The type I TGF-β receptor TβRI is sumoylated
To examine the sumoylation of TβRI or TβRII, we expressed Flag-tagged rat TβRI or TβRII with myc-tagged SUMO-1. Cell lysate immunoprecipitations using anti-Flag antibodies, followed by western blotting detected myc-tagged, sumoylated TGF-β receptors. As shown in , SUMO was conjugated to TβRI, but not TβRII, resulting in a >20 kd shift, similarly to other sumoylated proteins, indicating that TβRI is post-translationally sumoylated in vivo. TβRI sumoylation was increased when the E2 conjugating enzyme Ubc9 was co-expressed with SUMO-1, suggesting that Ubc9 is involved in sumoylation of TβRI (). Under conditions of Ubc9 overexpression and proportionally insufficient E3 SUMO ligase expression, up to three sumoylated TβRI forms were observed. Since only one SUMO can be linked to a Lys, we assume that, under these conditions, the initial, site-specific sumoylation can confer additional TβRI sumoylation at other sites.
Figure 1 The type I TGF-β receptor TβRI is sumoylated. (a) TβRI, but not TβRII, is sumoylated. Lysates of COS cells, expressing Flag-tagged TβRI or TβRII and myc-tagged SUMO-1, were subjected to immunoprecipitations (more ...)
We next evaluated whether TβRI can be sumoylated in vitro. Immunopurified TβRI was incubated with SUMO-1, the E1 activating SUMO enzyme, Aos1/Uba2, and Ubc9 in the presence of ATP. Western blotting detected a band with a size that was compatible with the attachment of a SUMO-1 protein to TβRI, and corresponded to the sumoylated TβRI in vivo. When the E1 or E2 enzyme, or SUMO-1, was absent, this band was not detected (). This result suggests that TβRI was sumoylated in vitro.
Immunoprecipitation of TβRI from Mv1Lu or MDA-231 cells, treated with or without TGF-β, and immunoblotting with antibodies against SUMO-1, revealed that endogenous TβRI was sumoylated and that TGF-β induced TβRI sumoylation (). These data indicate that receptor activation by TGF-β may induce sumoylation of TβRI.
We assessed if other type I TGF-β family receptors could be sumoylated. Each type I receptor was expressed in the presence or absence of SUMO-1 and Ubc9, and sumoylation was analyzed by immunoblotting. Whereas TβRI was sumoylated, other type I receptors were not ().
TβRI kinase activity and phosphorylation are required for sumoylation of the TβRI receptor
To further characterize whether activation of TβRI affects its sumoylation, as apparent by the TGF-β-induced TβRI sumoylation (), we compared the in vitro sumoylation efficacy of immunopurified wild-type TβRI and activated TβRI (caTβRI) with a Thr202
to Asp mutation (Thr202
in rat TβRI corresponds to Thr204
in human TβRI) resulting in elevated kinase activity8
. As shown in , caTβRI was sumoylated much more efficiently than wild-type TβRI, suggesting that TβRI activation, which normally occurs by TβRII-mediated phosphorylation in response to TGF-β, facilitates sumoylation of the receptor.
Figure 2 The kinase activities of TβRI and TβRII are required for TβRI sumoylation. (a) Activated TβRI is more sumoylated than wild-type TβRI. In vitro sumoylation of immunopurified Flag-tagged wild-type and activated (ca) (more ...)
Since the activated TβRI has elevated kinase activity and increased phosphorylation8
, we examined whether increased TβRI sumoylation resulted from increased TβRI kinase activity or phosphorylation. In vitro sumoylation of wild-type TβRI or caTβRI was decreased in the presence of SB431542, a specific TβRI kinase inhibitor, although this was more easily detected using caTβRI (). These data suggested that the TβRI kinase regulates TβRI sumoylation. Since TβRI did not phosphorylate SUMO-1, the E1 enzyme (Aos/Uba2) and Ubc9 (Supplementary Fig. S1a
), these data suggested that TβRI autophosphorylation plays a role in its sumoylation. To determine whether TβRI phosphorylation regulates TβRI sumoylation, we removed the Ser/Thr phosphorylation from TβRI using lambda phosphatase prior to in vitro sumoylation. The absence or reduction of TβRI phosphorylation decreased the sumoylation of wild-type or activated TβRI (). This again was more easily detected with activated than with wild-type TβRI, due to the difference in sumoylation level. These results suggest that increased kinase activity together with increased phosphorylation contribute remarkably to the efficiency of TβRI sumoylation.
TGF-β binding to TβRII results in stable complex formation of two TβRII and two TβRI receptors, in which TβRII phosphorylates the TβRI cytoplasmic domain and thereby activates the TβRI kinase1
. The activated receptor complex allows for autophosphorylation of the TβRII and TβRI dimers. To determine the roles of the TβRII and TβRI kinases in TβRI sumoylation, we used a cytoplasmic chimera that fuses the TβRI cytoplasmic domain to the TβRII cytoplasmic domain9
. In this complex, the TβRII kinase activates the TβRI kinase without the need to add TGF-β. The receptor chimera, expressed in the presence of SUMO and Ubc9, was sumoylated. Since TβRII is not sumoylated (), the sumoylation site is within the TβRI cytoplasmic domain. Inactivation of the TβRI kinase by Lys230
to Arg mutation decreased the chimera sumoylation (), consistent with the decreased TβRI sumoylation in the presence of SB431542 (). Similar inactivation of the TβRII kinase by Lys277
to Arg mutation also decreased the chimera sumoylation, when compared with the wild-type, kinase-active version (). Since the TβRII cytoplasmic domain is not targeted for sumoylation, this result indicates that TβRI cytoplasmic domain phosphorylation by the TβRII kinase plays an important role in the sumoylation of TβRI. Mutation of both kinase ATP binding sites in the chimera blocked sumoylation.
The requirement for both receptor kinase activities for sumoylation was also studied in vitro. The efficacies of in vitro sumoylation of wild-type, TβRI kinase-defective, TβRII kinase-defective and TβRI/II kinase-defective chimeras, immunopurified from transfected cells, were compared (). Inactivation of the kinase functions of TβRII or TβRI strongly decreased the sumoylation in vitro, while inactivation of both kinases abolished the chimera sumoylation ().
These observations indicate that the kinase activities of both TβRI and TβRII, and consequent phosphorylation of TβRI, are required for efficient TGF-β-induced sumoylation of TβRI in the receptor complex. This is consistent with the TGF-β-induced phosphorylation and consequent activation of TβRI by TβRII ().
The TβRI receptor is sumoylated on lysine 389
Sumoylation often occurs on lysine (K) within a consensus sequence ΨKx(D/E), in which Ψ represents a large hydrophobic residue10
. Since this consensus sequence is absent in the TβRI amino acid sequence, each of the 18 lysines in the cytoplasmic domain was singly replaced by arginine, and the effect of each mutation on TβRI sumoylation was tested. Lysates of cells co-expressing each Flag-tagged mutant lysine TβRI with SUMO-1 and Ubc9 were subjected to immunoprecipitation using anti-Flag antibody, and western blotting. TβRI was not sumoylated when Lys389
residue was replaced by arginine, whereas arginine replacements of other lysines did not affect sumoylation of TβRI (; data not shown), indicating that Lys389
is a major site for TβRI sumoylation. The Lys389
mutation also affected in vitro sumoylation, since no sumoylated TβRI was detected when Lys389
was replaced by arginine (). These results indicate that Lys389
is the only residue targeted for sumoylation.
Figure 3 The TβRI receptor is sumoylated on lysine 389. (a) Mutation of Lys389 abolishes TβRI sumoylation. 293T cells expressed Flag-tagged wild-type or mutant TβRI with the indicated lysine-to-arginine mutation, with myc-tagged SUMO-1 (more ...)
The proposed structure of the TβRI cytoplasmic domain () predicts that Lys389 is located in the hinge between the αEF helix and the αF helix, and is exposed at the surface of the C lobe of the kinase domain. The sumoylation site faces the same orientation as the GS region, which is phosphorylated by TβRII upon TGF-β binding, and the L45 loop, which specifies the Smad interaction, albeit in a separate protein domain. The GS region and L45 loop, both located in the N lobe, interact with the Smad for phosphorylation by TβRI. The exposure of Lys389 at the protein surface predicts that SUMO conjugation strongly affects the cytosolic surface of TβRI, and may regulate the Smad binding to the L45 loop and GS domain of TβRI, and interactions of additional proteins with the receptor complex.
Sequence comparisons () show that Lys389
is not conserved in other TGF-β family type I receptors, with the exception of the activin receptor ActRIB/ALK-4. This is consistent with the absence of sumoylation of these type I receptors in vivo (). The lack of ActRIB/ALK-4 sumoylation suggests that other determinants besides the target lysine are needed for sumoylation of TβRI. It is unlikely that this is due to the Ser versus Thr difference, four residues preceding the sumoylated Lys in TβRI compared to ActrIB/ALK-4 (), since S385T replacement did not affect the in vitro sumoylation of TβRI (Supplementary Fig. S1b
TβRI sumoylation regulates Smad interaction and activation
To evaluate whether sumoylation of the exposed Lys389
affects Smad activation, we examined the interaction of Smad3 with caTβRI. Since this interaction is hard to detect by immunoprecipitation, likely due to its low affinity and transient nature, we examined the interaction of the Smad3D407E mutant with caTβRI. The D407E mutation in the MH2 domain was identified in Smad2 in colorectal carcinoma, and affects the Smad interaction with TβRI and heteromerization with Smad411
. Increased caTβRI sumoylation by co-expressing Ubc9 and SUMO-1 enhanced the TβRI interaction with Smad3D407E. In contrast, coexpression of SUMO and Ubc9 did not enhance the interaction of Smad3D407E with caTβRI carrying the sumoylation-resistant K389R mutation (). We also incubated immobilized GST-Smad3D407E with a mixture of sumoylated and unsumoylated TβRI. Western blotting of purified GST-Smad3 - TβRI complexes showed preferential binding of Smad3 to sumoylated TβRI, compared with unsumoylated TβRI, even though the latter was in large excess (). This result, and the data in , indicates that sumoylation of TβRI enhances Smad3 recruitment and suggests that TβRI sumoylation enhances Smad activation.
Figure 4 TβRI sumoylation regulates Smad activation and TGF-β responses. (a) Interaction of Smad3 with TβRI. 293T cells were transfected to co-express activated (ca) TβRI or its K389R mutant, with myc-SUMO-1 and Ubc9, and/or Smad3D407E. (more ...)
Since upon TGF-β binding, TβRI phosphorylates Smad2 and Smad3, we investigated whether sumoylation of activated TβRI affects Smad3 phosphorylation. TβRI-defective mouse embryonic fibroblasts (MEFs) derived from Tgfbr1−/−
were retrovirally infected to express wild-type TβRI or sumoylation-resistant K389R TβRI. Stably selected cell populations, expressing either TβRI form at equal levels (), showed equivalent cell surface levels of wild-type or mutant TβRI (), suggesting that the K389R mutation did not affect cell surface transport or stability of TβRI. K389R TβRI also showed a similar phosphorylation level as wild-type TβRI, resulting primarily from the TβRI kinase activity, since SB431542 abolished this phosphorylation (). Fractionation of cell lysates did not reveal differences in subcellular compartmentalization of wild-type versus mutant TβRI (Supplementary Fig. S1c
). We then compared wild-type and K389R TβRI for their ability to phosphorylate Smad3 in response to TGF-β. In fibroblasts expressing wild-type TβRI, TGF-β induced Smad3 phosphorylation within 15 min, whereas, in cells expressing K389R TβRI, the Smad3 phosphorylation kinetics in response to TGF-β was slower, with first detection at 30 min. Furthermore, the overall level of Smad3 activation was lower in cells expressing K389R TβRI, compared to cells expressing wild-type TβRI (). Similar results were seen with TGF-β-induced activation of Smad2 (). Replacement of Lys393
in ActRIB/ALK-4, which is not sumoylated and corresponds to Lys389
in TβRI, with Arg did not affect Smad3 activation (Supplementary Fig. S1d
The differences in level and kinetics of Smad2 and Smad3 phosphorylation by wild-type versus K389R TβRI, together with the results of the Smad3 interaction with TβRI, suggest that TβRI sumoylation enhances the Smad interaction with TβRI, allowing more efficient Smad2/3 phosphorylation and activation in response to TGF-β.
TβRI sumoylation regulates functional responses to TGF-β
fibroblasts ectopically expressing wild-type or K389R TβRI, we characterized the effect of TβRI sumoylation on Smad-mediated transcription, i.e. the functional consequence of Smad activation. We used a reporter in which tandem Smad binding sites control luciferase transcription. Cells expressing K389R TβRI showed reduced transcription from the Smad3-responsive promoter compared to cells expressing wild-type TβRI (). We also compared the endogenous expression of the TGF-β-responsive Smad7 gene by RT-PCR. Cells expressing K389R TβRI showed reduced Smad7 mRNA expression in response to TGF-β compared to cells expressing wild-type TβRI (). Similar results were obtained with two additional populations of Tgfbr1−/−
fibroblasts ectopically expressing TβRI or K389R TβRI at similar levels (Supplementary Fig. S2
). These results suggest that TβRI sumoylation defines the TGF-β-induced transcriptional regulation.
We also examined the contribution of TβRI sumoylation to the antiproliferative response to TGF-β. We seeded the fibroblasts expressing wild-type or K389R TβRI in parallel with the parental Tgfbr1−/− cells as control cells, and determined the proliferative response after adding TGF-β. Cells lacking TβRI were not affected in their proliferation by TGF-β, whereas those expressing wild-type TβRI responded with decreased proliferation, as assessed by cell number (). In contrast to wild-type TβRI, cells expressing K389R TβRI showed a decreased growth inhibitory response to TGF-β. This result suggests that sumoylation regulates the TβRI-mediated antiproliferative response to TGF-β and renders the cells more responsive to TGF-β.
TβRI sumoylation enhances invasion and metastasis of Ras-transformed cells
Since autocrine TGF-β signaling regulates cancer progression13,14
we postulated that resistance to sumoylation, while suppressing TGF-β growth inhibitory activities, affects tumor progression. To address this issue, the Tgfbr1−/−
fibroblasts, carrying a control empty vector or ectopically expressing wild-type or K389R TβRI at similar levels, were transduced with a control vector or a vector expressing activated Ha-Ras (Leu-61) to generate tumorigenic cell populations. Mutant Ras was expressed and activated Erk MAP kinase to similar extents in all three Ras-transformed cell populations (). Cells expressing activated Ras had a transformed phenotype, apparent from the altered cell morphology and loss of contact inhibition (data not shown).
Figure 5 Lack of TβRI sumoylation decreases TGF-β-regulated invasion and metastasis. (a) Ras-transformed Tgfbr1−/− fibroblasts stably expressing wild-type or K389R TβRI, or with an empty vector, were subjected to western (more ...)
We examined the invasion of the Ras-transformed cells using a modified Boyden chamber assay whereby cells migrate through Matrigel toward serum. Ras-transformed Tgfbr1−/− cells ectopically expressing wild-type TβRI showed a higher invasion activity compared to Ras-transformed Tgfbr1−/− cells lacking TβRI (), indicating that the invasive capacity of Ras-transformed MEFs depends on TβRI signaling. Ras-transformed Tgfbr1−/− MEFs expressing K389R TβRI were less invasive than cells expressing wild-type TβRI, indicating that, in this system, lack of TβRI sumoylation impairs the TβRI-dependent invasion of transformed cells (). These observations are consistent with the role of TβRI sumoylation in TGF-β-induced gene expression and growth inhibition ().
Metastasis is a complex process, requiring cell growth, migration, invasion, intra- and extravasation, and cell survival in the circulatory system and at the metastatic site. Using a mouse tail vein injection model, autocrine TGF-β signaling was shown to enhance the ability of tumor cells to establish metastatic nodules within the lung15,16
. To determine the roles of TβRI and TβRI sumoylation in the formation of metastatic nodules in this model, we compared the ability of the Tgfbr1−/−
MEF derivatives to colonize the lung. Colonization of the lungs by MEFs was fully dependent on expression of activated Ha-Ras (; data not shown). Ras-transformed Tgfbr1−/−
cells gave rise to only few very small lung tumor nodules (). Tumor cells were proliferative with a high incidence of apoptosis (Supplementary Fig. S3h
). The tumors were morphologically heterogeneous, and large cells with massive nuclei were indicative of chromosomal instability (Supplementary Fig. S3e
). Remarkably, Ras-transformed Tgfbr1−/−
cells ectopically expressing wild-type or K389R TβRI developed numerous large metastatic nodules (). MEFs expressing sumoylation-defective TβRI gave rise to fewer tumor nodules than cells expressing the wild-type receptor (). These nodules were generally smaller than those from wild-type TβRI expressing MEFs (data not shown; ), although their histological appearance () and proliferative rates (Supplementary Fig. S3a
) were similar. These results suggest that, in this model of TGF-β-mediated metastasis, TβRI sumoylation contributes to tumor progression by enhancing tumor cell extravasation, survival and/or growth at the metastatic site.
The Ser385Tyr TβRI mutation, implicated in metastatic cancer, confers sumoylation resistance
Mutations in TGF-β signaling mediators, including TGFBR1
, have been associated with human cancers13
. Among these, a missense mutation, S387Y, in TGFBR1
was enriched in breast and head-and-neck cancer metastases, compared to corresponding primary tumors17,18
. This mutation confers diminished TGF-β signaling, and is the only mutation in TGFBR1
, known to specifically associate with tumor metastases17
. Since the corresponding residue in rat TβRI, i.e. Ser385
, localizes close to the Lys389
sumoylation site of TβRI (), we examined whether rat S385Y TβRI was sumoylated. In contrast to wild-type TβRI, this mutant was not sumoylated in cells overexpressing SUMO-1 and Ubc9 (). In vitro, replacement of Ser385
with Tyr () or Ala () also strongly decreased sumoylation. In contrast, replacement of Ser385
with Thr, which may, similarly to Ser, be targeted for phosphorylation by Ser/Thr kinases, did not affect sumoylation (Supplementary Fig. S1b
). These results indicate that this single amino acid substitution prevents TβRI sumoylation.
Figure 6 The Ser385Tyr mutation impairs TβRI sumoylation and function. (a, b) Ser385Tyr TβRI is not sumoylated. Panel (a) shows the rat TβRI sequence with Lys389 as sumoylation site four amino acids away from Ser385, which is equivalent (more ...)
We examined whether the S385Y mutation decreased the TGF-β responsiveness, as observed with the sumoylation-defective K389R TβRI (). Using the Smad3-responsive reporter of , cells expressing S385Y TβRI responded to TGF-β with a lower level transcription than cells expressing wild-type TβRI, but a higher level than cells expressing K389R TβRI (), indicating that the decreased TGF-β responsiveness associated with S385Y TβRI correlates with impaired TβRI sumoylation. We also evaluated the S385Y TβRI mutation in the lung colonization tumor model. Ras-transformed MEFs expressing S385Y TβRI formed fewer and smaller metastatic nodules than cells expressing wild-type TβRI (). While this difference in tumor nodule formation compares qualitatively with the K389R TβRI cells, the efficiency of metastatic nodule formation was more impaired in S385Y TβRI cells than in K389R TβRI cells. This quantitative difference between cells expressing K389R TβRI or S385Y TβRI, when compared with their differential activities in transcription assays, raises the possibility that the impaired TβRI sumoylation resulting from the S385Y mutation may be complemented with an additional defect of relevance to cancer progression.