membrane. Spot intensity qualitatively correlates with peptide binding

affinities (17) and allows the detection of interactions with KD values that are in the

high µM range. Figure 2 shows a strong requirement for proline at positions 6 and 7,

Downloaded from http://www.jbc.org/ by guest on April 14, 2019

while proline at position 4 and 5 can be substituted by most other amino acids. This

correlates well with the numerous van der Waals interactions observed between

prolines 6 and 7 and the aromatic side-chains of the domain in the complex (Fig.1). At

position 8, glycine results in the strongest signal, however several other amino acids

are allowed at this position. Surprisingly, a tryptophane instead of glycine at position

8 is tolerated. Since the bulky tryptophane side-chain would sterically clash with GYF

domain residues 8 and 15 if it were bound in the same conformation as the wild type

peptide, it is likely that an alternate mode of binding exists for this peptide. At

position 10, only positively charged amino acids are allowed, whereas position 3 also

tolerates a substitution to glycine. The contribution of positively charged amino acids

to peptide binding affinity support the notion of charge complementarity conferring

specificity to complex formation (Fig. 1). The remaining positions of the CD2 peptide

are indifferent to most substitutions, except for aspartate and glutamate which are

disfavoured. In general, the substitution analysis can be rationalized by the structure

of the CD2BP2-GYF domain in complex with the CD2 peptide (11) and allows us to

define the motif +/GxxPPGx+ as the ligand core motif for the CD2BP2-GYF domain.
CD2BP2-GYF interacts with the spliceosomal protein SmB/B' 9

Sequence motifs from SmB/B’ bind to the GYF domain of CD2BP2 in vitro.

We tested natural sequences that are known ligands for other proline-rich sequence

recognition domains for their interaction with CD2BP2-GYF. A diverse set of

sequences that represent binding sites for SH3, WW and EVH1 domains was

examined by peptide spot analysis. Only a few peptides resulted in detectable spot

intensities (Fig. 3b and Table 1). For comparison, CD2-derived peptides from Homo

sapiens and homologues from other species were included on the cellulose membrane

(Fig. 3a and Table 1). Aside from the CD2 sequences, the sequence with the highest

spot intensity arises from the SmB core splicing protein (position 20 in Fig. 3b). SmB

Downloaded from http://www.jbc.org/ by guest on April 14, 2019

and the alternative splicing product SmB’ contain two and three motifs, respectively,

that exactly match the suggested binding motif +/GxxPPGx+. To gain a more detailed

insight into the interaction of these peptides with the CD2BP2-GYF domain we

performed NMR shift mapping experiments. Increasing amounts of the SmB derived

peptides PMGRGAPPPGMMGPPPGMRPPM (SmB-1, data not shown) and

GTPMGMPPPGMRPPPPGMRGLL (SmB-2) or the previously characterized control

15
peptide SHRPPPPGHRV (CD2 peptide) (11) were added to the N-labelled

CD2BP2-GYF domain and 15N-1H correlation spectra were recorded. In Figure 4a, an

overlay of the spectra of the isolated CD2BP2-GYF domain and the CD2BP2-GYF

domain in the presence of equimolar amounts of either SmB-2 or CD2 peptides are

shown. The observed chemical shift changes are indicative of amino acid NH-groups

close in space to the ligand. Both peptides display a similar pattern of resonance shifts

within the spectrum (Fig. 4a), however the extent of the chemical shift changes at a

given peptide concentration is larger for SmB-2 in comparison to the CD2 peptide

(Fig. 4a, inset). Since SmB-2 contains two intercalated consensus motifs

(GMPPPGMRPPPPGMR) it is very likely that the presence of two motifs in SmB-2

CD2BP2-GYF interacts with the spliceosomal protein SmB/B' 10

increases the apparent affinity by enhancing the local concentration of interaction

sites. Residues displaying significant chemical shift changes were highlighted on the

surface of the CD2BP2-GYF domain (Fig. 4b). These residues define a contiguous

area with side-chains of the bulge-helix-bulge motif and Trp8 forming a hydrophobic

patch. The surface defined by the NMR chemical shift data is very similar for the two

peptides, confirming that the peptide motifs within SmB-2 bind to the GYF domain in

an analogous orientation and conformation as the CD2 peptide (Fig.4b). The binding

mode and apparent affinity for the SmB-1 peptide is very similar to that of the SmB-2

peptide (data not shown).

Downloaded from http://www.jbc.org/ by guest on April 14, 2019

SmB/B’ is a specific binding partner of the CD2BP2-GYF domain in vivo

GST-pulldown experiments were undertaken to establish that the interaction between

CD2BP2-GYF and SmB/B’ occurs under physiological conditions. It is shown in

Figure 5a that SmB/B’ could be specifically captured from a Jurkat cell lysate with

GST-GYF in contrast to GST alone or the unrelated GST-hSH3 fusion construct

which was shown not to bind to proline rich sequences (18). Furthermore, this

interaction could be effectively competed out by the addition of an excess of SmB-2

peptide to the cell lysate whereas the interaction remained unperturbed by the

presence of an unrelated peptide (compare lanes 3 and 4 in Fig. 5b). These results

demonstrate that a physiologically relevant interaction occurs between the C-terminal

proline-rich sequences of the SmB/B’ protein and the GYF domain of CD2BP2.
CD2BP2-GYF interacts with the spliceosomal protein SmB/B' 11

Colocalization of CD2BP2 and SmB/B’

Transient expression of EGFP tagged full-length CD2BP2 and EYFP-SmB fusion

proteins in Jurkat and HeLa cells were used to determine their intracellular

localization. The proteins localized predominantly to the nucleus with a residual

cytoplasmic pool 24 h after transfection, as verified by Hoechst staining of DNA (Fig.

6a, b). Co-transfection of EGFP-CD2BP2 and EYFP-SmB in HeLa cells shows

considerable colocalization of the two proteins in the nucleus in live cell laser

scanning microscopy imaging studies (Fig. 6c). Both CD2BP2 and SmB/B’ were

excluded from nucleoli (Fig. 6c).

Downloaded from http://www.jbc.org/ by guest on April 14, 2019

Interaction of CD2BP2-GYF with other splicing factors

The identified interaction of the CD2BP2-GYF domain with a proline-rich region of

SmB/B’ raised the question whether proline rich sequences of other splicing factors

that resemble the binding motif +/GxxPPGx+ are also able to bind to CD2BP2-GYF.

Sequences of splicing proteins containing the reduced core motif PPG were analysed

by spot synthesis (Fig. 7 and supplementary material). Peptides derived from SmB/B’

or SmB like proteins such as SNRPB protein (a variant of SmB), Sm-D and RT-LI

(snRPN with a novel exon 3) contain motifs that match the binding motif

+/GxxPPGx+ (spots 17, 27, 28, 31–33, 36–38 in Fig. 7). In addition to these peptides

sequence motifs from the proline- and glutamine-rich splicing factor (position 9 in

Fig. 7) and splicing factor 1 isoform (position 18 in Fig. 7) showed binding

comparable to the human CD2 peptide (position 1). In summary, the abundance of

proline-rich sequences within spliceosomal proteins that bind to the CD2BP2-GYF

domain sets the stage for further investigations of GYF domain mediated protein-

protein interactions within the spliceosomal complexes.

CD2BP2-GYF interacts with the spliceosomal protein SmB/B' 12

Discussion

Sequence recognition by the CD2BP2-GYF domain is not restricted to motifs that

precisely match the previously described PPPPGHR motifs in CD2 (6). Our peptide

analysis shows that the CD2BP2-GYF domain requires two adjacent prolines in the

ligand to contact the hydrophobic hot spot of the domain (Fig. 2). A glycine at the

position directly C-terminal to the second proline (position 8 in the CD2 peptide) is

favourable, and positively charged residues flanking the PPII helix are important for

high-affinity binding. Peptide spot analysis of diverse sequences that obey the

Downloaded from http://www.jbc.org/ by guest on April 14, 2019

sequence requirements confirm that this subset of proline-rich sequences binds to the

CD2BP2-GYF domain in vitro. Interestingly, the interacting peptides and putative

target sites are found in several splicing proteins. The most strongly interacting

sequences belong to the SmB/B’ core splicing protein which contains several binding

motifs within its C-terminal tail. Our experiments with EGFP-tagged CD2BP2 show a

localization of CD2BP2 primarily to the nucleus, possibly due to the presence of a

putative bipartite nuclear localization sequence at the N-terminus of the CD2BP2

protein. Nuclear localization of CD2BP2 and SmB/B’ is in agreement with a direct

interaction. The finding that CD2BP2 is stably associated with the pre-spliceosome

(21) supports this notion, however since CD2BP2 has not been found in snRNP’s

other than U2, regulatory mechanisms might determine the temporal and spatial

interaction of SmB/B’ with CD2BP2. Interestingly, the proline-rich sequences from

SmB/B’ identified to bind to CD2BP2-GYF in this study were shown to be target

sites for WW and SH3 domains of several proteins in vitro (22, 23), but in most cases

it is not known whether the corresponding proteins are physiological interaction

partners of SmB/B’. However in mouse, the WW domain of the FBP21 protein was
CD2BP2-GYF interacts with the spliceosomal protein SmB/B' 13

shown to directly bind to the PPPGMR motifs in SmB/B’ and to colocalize with

spliceosomal proteins in nuclear speckles (22). Since we show binding of CD2BP2 to

+/GxxPPGx+ containing sequences of SmB/B’ in this study, simultaneous and/or

competitive binding of FBP21 and CD2BP2 to SmB/B’ might occur. Assuming that

the interaction of SmB/B’ with FBP21 is characterized by a high off rate similar to the

interaction of SmB/B’ with CD2BP2, competitive binding of CD2BP2 and FBP21 to

SmB/B’ as a regulatory mechanism would require large relative differences in protein

concentration within the specific cell compartment. In contrast, the colocalization of

the two proteins with the SmB/B’ tail is likely under conditions where the

Downloaded from http://www.jbc.org/ by guest on April 14, 2019

concentrations of the interacting proteins and the KD’s of the interactions are in the

same order of magnitude. The latter scenario would support the hypothesis that the

proline-rich sequences of SmB/B’ mediate transient interactions that aid in the

formation of more stable protein assemblies during the complex biogenesis of the

snRNPs. The moderate specificity and affinity of proline-rich sequence recognition by

CD2BP2 might reflect the dynamic nature of molecular associations during

spliceosomal protein assembly.

CD2BP2-GYF interacts with the spliceosomal protein SmB/B' 14

Reference List

1. Carlsson, L., Nystrom, L. E., Sundkvist, I., Markey, F., and Lindberg, U. (1977)

J. Mol. Biol. 115, 465-483

2. Mayer, B. J., Hamaguchi, M., and Hanafusa, H. (1988) Nature 332, 272-275

3. Stahl, M. L., Ferenz, C. R., Kelleher, K. L., Kriz, R. W., and Knopf, J. L. (1988)

Nature 332, 269-272

4. Bork, P. and Sudol, M. (1994) Trends Biochem. Sci. 19, 531-533

Downloaded from http://www.jbc.org/ by guest on April 14, 2019

5. Niebuhr, K., Ebel, F., Frank, R., Reinhard, M., Domann, E., Carl, U. D., Walter,

U., Gertler, F. B., Wehland, J., and Chakraborty, T. (1997) EMBO J. 16, 5433-

5444

6. Nishizawa, K., Freund, C., Li, J., Wagner, G., and Reinherz, E. L. (1998) Proc.

Natl. Acad. Sci. U S A 95, 14897-14902

7. Freund, C., Dötsch, V., Nishizawa, K., Reinherz, E. L., and Wagner, G. (1999)

Nat. Struct. Biol. 6, 656-660

8. Sancho, E., Vila, M. R., Sanchez-Pulido, L., Lozano, J. J., Paciucci, R., Nadal,

M., Fox, M., Harvey, C., Bercovich, B., Loukili, N., Ciechanover, A., Lin, S. L.,

Sanz, F., Estivill, X., Valencia, A., and Thomson, T. M. (1998) Mol. Cell. Biol.

18, 576-589

9. Pornillos, O., Alam, S. L., Davis, D. R., and Sundquist, W. I. (2002) Nat. Struct.

Biol. 9, 812-817
CD2BP2-GYF interacts with the spliceosomal protein SmB/B' 15

10. Myllyharju, J. and Kivirikko, K. I. (1999) EMBO J. 18, 306-312

11. Freund, C., Kühne, R., Yang, H. L., Park, S., Reinherz, E. L., and Wagner, G.

(2002) EMBO J. 21, 5985-5995

12. Freund, C., Kühne, R., Park, S., Thiemke, K., Reinherz, E. L., and Wagner, G.

(2003) J. Biomol. NMR 27, 143-149

13. Chang, H. C., Moingeon, P., Pedersen, R., Lucich, J., Stebbins, C., and

Reinherz, E. L. (1990) J. Exp. Med. 172, 351-355

Downloaded from http://www.jbc.org/ by guest on April 14, 2019

14. Freund, C., Ross, A., Guth, B., Plückthun, A., and Holak, T. A. (1993) FEBS

Lett. 320, 97-100

15. Frank, R. (1992) Tetrahedron 48, 9217-9232

16. Kramer, A. and Schneider-Mergener J. (1998) Methods Mol. Biol. 87, 25-39

17. Kramer, A., Reineke, U., Dong, L., Hoffmann, B., Hoffmuller, U., Winkler, D.,

Volkmer-Engert, R., and Schneider-Mergener, J. (1999) J. Pept. Res. 54, 319-

327

18. Heuer, K., Kofler, M., Langdon, G., Thiemke, K., Freund, C. (2004) Structure

12, 603-10

19. Eccles, C., Guntert, P., Billeter, M., and Wuthrich, K. (1991) J. Biomol. NMR 1,

111-130

20. T.D.Goddard and D.G.Kneller, SPARKY 3 University of California San

Francisco. T. D. Goddard and D. G. Kneller, SPARKY 3,

University of California, San Francisco

CD2BP2-GYF interacts with the spliceosomal protein SmB/B' 16

21. Hartmuth, K., Urlaub, H., Vornlocher, H. P., Will, C. L., Gentzel, M., Wilm, M.,

and Lührmann, R. (2002) Proc. Natl. Acad. Sci. U.S.A 99, 16719-16724

22. Bedford, M. T., Reed, R., and Leder, P. (1998) Proc. Natl. Acad. Sci. U.S.A 95,

10602-10607

23. Espejo, A., Cote, J., Bednarek, A., Richard, S., and Bedford, M. T. (2002)

Biochem. J. 367, 697-702

Acknowledgements:

Downloaded from http://www.jbc.org/ by guest on April 14, 2019

The authors are grateful to Katharina Thiemke for excellent technical assistance,

Angelika Ehrlich and Rudolph Volkmer-Engert for membrane spot synthesis and

Michael Beyermann for peptide synthesis. We are also thankful to Torsten Meissner

for help with HeLa cell culturing. This work was supported by a grant from the

Deutsche Forschungsgemeinschaft (FR 1325/2-1) to C.F.

Figure Legends:

Figure 1:

3D view of the CD2BP2-GYF domain in complex with the CD2 peptide

SHRPPPPGHRV as derived from the pdb file 1L2Z. The peptide is shown as bonded

structure with peptide residues that are directly contacting the GYF domain coloured

in green (R3, P6, P7, G8 and R10) and residues pointing away from the binding

surface shown in grey. The amino acid side-chains of GYF domain residues that form
CD2BP2-GYF interacts with the spliceosomal protein SmB/B' 17

the hydrophobic binding pocket are coloured blue and the negatively charged GYF

domain side-chains in the vicinity of the peptide arginines are shown in red.

Figure 2:

Substitution analysis of the CD2 peptide SHRPPPPGHRV with CD2BP2-GYF. All

possible single substitution analogues of the peptide were synthesized on a membrane.

The single letter code above each column indicates the amino acid that replaces the

corresponding wild-type residue, the row defines the position of the substitution

within the peptide. Spots in the most left column (WT) are identical and represent the

Downloaded from http://www.jbc.org/ by guest on April 14, 2019

wild type peptide. The membrane was incubated with GST-GYF (CD2BP2). Bound

domain was detected with an anti-GST primary antibody and a horse-radish

peroxidase coupled secondary antibody. The relative spot intensities correlate

qualitatively with the binding affinities (17).

Figure 3:

Binding of the CD2BP2-GYF domain to different proline rich peptides. Peptides were

synthesized on cellulose membrane and incubated with GST-GYF (CD2BP2) as in

Figure 2. Numbers above or below the spots correspond to numbers in Table 1. CD2

from different species (A) and different classes of ligands for proline-rich binding

domains (B) are tested for binding to CD2BP2-GYF.

Figure 4:

Binding analysis of SmB-2 to the CD2BP2-GYF domain by NMR. (A) Overlap of the
15
N–1H correlation spectra of the isolated GYF domain (green), the GYF domain

upon equimolar addition of the CD2 peptide (blue) or SmB-2 peptide (red). The
CD2BP2-GYF interacts with the spliceosomal protein SmB/B' 18

eleven NH resonances with the largest chemical shift difference between the bound

and unbound state are labelled according to residue type and number. The NH

crosspeak of F34 broadens upon ligand binding and is too weak to be followed in

these spectra. The asterisk at W8 and W28 indicate side-chain NH resonances

whereas all other labelled resonances are backbone NH resonances. The inset shows

the sum of chemical shift changes for all assigned peaks of the CD2BP2-GYF domain

as a function of the ligand concentration of either the CD2 peptide (blue) or the SmB-

2 peptide (red). (B) Binding site of SmB-2 peptide and CD2 peptide on the CD2BP2-

GYF domain as derived from the NMR titration. The binding site for CD2 (left) and

Downloaded from http://www.jbc.org/ by guest on April 14, 2019

SmB-2 (right) is derived from the NMR titration experiment and mapped onto the

surface of the CD2BP2-GYF. The residues with the eleven strongest chemical shift

changes are colored dark green and additional eleven residues with significant

chemical shift changes are depicted in light green. W8 and W28 have been classified

according to the chemical shift change of their side-chain NH group. Residues P19
15
and P35 of the CD2BP2-GYF domain, which could not be observed in the N–1H

correlation experiment have also been included in the binding surface and are colored

in dark green, since both proline residues are flanked by residues that show a

significant chemical shift change upon binding. The CD2 peptide is depicted as

bonded structure. The surface representations of the CD2BP2-GYF domain were

created in Sybyl (Tripos, Inc; St. Louis, MO 63144-2319).

Figure 5:

Association of SmB/B' with the human CD2BP2-GYF domain in vivo. GST

pulldowns were performed with whole Jurkat cell lysates (~300 million cells per

pulldown) and bound proteins were separated by 15 % SDS-PAGE (A) or NuPAGE

CD2BP2-GYF interacts with the spliceosomal protein SmB/B' 19

(4–12 % gradient; Invitrogen, B), blotted and probed for SmB/B'. (A) Lane 1 contains

GST loaded beads, lane 2 GST-GYF loaded beads and lane 3 contains a further

negative control, GST-hSH3 loaded beads. (B) Lane 1 contains GST loaded beads,

lane 2 GST-GYF beads, lane 3 contains GST-GYF beads where an excess of the

SmB-2 peptide was incubated with the cell lysate, and lane 4 contains GST-GYF

loaded beads with an excess of a non-interacting peptide (Pep1) added to the cell

lysate.

Figure 6:

Downloaded from http://www.jbc.org/ by guest on April 14, 2019

CD2BP2 localizes to the nucleus of HeLa and Jurket cells. (A) HeLa S3 and (B) J77

Jurkat cells transiently expressing EGFP-CD2BP2 were fixed 24 h after transfection

and nuclei were stained with Hoechst 33258 dye. CD2BP2 predominantly localizes to

the nuclei of both cell lines. (C) Live cell laser confocal images of HeLa S3 cells

transiently expressing EGFP-CD2BP2 (green) and EYFP-SmB (red) 24 h after co-

transfection. Overlay of both images shows co-localization of both proteins in the

nucleus (yellow signal).

Figure 7:

Binding of CD2BP2-GYF to peptides from spliceosomal proteins containing the

proline-proline-glycine motif. Peptide synthesis, membrane incubation and ranking of

intensities are as in Figure 2. Numbers above or below the spots correspond to the

values given in the supplementary material.

CD2BP2-GYF interacts with the spliceosomal protein SmB/B' 20

Tabel 1

Binding of CD2BP2-GYF to sequences derived from CD2 of different species and

from different classes of ligands for proline-rich binding domains. The class of ligand

is indicated by ‘Description’. The SWISSPROT/TREMBL entry names are listed

under ‘Protein’. Peptide 20 is from SmB. Binding classification: (-): < 5 x back-

ground [BLU] (Boehringer Light Units); (+): 5–25 x background [BLU]; (++): 25–

125 x background [BLU]; (+++): > 125 x background [BLU].

Description Sequence Binding Protein

1 CD2 proteins HPPPPPGHRSQAPSHRPPPPGHRV +++ Human CD2 P06729
2 SHRPPPPGHR +++ Human CD2
3 homologoues QAPPPPGHHLQTPGHRPLPPGHRT +++ Mouse CD2 P08920

Downloaded from http://www.jbc.org/ by guest on April 14, 2019

4 region to QPPPPPSHRPQAPGHRPQVPGHRP + Horse CD2 P37998
5 CD2BP2 QAPGHRPQVPGHRPLPPGHRVQHQ ++ Horse CD2
6 binding site HRPQAPGHRPQVPGHRPLPPGHRV ++ Horse CD2
7 QAPPPPGHHLQTPGHRPLPPSHRN +++ Rat CD2 P08921
8 APPPPSHRPQAPGHRPPPPGHRAQ +++ Cat CD2 Q8WNV8
9 KRPPPSGTQIHQQKGPPLPRPRVQ - Mouse CD2
10 other proline PPSGTQIHQQKGPPLPRPRVQPKP - Mouse CD2
11 rich region KRPPAPSGTQVHQQKGPPLPRPRV - Human CD2
12 PSGTQVHQQKGPPLPRPRVQPKPP - Human CD2
13 Profilin PPPPPPPPPPPPPPPPPPPPPPPP -
14 WW-ClassI LPPAPGSTVPGTPPPRYNTLRLDR ++ NEDD4 P46934
15 DQQTLPIPGTPPPNYDSLRLQPLD -
16 SEQALPIPGTPPPNYDSLRLQPLD -
17 EDVVHHPGTPPPPYTVGPGYPWTT -
18 WW-ClassII EETLPLPPPPAPPLPPPLLHYQTS -
19 LPSPPLTPPVPSPPPPLPPPSTSP -
20 WW-ClassIII GRGTPMGMPPPGMRPPPPGMRGLL +++ SMB P14678-2
21 PPPPPPGSAGMMYAPPPPPPPPMD -
22 FVTMMGMGVAGMPPFGMPPAPPPP -
23 WW-Class IV CSVTFSPEQPLTPVTDLAVGFSNL -
24 EVH1-ClassI EVNASDFPPPPTDEELRLALPETP -
25 TSEPSSFEFPPPPTDEELRLALPE -
26 EVH1-ClassII GVGVTTSLRPPHHFSPPCFVAALP -
27 PETGDLNNPPKKFRDCLFKLCPMN -
28 SH3-1R QSAPVKQPPPLAPQSPQGGVMGGS -
29 ISTEWNERQPAPALPPKPPKPTTV -
30 SPPTPKPRPPRPLPVAPGSSKTEA -
31 PSEKKVAIIRTPPKSPATPKQLRL -
32 SH3-2R PAGSALGGAPPVPSRPGASPDPFG -
33 TQEMDLHSIAGPPVPPRQSTSQLI -
34 FSMAPQASLPPVPPRLDLLPQRVC -
35 SH3-1K SVGAESRPQRRPLPCTPGDCPSRD -
36 PSSRLGDSWLPRPIPKVPVSAPSS -
37 ATVTRGVPPPPTVRGAPAPRARTA ++ Sam68 Q07666
38 SH3-2K RGAVSTLLQAPELPTKTRTSRRAA -
39 SH3-1@ APTMPGPGYGGPPVPPPQQQAENP -
CD2BP2-GYF interacts with the spliceosomal protein SmB/B' 21

40 LVGTSLRAPTMPPPLPPVPPQPAR -
41 SH3-2D PMFDDSPPPPPPPPVDYEDEEAAV -
42 GSPKNGKLIIPPVDYLPDNRTWSE -
43 SH3-X PVPSLPIRPSRAPSRTPGPPSAQS -
44 PAGYSTARPTIPPRAGVISAPQSH -
45 QPLEPKRPPPPRPVAPPTRPAPPQ ++ Synaptojanin O43426
46 PAPPQRPPPPSGARSPAPTRKEFG ++ Synaptojanin
47 LETPPQPPPRSRSSHSLPSEASSQ -

Downloaded from http://www.jbc.org/ by guest on April 14, 2019

CD2BP2-GYF interacts with the spliceosomal protein SmB/B' 22
Figure 1

R10
P7
P6 G8
R3

Downloaded from http://www.jbc.org/ by guest on April 14, 2019

CD2BP2-GYF interacts with the spliceosomal protein SmB/B' 23

Figure 2

WT A C D E F G H I K L M N P Q R S T V W Y
S
H
R
P

P
P
P
G
H

R
V

Downloaded from http://www.jbc.org/ by guest on April 14, 2019

CD2BP2-GYF interacts with the spliceosomal protein SmB/B' 24

Figure 3

1 2 3 4 5 6 7 8 9 10 11 12

13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32

33 34 35 36 37 38 39 40 41 42 43 44 45 46 47

Downloaded from http://www.jbc.org/ by guest on April 14, 2019

Downloaded from http://www.jbc.org/ by guest on April 14, 2019
Downloaded from http://www.jbc.org/ by guest on April 14, 2019
CD2BP2-GYF interacts with the spliceosomal protein SmB/B' 27
Figure 6

EGFP-CD2BP2 Hoechst overlay

B.

Downloaded from http://www.jbc.org/ by guest on April 14, 2019

EGFP-CD2BP2 EYFP-SmB overlay
CD2BP2-GYF interacts with the spliceosomal protein SmB/B' 28

Figure 7

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15

16 17 18 19 20 21 22 23 24 25 26 27 28

29 30 31 32 33 34 35 36 37 38 39 40 41

Downloaded from http://www.jbc.org/ by guest on April 14, 2019

CD2BP2-GYF interacts with the spliceosomal protein SmB/B' 29

Supplementary Data

Binding of CD2BP2-GYF to sequences of splicing factors, containing the reduced

motif PPG. A protein description is given in column ‘Description’. The

SWISSPROT/TREMBL entry names are listed in the column ‘Protein’. Binding

classification: (-): < 5 x background [BLU] (Boehringer Light Units); (+): 5–

25 x background [BLU]; (++): 25–125 x background [BLU]; (+++): > 125 x

background [BLU].

Description Sequence Binding Protein

1 CD2BP2 interaction partner SHRPPPPGHR +++ Human CD2
2 Part of the U2 snRNP LGSGLPPPGMPPPGSF + Splicing factor 3B
3 PPPGMPPPGSFPPPVP - subunit 4

Downloaded from http://www.jbc.org/ by guest on April 14, 2019

4 FPPPVPPPGALPPGIP + (Q15427)
5 PPPGALPPGIPPAMPP -
6 MPPPPMPPGAAGHGPP -
7 GHPHAGPPGSGGQPPP -
8 QPPPRPPPGMPHPGPP +
9 Associates with HDFRSPPPGMGLNQNR +++ Splicing factor,
10 U4/U6.U5 tri-snPNP PPTSGAPPGSGPGPTP - proline-and glutamine-
11 PAVTSAPPGAPPPTPP + rich (P23246)
12 Vertebrate homologs to YCLAPPPPGIDVTTYY - Splicing factor,
13 Drosophila splicing regulator, TTTAPPPPGTTPPPPP - arginine/serine - rich 8
suppressor-of-white-apricot (Q12872)
14 Close homology to SmB GAPTQYPPGLGPPPPM - SNRPB protein
15 MGRGAPPPGMMGPPPG + (Q15182)
16 PGMMGPPPGMRPPMGP +
17 FLTGPPPPGMRPPRP +++
18 Mammalian branch point GMPTVIPPGLTREQER ++ Splicing factor 1
19 binding protein mBBP; MNQGPHPPGHHGPPPM - SF1-Bo isoform
20 interacts with U1 snRNA GVQPPLPPGAPPPPPP + (Q15637-2)
21 PPPPPPPPGSAGMMIP -
22 MNQGPHPPGHHGPPPM - (Q15637-1)
23 GVQPPLPPGAPPPPPP +
24 PPPPPPPPGSAGMMYA -
25 Human SNRPN gene with VGRATPPPGIMAPPPG + RT-LI
26 novel exon 3 PGIMAPPPGMRPPMGP + (Q9BPU5)
27 TPIGMPPPGMRPPPPG ++
28 PGMRPPPPGIRGEAFL ++
29 MGRGAPPPGMMGPPPG + SmB/B’
30 PGMMGPPPGMRPPMGP + (P14678)
31 TPMGMPPPGMRPPPPG +
32 PGMRPPPPGMRGPPPP +++
33 GMRGPPPPGMRPPRP ++
34 Tissue-specific splicing VGRATPPPGIMAPPPG + Sm-D
35 protein PGIMAPPPGMRPPMGP + (P14648)
36 TPIGMPPPGMRPPPPG ++
37 PGMRPPPPGIRGPPPP +++
38 GIRGPPPPGMRPPRP +++
CD2BP2-GYF interacts with the spliceosomal protein SmB/B' 30

39 QPPYMPPPGMIPPPGL + U1 small nuclear

40 PPGMIPPPGLAPGQIP - ribonucleoprotein
41 LAPGQIPPGAMPPQQL - A (P09012)

Downloaded from http://www.jbc.org/ by guest on April 14, 2019

Recognition sequences for the GYF domain reveal a possible spliceosomal function of
CD2BP2
Michael M. Kofler, Katja Heuer, Tobias Zech and Christian Freund
J. Biol. Chem. published online April 22, 2004

Access the most updated version of this article at doi: 10.1074/jbc.M402008200

Alerts:
• When this article is cited
• When a correction for this article is posted

Click here to choose from all of JBC's e-mail alerts

Supplemental material:

Downloaded from http://www.jbc.org/ by guest on April 14, 2019

http://www.jbc.org/content/suppl/2004/05/06/M402008200.DC1