Constitutive homo- and hetero-oligomerization of TβRII-B, an alternatively spliced variant of the mouse TGF-β type II receptor

BBRC Biochemical and Biophysical Research Communications 351 (2006) 651–657 www.elsevier.com/locate/ybbrc

Constitutive homo- and hetero-oligomerization of TbRII-B, an alternatively spliced variant of the mouse TGF-b type II receptor Manda S. Krishnaveni a, Jakob Lerche Hansen b, Werner Seeger a, Rory E. Morty a, Søren P. Sheikh b, Oliver Eickelberg a,* b

a Department of Medicine II, University of Giessen Lung Center, Justus-Liebig University, Giessen, Germany Laboratory of Molecular and Cellular Cardiology, The Heart Centre and The Danish National Research Foundation Centre for Cardiac Arrhythmia Copenhagen University Hospital, and the Faculty of Health, University of Copenhagen, Denmark

Received 13 October 2006 Available online 24 October 2006

Abstract Transforming growth factor (TGF)-b ligands signal through transmembrane type I and type II serine/threonine kinase receptors, which form heteromeric signalling complexes upon ligand binding. Type II TGF-b receptors (TbRII) are reported to exist as homodimers at the cell surface, but the oligomerization pattern and dynamics of TbRII splice variants in live cells has not been demonstrated thus far. Using co-immunoprecipitation and bioluminescence resonance energy transfer (BRET), we demonstrate that the mouse TbRII receptor splice variant TbRII-B is capable of forming ligand-independent homodimers and heterodimers with TbRII. The homomeric interaction of mouse (m)TbRII-B isoforms, however, is less robust than the heteromeric interactions of mTbRII-B with wild-type TbRII, which indicates that these receptors may be more likely to heterodimerize when both receptors are expressed. Moreover, we demonstrate that mTbRII-B is a signalling receptor with ubiquitous tissue expression. Our study thus demonstrates previously unappreciated complex formation of TGF-b type II receptors, and suggests that mTbRII-B can direct TGF-b-induced signalling in vitro and in vivo.  2006 Elsevier Inc. All rights reserved. Keywords: TbRII-B; Extracellular domain; BRET; Receptor oligomerization; Signalling

Transforming growth factor (TGF)-b belongs to a family of multifunctional proteins, which play key roles in cell growth, differentiation, and apoptosis [1]. TGF-b exerts its biological activities by binding to specific cell surface receptors, which include the type I, II, and III TGF-b receptors [2,3]. While type I and type II receptors have intrinsic signalling capacity by virtue of their serine/threonine kinase domains, betaglycan (the type III TGF-b receptor) and endoglin serve as accessory receptors and function predominantly as modulators of ligand availability [4]. Two different mechanisms of oligomerization and activation of TGF-b receptors have been proposed. The first model proposes that TGF-b ligands bind to the extracellu*

Corresponding author. Fax: +49 641 9942309. E-mail address: [email protected] (O. Eickelberg). 0006-291X/$ - see front matter  2006 Elsevier Inc. All rights reserved. doi:10.1016/j.bbrc.2006.10.083

lar domain of TGF-b receptor II (TbRII), which recruits the TGF-b receptor I (TbRI) and leads to the formation of a heterotetrameric signalling entity (presumably of two type I and two type II receptors) [5]. A second model emphasizes the existence of inactive preformed complexes of TbRI and TbRII, suggesting that ligand binding induces a conformational change leading to subsequent receptor activation of a pre-assembled receptor complex [6,7]. Furthermore, the existence of heterotetrameric complexes of different type I and type II receptors has been postulated [8,9]. The likelihood of this has been strengthened by the discovery of various splice variants for TGF-b receptors, which increases the number of receptor combinations possible for TGF-b ligands, thereby generating signal diversity. To date, TbRI, TbRII, and betaglycan have all been found to be present as homo-oligomers on the cell surface in the absence of TGF-b [10–12]. Homomeric preformed

652

M.S. Krishnaveni et al. / Biochemical and Biophysical Research Communications 351 (2006) 651–657

receptor complexes are not sufficient to transduce TGF-b responses, but are proposed to be functionally important for regulating receptor kinase activity. For instance, it has been reported that the TbRII is activated by intermolecular autophosphorylation, upon ligand binding [13,14]. Although their existence has been predicted, the presence of TGF-b receptor oligomers at the cell surface has never been experimentally demonstrated in living cells. In light of these observations, receptor splice variants may play a role in determining different oligomerization patterns, in addition to their specific ligand binding properties. In the present study, we took advantage of the recently developed technique of bioluminescence resonance energy transfer (BRET2, a second generation system of BRET, [15]) and co-immunoprecipitation to evaluate ligand-independent homo- and hetero-dimer formation of TbRII splice variants in living cells. We chose BRET2 analysis, as this method allowed us to monitor receptor interactions in living cells without the need for the fixation and/or permeabilization of cells. BRET is the product of nonradiative energy transfer from a luminescent donor to a fluorescent acceptor protein. This technique has been applied in oligomerization studies of G-protein-coupled receptors and calcium-sensing receptor [16,17]. Here we report that mouse (m)TbRII exists in different dimer combinations in live cells (TbRII/TbRII, TbRII/TbRII-B, and TbRII-B/TbRII-B). Our data also indicate that the TbRII-B variant prefers to form TbRII/TbRII-B heterodimers rather than homodimers, since the TbRII-B/TbRII-B interaction is less robust than TbRII/TbRII, or the TbRII/TbRII-B interactions. Thus, these results suggest an unexpected complexity of TGF-b signalling with respect to cell surface interactions of TGF-b receptor isoforms. Materials and methods Reverse transcription-polymerase chain reaction (RT-PCR). For all PCRs, first strand cDNA (Clontech, Mountain View, CA) derived from the respective tissues was used. For amplification of TbRII splice variants, primers that discriminate between the two isoforms were designed (forward: 5 0 -GAG AGG GCG AGG AGT AAA GG-3 0 reverse: 5 0 -GTG GTA GGT GAG CTT GGG GT-3 0 ) and the PCR was performed using Taq polymerase (Promega, Madison, WI). The gene encoding heat shock protein cognate 70 (forward, 5 0 -CAA GCG AAA GCA CAA GAA AGA CAT-3 0 ; reverse, 5 0 -ATA CCA AGC GAA AGA GGA GTG ACA TC-3 0 ) served as a loading control. Plasmids. Full-length cDNAs encoding mouse TbRII (GenBank Accession No. NM_029575) and TbRII-B (GenBank Accession No. NM_009371) were subcloned into codon-optimized pGFP2-N1 vectors at their EcoRI and BamHI sites (Perkin-Elmer, Wellesley, MA). Similarly, TbRII and TbRII-B coding sequences were subcloned into codon-optimized pRluc-N1 (Perkin-Elmer) at the MluI and BamHI sites, as well as into the BamHI and EcoRI sites of pCMV-Flag (Stratagene, La Jolla, CA). Cell culture and transfections. NIH-3T3 and COS-7 cells were maintained in Dulbecco’s modified Eagle’s medium (DMEM) supplemented with 10% (v/v) fetal bovine serum (FBS). NIH-3T3 cells were seeded in 10-cm dishes (1 · 105 cells/well) and transfected with Lipofectamine 2000 (Invitrogen, Carlsbad, CA) according to the manufacturer’s protocol. COS-7 cells were seeded into 10-cm dishes (2.5 · 106 cells/well) and transfected with Lipofectamine 2000. After 4 h of incubation, the transfection reagents were removed and replaced with fresh growth media.

Western-blotting. Mouse tissues were ground to powder under liquid nitrogen and solubilized in lysis buffer [20 mM Tris–Cl, pH 7.5, 150 mM NaCl, 1% (v/v) Triton X-100, including Complete protease inhibitor cocktail (Roche Molecular Biosciences, Basel, CH)] for 40 min at 4 C. Solubilized proteins were separated by SDS–PAGE and hybridized with a mouse-specific TbRII-B antibody (a kind gift of P. Knaus, Free University of Berlin). Immunoprecipitation. Twenty-four hours after transfection, cells were solubilized in immunoprecipitation lysis buffer [20 mM Tris–Cl, pH 7.5, 150 mM NaCl, 1% (v/v) Nonidet P-40, 2% (v/v) glycerol including Complete protease inhibitor cocktail] for 40 min on ice. The indicated receptors were immunoprecipitated using an anti-GFP mouse monoclonal antibody (Santa Cruz Biotechnology, Santa Cruz, CA) using Protein G– Sepharose beads (Amersham, Piscataway, NJ) for 4 h at 4 C. The bound protein fraction was eluted by boiling the beads in SDS–PAGE sample loading buffer containing b-mercaptoethanol for 5 min at 95 C. The proteins were analyzed on 10% SDS–PAGE gels, followed by immunoblotting with an anti-Renilla luciferase (Rluc) mouse monoclonal antibody (Chemicon, Temecula, CA). BRET2 assay. The BRET2 assay was performed as previously described [17,18], with minor modifications. Briefly, 2.5 million COS-7 cells were seeded into 10-cm dishes. The following day, cells were transfected using Lipofectamine 2000 (Invitrogen), according to the manufacturer’s protocol. For titration experiments, all transfections contained 1 lg of Rluc-tagged expression plasmid and 0.1–4 lg of the GFP2-tagged expression plasmid. In competition titration experiments using wild-type proteins, 4 lg of the relevant Flag-tagged expression plasmid was co-transfected with the Rlucor GFP2-tagged expression plasmid. After 48 h, the cells were washed twice with PBS prior to detachment in PBS supplemented with 1 mM EDTA. The cells were spun down, resuspended in PBS, and split into two pools of approximately 0.5 · 106 cells. The first pool was used to examine GFP2 expression levels by fluorescence analysis, and Rluc expression by luminescence analysis. The second pool was incubated with the Rluc substrate DeepBlueC (Perkin-Elmer), and luminescence at 515/30 nm and 410/80 nm was measured in a Fusion Reader (Packard Biosciences). The BRET2 ratio was determined according to the previously described calculation [16]: 1000 · [(emission (515/30 nm)  (emission (410/80 nm) · Cf)]/(emission (410/80 nm); where Cf denotes the Rluc luminescence crosstalk ratio defined as emission (515/30 nm)/emission (410/80 nm), when Rluc expressed alone is excited. Data were analyzed by one site binding/hyperbola and non-linear regression curve fitting (GraphPad Prism). Reporter gene assays. Cells were split into 48-well dishes and transfected using Lipofectamine 2000 (Invitrogen), according to the manufacturer’s protocol. Twenty-four hours after transfection, cells were stimulated with TGF-b1 (2 ng/ml; R&D Systems, Minneapolis, MN) for 8–12 h. Cells were homogenized in lysis buffer and luciferase activity was determined using the Dual Luciferase Assay system (Promega).

Results Isolation of the mouse TbRII-B-encoding gene RT-PCR was used to screen for variants of the TGF-b type II receptor exhibiting alterations in the extracellular (ligand-binding) domain. Upon amplification of cDNA from mouse total lung RNA, an additional PCR product with lower electrophoretic mobility was detected (Fig. 1). Sequence analysis revealed that this PCR product was identical to mouse TbRII-B, an alternatively spliced variant of TbRII described previously [19]. A human ortholog of this splice-variant has also been reported [20]. The insertion exhibits only 82% identity between the human and mouse variants. Moreover, the insertion in the human variant corresponds to an additional exon (exon 1a) within intron 1,

M.S. Krishnaveni et al. / Biochemical and Biophysical Research Communications 351 (2006) 651–657

A

aa

1

23

184

aa

1

23 31

214

57

653

569 592

209

239

594 617

Exon 3

Signal peptide

Transmembrane domain

Extracellular domain

Kinase domain

Insertion

Tail

B

Exon 3 30

40

50

Thymus

Testis

Sk. muscle

mTβRII-B

Lung

D Testis

Sk. muscle

Kidney

Spleen

Liver

Heart

Brain

M

Lung

C

mTβRII-B mTβRII

Stomach

20

Spleen

10

* MGRGLLRGLWPLHIVLWTRIASTIPPHVPKSDVEMEAQKDASIHVSCNRTIHPLKHFNS MGRGLLRGLWPLHIVLWTRIASTIPPHVPKS-------------------------VNS

mTβRII HSC

Fig. 1. Ubiquitous expression of mTbRII-B, a splice variant of mTbRII. (A) Sequence alignment of mTbRII and mTbRII-B, indicating the cysteine residues and a potential N-glycosylation site (bold and underlined). The single amino acid substitution is indicated by an asterisk. (B) PCR products were obtained from cDNAs of different tissues using isoform-spanning primers. HSC was used as the loading control. (C) Proteins were extracted from various tissues and probed with a mouse-specific aTbRII-B antibody.

whereas the insertion in the mouse variant corresponds to exon 3, as revealed by comparison of the exon–intron structure of TbRII from the two species. The alternative splicing causes an insertion of 75 nucleotides at the coding sequence for Val32 of the wild-type receptor. This insertion generates an amino acid change (Val32 to Phe57 of TbRII-B) in addition to the 25 amino acid insertion and harbors a cysteine residue, and a potential N-glycosylation site (Fig. 1A and B). Ubiquitous expression of TbRII-B in different mouse tissues Next we analyzed the expression pattern of the newly discovered mouse splice variant, as the human variant has been reported to show a restricted expression profile [19]. As illustrated in Fig. 1B, different tissues exhibited different expression levels of TbRII and TbRII-B. For example, the lung, heart, kidney, and spleen exhibited high expression levels of TbRII-B, whereas the liver, skeletal muscle, and testis exhibited a relatively low expression, with brain exhibiting the lowest expression level. Western blot analysis (Fig. 1C) of mouse tissues further confirmed the ubiquitous expression of the TbRII-B not only at the mRNA level, but also at the protein level. TbRII-B forms constitutive homo- and hetero-oligomeric complexes It has been demonstrated that TGF-b receptors form homodimers even in the absence of TGF-b, which later

may (in the case of signalling receptors) or may not (in the case of accessory receptors) form part of the signalling complex. In order to analyze receptor complex formation of the TbRII splice variants, we used two different approaches, co-immunoprecipitation and the BRET2 technology. Classic co-immunoprecipitation performed on extracts of transiently transfected NIH-3T3 cells was conducted with different combinations of the two TbRII splice variants that carried either a GFP2 or a Rluc-tag at their Cterminus. Co-immunoprecipitation of the Rluc antigen with an anti-GFP antibody demonstrated the physical association between the two variants. As indicated in Fig. 2, different combinations of dimers occurred: TbRII/ TbRII (lane 1), TbRII-B/TbRII (lane 2), and TbRII-B/ TbRII-B (lane 3). Proper expression of the constructs was confirmed by Western blotting with their respective anti-epitope tag antibodies performed on total cell lysates (lower panels). BRET2 experiments show that TbRII-B may prefer to enter heterodimeric complexes To further characterize the dimerization of the homoand hetero-dimeric interactions of TbRII splice variants, we next used the BRET2 experiments. TbRII, TbRII-B, and calcium-sensing receptors (CaR; as negative controls) were fused to either Rluc or GFP2, transfected into COS7 cells, and expressed in various combinations to generate BRET2 saturation curves (Fig. 3A). As depicted in

654

M.S. Krishnaveni et al. / Biochemical and Biophysical Research Communications 351 (2006) 651–657

IP: GFP; IB: Rluc

Whole cell lysate

WB: GFP

WB: Rluc

TβRII-Rluc TβRIIB-Rluc TβRII-GFP TβRIIB-GFP

Fig. 2. mTbRII-B forms homo- and hetero-oligomers. NIH-3T3 cells were transfected with the indicated combinations of GFP2- and Rluc-tagged mTbRII and mTbRII-B constructs. Cell lysates were immunoprecipitated with a-GFP antibody and then probed with a-Rluc antibody.

A 175

B

150

150 100

BRET

BRET

125 100 75

50

50 25 0 0.0

0

BRET

C

120 110 100 90 80 70 60 50 40 30 20 10 0 0.00

50

75 100 125 150 GFP2/Rluc-ratio

175 200

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

GFP2/Rluc-ratio

D

75

50

BRET

-25

25

25

0.25

0.50

0.75

1.00

1.25

GFP2/Rluc-ratio

0 0.00

0.25

0.50

0.75

1.00

GFP2/Rluc-ratio

2

Fig. 3. Oligomerization of TGF-b receptors using BRET technology. (A) COS-7 cells were transiently transfected with a fixed amount of mTbRII-Rluc or mTbRII-B-Rluc expression vectors and increasing amounts of TbRII-GFP2, TbRII-B-GFP2, or CaR-GFP2. BRET2-ratios, total luminescence, and total fluorescence were measured as described in Materials and methods. Data shown are representative from at least three independent experiments, and depicted as mBRET2 (mBRET2 = 1000 · BRET2) as a function of GFP2/Rluc ratio in arbitrary units. [(mTbRII-mTbRII (j); mTbRII-mTbRII-B (.); mTbRII-B-mTbRII-B (m); mTbRII-CaR (); mTbRIIB-CaR (d).] (B) The specificity of mTbRII receptor interactions was assessed by saturation experiments using ‘wild type competition’. Cells were transfected in the following combinations. (C) TbRII-Rluc and TbRII-GFP2 either with (m) or without (j) Flag-tagged TbRII receptor. (B) TbRII-Rluc and TbRII-B-GFP2 either with (m) or without (j) Flag-tagged TbRII. (D) TbRII-B-Rluc and TbRII-B-GFP2 with (m) or without (j) Flag-tagged mTbRII. Data shown are representative from at least three independent experiments and are reported as mBRET2 (mBRET2 = 1000 · BRET2) as a function of the GFP2/Rluc ratio in arbitrary units.

Fig. 3A, the TbRII-Rluc/TbRII-GFP2, TbRII-Rluc/ TbRII-B-GFP2, and TbRII-Rluc/TbRII-B-GFP2 BRET2 titration curves were all robust in slope and maximum response, but were dramatically different from those of the TbRII-Rluc/CaR-GFP2 titration curve. These experi-

ments indicated that TbRII-B formed both homo- and hetero-oligomers with TbRII (BRET values are reported in arbitrary units in Table 1). We observed no significant difference between the titration curves, when calculating the maximum BRET (BRETmax) and BRET50 values

M.S. Krishnaveni et al. / Biochemical and Biophysical Research Communications 351 (2006) 651–657

655

Table 1 Detailed BRET analysis of TbRII and TbRII-B dimerization Dimer

BRET values

Donor

Acceptor

BRET50

N

BRET50/break

N

BRETmax

N

TbRII-Rluc TbRII-B-Rluc TbRII-Rluc TbRII-Rluc TbRII-B-Rluc

TbRII-GFP2 TbRII-B-GFP2 TbRII-B-GFP2 CaR-GFP2 CaR-GFP2

0.105 ± 0.01 0.140 ± 0.02 0.087 ± 0.01 >1000 >1000

6 6 6 3 3

0.80 ± 0.2 0.74 ± 0.41 0.55 ± 0.15

3 3 3 ND ND

209.00 ± 23.4 90.15 ± 7.63 167.63 ± 31.6 ND ND

6 6 6

BRET50 values indicate at which GFP2/Rluc ratio 50% of the maximum BRET value is reached. Data shown are average values from N titration-curve experiments (as depicted in Fig. 3). Donor represents the Rluc-tagged receptor molecule and Acceptor represents the GFP2-tagged receptor molecule. The BRET50/break reports the BRET50 value in the BRET competition assays where unlabelled TbRII or TbRII-B receptor is present. The BRETmax value reports the maximum BRET value reached for a given dimer combination. Statistical analysis included two-tailed unpaired t test to determine significant differences between different forms of dimerization (see text for details). ND denotes not determined.

between the TbRII/TbRII homodimers and the TbRII-B/ TbRII heterodimers. The BRET50 values were 0.105 ± 0.01 and 0.087 ± 0.01 for the TbRII/TbRII homodimer and the TbRII-B/TbRII heterodimer, respectively. The BRETmax values for the same combinations were 209 ± 23 and 168 ± 31, respectively. On the other hand, the TbRII-B/TbRII-B homodimeric interaction exhibited a BRET50 value of 0.014 ± 0.02 and a BRETmax of 90 ± 8, which was significantly weaker than those of the TbRII/TbRII homodimers or TbRII-B/TbRII heterodimers (p < 0.01). These data suggest that the TbRII-B splice variant preferentially forms heterodimers when both receptors are present.

TbRII and CaR are not considered to dimerize due to the strong homodimeric properties of these receptors [17]. From this experiment, it is clear that all titration curves resulting from TbRII-B and TbRII homo- and hetero-dimeric interactions are significantly different both in curve and maximum values from those of the TbRII-B-Rluc and the TbRII-B-GFP2 BRET2-titration curve, both of which had a BRET50 value higher than 100. The second control experiment was performed to verify that the BRET2 signals were derived from specific homoand hetero-oligomerization. To do so, we overexpressed Flag-tagged receptors in combination with the Rluc- and GFP2-tagged receptors and investigated whether these receptors could reduce the BRET2 signal. To do so, we compared the saturation curves generated from the Rluc/GFP2tagged receptors in the absence or presence of the Flagtagged receptors. The BRET50 values are reported in Table 1 as BRET50/break values. Over-expression of Flag-tagged receptors significantly right-shifted the BRET50 values of the saturation curves for the TbRII-Rluc/TbRII-GFP2 interaction (Fig. 3B, right-shift of 7.6-fold, p < 0.01), TbRII-Rluc/TbRII-B-GFP2 interaction (Fig. 3C, right-

BRET2 curves are due to specific protein–protein interactions To test that the BRET signal was indeed a result of a specific protein–protein interaction, we performed two essential control experiments. First, we coexpressed TbRII-B/Rluc and TbRII/Rluc together with a receptor from the seven transmembrane receptor family C, CaRGFP2. We used CaR-GFP2 as a negative control, since

Unstimulated

Stimulated (TGF-β1)

** ** 40000

3000

Relative luciferase units

Relative luciferase units

**

*

3500

2500 2000 1500 1000 500 0

30000

20000

10000

0 EV

TβRII

TβRII-B

EV

TβRII

TβRII-B

Fig. 4. TbRII-B is a signalling receptor: NIH-3T3 cells were transfected with either mTbRII-Rluc, mTbRII-B Rluc expression vectors, or an empty expression vector (EV) as a control, as indicated, in combination with the reporter plasmid (CAGA)12-luc. After 24 h, cells were treated with 2 ng/ml TGFb1 (black bars) or without TGF-b1 (gray bars) for 8 h. Luciferase activity was measured by Luciferase assays. Data represent means of four independent experiments. *p<0.05; **p < 0.001.

656

M.S. Krishnaveni et al. / Biochemical and Biophysical Research Communications 351 (2006) 651–657

shift of 2-fold, p < 0.01), and TbRII-B-Rluc/TbRII-BGFP2 interaction (Fig. 3D, right-shift of 6.3-fold, p < 0.01), which further supported the finding that these interactions are specific for TGF-b receptors. In combination, these data suggest that both the TbRII and the TbRII-B variant form constitutive homo- and hetero-oligomeric complexes in living cells and that TbRII-B preferentially forms hetero-dimers. TbRII-B is a signalling receptor To assess whether the TbRII-B splice variant had a signalling potency similar to TbRII, we performed reporter gene assays. The induction of the TGF-b-responsive reporter gene (CAGA)12-luc was tested in NIH-3T3 cells transfected with either of the splice variants, where TGFb1 showed a twofold induction with both variants compared with an empty expression plasmid (Fig. 4). Thus, these results indicated that TbRII-B can also be a part of the signalling complex, and can regulate the transcription of downstream target genes. Discussion To date, various splice variants have been reported for TGF-b receptors [1,4]. Though the functional implications of these splice variants have not been fully elucidated, most of them contain insertions in their extracellular domains, implying a role in determining isoform specificity of the ligand. TbRII has been reported to bind TGF-b1 and TGF-b3 with a higher affinity, whereas it requires TbRIII for binding of TGF-b2 [4]. The human ortholog of TbRII-B (hTbRII-B) is reported to bind TGF-b2 with higher affinity independent of TbRIII [21]. The present isoform under study, mouse TbRII-B (mTbRII-B), also contains an insertion of 25 amino acids in the extracellular domain introducing a cysteine residue and a potential N-glycosylation site similar to the human variant, suggesting similar roles of these isoforms. However, in contrast to the predicted function, mTbRII-B exhibits ligand specificity similar to TbRII. Along these lines, our study with the mTbRII-B reporter assay (Fig. 4) demonstrated transcriptional activity in response to TGF-b1, confirming similar signalling capacity to that of TbRII. While mTbRII-B has been isolated from a brain cDNA library [19], we observed that brain exhibited the lowest expression of mTbRII-B. The sequence alignment of the mouse and human variants revealed only 82% homology of the insertion. While hTbRII-B exhibited an additional exon 1a within its intronic sequence, the insertion in mTbRII-B is generated by an additional exon 3. In this respect, it would be interesting to analyze in future the differential signalling of mTbRII-B in contrast to its wild type variant (Fig. 1). TGF-b receptors have been shown to exist as multimeric complexes on the cell surface [4]. It has been demonstrated that homodimerization as well as their particular localization at the cell surface have a strong impact on mediating

and regulating intracellular signals [1,4]. Therefore, the next possible significance apart from ligand specificity can be that mTbRII-B might exhibit a different oligomerization pattern. As such, we performed co-immunoprecipitation to assess the homo- and heteromeric interactions of mTbRII and mTbRII-B. As shown in Fig. 2, different combinatorial dimers (mTbRII-mTbRII; mTbRII-mTbRII-B; mTbRIIB-mTBRII-B) existed in the cell, even in the absence of ligand. Given that heteromeric ligand dimers exist for TGF-b isoforms-like TGF-b1 and b2, and that TGF-b signals via two different type I receptors [activin receptor-like kinases (ALK) 1 and 5] in co-operation with TbRII [22], these results are indicative of complex cell-surface receptor interactions in TGF-b signalling. Different TbRII dimer combinations might recruit different type I receptors, and as such different ligand dimer complexes and their conformational arrangements might influence cellular signalling and their downstream responses. Light energy transfer techniques have evolved as sensitive and sophisticated methods to analyze receptor oligomerization [23,24]. Therefore, we have taken advantage of a bioluminescence resonance energy transfer technique to assess the oligomerization patterns of TGF-b receptors. As such, our data from co-immunoprecipitations have been validated, and furthermore clarified preferential receptor complex formation, in that the interaction of the TbRIIB-TbRII-B combination is not as robust as that of the other two combinations (Table 1 and Fig. 4). Accordingly, it is likely that TbRII-B prefers to form heterodimers rather than homodimers when presented to TbRII, whereas TbRII does not discriminate between homo- and hetero-dimerization with respect to TbRII-B. In future studies, the BRET technology represents a promising technique for deciphering TGF-b receptor complexity with respect to more isoforms, as it has been in the case of G-protein-coupled receptors [23,24]. In summary, our data support a model of differential signalling activities with respect to TbRII splice variants. The complex nature of this receptor system would represent a significant departure from what is currently known about transmembrane TGF-b signalling by serine/threonine kinases.

Acknowledgments This study was supported by the German Research Foundation (DFG) Collaborative Research Center 547 (to W.S., O.E.), and a Molecular Biology and Medicine of the Lung (MBML) PhD student fellowship to M.S.K., sponsored by Altana Pharma. This work was also funded by the John Meyer Foundation (J.L.H., S.P.S.), the Danish Medical Research Council and the Danish National Research Foundation (J.L.H., S.P.S.). We are grateful for stimulating discussions provided by all members of the Eickelberg Lab, Jeanette Hansen for excellent technical assistance, and Petra Knaus for providing the TbRII-B antibody.

M.S. Krishnaveni et al. / Biochemical and Biophysical Research Communications 351 (2006) 651–657

References [1] Y. Shi, J. Massague, Mechanisms of TGF-beta signaling from cell membrane to the nucleus, Cell 113 (2003) 685–700. [2] J. Massague, Receptors for the TGF-beta family, Cell 69 (1992) 1067– 1070. [3] R. Derynck, TGF-beta-receptor-mediated signaling, Trends Biochem. Sci. 19 (1994) 548–553. [4] J. Massague, TGF-beta signal transduction, Annu. Rev. Biochem. 67 (1998) 753–791. [5] P. ten Dijke, H. Yamashita, H. Ichijo, P. Franzen, M. Laiho, K. Miyazono, C.H. Heldin, Characterization of type I receptors for transforming growth factor-beta and activin, Science 264 (1994) 101–104. [6] H.J. Zhu, A.M. Sizeland, Extracellular domain of the transforming growth factor-beta receptor negatively regulates ligand-independent receptor activation, J. Biol. Chem. 274 (1999) 29220–29227. [7] H.J. Zhu, A.M. Sizeland, A pivotal role for the transmembrane domain in transforming growth factor-beta receptor activation, J. Biol. Chem. 274 (1999) 11773–11781. [8] L. Gilboa, A. Nohe, T. Geissendorfer, W. Sebald, Y.I. Henis, P. Knaus, Bone morphogenetic protein receptor complexes on the surface of live cells: a new oligomerization mode for serine/threonine kinase receptors, Mol. Biol. Cell 11 (2000) 1023–1035. [9] H. Yamashita, P. ten Dijke, P. Franzen, K. Miyazono, C.H. Heldin, Formation of hetero-oligomeric complexes of type I and type II receptors for transforming growth factor-beta, J. Biol. Chem. 269 (1994) 20172–20178. [10] Y.I. Henis, A. Moustakas, H.Y. Lin, H.F. Lodish, The types II and III transforming growth factor-beta receptors form homo-oligomers, J. Cell Biol. 126 (1994) 139–154. [11] L. Gilboa, R.G. Wells, H.F. Lodish, Y.I. Henis, Oligomeric structure of type I and type II transforming growth factor beta receptors: homodimers form in the ER and persist at the plasma membrane, J. Cell Biol. 140 (1998) 767–777. [12] R.H. Chen, R. Derynck, Homomeric interactions between type II transforming growth factor-beta receptors, J. Biol. Chem. 269 (1994) 22868–22874. [13] K. Luo, H.F. Lodish, Signaling by chimeric erythropoietin-TGF-beta receptors: homodimerization of the cytoplasmic domain of the type I

[14] [15]

[16]

[17]

[18]

[19]

[20]

[21]

[22]

[23]

[24]

657

TGF-beta receptor and heterodimerization with the type II receptor are both required for intracellular signal transduction, EMBO J 15 (1996) 4485–4496. M. Lutz, P. Knaus, Integration of the TGF-beta pathway into the cellular signalling network, Cell Signal 14 (2002) 977–988. Y. Xu, D.W. Piston, C.H. Johnson, A bioluminescence resonance energy transfer (BRET) system: application to interacting circadian clock proteins, Proc. Natl. Acad. Sci. USA 96 (1999) 151–156. S. Angers, A. Salahpour, E. Joly, S. Hilairet, D. Chelsky, M. Dennis, M. Bouvier, Detection of beta 2-adrenergic receptor dimerization in living cells using bioluminescence resonance energy transfer (BRET), Proc. Natl. Acad. Sci. USA 97 (2000) 3684–3689. A.A. Jensen, J.L. Hansen, S.P. Sheikh, H. Brauner-Osborne, Probing intermolecular protein–protein interactions in the calcium-sensing receptor homodimer using bioluminescence resonance energy transfer (BRET), Eur. J. Biochem. 269 (2002) 5076–5087. J.L. Hansen, J. Theilade, S. Haunso, S.P. Sheikh, Oligomerization of wild type and nonfunctional mutant angiotensin II type I receptors inhibits gaq protein signaling but not ERK activation, J. Biol. Chem. 279 (2004) 24108–24115. A. Suzuki, N. Shioda, T. Maeda, M. Tada, N. Ueno, Cloning of an isoform of mouse TGF-beta type II receptor gene, FEBS Lett. 355 (1994) 19–22. R. Hirai, T. Fijita, A human transforming growth factor-beta type II receptor that contains an insertion in the extracellular domain, Exp. Cell Res. 223 (1996) 135–141. D. Rotzer, M. Roth, M. Lutz, D. Lindemann, W. Sebald, P. Knaus, Type III TGF-beta receptor-independent signalling of TGF-beta2 via TbetaRII-B, an alternatively spliced TGF-beta type II receptor, EMBO J 20 (2001) 480–490. M.J. Goumans, G. Valdimarsdottir, S. Itoh, F. Lebrin, J. Larsson, C. Mummery, S. Karlsson, P. ten Dijke, Activin receptor-like kinase (ALK)1 is an antagonistic mediator of lateral TGFbeta/ALK5 signaling, Mol. Cell 12 (2003) 817–828. K.D. Pfleger, K.A. Eidne, New technologies: bioluminescence resonance energy transfer (BRET) for the detection of real time interactions involving G-protein coupled receptors, Pituitary 6 (2003) 141–151. J.L. Hansen, S.P. Sheikh, Functional consequences of 7TM receptor dimerization, Eur. J. Pharm. Sci. 23 (2004) 301–317.