Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 7-Day Trial for You or Your Team.

Learn More →

Phosphorylation of Ser165 in TGF‐beta type I receptor modulates TGF‐beta1‐induced cellular responses.

Phosphorylation of Ser165 in TGF‐beta type I receptor modulates TGF‐beta1‐induced cellular... The EMBO Journal vol.15 no.22 pp.6231-6240, 1996 Phosphorylation of in TGF-13 type I receptor Ser165 modulates TGF-f1-induced cellular responses Serhiy Souchelnytskyil, Peter ten Dijke, TGF-,B type I and type II receptors (T,BR-I and T,R-II) are essential for signal transduction (Wrana et al., 1994). Kohei Miyazono2 and Carl-Henrik Heldin T3R-I and TfR-II are widely expressed on many cell Ludwig Institute for Cancer Research, Box 595, S-751 24, Uppsala, types and bind TGF-,B with dissociation constants in the Sweden and 2The Cancer Institute, Japanese Foundation for Cancer low picomolar range (Massague et al., 1990). Cloning Research, 1-37-1 Kami-Ikebukuro, Toshima-ku, Tokyo 170, Japan and sequencing of TIR-I and cDNAs revealed TOR-II 'Corresponding author that they are single transmembrane serine/threonine kin- ases with cysteine-rich extracellular domains (Lin et al., Transforming growth factor-,l (TGF-1) signals via 1992; Franzen et al., 1993; Bassing et al., 1994). The two of two serine/threonine kinase an oligomeric complex kinase domains have 41% amino acid sequence similarity receptors denoted type I receptor (T,R-I) and TGF-P and each contains two short kinase inserts. binds TPR-II type II receptor We investigated the in vivo (TPR-II). directly, whereas T,3R-I can recognize only ligand TGF-P phosphorylation sites in TjR-I and T3R-II after com- when complexed with T,3R-II (Wrana et al., 1992). Phosphorylation of TJR-II was plex formation. TfR-II is a constitutively active kinase, Overexpressed residues in the C-terminus (Ser549 and observed at the ligand-recruited T,R-I on which transphosphorylates at residues in the juxtamembrane domain Ser551) and residues, predominantly located in a serine and threonine TGF-,1 induced in vivo (Ser223, Ser226 and Ser227). juxtamembrane domain containing a region of the of serine and threonine residues in phosphorylation (Wrana et al., 1994). GSGSGS motif (the GS domain) the domain of TfR-I in a region rich juxtamembrane serine and threonine residues in the Mutation of multiple in serine and threonine residues (GS domain; glycine, illustrating their GS domain impairs TGF-f responses, Thrl86, Serl87, Serl89 and Serl91), and more Thrl85, importance in signal transduction (Wrana et al., 1994; (Serl65). Phosphorylation in N-terminal of this region Franzen et al., 1995; Wieser et al., 1995). Serl72 and has been shown previously to be the GS domain Thrl76 in TfAR-I were found to be dispensable for extra- in activation of the T,BR-I kinase. We show involved cellular matrix (ECM) formation, but essential for the of TIR-I at Serl65 is here that phosphorylation inhibition by (Saitoh et al., 1996). However, growth TGF-P in modulation of TGF-j1 signaling. Mutations involved of Serl72 and Thrl76 has not been phosphorylation in led to an increase in of Serl65 TGF-p1- TOR-I and a molecular mechanism remains to be reported, mediated inhibition and extracellular matrix growth elucidated. The notion that acts downstream of TPR-I but, in contrast, to decreased TGF-f1- formation, and that TfR-I phosphorylation and activation is T,R-II induced A transcriptional activation signal apoptosis. essential and sufficient for most TGF-,-mediated was not affected. Mutations of Serl65 changed the is by data showing that a mutation responses supported pattern of TI3R-I. These observations phosphorylation which inactivates its ability to recognize T,R-I in T,BR-II that TGF-J receptor signaling specificity is suggest does not signal (Caircamo et al., 1995), and as a substrate of Serl65 of T,R-I. modulated by phosphorylation active mutant (threonine residue that a constitutively T,BR-I serine-threonine kinase/signal transduction/ Keywords: an acid residue) can signal in at 204 replaced by aspartic growth factor-,B receptor/transforming of and et al., 1995). the absence ligand (Wieser TPR-II the formation of a heteromeric complex TGF-4 induces T3R-II et Franzen et al., of ThR-I and (Wrana al., 1992; a heterotetramer of two molecules each 1993), most likely Introduction et Recent of and (Yamashita al., 1994b). TPR-I TPR-II studies between kinase-defective TfR-I isoforms complementation factor-j, (TGF-a) (TGF- Transforming growth indicate a and activation-defective TiR-I cooperative of structur- and belong to a large superfamily -,2 -P3) P1, molecules that is interaction between multiple T,R-I that cell related dimeric proteins regulate proliferation, ally transduction and essential for (Weiss-Garcia and of different signal differentiation motility many apoptosis, Massague, 1996). et Roberts and cell al., 1990; Sporn, 1990). types (Moses intrinsic for each have an and affinity is on cell The effect induced by TGF-,B dependent type TPR-I TPR-II et other al., 1994) and, upon overexpression, microenvironment the cell. TGF-,B (Ventura and on the surrounding heteromeric a complex may form ligand-independent roles in they is thought to play important many physiological are et al., in which both receptors phosphorylated (Ventura as the immune inflammation, such response, processes it has been and Moreover, Chen Weinberg, 1995). and wound healing. 1994; embryogenesis, angiogenesis of alone to a that high interaction with overexpression T,R-I its effects reported TGF-1 exerts through its constutive can lead to (Chen cell surface level autophosphorylation and binding proteins (Lin multiple receptors and et Weinberg, 1995). al., 1994; and Lodish, 1993; Derynck, 1994; Massague of the mechanisms that ten et of which Our understanding signaling et Yingling al., 1995; Dijke al., 1996), © Press Oxford University et aL S.Souchelnytskyi T. R-lI R-l1 s . H-=L HA H is is H sHHis Ti) R-l - 2 To R-11 + Ni-NTA HAHai HA A X 1.5 IP -HA VPN E. H; A *- i no-omrnpIleCxe: T R--- HA ~\? ; ,,, AA A .2 0.5 IP DRL I+TGF-3" : 0 nInr.- COmpIes x-Li' - TGF- _+ + fil is Hi s - 'I Hi,- lII IFi 1.lk uflT 141'w1 1.S1z II I l in V -1 TR-1T 1 i11TlITi] \L c\.1k .I A IC l Tl'L11d d Ti% -1 arld T. R-111 ( COrIY1plexe HA ,I IF1I<R 1.II I lb \ s Illi IICI I\. ","II 1 tItIl TLItLIv Ciopi-'. C i'i ll l M -NTA ;L L Ill t1II Tc1Tle \ Ilf T 1]T IIII iT LILICH PCI LI1T I1i .. l Il ldLI1 i-1 I .\LiI 111 11 1ll 1'Ti [)II T T \111ll 1i L:ll LIT .1. ; i TLiI 11I I R- IIl1II\111 IItIX ITiLl L L 1LJiIu1 ItL (C ti C TIT-1 t11i5cl \.tLtl I \cTi - )IlL il I 'ILL1 i r li .K.I;l cmli1cii IcIII IITji| D,r I hkt |cl e d L u I iii I IITtl U I F t tI ) l 1 L i 414} [|§g'-I~I'v ill111 t\|) -\TA-[ i']] n L11111i_ 'tllt' HA] MM2i15Wtlk ;111 nC> 1 .;! l1c T 11i- 1L ' F l 11 I 11I 1 h1 t1 l i ['ijli )1 i i L plIc\i1iL v}.1i11 ... .:--Ikkl.. ;- /7- -AE 4; ---Ipq .. W Lii( il -CT . IlL .|riLe)tI11(j)c L\LL TI R * LI fUR -J \\ l ill I mIll .: iC1 iiIlI).I-I ( Ci'i LiIII1LCLUiIIV F\LFL LIiliiIRlII \h.FC [-112'[ : ;-, l si l l[IC l i l l;l l jI I'; lll 1ii d \ -t )1l '11 ~~~~~~~~l IIj['k1 P] Uit P|I-OC2 LI C. hC - )-- llt.' aL 11 5\ ) t 'i< 1 511 , h S% 1)0 ' t.'d t.. d1 1VCl LI SI (1 c ) I.i " t. C C\ ]>' l t 1 t1 I Iji - I.L 'I. , ;HI' C( c\ c i- 'i /cd M I p li I )f 3 Ii, . 01 Ie Ii. R-I , ;IIt ;, I;ei4llxt I5 vttlr^ l) v t ;. [1x t' ; \'i it \\. } t 1 T h II',, 4,' )LYl1 i:Lt,[ d ) 11d B 1VIRI1]C II p tco' . t) .S t tl '' 2 e 1111 t|1 \ I C' IIL, ort] or 11kx1c cd C (IP> \\ IV ( h I1 I Pt LIl1 LI. 35S 32p | lji tII . I<i cl r. a link between activated receptors and TGF-f3 TGF-13 receptors, we used a sequential immunoprecipit- provide downsteam effects of inhibition and e.g. growth ation approach, by which and TfR-II can be isolated TGF-,3, TPR-I increased ECM is Using a yeast in incomplete. a heteromeric complex and in non-complexed forms formation, interaction FKBP-12 et al., from a screen, (Wang cell extract (Wrana et al., 1994) (Figure IA). In two-hybrid and transferase (Kawabata et al., with ox agreement previous data (Chen and Weinberg, 1994) famesyl-protein 1995), have been isolated as molecules that interact with ligand-independent complex formation was 1995) observed In TRIP-1 has been isolated as a protein between and addition, TIR-I TfZR-II when both receptor types were TPR-I. with TrR-II using a similar method (Chen overexpressed in COS- 1 cells (Figure IB). Endogenously interacting et the functional of these significance produced TGF-f did not contribute significantly al., However, to com- 1995). interactions remains to be determined. plex formation as addition of anti-TGF-f3 neutralizing In the we describe that one of the major antibodies had no effect on complex formation. present study Over- sites in TfAR-I is Serl65. expression of receptors in Mv1Lu cells also phosphorylation promoted ligand-dependent Whereas the of Serl65 is not essential complex formation in the absence of ligand (data phosphorylation not for its mutation increases the ECM shown). signaling, and T3R-II in complex were found to TPR-I be TGF-P1 formation and inhibition, but decreases the apop- phosphorylated, growth as revealed by isolation of the complex totic induced TGF- 1 in mink lung epithelial by from [32P]orthophosphate-labeled response cells; addition of ligand cells. of Serl65 therefore Phosphorylation led to an increase in complex formation and to a (MvlLu) concomit- to modulate TGF-,B1 signaling. ant increase in phosphorylation appears of the receptors, as deter- mined by comparison with parallel immunoprecipitation of 35S-labeled receptors. Ligand addition led to an increase Results of the phosphorylation level of but not of TI3R-II TPR-I of T/3R-I and TI3R-ll (Figure IC). Phosphorylation TGF-1 a heteromeric complex of TfR-I signals through Consistent with previous reports (Wrana et al., 1994; and and Lodish, 1993; Derynck, 1994; (Lin Chen and Weinberg, T,BR-II 1995), we found that the non- et 1994; Yingling et al., 1995; ten Dijke Massague al., complexed overexpressed Tf3R-II was phosphorylated and et To determine the phosphorylation sites in al., 1996). that this phosphorylation was not dependent on the pres- 6232 TIR-1 Serl65 phosphorylation modulates signaling plexed with T,BR-I also gave a similar phosphopeptide map (data not shown). Phosphoamino acid analysis of the peptides shown in gray (Figure 2C; spots 2, 3, 4, 5, 6, 7 :r A and 9) revealed that these peptides were phosphorylated :-. only on serine residues (data not shown). The positions of the phosphorylated serine residues were determined by release of radioactivity upon Edman degradation; alignment with serines that are present in tryptic peptides, as predicted from the TfR-II cDNA sequence, indicated that Ser549 and Ser551 in the C-terminal tail of TfR-II ~~~~~~~~~~~I t (Figure 2C; spot 4), and Ser223, Ser226 and Ser227 in the juxtamembrane region of T3R-II (Figure 2C; spots 6 and 7), were phosphorylated. Two phosphorylation patterns .i *: I. in the juxtamembrane region were observed, either Ser223, a, Ser226 and Ser227 were phosphorylated (spot 7), or Ser223 and Ser227 (spot 6; see Figure 2G). Tryptic phosphopeptide maps of TfR-I phosphorylated in l'ivo in COS- 1 cells and isolated in a complex with -I ES .-iW.- T3R-II revealed 16 spots of different intensity. In contrast to what was found for ligand binding and complex TPR-II, - -,0 -.; .-- formation induced changes in the phosphopeptide maps c -. ap -- -I lim la.-C, '- r,-m of T3R-I. Whereas non-complexed was not phos- 0 TPR-I ..: 4) @. phorylated appreciably, TfR-I in complex with Tf3R-II on several sites (Figure 2D, E and was phosphorylated of five peptides (spots 1, 2, 3, 4 F). The phosphorylation upon stimulation with TGF-51. and 5) increased further tryptic phosphopeptides were subjected to The different acid analysis and Edman degradation; by phosphoamino .. IE I - ... {i the results with the predicted tryptic peptides comparing from the Tf3R-I cDNA sequence, the location of several Fig. 2. In vivo phosphorylation sites in TIR-I and T3R-11. Epitope- phosphorylated serine and threonine residues in the juxta- tagged T3R-I and were transiently expressed in COS-1 cells. TPR-II membrane region of TfR-I could be determined (Figure The complexed receptors were purified, digested with trypsin and high-voltage electrophoresis and thin-layer peptides were resolved by 2G). and analyzed by using a FujiX chromatography. Plates were exposed Edman on the release of 32P radioactivity upon Based The maps of complexed T,BR-II from Bio-Imager. phosphopeptide the peptide of the major broad migrating degradation, or cells treated with TGF-f1 (B) and a schematic non-treated cells (A) 1) could be aligned to the GS ligand-stimulated spot (spot of the map of T,BR-II (C) are shown, as presentation phosphopeptide This broad 1 consisted of multiple maps of TfR-I purified from non-treated domain of T3R-I. spot well as the phosphopeptide treated with TGF- and a schematic cells (D) or cells 1 (E) Phosphoamino acid ana- slightly overlapping components. of the maps of T3R-I (F). The presentation phosphopeptide that of spot 1 were phosphorylated lysis revealed peptides I and II receptors used for Edman phosphopeptide spots of type threonine residues not shown). both on serine and (data acid analysis are shown in gray. degradation and phosphoamino different of phosphorylation in We found three patterns shown small black squares. Sample application points are by illustration of the of TI3R-1 and TIR-II with i.e. of Thrl86, of Serl89 (G) Schematic sequences the GS domain, phosphorylation in i(v phosphorylation sites the juxtamembrane domains (JM) and of Serl87, Serl89 and and Serl91, Thrl85, ThrI86, domains (TM) and kinase domains (KD) of indicated. Transmembrane It remains to be determined if the Serl9l (Figure 2G). are indicated. Serine and threonine residues that are the receptors sites are present in one peptide multiple phosphorylation are indicated by bold lines. phosphorylated at different positions or if several peptides phosphorylated acid in one The phosphoamino analysis co-migrate spot. of another site of ligand-stimulated phosphorylation ence of ligand. In contrast, non-complexed T3R-I major on serine residues (Figure showed phosphorylation in transfected COS-l and MvlLu cells was (spot 2) expressed of a residue after The appearance phosphorylated not to be in most experiments. 3A). found phosphorylated Edman indicates in degradation sequencing in some experiments when Tf3R-1 was over- eight cycles However, in the of amino acid that Ser165 tryptic peptide consisting in COS-1 cells to very high levels, non-com- expressed was since this is the in a form residues 158-178 phosphorylated, T3R-I was found phosphorylated (data plexed residue residues downstream of a lysine only serine eight not shown). from the TrR-I cDNA as sequence or an arginine predicted data of Ser 165 as a The phosphoryl- in vivo sites in Tf3R-I (Figure 3B). alignment of phosphorylation Mapping the of site were confirmed immunoprecipitation ation by and T,JR-II from this with VPN antibodies, which spot of T3R-II phosphoryl- radioactivity phosphopeptide mapping Tryptic to amino acid are made corresponding in COS- 1 cells and isolated in a complex against peptide ated in vivo in T3R-I 158-178 3C). residues (Figure revealed 17 of different intensities with TIR-I spots to at Serl65; was found We found no in the distribution TIR-II phosphorylate 2A and change TP3R-I (Figure B). or complexed wild-type in the absence when baculovirally expressed T3R-II-derived (Figure of phosphopeptides TfR-1 were and to subjected of not com- kinase-inactive purified or 2B) ligand; T,R-II 2A) presence (Figure 6233 S.Souchelnytskyi et aL fected with and mutants in wild-type receptor which TPR-I Serl65 was to alanine changed (T,3R-I/S165A), glutamic | 7 ,e.-1 acid or acid (T,R-I/S165E) aspartic (TPR-I/S165D) with to residues, formation with T,1R-II, respect complex overall intensities of (in vivo and in vitro) phosphorylation or in their abilities to bind (data not shown). ligand of [32P]Orthophosphate-labeled with complexes TPR-II T,BR-I or wild-type mutants, formed in the TPR-I/S165 or absence of were presence the ligand, purified using method and B sequential precipitation to subjected tryptic followed digestion phosphopeptide by mapping. Figure 4C-F shows that to a 2, with spot corresponding peptide was not seen in phosphorylated Serl65, the phospho- of the mutant. This peptide map was spot TfPR-I/S165A 5t:i also not seen in the derived from the map T,3R-I/S165E mutant not The low (data shown). detected intensity spot in this area of the 165A mutant be a T,3R-IIS may co- from another of the migrating peptide part of receptor, _ 1'iUhhIIIIbn which the is too low to be detected ~~~~Numunt intensity radio- 1 by 4 7 a 1 l t 6 9 3 4 1f cvc.!r-Z V P N E E D P L chemical We S D R P F... also found that sequencing. 4, which spot in the of after C appeared maps wild-type T,3R-I ligand stimulation 4C and was detected in (Figure D), the maps of the 165A mutant even in the absence of TPR-I/S 4E and ligand (Figure the low amount F); however, of in 4 radioactivity determination of peptide precluded the corresponding phosphorylation site. Other differences are also seen between the of maps wild-type T,R-I (Figure 4C and and the 165A D) 4E and or TfR-I/S (Figure F) 165E not mutants. TfR-I/S (data shown) in a However, series of four different in experiments which the separation conditions were these other varied, differences were not found to be reproducible. Mutation of Ser165 in modulates TPR-1 TGF-fi1 signaling Fig. 3. Identification of Serl65 as a site in To determine the role of phosphorylation TJR-I. Serl65 phosphorylation in TGF-, (A) The peptide to corresponding spot 2 of TfR-I was subjected to we signal transduction, tested TGF-f31 responses in a cell phosphoamino acid The analysis. migration of phosphorylated serine line a functional lacking endogenous T,3R-I (clone R4.2 (S), threonine (T) and tyrosine (Y), used as is standards, shown. of MvlLu Laiho (B) The same after stable peptide was also cells; etal., 1990), transfection subjected to Edman degradation; the with elution positions of 32P-labeled or mutant amino acids are shown and wild-type T,R-I. In mutants aligned to T,R-I Serl65 the of sequence a putative tryptic peptide (VPN), the only one with a was with other replaced amino acid residues (TPR-I/ serine eight residues downstream of a lysine or an arginine residue. T,R-I/S 165E or 165A, or with a 165D) kinase- (C) The TOR-I/S peptide corresponding to spot 2 was precipitated by VPN inactive variant of TfR-I (T3R-I/K232R) under antibodies, directed against the VPN peptide transcrip- (residues 158-178 in tional control TfR-I), spotted onto a of the inducible thin-layer chromatography human metallothionein IIA plate and quantified by using a FujiX Bio-Imager. Radioactivities, All immunoprecipitated promoter. transfectants that were used expressed T,3R-I from of peptide spot 2 by VPN antibodies (1), VPN antibodies in the or TfR-I mutants to the same level (Figure 5A). We found presence of excess VPN peptide (2), and non-immune serum (3), and that phosphorylation of is not Serl65 essential for the their quantitations are shown. transduction of the effect of TGF-f1 on cell growth, since cells transfected with 165A or Tf.R-hIS 165E TOR-I/S in vitro kinase assay followed by phosphopeptide mapping, responded to TGF-j1 addition with growth inhibition Serl65 was phosphorylated in both cases (Figure 4A and (Figure SB). In the fact, TGF-f81-induced inhibition of B). That Serl65 was phosphorylated was confirmed by cell proliferation was more pronounced in cells transfected the observation that radioactivity from the predicted spot with and especially Tf3R-IIS165E, than in TOR-I/S165A, could be immunoprecipitated with VPN antibodies, but cells with wild-type values of 0.41, 0.06 TPR-I (ED50 and not by pre-immune serum. Phosphorylation of Serl65 0.54 ng/ml, respectively). in kinase-inactive T3R-I (Figure 4B) indicates that the TGF-,1 stimulated the biosynthesis of plasminogen phosphorylation of this residue is not dependent on T,3R-I activator inhibitor-i (PAI-1) and fibronectin and their kinase activity, and thus most likely is phosphorylated in deposition ECM of wild-type T3R-I transfected cells directly by T,R-II. The (Figure 6). level of PAI-1 induction was higher in In order to determine the functional importance of the cells expressing Tf3R-I/S165A or than in T,PR-I/S165E phosphorylation of Serl65 in Tf3R-I, this residue was cells expressing wild-type Tf3R-I (Figure 6A and B). mutated to alanine, glutamic acid or aspartic acid residues. These differences were detected already after 4-5 h of No differences were 1 found between COS- cells trans- TGF-f1 treatment (data not shown), but the effect was 6234 TIR-1 Serl65 phosphorylation modulates signaling -_.1\\j4 o. AV _F. r-i ,1 R *. .e, .T' TGF-:!- Fig. 4. Phosphopeptide maps of complexed mutated and wild-type TfR-I. Baculovirally expressed wild-type (A) and kinase-inactive (B) TIR-I, complexed with T3R-II, were purified, subjected to in vitro kinase assay, digested with trypsin and phosphopeptide maps were obtained. Complexes of T,R-II with wild-type T3R-I (C and D) or with TIR-I/S165A (E and F), phosphorylated in vivo were purified from COS cells, digested by trypsin and subjected to two-dimensional analysis. The arrows show the migration position of the Serl65-containing peptide. The arrowheads show the migration positions of a peptide, the phosphorylation of which is dependent on mutation of Serl65. Cells were treated (A, B, D and F) or not (C and E) with 10 ng/ml of TGF-431. Sample application points are marked with open arrowheads. shape of the structures is not regular and it is not The more pronounced after 16-18 h of ligand treatment (Figure to be derived from a single cell (Figure 7G). The likely 6A and B). A similar pattern of response, albeit much of these structures may be related to the effect appearence weaker, was found for stimulation of newly TGF-P1 mutants on ECM formation. of Serl65 synthesized fibronectin, incorporated in ECM (Figure 6C The effects of TGF-, 1 on growth inhibition, fibronectin and D); the cells with Serl65 of mutated incorpora- TPR-1 and deposition in ECM, and PAI- I induction, biosynthesis ted 1.8-2.0 times more fibronectin into ECM than did that the mutations of Serl65 to alanine, and suggest wild-type T,R-I cells. In this assay, no difference between to glutamic acid, give a higher signal especially TPR-I T,R-I/S 165A and T,R-I/S 165E cells was observed. TGF- transducing potential in these assays. However, no differ- had no effect on ECM formation in non-transfected P1 was found between wild-type T,BR-1, TIR-I/S165A ence cells or cells (Figure 6 and data not shown). TPR-I/K232R and 165E cells in the 1 induction of lucifer- TGF-P To investigate the effect of the Serl65 mutations in TPR-I TPR-I/S ase expression, controlled by PAI-I promoter and TPA- on cell behavior in dense cultures, we cultured cells elements in transfected MvlLu and COS-1 responsive transfected with inducible vectors encoding wild-type 8), and in TGF-,Bl-mediated inhibition of cells (Figure T,R-I/S165D or TPR-IIS165A, TPR-I/K232R, TPR-I, as measured by the migration of cells into cell migration 165E for 3 days to confluency. Thereafter, cells TPR-I/S area in dense cultures from which cells have a defined were cultured with or without ZnCl2 to induce the expres- scraped off (data not shown). been sion of the receptors, and with various concentrations of of wild-type T,R-1 in R4.2 cells restored Expression TGF- 1 for 7-10 days with fresh media replacement every (Figure 9, lanes 13-16). Interest- apoptosis ligand-induced day. No differences were observed for the wild-type TGF-lI-stimulated accumulation of apoptotic the ingly, and mutant receptor-transfected cells in the absence of DNA ladder formation occurred less efficiently and bodies induction of expression (data not shown). Upon TPR-I cells and especially for T,R-I/S165E for induction of receptor expression, TGF-P1 showed no TPR-I/S165A when with wild-type cells, and was compared TPR-1 cells, effect on cells with wild-type TPR-I (Figure 7A and B), in cells (Figure 9, lanes 1- at all not seen TPR-I/K232R T,R-I/S165A (Figure 7C and D) or TPR-I/K232R cells mutation of Serl65 to alanine or glutamic the 12). Thus, (data not shown). However, cells with TPR-I/S165E weakens the effect of on apoptosis, residues TGF-PI acid (Figure 7E-G) or (data not shown) showed TPR-I/S165D effect on growth inhibition and ECM forma- the whereas TGF-,B 1-dependent cell overgrowth and formation of three- is strengthened. tion dimensional structures after 4-5 days in dense cultures. 6235 et aL S.Souchelnytskyi Discussion exerts its cellular TGF-P3 signals of through cooperation two related serine/threonine kinase distantly K'; receptors, .-~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~...... and and TPR-I TO3R-11 (Lin Lodish, 1993; Derynck, 1994; et al., et ten TGF- 1: Massague' 1994; al., 1995; ++ + Yingling Dijke et al., to the cDNA of T RI 232R 1996). Subsequent both K T F.-I S 1 65E T. R- Si I jA cloning id -tYPe our of the mechanism receptor types, of understanding transduction of has increased signal TGF-03 dramatically. The model for activation is B proposed distinct receptor from that of other classes of in that cytokine receptors occurs in a manner with phosphorylation sequential Tf3R-1 as a substrate for et acting Tj3R-11 (Wrana al., 1994; 4' 0 Wieser et al., little is known I 1995). However, about yet 2,, the sites that are in phosphorylation present Tf3R-1 and w. T. iI C: and their in activation TO3R-11 significance receptor and transduction. In the signal we present the study, report identification of several in vivo sites phosphorylation in and 1 We found that Serl.65 Tf3R-1 in is TfOR-11. Tf3R-1 one of the sites of major ligand-dependent in phosphorylation K 232R and that it has an role in TPR-1 important signal regulation of TGF-f31. In with agreement we observed previous reports, that TGF- of and overexpression induced TOR-1 TO3R-11 ligand- of in cells heteromeric Fig. Analysis transfected with independent which TGF-01 signaling TI3R-1 complex formation, has and mutants. of in TOR-1 (A) Expression been stably transfected attributed to an intrinsic TPR-1 of and Tj3R-1 affinity TOR-11 cells. R4.2 cells were transfected with stably different constructs of for each other et (Ventura Chen and al., 1994; in vector. Weinberg, Chosen cell were TP3R-I pMEP4 35S-labeled, pools treated addition led 1995). to a further increase with 50 ltM and/or TGF-13I and TGF-01 in ZnCl2 (10 Tf3R-1 were ng/ml), with anti-HA immunoprecipitated antibodies. complex formation, the probably Immunoprecipitates were by stabilizing interaction resolved by followed visualization SDS-gel electrophoresis, between the in by and the In receptors the complex. complex, quantitation using of FujiX Bio-Imager. (B) Analysis both growth and Tf3R-1 were found to be TJOR-11 inhibition in to TGF-j31. R4.2 phosphorylated, response cells transfected with stably whereas was found non-complexed not to and mutants be wild-type were to a TPR-1 TfOR-1 TP3R-I subjected [3H]thymidine in the unless or absence of different phosphorylated it was assay presence concentrations to of overexpressed very high TGF-f31. The of percentage inhibition was calculated levels. In growth contrast, Tf3R-11 was found to be by phosphorylated [3 decrease relative to the measuring H]thymidine incorporation into also in form. These non-complexed observations are cells in the absence of P<0.001. TGF-P1. consistent with data previous that Tj3R-II showing forms a homodimer and (Chen Henis et Derynck, 1994; al., A~~~~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~~~~~~A IdIy 'T' Fi- A T R CF- f., T R K ~ T'~~~~ ~ ~ ~ ~ ~ -- -1 R4 y w R 2 Id- I" T fj3A I I'i S 165E 2 2. IZ) .) S165A L) c - S165E LLJ 'Z LO) c :t :3 0 r - -a z --I -; 7 w .; 0 -0 = -~~~~~~~ T. D- 1 4) -~~W. 5A SiSS 'O ;., .S2 W. T. Z.0 .- ei !. ~~~~~K232R 2 R 4.2 LL TGF- (nrg,;mII TGF- (ng/ml) Mutation of Serl65 Fig. in leads to an increase Tf3R-1 of extracellular matrix TGF-j3I-stimulated formation. Cells transfected with or stably mutant wild- type constructs were treated Tf3R-1 with 50 j.M to induce ZnCl2 of expression and then receptors treated with for 16-18 were TGF-f3I h. 35S-labeled with 'ProMix' Cells for the last 2 h of incubation and in proteins extracellular matrix were resolved by visualized SDS-gel and electrophoresis, quantitated a using FujiX Bio-Imager. TGF-f31 induction of and PAI-I1 biosynthesis (A) fibronectin incorporated into matrix extracellular (C) cells by Tf3R-I expressing or Tf3R-I mutants wild-type are shown; of quantitation data is presented in and (B) (D), respectively. 6236 TIR-1 Ser165 phosphorylation modulates signaling * wild-type A B 100 r- K232R [1H C.) S165A 7- Z S165E CD (A L. L)- c -j TGF- (ng/ml) P1 e. Fig. 8. Effect of TfR-I Serl65 mutations on gene expression. R4.2 E F cells, stably transfected with wild-type TjR-1 or T3R-I mutants, were ..: I. transiently transfected with p3TP-Lux plasmid, and receptor expression was induced by treatment with ZnCl,. Luciferase activity was determined after stimulation with different concentrations of r, as described in Materials and methods (representative of five TGF-p1, experiments). JI t. -, r, . - r. T n T F-, Sz j--I A F ,S ;4 p 7nl,c T CF.- 1 |~~~~~~~~~~~~~~~~~~~~~ ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~.. < < K''. r.~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~. .s __. .. -- l~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~. 7. TGF-f31 induces cell overgrowth in cells. Cells Fig. TPR-I/S165E stably transfected with wild-type T,BR-I (A and B), (C TPR-I/S165A and D) or (E, F and G) were kept in dense cultures; the TPR-I/S165E aI medium was changed every day in the absence (A, C and E) or (B. D, F and G) of TGF-.1 (5 ng/ml) and 50 ,uM ZnCIl. presence Photographs were taken after 5 days of culturing. Representative from four experiments are shown. (G) represents a picture photographs 9. Mutation of Serl65 in T3R-1 decreases the ability of TGF-,B1 Fig. of one of the structures, formed by cell overgrowth, at higher to stimulate apoptosis. R4.2 cells expressing wild-type TIR-I (lanes magnification. 13-16), T,BR-I/K232R (lanes 9-12), TPR-I/S165A (lanes 5-6) or TIR-I/S165E (lanes 1-4) were treated or not with ZnCl, to induce receptor expression. Cells were then stimulated or not with 5 ng/ml 1994) and is a constitutively active kinase in the absence TGF-f1. DNA prepared from apoptotic bodies was analyzed by of ligand (Wrana et al., 1994). Identical tryptic phospho- agarose gel electrophoresis and DNA was visualized by ethidium peptide maps were found for T3R-II whether it was Migration of DNA molecular weight standards is bromide staining. in lane 17. Essentially the same results were obtained in six shown stimulated or not with ligand, and whether it occurred in and in experiments with two different cell different experiments complex with TIR-I or not. Phosphoamino acid analysis preparations. of isolated tryptic fragments from TIR-II showed phos- phorylation only on serine residues, consistent with pre- of the kinase domain of T3R-I by charge- in the activation vious findings (Wrana et al., 1994). In T3R-II, Ser549 conformational changes rather than providing induced and Ser55 1, located in the C-terminal tail, were identified sites for downstream elements. Consistent with docking as phosphorylation sites. Wieser et al. (1993) found that notion, we found three patterns of GS domain phos- this the T,BR-II C-terminal tail was dispensable for signal in vivo, which may correspond to different phorylations as a TfR-II mutant without this domain, and transduction, kinase activation. Phosphorylation in the levels of T3R-I both Ser549 and Ser551, was capable of thus lacking domain of T3R-II also occurs in multiple juxtamembrane TGF-j sensitivity to cells lacking TfR-II (DR restoring also have a role in activation of the kinase. and may patterns cells) after transfection (Wieser et al., 1993). mutant treatment of cells not only led to an increase in Ligand Phosphorylation sites were also identified in the juxta- through increased complex forma- T3R-I phosphorylation domain of both T3R-I and In T3R-II, membrane TPR-II. also induced qualitative changes in the tryptic but tion, was detected at Ser223, Ser226 and phosphorylation pattern of T3R-I (Figure 2D-F). The latter phosphopeptide Ser227, which are located a similar distance from the be of importance for full activation particular changes may kinase domain as Thrl86, Serl87, Serl89 and Serl91 that the serine and threonine residues in For T3R-I, of T,R-I. were identified as phosphorylation sites in the GS domain as well as Serl65, were identified as GS domain, the of (Figure 2G). The number of phosphorylated TPR-I in vivo phosphorylation sites ligand-dependent major residues, as opposed to their particular position in the GS mechanism behind this change and the The 2G). (Figure domain, appears to be of importance for T3R-I signal TGF-I receptor signaling remain to be for significance transduction (Franzen et al., 1995; Wieser et al., 1995). mutation of Serl65 also led to In addition, elucidated. This suggests that the phosphorylated residues have a role 6237 S.Souchelnytskyi et aL of effects of mutants on PAI-I appearence Serl65 Table I. protein Comparison of the effects biological of and wild-type Serl65 and the absence of mutants on TfR-I differences in signal transduction synthesis luciferase from the construct expression p3TP-Lux be reporter may Assay TfR-I differences in explained by respective gene promoters. The fact that Wild-type mutations affects Serl65Ala Serl65 Serl65Glu/ signaling, TGF-f1 without Serl65Asp kinase makes it affecting TfR-I activity, unlikely that these mutations have led to of the perturbation Growth inhibition + + + ++ conformation of and the TPR-I supports that ECM production possibility + + ++ of (fibronectin, PAI- phosphorylation-dephosphorylation has a 1) Serl65 regu- role in Cell overgrowth the interaction with - - latory downstream + signaling Induction of gene + expression + molecules. Cell migration + + is + multifunctional and its effects on TGF-0 depend Apoptosis + (+) (+) the cellular context, a need for coordinate implicating of its actions. Response: -, none; (+), Control of weak; +, regulation action intermediate; ++, strong; +++, TGF-13 may very strong. exist at different levels in the transduction signal pathway, from of to regulation TGF-,3 modulation gene expression of Our TGF-f-induced transcriptional results responses. ligand-independent appearance of 4 in the that spot Serl65 tryptic suggest an phosphorylation plays important phosphopeptide of role maps T,BR-I, which for in this wild-type T,R-I process by modulating multiple biological appeared after only treatment. effects ligand Whether this of TGF-, 1. change reflects an influence of Serl65 on the T,R-I phosphoryl- ation pattern or is due to structural as a result changes of Materials the and introduced mutation is methods not known. The Serl65-dependent changes in the phosphorylation lines Cell patterns of TfAR-I (Figure and 4C-F) the COS-1 and strong ligand cells were obtained from MvlLu American Culture Type dependence of Collection. phosphorylation of Serl65 cells that lack increased our MvlLu active cells; functionally (R4.2 TfR-I Laiho et interest in al., were elucidating its Dr involvement in the 1990), Cells were regulation of provided by cultured J.Massagu6. in Dulbecco's modified medium TGF-,81 signaling. We mutated Eagle's (DMEM; with Serl65 of TfR-I Gibco-BRL) to alanine, 10% fetal bovine serum (FBS; of Gibco-BRL), 100 U/ml and glutamic acid or aspartic acid penicillin residues, thereby 50 abrogating of H5 gg/ml streptomycin. cells were Spodopterafrugiperda cultured the phosphorylation of this residue. in Mutations IPL-41 medium to glutamic 10% FBS (JRH) containing with supplemented acid or aspartic acid residues were broth. made with the tryptose intention phosphate of mimicking the negative of the charge phosphate group. Constructs and We transfection studied the effect cell of Serl65 mutations in T,BR-I on cDNAs for with and growth TO3R-I tagged hemagglutinin inhibition, epitope gene T,BR-II expression, ECM production, with cloned in the tagged vector et (His)6, pCMV-5 (Wrana al., contact 1994), inhibition, cell migration and apoptosis were obtained from induced Dr Mutant forms of J.Massague. were receptors by TGF-,1 using MvlLu cells. In PCR with none of these generated assays by mutagenic DNAs as primers using receptor did All mutant we find an constructs were abrogation of the template. over the TGF-f1 sequenced that response or regions were with the exchanged COS-1 cells ligand-independent in parental receptor were signal any of the plasmids. Serl65 mutations. transfected with of vector transiently 1 T,BR-I and 7 of However, we jg observed that ,ug Serl65 mutations can modulate vector TfR-II 40 mm dish the calcium per by phosphate precipitation method the effects of TGF-31 in a positive or the MBS negative mammalian way, using transfection kit (Stratagene), the following depending on the manufacturer's particular For stable response that transfection in was measured protocol. R4.2 all cells, tagged cDNAs were (Table I). cloned in The receptor mutation of Serl65 under led to pMEP4 (Invitrogen) higher signaling transcriptional control of the metallothionein potential Transfection was done for TGF-f31 promoter. in the growth inhibition assay (Figure using the calcium method. phosphate Selection of precipitation transfectants 5B) and in ECM production (Figure 6). In was addition, in in the of 100 performed U/ml of presence We hygromycin. obtained cells with TfR-IS 165E or four cell but not transfected with in cells pool cultures, TPR-I/S165D, wild-type TOR-I, TjR-I/K232R, with T,BR-I/S165A wild-type and T,R-I, TGF-j1 was 165E, and 60 able to induce TPR-I/S individual cell respectively, clones, 12 for above-mentioned overgrowth (Figure every variant and 12 7). A TfR-I for certain threshold appears T,R-IS165D. neces- All these cultures and clones were tested for sary for induction of this response, because in T,3R-I/S 165A cells TPR-I TGF- treatment. The one expression upon cell ZnCl2 culture pool and two f I showed no effect. The observed cell clones for overgrowth may mutant with the same every level of the receptor expression be related to the were used in stimulation of substrate-independent biological assays. proliferation of NRK cells by TGF-f (Roberts et al., Purification of 1981), in which induction TGF-,8 of ECM formation plays receptors an Cells were washed three times with important role. phosphate-buffered saline (PBS) and in buffer 20 mM lysed lysis [LB; In Tris-HCl, pH 7.4, 150 mM contrast, we found a decrease NaCl, in TGF-f31-stimulated 0.5% Triton 50 mM X-100, 10 mM NaF, 1 Na4P207, mM apoptosis in Na3VO4, cells with a Serl65 mutant 1 mM of TfR-I compared phenylmethylsulfonyl fluoride (PMSF), 100 U/ml aprotinin]. with wild-type and TfR-I heteromeric cells (Figure 9). TI3R-I TtR-II and Thus, complex non-complexed T,BR-I (His)6- T,BR-II were phosphorylation at tagged from Serl65 precipitated clarified appears extract with important for signal Ni2+-NTA agarose (Qiagen). Non-precipitated regulation non-complexed HA-tagged of the activated T,BR-I TGF-, receptor. We did not was subsequently immunoprecipitated from the unadsorbed detect fraction any differences between wild-type anti-HA TfBR-I and antibodies using (12CA5, or Babco) anti-T,BR-I (VPN) anti- TfR-I/Serl65 mutants bodies et in their cell migration and (Franzen induction al., Immunocomplexes 1993). were collected on of protein the A-Sepharose beads gene EC expression (Immunosorb; Diagnostics using construct AB). Ni p3TP-Lux +- reporter NTA agarose beads which were washed three times with 25 mM that imidazole (Figure 8), the in of suggests these signaling LB and bound receptors were eluted in 250 effects is mM imidazole in LB. not The Serl65 regulated by The eluate was subjected to with phosphorylation. immunoprecipitation anti-HA antibodies to 6238 T3R-1 Ser165 phosphorylation modulates signaling enrich for the TfR-I and T,R-II heteromeric complex and with rabbit cells were washed with PBS to remove calcium extensively phosphate peptide antisera against TfR-II (DRL) (Yamashita et al., 1994a) to the cells were incubated in DMEM with 10% precipitates. Subsequently, enrich for the non-complexed TPR-II. FBS for 16-20 h. the was induced Thereafter, receptor expression by treatment of the cells with 50 tM in DMEM with ZnCl2 supplemented labeling of cells, tryptic 132PlOrthophosphate 0.1% FBS for after which TGF-31 was added. Luciferase h, activity phosphopeptide mapping and two-dimensional in the cell was measured after 22-24 h the luciferase lysate using assay phosphoamino acid analysis to the manufacturer's system (Promega Biotec), according protocol using Cells were labeled in phosphate-free medium supplemented with 0.5% an LKB Luminometer (LKB-Bromma). pH 7.2, and 1.0 dialyzed FBS, 15 mM HEPES, of [32P]ortho- mCi/ml phosphate for 3 h. During the last 15 min of incubation, TGF-,B1 was added as mentioned in the figure legends. Cells were lysed and receptors Cell overgrowth assay isolated as described above, separated by SDS-gel electrophoresis and Cells transfected with TfR-I/S165A, stably wild-type T,BR-I, TPR-I/ electrotransferred to a nitrocellulose filter (Hybond C-extra; Amersham). 165D or were cultured for 3 to reach S165E, TfR-I/K232R T,R-I/S days For tryptic phosphopeptide mapping, receptor bands were localized by in the or absence of 50 lM and, thereafter, confluency presence ZnCl2 exposure on a FujiX Bio-Imager (Fuji), excised from the filter and for 7-10 Cells and TGF-f1. The medium was day days. changed every digested in situ with trypsin (modified sequencing grade; Promega). were examined regularly by microscope. Two-dimensional phosphopeptide mapping was done using the Hunter thin-layer electrophoresis apparatus (HTLE-7000; CBS Scientific), essen- tially as described by Boyle et al. (1991). First dimension electrophoresis Apoptosis assay Cell induction of and treatment with was performed in pH 1.9 buffer (formic acid/glacial acetic acid/double- TGF-P1 seeding, receptor expression done as defined for the inhibition that cells were distilled water; 44:156:1800 by vol.) for 35 min at 2000 and second V, growth assay, except in medium with 0.1% FBS for 13 h before were incubated dimension ascending thin-layer chromotography in isobutyric acid buffer harvesting bodies were treated with 0.5 the detached material. (isobutyric acid/n-butanoUpyridine/glacial acetic acid/double-distilled mg/ml Apoptotic 0.2 RNase for K for 1 h at and water; 1250:38:96:58:558 by vol.). After exposure, phosphopeptides (Sigma) proteinase (Sigma) 50°C mg/ml at to remove and were eluted from the plates in pH 1.9 buffer and lyophilyzed. The 20 RNA, min Apoptotic 60°C protein respectively. 2% and ethidium fractions were then subjected to two-dimensional phosphoamino acid DNA was analyzed by agarose gel electrophoresis analysis and, in parallel, automated Edman degradation. For Edman bromide staining. degradation, phosphopeptides were coupled to Sequelon-AA membranes (Millipore) according to the manufacturer's instructions and sequenced on an Applied Biosystems Gas Phase Sequencer Model 470A. Released phenylthiohydantoin amino acid derivatives from each cycle were spotted Acknowledgements onto thin-layer chromatography plates. The radioactivity in each spot for excellent technical Anita and Susanne We thank Grimsby Moren was quantitated by exposure to a FujiX Bio-Imager. for Christer for Ulla Wernstedt assistance, peptide synthesis, Engstrom For phosphopeptide mapping, H5 cells at -80% confluency were co- and radiochemical Lars Ronnstrand sequencing, oligonucleotide synthesis infected with and TfR-II baculoviruses at an m.o.i. of 10, and 5 Ohashi for TPR-I for with Hideya providing help phosphopeptide mapping, p.f.u./cell. Cells were used 48 h post-infection. Purification of receptors for R mutant MvlLu recombinant human TGF-41 and Joan Massague was done as described above, purified receptors were subjected to T3R-I and baculoviruses and recombinant p3TP-Lux cells, TPR-II in vitro kinase assay as described (Franzen et al., 1995), and tryptic plasmid. phosphopeptide mapping was done. Growth inhibition assay X well in 24-well Cells were seeded at a density of 1 104 cells per plates References in DMEM with 10% FBS. was stimulated Receptor expression by incubation of the cells in medium with 0.1% FBS with 50 ltM ZnCI2 Gustafson, He,W.W., Howe,D.J., Wang,T., Bassing,C.H., Yingling,J.M., for 5 h. Subsequently, medium containing 3% FBS with or without 50 tM and (1994) transforming Shah,P., Donahoe,P.K. M.L., Wang,X.-F. ZnCl2 was added and cells were with different concentrations of incubated f that to activate factor I gene expression. signals growth type receptor TGF-,B1 for 22-24 h; during the last were labeled with 1 2 h cells ,tCi/ml 87-89. 263, Science, [3H]thymidine (Amersham Corp.). the cells were fixed in Thereafter, der and Hunter,T. Van (1991) Geer,P. Phosphopeptide Boyle,W.J., 5% ice-cold trichloroacetic acid (TCA) for 20 washed three times min, acid two-dimensional and by analysis mapping phosphoamino with TCA, and solubilized 1 M NaOH. The cell extract was with cellulose Methods 201, on Enzymol., plates. separation thin-layer neutralized with 1 M and 3H was measured in a HCI radioactivity liquid 110-149. scintillation ,B-counter using Ecoscint (National Diagnostics). of and (1995) Zentella,A. Disruption Carcamo,J., Massague,J. factor a mutation that prevents Extracellular matrix formation assay signaling by transforming growth ,B within the Mol. Cell. Biol., Cells were seeded in 6-well plates at a of 5X 104 cells well. receptor complex. density per transphosphorylation 1573-1581. After 18-24 h, the medium was changed to DMEM supplemented with 15, Biochemical evidence for the and 0.1% FBS, with or without 50 .tM After 5 TGF-f31 was added (1995) h, Chen,F. Weinberg,R.A. ZnCl2. of and growth to the cells and incubation prolonged for 16-18 h. To label the transforming newly transphosphorylation autophosphorylation Proc. Acad. Sci. 92, 1565-1569. kinases. USA, synthesized proteins, the cells were incubated in methionine-free medium factor Natl receptor Homomeric interactions between MCDB (SVA, Sweden) with 50,Ci/ml of mixture 'ProMix' and 35S-labeling (1994) Chen,R.-H. Derynck,R. J. Biol. Chem., 269, h of treatment. of (Amersham Corp.) during the last 3 TGF-41 II Aliquots receptors. growth type transforming factor-5 the medium (100 or secreted proteins bound to 22868-22874. gelatin-Sepharose gl) For ECM and (Pharmacia) were analyzed by SDS-gel electrophoresis. Derynck,R. Maruoka,E.M., Choy,L. Miettinen,P.J., Chen,R.-H., on ice once in PBS, three that is associated with and isolation, the cells were removed by washing A WD-domain protein (1995) 548-552. times in 10 mM Tris-HCl, pH 8.0, 0.5% sodium and Nature, 377, deoxycholate, the II receptor. phosphorylated by type TGF-,B 1 mM PMSF, twice in 20 mM Tris-HCl, 8.0, and once in PBS. Trends Biochem. pH signaling. TGF-f-receptor-mediated (1994) Derynck,R. ECM off and extracted into SDS buffer proteins were scraped sample 548-553. 19, Sci., Secreted and ECM containing 10 mM dithiothreitol. proteins proteins Heldin,C.-H. Yamashita,H., Schulz,P., ten Ichijo,H., Franzen,P., Dijke,P., were analyzed followed by SDS-gel electrophoresis, by fluorography of a I that forms receptor and type (1993) Cloning Miyazono,K. TGF,B using Amplify The were and (Amersham Corp.). gels exposed quantified Cell, 75, with the TGFf receptor. a heteromeric type II complex using a FujiX PAI-i was identified as a 45 kDa in Bio-Imager. protein 681-692. the ECM et Fibronectin was identified as a fraction (Laiho al., 1991). The GS domain of and (1995) Heldin,C.-H. Miyazono,K. Franzen,P., 230 kDa protein (Laiho et and was both in the soluble al., 1991) present I is in factor-f signal receptor important the type transforming growth fraction and in the ECM fraction. Res. Commun., 207, 682-689. Biochem. transduction. Biophys. and The Lodish,H.F. (1994) types II Moustakas,A., Lin,H.Y. Transcriptional response assay Henis,Y.I., form homo-oligomers. and III receptors Stable transfectants of TfiR-I were transiently transfected with p3TP- transforming growth factor-,B 139-154. J. Cell Biol., 126, Lux (Carcamo et al., 1995), as described above. The following day, the 6239 S.Souchelnytskyi et aL Kawabata,M., ImamuraT., Miyazono,K., Engel,M.E. and Moses,H.L. (1995) Interaction of the transforming growth I factor-5 type receptor with farnesyl-protein transferase-ax. J. Biol. Chem., 270, 29628-29631. Laiho,M., Weis,F.M.B. and Concomitant loss Massague,J. (1990) of transforming growth factor (TGF)-f I and II in receptor types TGF- f3-resistant cell mutants both implicates in receptor types signal transduction. J. Biol. Chem., 265, 18518-18524. Laiho,M., Ronnstrand,L., Heino,J., DeCaprio,J.A., Ludlow,J.W., Livingston,D.M. and Control Massague,J. (1991) of JunB and extracellular matrix protein expression by transforming growth factor- is of independent SV40 T antigen-sensitive PI growth-inhibitory events. Mol. Cell. Biol., 11, 972-978. Lin,H.Y. and Lodish,H.F. (1993) for the Receptors TGF-,B superfamily. Trends Cell Biol., 3, 14-19. Lin,H.Y., Wang,X.-F., and Ng-Eaton,E., Weinberg,R.A. Lodish,H.F. (1992) Expression of cloning the II a TGF-,B type receptor, functional transmembrane serine/threonine kinase. Cell, 68, 775-785. Massagu6,J., Cheifetz,S., Boyd,F.T. and Andres,J.L. (1990) TGF-f receptors and binding proteoglycans: recent progress TGF-P in identifying their functional Y properties. Annii. N. Acad. Sci., 593,59-72. Massague,J., Attisano,L. and Wrana,J.L. (1994) The TGF-f family and its composite receptors. Trends Cell Biol., 4, 172-178. Moses,H.L., Yang,E.Y. and Pietenpol,J.A. (1990) TGF-f stimulation and inhibition of cell proliferation: new mechanistic insights. Cell, 63, 245-247. Roberts,A.B. and Sporn,M.B. (1990) The transforming factor- growth In Sporn,M.B. and Roberts,A.B. (eds), Peptide Groitth Ps. Factors and Their Receptors., Part l. Springer-Verlag, Berlin, pp. 419-472. Roberts,A.B., Anzano,M.A., Lamb,L.C., Smith,J.M. and Sporn,M.B. (1981) New class of transforming growth factors potentiated by epidermal growth factor. Proc. Natl Acad. Sci. USA, 78, 5339-5343. Saitoh,M., Nishitoh,H., Amagasa,T., Miyazono,K., Takagi,M and Ichijo,H. (1996) Identification of important regions in the cytoplasmic juxtamembrane domain of I type receptor that separate signaling of pathways transforming growth J. Biol. Chem., 271, factor-P3. 2769-2775. ten Dijke,P., Miyazono,K. and Heldin,C.-H. (1996) Signaling via hetero- oligomeric complex of type I and II type serine/threonine kinase receptors. Clirr Opin. Cell Biol., 8, 139-145. Ventura,F., Doody,J., Liu,F., Wrana,J.L. and Massagu6,J. (1994) Reconstitution and transphosphorylation of TGF-f receptor complexes. EMBO J., 13, 5581-5589. Wang,T., Donahoe,P.K. and Zervos,A.S. (1994) Specific interaction of type I receptors of the TGF-f family with the immunophilin FKBP- Science, 265, 674-676. Weiss-Garcia,F. and Massagu6,J. (1996) Complementation between kinase-defective and activation-defective TGF-f receptors reveals a novel form of receptor cooperativity. EMBO J., 15, 276-289. Wieser,R., Attisano,L., Wrana,J.L. and Massague,J. (1993) Signalino activity of transforming growth II factor type receptors lacking specific domains in the cytoplasmic region. Mol. Cell. Biol., 13, 7239-7247. Wieser,R., Wrana,J.L. and Massagu6,J. (1995) GS domain mutations that constitutively activate TfR-I, the downstream signaling component in the TGF-1 receptor complex. EMBO J., 14, 2199-2208. Wrana,J.L., Attisano,L., Carcamo,J., Zentella,A., Doody,J., Laiho,M., Wang,X.-F. and Massague,J. (1992) signals through a TGFP heteromeric protein kinase receptor complex. Cell, 71, 1003-1014. Wrana,J.L., Attisano,L., Wieser,R., Ventura,F. and Massague,J. (1994) Mechanism of activation of the TGF-,B receptor. Natiure, 370, 341-347. Yamashita,H., Ichijo,H., Grimsby,S., Moren,A., ten Dijke,P. and Miyazono,K. (1994a) Endoglin forms a heteromeric complex with the signaling receptors for transforming growth Biol. J. Chem., factor-P. 269, 1995-2001. Yamashita,H., ten Dijke,P., Franzen,P., Miyazono,K. and Heldin,C.-H. (1994b) Formation of hetero-oligomeric complexes of type I and type II receptors for transforming growth factor-5. J. Biol. Chemn., 269, 20172-20178. Yingling,J.M., Wang,X.-F. and Bassing,C.H. (1995) Signaling by the transforming growth factor-f receptors. Biochirn. Biophys. Acta, 1242, 115-136. Received on March 4, 1996; revised 13, oni Auigust 1996 http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png The EMBO Journal Springer Journals

Phosphorylation of Ser165 in TGF‐beta type I receptor modulates TGF‐beta1‐induced cellular responses.

Download PDF
10 pages

Loading next page...
 
/lp/springer-journals/phosphorylation-of-ser165-in-tgf-beta-type-i-receptor-modulates-tgf-NUalVmZxPD

References (31)

Publisher
Springer Journals
Copyright
Copyright © European Molecular Biology Organization 1996
ISSN
0261-4189
eISSN
1460-2075
DOI
10.1002/j.1460-2075.1996.tb01013.x
Publisher site
See Article on Publisher Site

Abstract

The EMBO Journal vol.15 no.22 pp.6231-6240, 1996 Phosphorylation of in TGF-13 type I receptor Ser165 modulates TGF-f1-induced cellular responses Serhiy Souchelnytskyil, Peter ten Dijke, TGF-,B type I and type II receptors (T,BR-I and T,R-II) are essential for signal transduction (Wrana et al., 1994). Kohei Miyazono2 and Carl-Henrik Heldin T3R-I and TfR-II are widely expressed on many cell Ludwig Institute for Cancer Research, Box 595, S-751 24, Uppsala, types and bind TGF-,B with dissociation constants in the Sweden and 2The Cancer Institute, Japanese Foundation for Cancer low picomolar range (Massague et al., 1990). Cloning Research, 1-37-1 Kami-Ikebukuro, Toshima-ku, Tokyo 170, Japan and sequencing of TIR-I and cDNAs revealed TOR-II 'Corresponding author that they are single transmembrane serine/threonine kin- ases with cysteine-rich extracellular domains (Lin et al., Transforming growth factor-,l (TGF-1) signals via 1992; Franzen et al., 1993; Bassing et al., 1994). The two of two serine/threonine kinase an oligomeric complex kinase domains have 41% amino acid sequence similarity receptors denoted type I receptor (T,R-I) and TGF-P and each contains two short kinase inserts. binds TPR-II type II receptor We investigated the in vivo (TPR-II). directly, whereas T,3R-I can recognize only ligand TGF-P phosphorylation sites in TjR-I and T3R-II after com- when complexed with T,3R-II (Wrana et al., 1992). Phosphorylation of TJR-II was plex formation. TfR-II is a constitutively active kinase, Overexpressed residues in the C-terminus (Ser549 and observed at the ligand-recruited T,R-I on which transphosphorylates at residues in the juxtamembrane domain Ser551) and residues, predominantly located in a serine and threonine TGF-,1 induced in vivo (Ser223, Ser226 and Ser227). juxtamembrane domain containing a region of the of serine and threonine residues in phosphorylation (Wrana et al., 1994). GSGSGS motif (the GS domain) the domain of TfR-I in a region rich juxtamembrane serine and threonine residues in the Mutation of multiple in serine and threonine residues (GS domain; glycine, illustrating their GS domain impairs TGF-f responses, Thrl86, Serl87, Serl89 and Serl91), and more Thrl85, importance in signal transduction (Wrana et al., 1994; (Serl65). Phosphorylation in N-terminal of this region Franzen et al., 1995; Wieser et al., 1995). Serl72 and has been shown previously to be the GS domain Thrl76 in TfAR-I were found to be dispensable for extra- in activation of the T,BR-I kinase. We show involved cellular matrix (ECM) formation, but essential for the of TIR-I at Serl65 is here that phosphorylation inhibition by (Saitoh et al., 1996). However, growth TGF-P in modulation of TGF-j1 signaling. Mutations involved of Serl72 and Thrl76 has not been phosphorylation in led to an increase in of Serl65 TGF-p1- TOR-I and a molecular mechanism remains to be reported, mediated inhibition and extracellular matrix growth elucidated. The notion that acts downstream of TPR-I but, in contrast, to decreased TGF-f1- formation, and that TfR-I phosphorylation and activation is T,R-II induced A transcriptional activation signal apoptosis. essential and sufficient for most TGF-,-mediated was not affected. Mutations of Serl65 changed the is by data showing that a mutation responses supported pattern of TI3R-I. These observations phosphorylation which inactivates its ability to recognize T,R-I in T,BR-II that TGF-J receptor signaling specificity is suggest does not signal (Caircamo et al., 1995), and as a substrate of Serl65 of T,R-I. modulated by phosphorylation active mutant (threonine residue that a constitutively T,BR-I serine-threonine kinase/signal transduction/ Keywords: an acid residue) can signal in at 204 replaced by aspartic growth factor-,B receptor/transforming of and et al., 1995). the absence ligand (Wieser TPR-II the formation of a heteromeric complex TGF-4 induces T3R-II et Franzen et al., of ThR-I and (Wrana al., 1992; a heterotetramer of two molecules each 1993), most likely Introduction et Recent of and (Yamashita al., 1994b). TPR-I TPR-II studies between kinase-defective TfR-I isoforms complementation factor-j, (TGF-a) (TGF- Transforming growth indicate a and activation-defective TiR-I cooperative of structur- and belong to a large superfamily -,2 -P3) P1, molecules that is interaction between multiple T,R-I that cell related dimeric proteins regulate proliferation, ally transduction and essential for (Weiss-Garcia and of different signal differentiation motility many apoptosis, Massague, 1996). et Roberts and cell al., 1990; Sporn, 1990). types (Moses intrinsic for each have an and affinity is on cell The effect induced by TGF-,B dependent type TPR-I TPR-II et other al., 1994) and, upon overexpression, microenvironment the cell. TGF-,B (Ventura and on the surrounding heteromeric a complex may form ligand-independent roles in they is thought to play important many physiological are et al., in which both receptors phosphorylated (Ventura as the immune inflammation, such response, processes it has been and Moreover, Chen Weinberg, 1995). and wound healing. 1994; embryogenesis, angiogenesis of alone to a that high interaction with overexpression T,R-I its effects reported TGF-1 exerts through its constutive can lead to (Chen cell surface level autophosphorylation and binding proteins (Lin multiple receptors and et Weinberg, 1995). al., 1994; and Lodish, 1993; Derynck, 1994; Massague of the mechanisms that ten et of which Our understanding signaling et Yingling al., 1995; Dijke al., 1996), © Press Oxford University et aL S.Souchelnytskyi T. R-lI R-l1 s . H-=L HA H is is H sHHis Ti) R-l - 2 To R-11 + Ni-NTA HAHai HA A X 1.5 IP -HA VPN E. H; A *- i no-omrnpIleCxe: T R--- HA ~\? ; ,,, AA A .2 0.5 IP DRL I+TGF-3" : 0 nInr.- COmpIes x-Li' - TGF- _+ + fil is Hi s - 'I Hi,- lII IFi 1.lk uflT 141'w1 1.S1z II I l in V -1 TR-1T 1 i11TlITi] \L c\.1k .I A IC l Tl'L11d d Ti% -1 arld T. R-111 ( COrIY1plexe HA ,I IF1I<R 1.II I lb \ s Illi IICI I\. ","II 1 tItIl TLItLIv Ciopi-'. C i'i ll l M -NTA ;L L Ill t1II Tc1Tle \ Ilf T 1]T IIII iT LILICH PCI LI1T I1i .. l Il ldLI1 i-1 I .\LiI 111 11 1ll 1'Ti [)II T T \111ll 1i L:ll LIT .1. ; i TLiI 11I I R- IIl1II\111 IItIX ITiLl L L 1LJiIu1 ItL (C ti C TIT-1 t11i5cl \.tLtl I \cTi - )IlL il I 'ILL1 i r li .K.I;l cmli1cii IcIII IITji| D,r I hkt |cl e d L u I iii I IITtl U I F t tI ) l 1 L i 414} [|§g'-I~I'v ill111 t\|) -\TA-[ i']] n L11111i_ 'tllt' HA] MM2i15Wtlk ;111 nC> 1 .;! l1c T 11i- 1L ' F l 11 I 11I 1 h1 t1 l i ['ijli )1 i i L plIc\i1iL v}.1i11 ... .:--Ikkl.. ;- /7- -AE 4; ---Ipq .. W Lii( il -CT . IlL .|riLe)tI11(j)c L\LL TI R * LI fUR -J \\ l ill I mIll .: iC1 iiIlI).I-I ( Ci'i LiIII1LCLUiIIV F\LFL LIiliiIRlII \h.FC [-112'[ : ;-, l si l l[IC l i l l;l l jI I'; lll 1ii d \ -t )1l '11 ~~~~~~~~l IIj['k1 P] Uit P|I-OC2 LI C. hC - )-- llt.' aL 11 5\ ) t 'i< 1 511 , h S% 1)0 ' t.'d t.. d1 1VCl LI SI (1 c ) I.i " t. C C\ ]>' l t 1 t1 I Iji - I.L 'I. , ;HI' C( c\ c i- 'i /cd M I p li I )f 3 Ii, . 01 Ie Ii. R-I , ;IIt ;, I;ei4llxt I5 vttlr^ l) v t ;. [1x t' ; \'i it \\. } t 1 T h II',, 4,' )LYl1 i:Lt,[ d ) 11d B 1VIRI1]C II p tco' . t) .S t tl '' 2 e 1111 t|1 \ I C' IIL, ort] or 11kx1c cd C (IP> \\ IV ( h I1 I Pt LIl1 LI. 35S 32p | lji tII . I<i cl r. a link between activated receptors and TGF-f3 TGF-13 receptors, we used a sequential immunoprecipit- provide downsteam effects of inhibition and e.g. growth ation approach, by which and TfR-II can be isolated TGF-,3, TPR-I increased ECM is Using a yeast in incomplete. a heteromeric complex and in non-complexed forms formation, interaction FKBP-12 et al., from a screen, (Wang cell extract (Wrana et al., 1994) (Figure IA). In two-hybrid and transferase (Kawabata et al., with ox agreement previous data (Chen and Weinberg, 1994) famesyl-protein 1995), have been isolated as molecules that interact with ligand-independent complex formation was 1995) observed In TRIP-1 has been isolated as a protein between and addition, TIR-I TfZR-II when both receptor types were TPR-I. with TrR-II using a similar method (Chen overexpressed in COS- 1 cells (Figure IB). Endogenously interacting et the functional of these significance produced TGF-f did not contribute significantly al., However, to com- 1995). interactions remains to be determined. plex formation as addition of anti-TGF-f3 neutralizing In the we describe that one of the major antibodies had no effect on complex formation. present study Over- sites in TfAR-I is Serl65. expression of receptors in Mv1Lu cells also phosphorylation promoted ligand-dependent Whereas the of Serl65 is not essential complex formation in the absence of ligand (data phosphorylation not for its mutation increases the ECM shown). signaling, and T3R-II in complex were found to TPR-I be TGF-P1 formation and inhibition, but decreases the apop- phosphorylated, growth as revealed by isolation of the complex totic induced TGF- 1 in mink lung epithelial by from [32P]orthophosphate-labeled response cells; addition of ligand cells. of Serl65 therefore Phosphorylation led to an increase in complex formation and to a (MvlLu) concomit- to modulate TGF-,B1 signaling. ant increase in phosphorylation appears of the receptors, as deter- mined by comparison with parallel immunoprecipitation of 35S-labeled receptors. Ligand addition led to an increase Results of the phosphorylation level of but not of TI3R-II TPR-I of T/3R-I and TI3R-ll (Figure IC). Phosphorylation TGF-1 a heteromeric complex of TfR-I signals through Consistent with previous reports (Wrana et al., 1994; and and Lodish, 1993; Derynck, 1994; (Lin Chen and Weinberg, T,BR-II 1995), we found that the non- et 1994; Yingling et al., 1995; ten Dijke Massague al., complexed overexpressed Tf3R-II was phosphorylated and et To determine the phosphorylation sites in al., 1996). that this phosphorylation was not dependent on the pres- 6232 TIR-1 Serl65 phosphorylation modulates signaling plexed with T,BR-I also gave a similar phosphopeptide map (data not shown). Phosphoamino acid analysis of the peptides shown in gray (Figure 2C; spots 2, 3, 4, 5, 6, 7 :r A and 9) revealed that these peptides were phosphorylated :-. only on serine residues (data not shown). The positions of the phosphorylated serine residues were determined by release of radioactivity upon Edman degradation; alignment with serines that are present in tryptic peptides, as predicted from the TfR-II cDNA sequence, indicated that Ser549 and Ser551 in the C-terminal tail of TfR-II ~~~~~~~~~~~I t (Figure 2C; spot 4), and Ser223, Ser226 and Ser227 in the juxtamembrane region of T3R-II (Figure 2C; spots 6 and 7), were phosphorylated. Two phosphorylation patterns .i *: I. in the juxtamembrane region were observed, either Ser223, a, Ser226 and Ser227 were phosphorylated (spot 7), or Ser223 and Ser227 (spot 6; see Figure 2G). Tryptic phosphopeptide maps of TfR-I phosphorylated in l'ivo in COS- 1 cells and isolated in a complex with -I ES .-iW.- T3R-II revealed 16 spots of different intensity. In contrast to what was found for ligand binding and complex TPR-II, - -,0 -.; .-- formation induced changes in the phosphopeptide maps c -. ap -- -I lim la.-C, '- r,-m of T3R-I. Whereas non-complexed was not phos- 0 TPR-I ..: 4) @. phorylated appreciably, TfR-I in complex with Tf3R-II on several sites (Figure 2D, E and was phosphorylated of five peptides (spots 1, 2, 3, 4 F). The phosphorylation upon stimulation with TGF-51. and 5) increased further tryptic phosphopeptides were subjected to The different acid analysis and Edman degradation; by phosphoamino .. IE I - ... {i the results with the predicted tryptic peptides comparing from the Tf3R-I cDNA sequence, the location of several Fig. 2. In vivo phosphorylation sites in TIR-I and T3R-11. Epitope- phosphorylated serine and threonine residues in the juxta- tagged T3R-I and were transiently expressed in COS-1 cells. TPR-II membrane region of TfR-I could be determined (Figure The complexed receptors were purified, digested with trypsin and high-voltage electrophoresis and thin-layer peptides were resolved by 2G). and analyzed by using a FujiX chromatography. Plates were exposed Edman on the release of 32P radioactivity upon Based The maps of complexed T,BR-II from Bio-Imager. phosphopeptide the peptide of the major broad migrating degradation, or cells treated with TGF-f1 (B) and a schematic non-treated cells (A) 1) could be aligned to the GS ligand-stimulated spot (spot of the map of T,BR-II (C) are shown, as presentation phosphopeptide This broad 1 consisted of multiple maps of TfR-I purified from non-treated domain of T3R-I. spot well as the phosphopeptide treated with TGF- and a schematic cells (D) or cells 1 (E) Phosphoamino acid ana- slightly overlapping components. of the maps of T3R-I (F). The presentation phosphopeptide that of spot 1 were phosphorylated lysis revealed peptides I and II receptors used for Edman phosphopeptide spots of type threonine residues not shown). both on serine and (data acid analysis are shown in gray. degradation and phosphoamino different of phosphorylation in We found three patterns shown small black squares. Sample application points are by illustration of the of TI3R-1 and TIR-II with i.e. of Thrl86, of Serl89 (G) Schematic sequences the GS domain, phosphorylation in i(v phosphorylation sites the juxtamembrane domains (JM) and of Serl87, Serl89 and and Serl91, Thrl85, ThrI86, domains (TM) and kinase domains (KD) of indicated. Transmembrane It remains to be determined if the Serl9l (Figure 2G). are indicated. Serine and threonine residues that are the receptors sites are present in one peptide multiple phosphorylation are indicated by bold lines. phosphorylated at different positions or if several peptides phosphorylated acid in one The phosphoamino analysis co-migrate spot. of another site of ligand-stimulated phosphorylation ence of ligand. In contrast, non-complexed T3R-I major on serine residues (Figure showed phosphorylation in transfected COS-l and MvlLu cells was (spot 2) expressed of a residue after The appearance phosphorylated not to be in most experiments. 3A). found phosphorylated Edman indicates in degradation sequencing in some experiments when Tf3R-1 was over- eight cycles However, in the of amino acid that Ser165 tryptic peptide consisting in COS-1 cells to very high levels, non-com- expressed was since this is the in a form residues 158-178 phosphorylated, T3R-I was found phosphorylated (data plexed residue residues downstream of a lysine only serine eight not shown). from the TrR-I cDNA as sequence or an arginine predicted data of Ser 165 as a The phosphoryl- in vivo sites in Tf3R-I (Figure 3B). alignment of phosphorylation Mapping the of site were confirmed immunoprecipitation ation by and T,JR-II from this with VPN antibodies, which spot of T3R-II phosphoryl- radioactivity phosphopeptide mapping Tryptic to amino acid are made corresponding in COS- 1 cells and isolated in a complex against peptide ated in vivo in T3R-I 158-178 3C). residues (Figure revealed 17 of different intensities with TIR-I spots to at Serl65; was found We found no in the distribution TIR-II phosphorylate 2A and change TP3R-I (Figure B). or complexed wild-type in the absence when baculovirally expressed T3R-II-derived (Figure of phosphopeptides TfR-1 were and to subjected of not com- kinase-inactive purified or 2B) ligand; T,R-II 2A) presence (Figure 6233 S.Souchelnytskyi et aL fected with and mutants in wild-type receptor which TPR-I Serl65 was to alanine changed (T,3R-I/S165A), glutamic | 7 ,e.-1 acid or acid (T,R-I/S165E) aspartic (TPR-I/S165D) with to residues, formation with T,1R-II, respect complex overall intensities of (in vivo and in vitro) phosphorylation or in their abilities to bind (data not shown). ligand of [32P]Orthophosphate-labeled with complexes TPR-II T,BR-I or wild-type mutants, formed in the TPR-I/S165 or absence of were presence the ligand, purified using method and B sequential precipitation to subjected tryptic followed digestion phosphopeptide by mapping. Figure 4C-F shows that to a 2, with spot corresponding peptide was not seen in phosphorylated Serl65, the phospho- of the mutant. This peptide map was spot TfPR-I/S165A 5t:i also not seen in the derived from the map T,3R-I/S165E mutant not The low (data shown). detected intensity spot in this area of the 165A mutant be a T,3R-IIS may co- from another of the migrating peptide part of receptor, _ 1'iUhhIIIIbn which the is too low to be detected ~~~~Numunt intensity radio- 1 by 4 7 a 1 l t 6 9 3 4 1f cvc.!r-Z V P N E E D P L chemical We S D R P F... also found that sequencing. 4, which spot in the of after C appeared maps wild-type T,3R-I ligand stimulation 4C and was detected in (Figure D), the maps of the 165A mutant even in the absence of TPR-I/S 4E and ligand (Figure the low amount F); however, of in 4 radioactivity determination of peptide precluded the corresponding phosphorylation site. Other differences are also seen between the of maps wild-type T,R-I (Figure 4C and and the 165A D) 4E and or TfR-I/S (Figure F) 165E not mutants. TfR-I/S (data shown) in a However, series of four different in experiments which the separation conditions were these other varied, differences were not found to be reproducible. Mutation of Ser165 in modulates TPR-1 TGF-fi1 signaling Fig. 3. Identification of Serl65 as a site in To determine the role of phosphorylation TJR-I. Serl65 phosphorylation in TGF-, (A) The peptide to corresponding spot 2 of TfR-I was subjected to we signal transduction, tested TGF-f31 responses in a cell phosphoamino acid The analysis. migration of phosphorylated serine line a functional lacking endogenous T,3R-I (clone R4.2 (S), threonine (T) and tyrosine (Y), used as is standards, shown. of MvlLu Laiho (B) The same after stable peptide was also cells; etal., 1990), transfection subjected to Edman degradation; the with elution positions of 32P-labeled or mutant amino acids are shown and wild-type T,R-I. In mutants aligned to T,R-I Serl65 the of sequence a putative tryptic peptide (VPN), the only one with a was with other replaced amino acid residues (TPR-I/ serine eight residues downstream of a lysine or an arginine residue. T,R-I/S 165E or 165A, or with a 165D) kinase- (C) The TOR-I/S peptide corresponding to spot 2 was precipitated by VPN inactive variant of TfR-I (T3R-I/K232R) under antibodies, directed against the VPN peptide transcrip- (residues 158-178 in tional control TfR-I), spotted onto a of the inducible thin-layer chromatography human metallothionein IIA plate and quantified by using a FujiX Bio-Imager. Radioactivities, All immunoprecipitated promoter. transfectants that were used expressed T,3R-I from of peptide spot 2 by VPN antibodies (1), VPN antibodies in the or TfR-I mutants to the same level (Figure 5A). We found presence of excess VPN peptide (2), and non-immune serum (3), and that phosphorylation of is not Serl65 essential for the their quantitations are shown. transduction of the effect of TGF-f1 on cell growth, since cells transfected with 165A or Tf.R-hIS 165E TOR-I/S in vitro kinase assay followed by phosphopeptide mapping, responded to TGF-j1 addition with growth inhibition Serl65 was phosphorylated in both cases (Figure 4A and (Figure SB). In the fact, TGF-f81-induced inhibition of B). That Serl65 was phosphorylated was confirmed by cell proliferation was more pronounced in cells transfected the observation that radioactivity from the predicted spot with and especially Tf3R-IIS165E, than in TOR-I/S165A, could be immunoprecipitated with VPN antibodies, but cells with wild-type values of 0.41, 0.06 TPR-I (ED50 and not by pre-immune serum. Phosphorylation of Serl65 0.54 ng/ml, respectively). in kinase-inactive T3R-I (Figure 4B) indicates that the TGF-,1 stimulated the biosynthesis of plasminogen phosphorylation of this residue is not dependent on T,3R-I activator inhibitor-i (PAI-1) and fibronectin and their kinase activity, and thus most likely is phosphorylated in deposition ECM of wild-type T3R-I transfected cells directly by T,R-II. The (Figure 6). level of PAI-1 induction was higher in In order to determine the functional importance of the cells expressing Tf3R-I/S165A or than in T,PR-I/S165E phosphorylation of Serl65 in Tf3R-I, this residue was cells expressing wild-type Tf3R-I (Figure 6A and B). mutated to alanine, glutamic acid or aspartic acid residues. These differences were detected already after 4-5 h of No differences were 1 found between COS- cells trans- TGF-f1 treatment (data not shown), but the effect was 6234 TIR-1 Serl65 phosphorylation modulates signaling -_.1\\j4 o. AV _F. r-i ,1 R *. .e, .T' TGF-:!- Fig. 4. Phosphopeptide maps of complexed mutated and wild-type TfR-I. Baculovirally expressed wild-type (A) and kinase-inactive (B) TIR-I, complexed with T3R-II, were purified, subjected to in vitro kinase assay, digested with trypsin and phosphopeptide maps were obtained. Complexes of T,R-II with wild-type T3R-I (C and D) or with TIR-I/S165A (E and F), phosphorylated in vivo were purified from COS cells, digested by trypsin and subjected to two-dimensional analysis. The arrows show the migration position of the Serl65-containing peptide. The arrowheads show the migration positions of a peptide, the phosphorylation of which is dependent on mutation of Serl65. Cells were treated (A, B, D and F) or not (C and E) with 10 ng/ml of TGF-431. Sample application points are marked with open arrowheads. shape of the structures is not regular and it is not The more pronounced after 16-18 h of ligand treatment (Figure to be derived from a single cell (Figure 7G). The likely 6A and B). A similar pattern of response, albeit much of these structures may be related to the effect appearence weaker, was found for stimulation of newly TGF-P1 mutants on ECM formation. of Serl65 synthesized fibronectin, incorporated in ECM (Figure 6C The effects of TGF-, 1 on growth inhibition, fibronectin and D); the cells with Serl65 of mutated incorpora- TPR-1 and deposition in ECM, and PAI- I induction, biosynthesis ted 1.8-2.0 times more fibronectin into ECM than did that the mutations of Serl65 to alanine, and suggest wild-type T,R-I cells. In this assay, no difference between to glutamic acid, give a higher signal especially TPR-I T,R-I/S 165A and T,R-I/S 165E cells was observed. TGF- transducing potential in these assays. However, no differ- had no effect on ECM formation in non-transfected P1 was found between wild-type T,BR-1, TIR-I/S165A ence cells or cells (Figure 6 and data not shown). TPR-I/K232R and 165E cells in the 1 induction of lucifer- TGF-P To investigate the effect of the Serl65 mutations in TPR-I TPR-I/S ase expression, controlled by PAI-I promoter and TPA- on cell behavior in dense cultures, we cultured cells elements in transfected MvlLu and COS-1 responsive transfected with inducible vectors encoding wild-type 8), and in TGF-,Bl-mediated inhibition of cells (Figure T,R-I/S165D or TPR-IIS165A, TPR-I/K232R, TPR-I, as measured by the migration of cells into cell migration 165E for 3 days to confluency. Thereafter, cells TPR-I/S area in dense cultures from which cells have a defined were cultured with or without ZnCl2 to induce the expres- scraped off (data not shown). been sion of the receptors, and with various concentrations of of wild-type T,R-1 in R4.2 cells restored Expression TGF- 1 for 7-10 days with fresh media replacement every (Figure 9, lanes 13-16). Interest- apoptosis ligand-induced day. No differences were observed for the wild-type TGF-lI-stimulated accumulation of apoptotic the ingly, and mutant receptor-transfected cells in the absence of DNA ladder formation occurred less efficiently and bodies induction of expression (data not shown). Upon TPR-I cells and especially for T,R-I/S165E for induction of receptor expression, TGF-P1 showed no TPR-I/S165A when with wild-type cells, and was compared TPR-1 cells, effect on cells with wild-type TPR-I (Figure 7A and B), in cells (Figure 9, lanes 1- at all not seen TPR-I/K232R T,R-I/S165A (Figure 7C and D) or TPR-I/K232R cells mutation of Serl65 to alanine or glutamic the 12). Thus, (data not shown). However, cells with TPR-I/S165E weakens the effect of on apoptosis, residues TGF-PI acid (Figure 7E-G) or (data not shown) showed TPR-I/S165D effect on growth inhibition and ECM forma- the whereas TGF-,B 1-dependent cell overgrowth and formation of three- is strengthened. tion dimensional structures after 4-5 days in dense cultures. 6235 et aL S.Souchelnytskyi Discussion exerts its cellular TGF-P3 signals of through cooperation two related serine/threonine kinase distantly K'; receptors, .-~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~...... and and TPR-I TO3R-11 (Lin Lodish, 1993; Derynck, 1994; et al., et ten TGF- 1: Massague' 1994; al., 1995; ++ + Yingling Dijke et al., to the cDNA of T RI 232R 1996). Subsequent both K T F.-I S 1 65E T. R- Si I jA cloning id -tYPe our of the mechanism receptor types, of understanding transduction of has increased signal TGF-03 dramatically. The model for activation is B proposed distinct receptor from that of other classes of in that cytokine receptors occurs in a manner with phosphorylation sequential Tf3R-1 as a substrate for et acting Tj3R-11 (Wrana al., 1994; 4' 0 Wieser et al., little is known I 1995). However, about yet 2,, the sites that are in phosphorylation present Tf3R-1 and w. T. iI C: and their in activation TO3R-11 significance receptor and transduction. In the signal we present the study, report identification of several in vivo sites phosphorylation in and 1 We found that Serl.65 Tf3R-1 in is TfOR-11. Tf3R-1 one of the sites of major ligand-dependent in phosphorylation K 232R and that it has an role in TPR-1 important signal regulation of TGF-f31. In with agreement we observed previous reports, that TGF- of and overexpression induced TOR-1 TO3R-11 ligand- of in cells heteromeric Fig. Analysis transfected with independent which TGF-01 signaling TI3R-1 complex formation, has and mutants. of in TOR-1 (A) Expression been stably transfected attributed to an intrinsic TPR-1 of and Tj3R-1 affinity TOR-11 cells. R4.2 cells were transfected with stably different constructs of for each other et (Ventura Chen and al., 1994; in vector. Weinberg, Chosen cell were TP3R-I pMEP4 35S-labeled, pools treated addition led 1995). to a further increase with 50 ltM and/or TGF-13I and TGF-01 in ZnCl2 (10 Tf3R-1 were ng/ml), with anti-HA immunoprecipitated antibodies. complex formation, the probably Immunoprecipitates were by stabilizing interaction resolved by followed visualization SDS-gel electrophoresis, between the in by and the In receptors the complex. complex, quantitation using of FujiX Bio-Imager. (B) Analysis both growth and Tf3R-1 were found to be TJOR-11 inhibition in to TGF-j31. R4.2 phosphorylated, response cells transfected with stably whereas was found non-complexed not to and mutants be wild-type were to a TPR-1 TfOR-1 TP3R-I subjected [3H]thymidine in the unless or absence of different phosphorylated it was assay presence concentrations to of overexpressed very high TGF-f31. The of percentage inhibition was calculated levels. In growth contrast, Tf3R-11 was found to be by phosphorylated [3 decrease relative to the measuring H]thymidine incorporation into also in form. These non-complexed observations are cells in the absence of P<0.001. TGF-P1. consistent with data previous that Tj3R-II showing forms a homodimer and (Chen Henis et Derynck, 1994; al., A~~~~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~~~~~~A IdIy 'T' Fi- A T R CF- f., T R K ~ T'~~~~ ~ ~ ~ ~ ~ -- -1 R4 y w R 2 Id- I" T fj3A I I'i S 165E 2 2. IZ) .) S165A L) c - S165E LLJ 'Z LO) c :t :3 0 r - -a z --I -; 7 w .; 0 -0 = -~~~~~~~ T. D- 1 4) -~~W. 5A SiSS 'O ;., .S2 W. T. Z.0 .- ei !. ~~~~~K232R 2 R 4.2 LL TGF- (nrg,;mII TGF- (ng/ml) Mutation of Serl65 Fig. in leads to an increase Tf3R-1 of extracellular matrix TGF-j3I-stimulated formation. Cells transfected with or stably mutant wild- type constructs were treated Tf3R-1 with 50 j.M to induce ZnCl2 of expression and then receptors treated with for 16-18 were TGF-f3I h. 35S-labeled with 'ProMix' Cells for the last 2 h of incubation and in proteins extracellular matrix were resolved by visualized SDS-gel and electrophoresis, quantitated a using FujiX Bio-Imager. TGF-f31 induction of and PAI-I1 biosynthesis (A) fibronectin incorporated into matrix extracellular (C) cells by Tf3R-I expressing or Tf3R-I mutants wild-type are shown; of quantitation data is presented in and (B) (D), respectively. 6236 TIR-1 Ser165 phosphorylation modulates signaling * wild-type A B 100 r- K232R [1H C.) S165A 7- Z S165E CD (A L. L)- c -j TGF- (ng/ml) P1 e. Fig. 8. Effect of TfR-I Serl65 mutations on gene expression. R4.2 E F cells, stably transfected with wild-type TjR-1 or T3R-I mutants, were ..: I. transiently transfected with p3TP-Lux plasmid, and receptor expression was induced by treatment with ZnCl,. Luciferase activity was determined after stimulation with different concentrations of r, as described in Materials and methods (representative of five TGF-p1, experiments). JI t. -, r, . - r. T n T F-, Sz j--I A F ,S ;4 p 7nl,c T CF.- 1 |~~~~~~~~~~~~~~~~~~~~~ ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~.. < < K''. r.~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~. .s __. .. -- l~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~. 7. TGF-f31 induces cell overgrowth in cells. Cells Fig. TPR-I/S165E stably transfected with wild-type T,BR-I (A and B), (C TPR-I/S165A and D) or (E, F and G) were kept in dense cultures; the TPR-I/S165E aI medium was changed every day in the absence (A, C and E) or (B. D, F and G) of TGF-.1 (5 ng/ml) and 50 ,uM ZnCIl. presence Photographs were taken after 5 days of culturing. Representative from four experiments are shown. (G) represents a picture photographs 9. Mutation of Serl65 in T3R-1 decreases the ability of TGF-,B1 Fig. of one of the structures, formed by cell overgrowth, at higher to stimulate apoptosis. R4.2 cells expressing wild-type TIR-I (lanes magnification. 13-16), T,BR-I/K232R (lanes 9-12), TPR-I/S165A (lanes 5-6) or TIR-I/S165E (lanes 1-4) were treated or not with ZnCl, to induce receptor expression. Cells were then stimulated or not with 5 ng/ml 1994) and is a constitutively active kinase in the absence TGF-f1. DNA prepared from apoptotic bodies was analyzed by of ligand (Wrana et al., 1994). Identical tryptic phospho- agarose gel electrophoresis and DNA was visualized by ethidium peptide maps were found for T3R-II whether it was Migration of DNA molecular weight standards is bromide staining. in lane 17. Essentially the same results were obtained in six shown stimulated or not with ligand, and whether it occurred in and in experiments with two different cell different experiments complex with TIR-I or not. Phosphoamino acid analysis preparations. of isolated tryptic fragments from TIR-II showed phos- phorylation only on serine residues, consistent with pre- of the kinase domain of T3R-I by charge- in the activation vious findings (Wrana et al., 1994). In T3R-II, Ser549 conformational changes rather than providing induced and Ser55 1, located in the C-terminal tail, were identified sites for downstream elements. Consistent with docking as phosphorylation sites. Wieser et al. (1993) found that notion, we found three patterns of GS domain phos- this the T,BR-II C-terminal tail was dispensable for signal in vivo, which may correspond to different phorylations as a TfR-II mutant without this domain, and transduction, kinase activation. Phosphorylation in the levels of T3R-I both Ser549 and Ser551, was capable of thus lacking domain of T3R-II also occurs in multiple juxtamembrane TGF-j sensitivity to cells lacking TfR-II (DR restoring also have a role in activation of the kinase. and may patterns cells) after transfection (Wieser et al., 1993). mutant treatment of cells not only led to an increase in Ligand Phosphorylation sites were also identified in the juxta- through increased complex forma- T3R-I phosphorylation domain of both T3R-I and In T3R-II, membrane TPR-II. also induced qualitative changes in the tryptic but tion, was detected at Ser223, Ser226 and phosphorylation pattern of T3R-I (Figure 2D-F). The latter phosphopeptide Ser227, which are located a similar distance from the be of importance for full activation particular changes may kinase domain as Thrl86, Serl87, Serl89 and Serl91 that the serine and threonine residues in For T3R-I, of T,R-I. were identified as phosphorylation sites in the GS domain as well as Serl65, were identified as GS domain, the of (Figure 2G). The number of phosphorylated TPR-I in vivo phosphorylation sites ligand-dependent major residues, as opposed to their particular position in the GS mechanism behind this change and the The 2G). (Figure domain, appears to be of importance for T3R-I signal TGF-I receptor signaling remain to be for significance transduction (Franzen et al., 1995; Wieser et al., 1995). mutation of Serl65 also led to In addition, elucidated. This suggests that the phosphorylated residues have a role 6237 S.Souchelnytskyi et aL of effects of mutants on PAI-I appearence Serl65 Table I. protein Comparison of the effects biological of and wild-type Serl65 and the absence of mutants on TfR-I differences in signal transduction synthesis luciferase from the construct expression p3TP-Lux be reporter may Assay TfR-I differences in explained by respective gene promoters. The fact that Wild-type mutations affects Serl65Ala Serl65 Serl65Glu/ signaling, TGF-f1 without Serl65Asp kinase makes it affecting TfR-I activity, unlikely that these mutations have led to of the perturbation Growth inhibition + + + ++ conformation of and the TPR-I supports that ECM production possibility + + ++ of (fibronectin, PAI- phosphorylation-dephosphorylation has a 1) Serl65 regu- role in Cell overgrowth the interaction with - - latory downstream + signaling Induction of gene + expression + molecules. Cell migration + + is + multifunctional and its effects on TGF-0 depend Apoptosis + (+) (+) the cellular context, a need for coordinate implicating of its actions. Response: -, none; (+), Control of weak; +, regulation action intermediate; ++, strong; +++, TGF-13 may very strong. exist at different levels in the transduction signal pathway, from of to regulation TGF-,3 modulation gene expression of Our TGF-f-induced transcriptional results responses. ligand-independent appearance of 4 in the that spot Serl65 tryptic suggest an phosphorylation plays important phosphopeptide of role maps T,BR-I, which for in this wild-type T,R-I process by modulating multiple biological appeared after only treatment. effects ligand Whether this of TGF-, 1. change reflects an influence of Serl65 on the T,R-I phosphoryl- ation pattern or is due to structural as a result changes of Materials the and introduced mutation is methods not known. The Serl65-dependent changes in the phosphorylation lines Cell patterns of TfAR-I (Figure and 4C-F) the COS-1 and strong ligand cells were obtained from MvlLu American Culture Type dependence of Collection. phosphorylation of Serl65 cells that lack increased our MvlLu active cells; functionally (R4.2 TfR-I Laiho et interest in al., were elucidating its Dr involvement in the 1990), Cells were regulation of provided by cultured J.Massagu6. in Dulbecco's modified medium TGF-,81 signaling. We mutated Eagle's (DMEM; with Serl65 of TfR-I Gibco-BRL) to alanine, 10% fetal bovine serum (FBS; of Gibco-BRL), 100 U/ml and glutamic acid or aspartic acid penicillin residues, thereby 50 abrogating of H5 gg/ml streptomycin. cells were Spodopterafrugiperda cultured the phosphorylation of this residue. in Mutations IPL-41 medium to glutamic 10% FBS (JRH) containing with supplemented acid or aspartic acid residues were broth. made with the tryptose intention phosphate of mimicking the negative of the charge phosphate group. Constructs and We transfection studied the effect cell of Serl65 mutations in T,BR-I on cDNAs for with and growth TO3R-I tagged hemagglutinin inhibition, epitope gene T,BR-II expression, ECM production, with cloned in the tagged vector et (His)6, pCMV-5 (Wrana al., contact 1994), inhibition, cell migration and apoptosis were obtained from induced Dr Mutant forms of J.Massague. were receptors by TGF-,1 using MvlLu cells. In PCR with none of these generated assays by mutagenic DNAs as primers using receptor did All mutant we find an constructs were abrogation of the template. over the TGF-f1 sequenced that response or regions were with the exchanged COS-1 cells ligand-independent in parental receptor were signal any of the plasmids. Serl65 mutations. transfected with of vector transiently 1 T,BR-I and 7 of However, we jg observed that ,ug Serl65 mutations can modulate vector TfR-II 40 mm dish the calcium per by phosphate precipitation method the effects of TGF-31 in a positive or the MBS negative mammalian way, using transfection kit (Stratagene), the following depending on the manufacturer's particular For stable response that transfection in was measured protocol. R4.2 all cells, tagged cDNAs were (Table I). cloned in The receptor mutation of Serl65 under led to pMEP4 (Invitrogen) higher signaling transcriptional control of the metallothionein potential Transfection was done for TGF-f31 promoter. in the growth inhibition assay (Figure using the calcium method. phosphate Selection of precipitation transfectants 5B) and in ECM production (Figure 6). In was addition, in in the of 100 performed U/ml of presence We hygromycin. obtained cells with TfR-IS 165E or four cell but not transfected with in cells pool cultures, TPR-I/S165D, wild-type TOR-I, TjR-I/K232R, with T,BR-I/S165A wild-type and T,R-I, TGF-j1 was 165E, and 60 able to induce TPR-I/S individual cell respectively, clones, 12 for above-mentioned overgrowth (Figure every variant and 12 7). A TfR-I for certain threshold appears T,R-IS165D. neces- All these cultures and clones were tested for sary for induction of this response, because in T,3R-I/S 165A cells TPR-I TGF- treatment. The one expression upon cell ZnCl2 culture pool and two f I showed no effect. The observed cell clones for overgrowth may mutant with the same every level of the receptor expression be related to the were used in stimulation of substrate-independent biological assays. proliferation of NRK cells by TGF-f (Roberts et al., Purification of 1981), in which induction TGF-,8 of ECM formation plays receptors an Cells were washed three times with important role. phosphate-buffered saline (PBS) and in buffer 20 mM lysed lysis [LB; In Tris-HCl, pH 7.4, 150 mM contrast, we found a decrease NaCl, in TGF-f31-stimulated 0.5% Triton 50 mM X-100, 10 mM NaF, 1 Na4P207, mM apoptosis in Na3VO4, cells with a Serl65 mutant 1 mM of TfR-I compared phenylmethylsulfonyl fluoride (PMSF), 100 U/ml aprotinin]. with wild-type and TfR-I heteromeric cells (Figure 9). TI3R-I TtR-II and Thus, complex non-complexed T,BR-I (His)6- T,BR-II were phosphorylation at tagged from Serl65 precipitated clarified appears extract with important for signal Ni2+-NTA agarose (Qiagen). Non-precipitated regulation non-complexed HA-tagged of the activated T,BR-I TGF-, receptor. We did not was subsequently immunoprecipitated from the unadsorbed detect fraction any differences between wild-type anti-HA TfBR-I and antibodies using (12CA5, or Babco) anti-T,BR-I (VPN) anti- TfR-I/Serl65 mutants bodies et in their cell migration and (Franzen induction al., Immunocomplexes 1993). were collected on of protein the A-Sepharose beads gene EC expression (Immunosorb; Diagnostics using construct AB). Ni p3TP-Lux +- reporter NTA agarose beads which were washed three times with 25 mM that imidazole (Figure 8), the in of suggests these signaling LB and bound receptors were eluted in 250 effects is mM imidazole in LB. not The Serl65 regulated by The eluate was subjected to with phosphorylation. immunoprecipitation anti-HA antibodies to 6238 T3R-1 Ser165 phosphorylation modulates signaling enrich for the TfR-I and T,R-II heteromeric complex and with rabbit cells were washed with PBS to remove calcium extensively phosphate peptide antisera against TfR-II (DRL) (Yamashita et al., 1994a) to the cells were incubated in DMEM with 10% precipitates. Subsequently, enrich for the non-complexed TPR-II. FBS for 16-20 h. the was induced Thereafter, receptor expression by treatment of the cells with 50 tM in DMEM with ZnCl2 supplemented labeling of cells, tryptic 132PlOrthophosphate 0.1% FBS for after which TGF-31 was added. Luciferase h, activity phosphopeptide mapping and two-dimensional in the cell was measured after 22-24 h the luciferase lysate using assay phosphoamino acid analysis to the manufacturer's system (Promega Biotec), according protocol using Cells were labeled in phosphate-free medium supplemented with 0.5% an LKB Luminometer (LKB-Bromma). pH 7.2, and 1.0 dialyzed FBS, 15 mM HEPES, of [32P]ortho- mCi/ml phosphate for 3 h. During the last 15 min of incubation, TGF-,B1 was added as mentioned in the figure legends. Cells were lysed and receptors Cell overgrowth assay isolated as described above, separated by SDS-gel electrophoresis and Cells transfected with TfR-I/S165A, stably wild-type T,BR-I, TPR-I/ electrotransferred to a nitrocellulose filter (Hybond C-extra; Amersham). 165D or were cultured for 3 to reach S165E, TfR-I/K232R T,R-I/S days For tryptic phosphopeptide mapping, receptor bands were localized by in the or absence of 50 lM and, thereafter, confluency presence ZnCl2 exposure on a FujiX Bio-Imager (Fuji), excised from the filter and for 7-10 Cells and TGF-f1. The medium was day days. changed every digested in situ with trypsin (modified sequencing grade; Promega). were examined regularly by microscope. Two-dimensional phosphopeptide mapping was done using the Hunter thin-layer electrophoresis apparatus (HTLE-7000; CBS Scientific), essen- tially as described by Boyle et al. (1991). First dimension electrophoresis Apoptosis assay Cell induction of and treatment with was performed in pH 1.9 buffer (formic acid/glacial acetic acid/double- TGF-P1 seeding, receptor expression done as defined for the inhibition that cells were distilled water; 44:156:1800 by vol.) for 35 min at 2000 and second V, growth assay, except in medium with 0.1% FBS for 13 h before were incubated dimension ascending thin-layer chromotography in isobutyric acid buffer harvesting bodies were treated with 0.5 the detached material. (isobutyric acid/n-butanoUpyridine/glacial acetic acid/double-distilled mg/ml Apoptotic 0.2 RNase for K for 1 h at and water; 1250:38:96:58:558 by vol.). After exposure, phosphopeptides (Sigma) proteinase (Sigma) 50°C mg/ml at to remove and were eluted from the plates in pH 1.9 buffer and lyophilyzed. The 20 RNA, min Apoptotic 60°C protein respectively. 2% and ethidium fractions were then subjected to two-dimensional phosphoamino acid DNA was analyzed by agarose gel electrophoresis analysis and, in parallel, automated Edman degradation. For Edman bromide staining. degradation, phosphopeptides were coupled to Sequelon-AA membranes (Millipore) according to the manufacturer's instructions and sequenced on an Applied Biosystems Gas Phase Sequencer Model 470A. Released phenylthiohydantoin amino acid derivatives from each cycle were spotted Acknowledgements onto thin-layer chromatography plates. The radioactivity in each spot for excellent technical Anita and Susanne We thank Grimsby Moren was quantitated by exposure to a FujiX Bio-Imager. for Christer for Ulla Wernstedt assistance, peptide synthesis, Engstrom For phosphopeptide mapping, H5 cells at -80% confluency were co- and radiochemical Lars Ronnstrand sequencing, oligonucleotide synthesis infected with and TfR-II baculoviruses at an m.o.i. of 10, and 5 Ohashi for TPR-I for with Hideya providing help phosphopeptide mapping, p.f.u./cell. Cells were used 48 h post-infection. Purification of receptors for R mutant MvlLu recombinant human TGF-41 and Joan Massague was done as described above, purified receptors were subjected to T3R-I and baculoviruses and recombinant p3TP-Lux cells, TPR-II in vitro kinase assay as described (Franzen et al., 1995), and tryptic plasmid. phosphopeptide mapping was done. Growth inhibition assay X well in 24-well Cells were seeded at a density of 1 104 cells per plates References in DMEM with 10% FBS. was stimulated Receptor expression by incubation of the cells in medium with 0.1% FBS with 50 ltM ZnCI2 Gustafson, He,W.W., Howe,D.J., Wang,T., Bassing,C.H., Yingling,J.M., for 5 h. Subsequently, medium containing 3% FBS with or without 50 tM and (1994) transforming Shah,P., Donahoe,P.K. M.L., Wang,X.-F. ZnCl2 was added and cells were with different concentrations of incubated f that to activate factor I gene expression. signals growth type receptor TGF-,B1 for 22-24 h; during the last were labeled with 1 2 h cells ,tCi/ml 87-89. 263, Science, [3H]thymidine (Amersham Corp.). the cells were fixed in Thereafter, der and Hunter,T. Van (1991) Geer,P. Phosphopeptide Boyle,W.J., 5% ice-cold trichloroacetic acid (TCA) for 20 washed three times min, acid two-dimensional and by analysis mapping phosphoamino with TCA, and solubilized 1 M NaOH. The cell extract was with cellulose Methods 201, on Enzymol., plates. separation thin-layer neutralized with 1 M and 3H was measured in a HCI radioactivity liquid 110-149. scintillation ,B-counter using Ecoscint (National Diagnostics). of and (1995) Zentella,A. Disruption Carcamo,J., Massague,J. factor a mutation that prevents Extracellular matrix formation assay signaling by transforming growth ,B within the Mol. Cell. Biol., Cells were seeded in 6-well plates at a of 5X 104 cells well. receptor complex. density per transphosphorylation 1573-1581. After 18-24 h, the medium was changed to DMEM supplemented with 15, Biochemical evidence for the and 0.1% FBS, with or without 50 .tM After 5 TGF-f31 was added (1995) h, Chen,F. Weinberg,R.A. ZnCl2. of and growth to the cells and incubation prolonged for 16-18 h. To label the transforming newly transphosphorylation autophosphorylation Proc. Acad. Sci. 92, 1565-1569. kinases. USA, synthesized proteins, the cells were incubated in methionine-free medium factor Natl receptor Homomeric interactions between MCDB (SVA, Sweden) with 50,Ci/ml of mixture 'ProMix' and 35S-labeling (1994) Chen,R.-H. Derynck,R. J. Biol. Chem., 269, h of treatment. of (Amersham Corp.) during the last 3 TGF-41 II Aliquots receptors. growth type transforming factor-5 the medium (100 or secreted proteins bound to 22868-22874. gelatin-Sepharose gl) For ECM and (Pharmacia) were analyzed by SDS-gel electrophoresis. Derynck,R. Maruoka,E.M., Choy,L. Miettinen,P.J., Chen,R.-H., on ice once in PBS, three that is associated with and isolation, the cells were removed by washing A WD-domain protein (1995) 548-552. times in 10 mM Tris-HCl, pH 8.0, 0.5% sodium and Nature, 377, deoxycholate, the II receptor. phosphorylated by type TGF-,B 1 mM PMSF, twice in 20 mM Tris-HCl, 8.0, and once in PBS. Trends Biochem. pH signaling. TGF-f-receptor-mediated (1994) Derynck,R. ECM off and extracted into SDS buffer proteins were scraped sample 548-553. 19, Sci., Secreted and ECM containing 10 mM dithiothreitol. proteins proteins Heldin,C.-H. Yamashita,H., Schulz,P., ten Ichijo,H., Franzen,P., Dijke,P., were analyzed followed by SDS-gel electrophoresis, by fluorography of a I that forms receptor and type (1993) Cloning Miyazono,K. TGF,B using Amplify The were and (Amersham Corp.). gels exposed quantified Cell, 75, with the TGFf receptor. a heteromeric type II complex using a FujiX PAI-i was identified as a 45 kDa in Bio-Imager. protein 681-692. the ECM et Fibronectin was identified as a fraction (Laiho al., 1991). The GS domain of and (1995) Heldin,C.-H. Miyazono,K. Franzen,P., 230 kDa protein (Laiho et and was both in the soluble al., 1991) present I is in factor-f signal receptor important the type transforming growth fraction and in the ECM fraction. Res. Commun., 207, 682-689. Biochem. transduction. Biophys. and The Lodish,H.F. (1994) types II Moustakas,A., Lin,H.Y. Transcriptional response assay Henis,Y.I., form homo-oligomers. and III receptors Stable transfectants of TfiR-I were transiently transfected with p3TP- transforming growth factor-,B 139-154. J. Cell Biol., 126, Lux (Carcamo et al., 1995), as described above. The following day, the 6239 S.Souchelnytskyi et aL Kawabata,M., ImamuraT., Miyazono,K., Engel,M.E. and Moses,H.L. (1995) Interaction of the transforming growth I factor-5 type receptor with farnesyl-protein transferase-ax. J. Biol. Chem., 270, 29628-29631. Laiho,M., Weis,F.M.B. and Concomitant loss Massague,J. (1990) of transforming growth factor (TGF)-f I and II in receptor types TGF- f3-resistant cell mutants both implicates in receptor types signal transduction. J. Biol. Chem., 265, 18518-18524. Laiho,M., Ronnstrand,L., Heino,J., DeCaprio,J.A., Ludlow,J.W., Livingston,D.M. and Control Massague,J. (1991) of JunB and extracellular matrix protein expression by transforming growth factor- is of independent SV40 T antigen-sensitive PI growth-inhibitory events. Mol. Cell. Biol., 11, 972-978. Lin,H.Y. and Lodish,H.F. (1993) for the Receptors TGF-,B superfamily. Trends Cell Biol., 3, 14-19. Lin,H.Y., Wang,X.-F., and Ng-Eaton,E., Weinberg,R.A. Lodish,H.F. (1992) Expression of cloning the II a TGF-,B type receptor, functional transmembrane serine/threonine kinase. Cell, 68, 775-785. Massagu6,J., Cheifetz,S., Boyd,F.T. and Andres,J.L. (1990) TGF-f receptors and binding proteoglycans: recent progress TGF-P in identifying their functional Y properties. Annii. N. Acad. Sci., 593,59-72. Massague,J., Attisano,L. and Wrana,J.L. (1994) The TGF-f family and its composite receptors. Trends Cell Biol., 4, 172-178. Moses,H.L., Yang,E.Y. and Pietenpol,J.A. (1990) TGF-f stimulation and inhibition of cell proliferation: new mechanistic insights. Cell, 63, 245-247. Roberts,A.B. and Sporn,M.B. (1990) The transforming factor- growth In Sporn,M.B. and Roberts,A.B. (eds), Peptide Groitth Ps. Factors and Their Receptors., Part l. Springer-Verlag, Berlin, pp. 419-472. Roberts,A.B., Anzano,M.A., Lamb,L.C., Smith,J.M. and Sporn,M.B. (1981) New class of transforming growth factors potentiated by epidermal growth factor. Proc. Natl Acad. Sci. USA, 78, 5339-5343. Saitoh,M., Nishitoh,H., Amagasa,T., Miyazono,K., Takagi,M and Ichijo,H. (1996) Identification of important regions in the cytoplasmic juxtamembrane domain of I type receptor that separate signaling of pathways transforming growth J. Biol. Chem., 271, factor-P3. 2769-2775. ten Dijke,P., Miyazono,K. and Heldin,C.-H. (1996) Signaling via hetero- oligomeric complex of type I and II type serine/threonine kinase receptors. Clirr Opin. Cell Biol., 8, 139-145. Ventura,F., Doody,J., Liu,F., Wrana,J.L. and Massagu6,J. (1994) Reconstitution and transphosphorylation of TGF-f receptor complexes. EMBO J., 13, 5581-5589. Wang,T., Donahoe,P.K. and Zervos,A.S. (1994) Specific interaction of type I receptors of the TGF-f family with the immunophilin FKBP- Science, 265, 674-676. Weiss-Garcia,F. and Massagu6,J. (1996) Complementation between kinase-defective and activation-defective TGF-f receptors reveals a novel form of receptor cooperativity. EMBO J., 15, 276-289. Wieser,R., Attisano,L., Wrana,J.L. and Massague,J. (1993) Signalino activity of transforming growth II factor type receptors lacking specific domains in the cytoplasmic region. Mol. Cell. Biol., 13, 7239-7247. Wieser,R., Wrana,J.L. and Massagu6,J. (1995) GS domain mutations that constitutively activate TfR-I, the downstream signaling component in the TGF-1 receptor complex. EMBO J., 14, 2199-2208. Wrana,J.L., Attisano,L., Carcamo,J., Zentella,A., Doody,J., Laiho,M., Wang,X.-F. and Massague,J. (1992) signals through a TGFP heteromeric protein kinase receptor complex. Cell, 71, 1003-1014. Wrana,J.L., Attisano,L., Wieser,R., Ventura,F. and Massague,J. (1994) Mechanism of activation of the TGF-,B receptor. Natiure, 370, 341-347. Yamashita,H., Ichijo,H., Grimsby,S., Moren,A., ten Dijke,P. and Miyazono,K. (1994a) Endoglin forms a heteromeric complex with the signaling receptors for transforming growth Biol. J. Chem., factor-P. 269, 1995-2001. Yamashita,H., ten Dijke,P., Franzen,P., Miyazono,K. and Heldin,C.-H. (1994b) Formation of hetero-oligomeric complexes of type I and type II receptors for transforming growth factor-5. J. Biol. Chemn., 269, 20172-20178. Yingling,J.M., Wang,X.-F. and Bassing,C.H. (1995) Signaling by the transforming growth factor-f receptors. Biochirn. Biophys. Acta, 1242, 115-136. Received on March 4, 1996; revised 13, oni Auigust 1996

Journal

The EMBO Journal – Springer Journals

Published: Nov 1, 1996

There are no references for this article.