Molecular Cancer Research CTRC-AACR San Antonio Breast Cancer Symposium Chemical and Biological Aspects of Inflammation and Cancer
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Cancer Research Clinical Cancer Research
Cancer Epidemiology Biomarkers & Prevention Molecular Cancer Therapeutics
Molecular Cancer Research Cancer Prevention Research
Cancer Prevention Journals Portal Cancer Reviews Online
Annual Meeting Education Book Meeting Abstracts Online
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Connor, M. K.
Right arrow Articles by Seth, A.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Connor, M. K.
Right arrow Articles by Seth, A.
Molecular Cancer Research 3:453-461 (2005)
© 2005 American Association for Cancer Research

Signaling and Regulation

Molecular Characterization of Ring Finger Protein 11

Michael K. Connor1, Peter B. Azmi1,3, Venkateswaran Subramaniam1, Hoaxia Li{dagger} and Arun Seth1,2,3

1 Molecular and Cellular Biology and 2 Laboratory of Molecular Pathology, Sunnybrook and Women's College Health Sciences Centre; and 3 Department of Laboratory Medicine and Pathobiology, University of Toronto, Toronto, Ontario, Canada

Requests for reprints: Arun Seth, Sunnybrook and Women's College Health Sciences Centre, 2075 Bayview Avenue, Room E-423B Toronto, Ontario, M4N 3M5. Phone: 416-480-6100; Fax: 416-480-5703. E-mail: arun.seth{at}utoronto.ca


   Abstract
Top
 Notes
 Abstract
Introduction
 Results
 Discussion
 Materials and Methods
 Acknowledgements
 References
 
Ring finger proteins serve many vital functions within the cell. We have identified RNF11, a novel 154-amino acid ring finger–containing protein, which is elevated in breast cancer. Within its ring finger domain, RNF11 contains an AKT phosphorylation site (T135) that is situated within a 14-3-3 binding domain. In WM239 cells with constitutively active AKT, RNF11 exhibits seven distinct phosphopeptides as measured using two-dimensional phosphopeptide mapping. Upon inhibition of the AKT pathway or mutation of T135, the phosphorylation at one of these sites is virtually eliminated, suggesting that AKT may phosphorylate RNF11 at T135. Moreover, RNF11 is phosphorylated by AKT in vitro and is recognized by phospho-AKT substrate antibodies. RNF11 shows enhanced binding to 14-3-3 in WM239 cells compared with that seen in the parental WM35 cells which have low AKT activity. Furthermore, treatment of WM239 cells with LY294002 reduces RNF11/14-3-3 interactions suggesting that RNF11/14-3-3 binding is regulated by AKT. In addition, RNF11/14-3-3 binding is enhanced by constitutively active AKT and is diminished by dominant-negative AKT. There is also reduced 14-3-3 binding to T135E RNF11. RNF11 localization was altered from the cytoplasm to the nucleus by activated AKT. Thus, phosphorylation of RNF11 by AKT either causes its nuclear localization or induces degradation of cytoplasmic RNF11. In addition, T135E RNF11, which does not bind 14-3-3 and is not phosphorylated by AKT, causes a greater enhancement of transforming growth factor-ß signaling than wild-type RNF11. It is clear that RNF11 function, localization, and potentially, degradation are regulated by AKT. Disregulation of proper RNF11 function by AKT may prove to be detrimental to patient outcomes, making RNF11 a potential target for novel cancer therapeutics.


   Introduction
Top
 Notes
 Abstract
 Introduction
Results
 Discussion
 Materials and Methods
 Acknowledgements
 References
 
The identification and characterization of new proteins has never been more prevalent than it is in today's research environment. Many debilitating diseases, including cancer, rely on therapies that are ineffective at managing symptoms and controlling disease progression. In addition, cancer provides an additional challenge in that tumors respond initially to treatments and often become resistant to therapy and the disease continues its deadly progression (1-3). Thus, the identification of new therapeutic strategies and novel potential drug targets is essential if cancer management and treatment is to improve. To that end, we have identified that ring finger protein 11 (RNF11), a 154–amino acid protein, is elevated in 90% of invasive ductal carcinomas of the breast (4, 5). RNF11 contains a PY motif in its NH2 terminus and a ring finger domain near the COOH-terminal end (ref. 6; Fig. 1). PY motifs are known to interact with WW-domains in partner proteins (7, 8). We have shown this for RNF11, which interacts with the E3 ligases, Smurf2 and AIP4 (9, 10), both of which are WW-containing proteins. In fact, the RNF11 PY domain is identical to that found in Smad7, a protein that binds to Smurf2 and acts to inhibit the transforming growth factor-ß (TGF-ß) signaling pathway by degrading the TGF-ß receptor (11) and preventing the activation of the receptor activated Smads (2 and 3). This gives RNF11 a potentially important role in the TGF-ß signaling pathway. RNF11 may compete with Smad7 for binding to Smurf2, thereby altering or possibly inhibiting Smad7 activity. In addition, ring finger domains are also important mediators of protein-protein interactions (12, 13), giving another potential region for the regulation of RNF11 function. However, the importance of this domain in mediating RNF11 function is as yet undefined.



View larger version (30K):
[in this window]
[in a new window]
  		FIGURE 1. The RNF11 sequence contains multiple regulatory elements. The primary amino acid sequence of RNF11 showing potential regulatory regions.

 
AKT is a serine/threonine kinase that controls a signaling pathway which plays an integral role in many cellular functions (14-16). AKT has also been shown to be important in breast cancer. Activated AKT phosphorylates the cyclin-dependent kinase inhibitor p27KIP1 on threonine 157 and alters its subcellular localization from the nucleus, where it acts to suppress the cell cycle and maintain cells in a quiescent state, to the cytoplasm which causes disruption of the normal regulation of the cell cycle (17). This permits cell cycle entry and likely plays a role in the transformation of normal cells to a malignant phenotype. Elevated AKT levels are associated with poor patient prognosis in multiple cancers, including breast and prostate cancer (17). In addition, AKT phosphorylation of the BAD protein prevents apoptosis by sequestering BAD away from the antiapoptotic protein BCL-2. Phosphorylation of BAD by AKT promotes its interaction with 14-3-3 (18, 19). This prevents the BAD-dependent regulation of BCL-2 activity preventing the induction of apoptosis in response to cellular insults. This would make a cell with elevated AKT activity susceptible to malignant transformation due to the altered phosphorylation status of target proteins.

We have identified an optimal 14-3-3 binding sequence within the RING finger domain of RNF11. 14-3-3 molecules are a family of proteins that bind phosphorylated serine/threonine residues within a consensus binding sequence (20). Several members of this gene family are down-regulated in breast cancer and serve as molecular markers of potential clinical interest (20, 21). The 14-3-3 proteins have roles in signal transduction pathways that control cell cycle checkpoints, mitogen-activated protein kinase activation, apoptosis, and programs of gene expression (22). Biochemical, structural, and genetic data have revealed some of the molecular basis of 14-3-3 function resulting in a model in which 14-3-3 stabilizes conformations of bound ligands to promote their interactions with downstream targets, or facilitate their subsequent modification by kinases and phosphatases (23). Although just seven genes form the 14-3-3 family, their function in signal transduction has remained obscure due to the plethora of interacting proteins. 14-3-3 binding has been shown to regulate its partners in a positive and/or negative fashion. Interaction with the partner proteins is through specific interactions between phosphorylated serine/threonine residues (24). One of the two optimal binding sequences of the 14-3-3 proteins, RxxS/T(p)xP, generates a sequence partially overlapping an optimal RxRxxS/T sequence, where the serine or threonine is phosphorylated by the AKT protein (15, 25, 26). This is the case for RNF11, where we observed a potential AKT site at T135 nestled within a putative 14-3-3 binding domain. Thus, AKT-dependent 14-3-3 binding may also be important in mediating RNF11 localization and function, and may provide insight into the relevance of elevations in RNF11 expression observed in breast cancer.

This study was designed to characterize the function of the novel protein RNF11 which is elevated in cancer. We show that RNF11 is phosphorylated by AKT and that this phosphorylation mediates RNF11 binding to 14-3-3. AKT phosphorylation also seems to promote an alteration of the subcellular distribution of RNF11, possibly by the selective proteasomal degradation of cytoplasmic RNF11. In support of this, mutant RNF11 that does not bind 14-3-3 and is not phosphorylated by AKT enhances TGF-ß signaling to a greater extent than does wild-type RNF11.


   Results
Top
 Notes
 Abstract
 Introduction
 Results
Discussion
 Materials and Methods
 Acknowledgements
 References
 
Phosphorylation of RNF11 by AKT In vitro
Bioinformatical analyses of the RNF11 amino acid sequence identified a potential AKT phosphorylation site located at T135 located within a 14-3-3 binding domain (Fig. 1). To determine whether or not RNF11 was in fact phosphorylated by AKT, recombinant RNF11 protein was incubated with recombinant constitutively active AKT and 32P-{gamma}ATP (Fig. 2A). Under these conditions, a phosphorylated RNF11 band is clearly evident (Fig. 2A, lane 3). This band was not present when the reactions were conducted in the presence of a 100 molar excess of cold ATP or at 4°C (Fig. 2A, lanes 1 and 2). In addition, we used another recombinant protein, the breast cancer–associated protein 2 (BCA2), in these experiments and observed phosphorylated bands that migrated with a much slower mobility as expected (lanes 4 and 5). As a positive control, reactions were conducted using recombinant GSK3ß (Fig. 2A, lane 6). Because the migration of the phosphorylated RNF11 band was similar to that of GSK3ß, we wanted to ensure that we were actually detecting phosphorylated RNF11 in lane 3 and not detecting a contamination with GSK3ß. A subset of reactions was carried out in parallel where one sample was incubated with 32P-{gamma}ATP and another was incubated with cold ATP. The radioactive sample was run on a gel (Fig. 2A), whereas the nonradioactive sample was subjected to MS/MS mass spectrometry. Although we were unable to detect phosphorylation using this method, we did conclusively identify the presence of numerous RNF11 trypsin fragments in our sample with high confidence (Fig. 2B), which were not present in the GSK3ß sample.



View larger version (46K):
[in this window]
[in a new window]
  		FIGURE 2. RNF11 is phosphorylated by AKT. A. His-tagged RNF11 was amplified and isolated from bacteria using nickel columns. Purified His-RNF11 was incubated with recombinant Akt and 32P-{gamma}ATP (lane 3). As negative controls, reactions were run in the presence of excess cold ATP (lane 1) or at 4°C (lane 2). Similar experiments were conducted using BCA2, a cancer protein that contains an AKT site within a 14-3-3 binding domain (lanes 3 and 4). GSK3ß was run as a positive control (lane 6). B. Parallel reactions with those in (A) were conducted and subjected to MS/MS mass spectrometry to verify the presence of purified HIS-tagged RNF11. Sequences of trypsin fragments that were identified as RNF11 are shown and their corresponding amino acid positions indicated with superscript numbers.

 
In vivo Phosphorylation of RNF11
In order to validate our observed in vitro AKT phosphorylation of RNF11, we attempted to generate two-dimensional phosphopeptide maps for RNF11. Wild-type RNF11 vectors were transfected into WM239 cells which possess constitutively elevated AKT activity. Upon incubation with 32P-orthophosphate and subsequent trypsinization, RNF11 yielded seven distinct phospho fragments (Fig. 3A). This pattern was reproducible and fragment 4 was the most intensely phosphorylated in the WM239 cells. In the presence of the phosphoinositide-3-kinase inhibitor, LY2940002, the pattern of RNF11 phosphorylation was altered. The resultant inhibition of AKT activity drastically reduced the intensity of, but did not entirely eliminate, fragment number 4 (Fig. 3B). When T135 was mutated to glutamate (T135E), fragment 4 was lost, in addition to fragments 1, 2, 3 and 5, however, fragments 6 and 7 were strongly labeled. Thus, it may be that the T135E mutation may cause changes that also affect phosphorylation of other sites. Nonetheless, the loss of spot 4 in Fig. 3C coupled with the reduction in intensity of fragment 4 in Fig. 3B is consistent with that fragment containing T135 and being phosphorylated by AKT. To further corroborate our findings, we evaluated the phosphorylation pattern of RNF11 in the parental WM35 cell line (Fig. 3D). In WM35 cells, there is a loss of two phosphorylation sites compared with the pattern observed in WM239 cells. Consistent with T135 being contained within fragment 4 and being phosphorylatable by AKT, spot 4 is not evident in the phosphopeptide map of RNF11 from WM35 cells (Fig. 3D versus Fig. 3A).



View larger version (138K):
[in this window]
[in a new window]
  		FIGURE 3. RNF11 is phosphorylated at multiple sites. A. FLAG-RNF11 was labeled with 32P in WM239 cells, which possess elevated AKT activity, and subjected to digestion with trypsin. RNF11 fragments were then subjected to two-dimensional electrophoresis. B. RNF11 was prepared as in (A) except that cells were treated with LY294002 to inhibit AKT activity. C. Same as (A) except T135E mutant RNF11 was used instead of wild-type cells. D. Similar analyses were conducted using WM35 cells, the parental line of WM239, in which AKT activity is normal.

 
AKT-Mediated Binding of RNF11 to 14-3-3
RNF11 contains a perfect 14-3-3 binding site within its COOH-terminal region that contains a potential AKT phosphorylation site (T135; Fig. 1). AKT-mediated regulation of 14-3-3 binding has been shown for other proteins (18, 19). We wanted to test whether 14-3-3 bound to RNF11 and if this binding was regulated by AKT. WM35 and WM239 cells were transfected with glutathione S-transferase (GST)-RNF11 expression vectors. Cell lysates were incubated with glutathione-linked sepharose beads and GST-RNF11 was isolated by centrifugation. When RNF11-associated proteins were probed with pan 14-3-3 antibodies, RNF11/14-3-3 association was evident (Fig. 4A). Binding of 14-3-3 to RNF11 in WM35 cells was easily detectable despite modest amounts of phosphorylated (active) AKT (Fig. 4B). However, when corrected for the amount of RNF11 in the GST-pulldowns, the level of RNF11/14-3-3 interaction was lower than in WM239 cells in the presence of LY2940002. RNF11/14-3-3 binding was enhanced upon activation of AKT in WM239 cells (Fig. 4A). When corrected for the lower amount of RNF11 present in the GST-pulldowns from WM239 cells, 14-3-3/RNF11 binding was >6-fold higher when AKT was active. When the GST-pulldowns were probed with an antibody that was specific for phosphorylated AKT substrates (Cell Signaling Technologies, Pickering, Ontario, Canada), there was positive staining which corresponded with GST-RNF11 and followed the pattern of AKT activity (Fig. 4A). WM239 cells showed a much higher staining with the AKT substrate antibody than did WM35 cell lysates. This was consistent with a higher AKT activity in WM239 cells as indicated by increased levels of phosphorylated AKT in WM239 cells compared with WM35 cells (Fig. 4B). Treatment with LY2940002 reduced the levels of phosphorylated AKT (Fig. 4B), which coincided with both a reduction in the amount of RNF11-associated 14-3-3 and a somewhat lower staining with the AKT substrate antibody (Fig. 4A). Notably, LY2940002 treatment of WM239 cells failed to completely inhibit the phosphorylation and activation of AKT (Fig. 4B). This may explain the residual 14-3-3/RNF11 binding in LY2940002-treated WM239 cells (Fig. 4A) and the persistence of phosphorylation of RNF11 peptide fragment number 4 following LY2940002-mediated AKT inhibition (Fig. 3B).



View larger version (42K):
[in this window]
[in a new window]
  		FIGURE 4. RNF11 binds 14-3-3 in an AKT-dependent manner. A. Cells were transfected with wild-type GST-RNF11 vectors and subjected to GST-pulldowns. RNF11-bound proteins were probed with 14-3-3, GST, and phospho-AKT substrate antibodies. Some cells were treated with TGF-ß or LY2940002. B. Whole cell extracts from cells in (A) were probed with antibodies against phosphorylated (activated) AKT.

 
To confirm that the results in Fig. 3 were directly an effect of AKT, we employed a slightly different approach. MCF-7 cells were transiently transfected with FLAG-RNF11 expression vectors and cDNAs for either dominant-negative or constitutively active AKT (Fig. 5). Equal amounts of constitutively active and dominant-negative AKT were expressed following transfection and these vectors had no effect on the levels of FLAG-RNF11 expression (Fig. 5A). In addition, expression of these AKT expression vectors altered AKT activity. AKT activity, as measured using a phospho-GSK3ß antibody, was low in cells expressing dominant-negative AKT, whereas transfection with constitutively active AKT resulted in a much higher AKT activity (Fig. 5A). However, the presence of dominant-negative AKT resulted in a very limited interaction of RNF11 and 14-3-3 as indicated by the amount of 14-3-3 present in FLAG-RNF11 immunoprecipitations (Fig. 5B). In contrast, the presence of constitutively active AKT resulted in a much higher amount of RNF11 bound to 14-3-3 than did dominant-negative AKT, despite similar levels of RNF11 in each immunoprecipitation (Fig. 5B). To determine the role of T135 in this AKT-dependent RNF11/14-3-3 interaction, WM239 cells were transfected with either wild-type or T135E FLAG-RNF11 expression vectors in the presence or absence of LY2940002. As shown previously (Fig. 4A), there was a strong association between wild-type RNF11 and 14-3-3 in WM239 cells (Fig. 5C). In contrast, the interaction between T135E RNF11 and 14-3-3 was virtually eliminated (Fig. 5C). Treatment with LY2940002 dramatically reduced wild-type RNF11/14-3-3 binding (Fig. 5C). Interestingly, it seems that RNF11 binding to 14-3-3 requires not only the negative charge that phosphorylation confers to an amino acid, but also the presence of the phosphate group itself. Mutation of phosphorylatable S or T to D or E is routinely used as a tool to mimic phosphorylation, as D and E substitutions will confer a negative charge at the substituted amino acid site. Thus, the lack of binding of the T135E mutant RNF11 to 14-3-3 (Fig. 5C) suggests that charge alone is insufficient for 14-3-3 binding to RNF11.



View larger version (33K):
[in this window]
[in a new window]
  		FIGURE 5. 14-3-3/RNF11 binding needs the phosphate group. A. Whole cell extracts from MCF-7 cells transfected with dominant-negative or constitutively active AKT and probed with AKT, FLAG, and phospho-specific GSK3ß antibodies. B. FLAG-RNF11 was immunoprecipitated from lysates in (A) and bound proteins probed with 14-3-3 antibodies. C. MCF-7 cells were transfected with wild-type and T135E RNF11 expression vectors in the presence or absence of LY2940002 and treated as in (B).

 
AKT and RNF11 Localization
14-3-3 proteins are known to mediate cellular localization/sequestration of their binding partners (27). To evaluate whether the observed AKT-mediated 14-3-3 binding to RNF11 affected RNF11 subcellular localization. MCF-7 cells were transfected with wild-type RNF11 expression vectors in the absence or presence of constitutively active AKT. Cells were fractionated into their nuclear and cytoplasmic components using a digitonin-permeabilization technique (28). When nuclear (N) and cytoplasmic (C) fractions were probed with anti-FLAG antibodies, wild-type RNF11 was located predominantly in the cytoplasmic compartment (Fig. 6A). Upon expression of constitutively active AKT, RNF11 subcellular compartmentalization was altered, with the majority of the protein being sequestered in the nuclear fraction (Fig. 6A). This result may be due to either a translocation of RNF11 from the cytoplasm to the nucleus or an accelerated degradation of the cytoplasmic RNF11 fraction.



View larger version (43K):
[in this window]
[in a new window]
  		FIGURE 6. Mechanism of RNF11 protein degradation. A. MCF-7 cells were transfected with wild-type FLAG-RNF11 either with or without transfected constitutively active AKT. Cells were separated into nuclear and cytoplasmic components by digitonin permeabilization and probed for FLAG-RNF11. B. HEK293 and WM239 cells were transfected with either wild-type or T135E mutant FLAG-RNF11 in the presence or absence of the proteasome inhibitor MG132. C. HEK293 cells were cotransfected with Smurf2 and various amounts of GST-RNF11 in the absence or presence of different amounts of HA-Smad7 expression vectors.

 
Degradation of RNF11 Protein
It was noticed that RNF11 expression was consistently lower when vectors were transfected into WM239 cells compared with WM35 cells (Fig. 4A). This suggested that the higher AKT activity in WM239 cells may be accelerating RNF11 degradation. In order to determine whether RNF11 proteolysis was indeed elevated by AKT, WM239 cells were transfected with wild-type and T135E RNF11 expression vectors (Fig. 6B). Wild-type RNF11 was expressed at low levels, whereas the expression of T135E RNF11 was much greater. Thus, preventing phosphorylation of T135 may inhibit RNF11 degradation. When ubiquitin-mediated proteolysis was inhibited by treatment of WM239 cells with the proteasome inhibitor MG132 (MG) resulted in an increase in wild-type RNF11, reaching levels approaching that of T135E RNF11 (Fig. 6B). In contrast, in normal cells, where AKT is only intermittently activated, expression of RNF11 is very high. In addition, when normal cells were treated with the protease inhibitor, MG132, no elevation in RNF11 protein levels was evident (Fig. 6B). This suggested that there is another AKT-independent mechanism of RNF11 degradation. Because RNF11 interacts with the E-3 ligase Smurf2, we expressed RNF11 with Smurf2 and its binding partner Smad7 in HEK293 cells (Fig. 6C). Smurf2 expression alone (2 µg) had no effect on RNF11 degradation (lanes 1-3). However, adding only a minimal amount of Smad7 (0.2 µg) dramatically reduced the level of RNF11 (lane 1 versus 7). Increasing Smad7 input further reduced RNF11 expression, with 4 µg of Smad7 cDNA eliminating RNF11 expression entirely (lane 5). Surprisingly, the expression of Smurf2 was also inhibited following the addition of Smad7 (lane 1 versus lanes 5-7). This suggested the formation of a tripartite RNF11/Smad7/Smurf2 complex which resulted in the degradation of both RNF11 and Smurf2.

RNF11 Function in TGF-ß Signaling
In order to test the cellular impact of 14-3-3 binding to RNF11, we used a TGF-ß-responsive system which we have already shown that RNF11 can enhance TGF-ß signaling and relieve Smurf2-mediated inhibition of TGF-ß signaling (5). Expression of Smurf2 represses TGF-ß signaling as measured using a TGF-ß-responsive luciferase assay. The coexpression of RNF11 with Smurf2 relieves Smurf2-mediated repression on TGF-ß signaling (Fig. 7, lanes 3 versus 4). When we coexpressed the RNF11 mutant (T135E) with Smurf2, we found that the T135E mutant induces a moderately better relief of the Smurf2 inhibition of the 3TPLux reporter (Fig. 7, lanes 4 versus 5). This trend is more pronounced when we look at the effects of the expression of RNF11 or the T135E mutant on TGF-ß signaling in the absence of Smurf2. There is a significant 35% increase in TGF-ß signaling (P = 0.007) when T135E RNF11 is expressed compared with wild-type RNF11 (Fig. 7, lanes 6 versus 7).



View larger version (16K):
[in this window]
[in a new window]
  		FIGURE 7. Disruption of 14-3-3 binding and AKT phosphorylation at T135 enhances RNF11 function. TGF-ß signaling assays were conducted using a TGF-ß-responsive (3TPLux) reporter system. HepG2 cells were transiently transfected with 3TPLux reporter constructs alone or in combination with expression vectors for Smurf2, wild-type RNF11, and mutant RNF11 (T135E), which does not bind to 14-3-3 and is not phosphorylated by AKT. Cells were cultured in the presence or absence of 10 ng/mL TGFß-1 and reporter activity was assayed using a dual luciferase activity kit. Statistical differences between experimental conditions was determined using unpaired t tests; *, P = 0.007.

 

   Discussion
Top
 Notes
 Abstract
 Introduction
 Results
 Discussion
Materials and Methods
 Acknowledgements
 References
 
The initial observation that RNF11 was elevated in 90% of invasive ductal carcinomas suggested that RNF11 may be important in cell transformation. Ring finger–containing proteins, such as CBL, MDM2, and ROC1, are involved in the ubiquitin-mediated proteolysis pathway (13, 29-34). This pathway has been implicated as a mechanism behind cancer development. For example, p27KIP1 protein levels are reduced in >60% of all cancers, despite no known genetic mutations and no reduction in protein synthesis rates (35). Thus, an increase in p27 proteolysis almost certainly underlies the decrease in cellular p27 protein levels. In fact, mechanisms of p27 degradation are continually being elucidated and involve signaling pathways that are known to be activated in cancer. This type of regulation of protein expression led to the hypothesis that the overexpression of RNF11 was an oncogenic and deleterious occurrence.

The effects of the TGF-ß signaling pathway are regulated by numerous cellular proteins (36-38). Binding of TGF-ß to the receptors at the cell surface leads to the phosphorylation/activation of the receptor-activated Smads 2 and 3 (R-Smad).

Activated R-Smads bind to Smad4 and the heterodimer translocates to the nucleus where it regulates the transcription of TGF-ß-responsive genes. Smad7 can bind to the TGF-ß receptor which prevents phosphorylation of the R-Smads. In addition, when complexed with the E3 ligase Smurf2, Smad7 acts to degrade the TGF-ß receptor (11). These two actions suppress the TGF-ß signaling pathway and are part of the normal feedback loop that regulates the activity of the TGF-ß pathway. However, inappropriate activation of these inhibitory mechanisms can render a cell insensitive to TGF-ß, effectively shutting down this pathway. Because one of the main ramifications of activation of the TGF-ß pathway is cell cycle arrest (37, 39, 40), hyperactivation of the Smad7/Smurf2 complex may lead to inappropriate initiation of the cell cycle. In fact, TGF-ß insensitivity is a common characteristic of many breast cancers. Thus, overexpression of RNF11, a protein that binds Smurf2, could alter Smurf2 activity. RNF11 contains a PY motif that is identical to that found in the TGF-ß inhibitory Smad7 and seems to actually inhibit Smuf2 activity and restore TGF-ß sensitivity when Smurf2 is overexpressed (5). However, unlike Smad7, RNF11 cannot bind Smurf2 when its PY domain is altered (5, 11). This implies that Smad7 could retain association with Smurf2 in a manner that is implicitly different than RNF11/Smurf2 association, thereby allowing for the formation of a tripartite complex. Indeed, we observe that RNF11, Smurf2 and Smad7 interact at the molecular level and influence steady state protein levels (Fig. 6C). Interestingly, in an assay looking at the activity of the tripartite complex on TGF-ß signaling, we observe that the tripartite complex impacts negatively on TGF-ß signaling. However, when a PY mutant Smad7 is substituted in the tripartite complex, there is slight, but significant increase in TGF-ß responsiveness (P < 0.001; data not shown). We were likely unable to achieve a complete restoration of function with the coexpression of the Smad7 PY mutant with RNF11 and Smurf2, due to the fact that although the Smad7 PY mutant is still capable of binding to Smurf2, it binds at a lowered efficiency (11). However, these observations indicate that Smad7/Smurf2 complexes can affect RNF11 protein levels and consequently alter RNF11 function. If RNF11 can act to restore TGF-ß signaling by competitively inhibiting/altering Smurf2 interaction and function, why doesn't RNF11 overexpression prevent or oppose malignant transformation? In some cell types and disease scenarios (41), the constitutive expression of Smad7 may down-regulate steady state RNF11 levels and disrupt protein function, thereby rendering RNF11 incapable of restoring TGF-ß responsiveness. Also, it may be that RNF11 has cell type–specific functions and depends on the partner proteins to which it is bound. RNF11 complexes may inhibit/degrade a tumor suppressor in one cell type, making it oncogenic, whereas in another cell, RNF11 may degrade/inhibit an oncogenic factor, giving RNF11 a tumor suppressor function. This potential for differential RNF11 function may be determined by its amino acid sequence (Fig. 1). We identified multiple potential RNF11 regulatory domains, the most attractive of which being an AKT phosphorylation site (T135) contained within a 14-3-3 binding domain. Phosphorylation by AKT has been shown to promote binding to 14-3-3 (18, 19), and this study was undertaken to establish whether this is the case for RNF11. In addition, constitutive activation of AKT confers TGF-ß resistance to epithelial cells (42). Inhibition of AKT activity with LY2940002 restores TGF-ß sensitivity and represents the rationale behind our choosing WM35 and WM239 cells as model systems to evaluate RNF11 regulation by AKT. RNF11 is indeed phosphorylated by AKT in vitro (Fig. 2A) and in vivo at T135 (Fig. 3) by AKT. This phosphorylation at T135 enhances RNF11 binding to 14-3-3 and redistributes RNF11 from being predominantly cytoplasmic to mostly nuclear. Thus, 14-3-3 binding to RNF11, mediated by phosphorylation at T135 by AKT, is likely an important regulatory mechanism for RNF11 function. This may help explain why RNF11 is elevated in so many breast cancers. AKT has also been shown to be elevated in breast cancer, which, likely not coincidentally, is a cancer that often becomes insensitive to TGF-ß. TGF-ß insensitivity can be brought about by activated AKT in mammary epithelial cells (17, 42), and TGF-ß responsiveness can be restored subsequently by pharmacologically inhibiting AKT activity. We propose that one of the ramifications of the phosphorylation of RNF11 by AKT is that RNF11 is sequestered away from Smurf2 by 14-3-3. This sequestration may explain why RNF11 is elevated in breast cancer and likely unable to restore the integrity of the TGF-ß signaling pathway and prevent tumor development. This hypothesis is further supported by our observations that the RNF11 (T135E) mutant, which is not phosphorylated by AKT and does not bind 14-3-3, is in fact better at enhancing TGF-ß signaling than is wild-type RNF11.

AKT activation not only promotes RNF11/14-3-3 interaction, but leads to a nuclear predominance of RNF11. This may affect RNF11 interaction with Smurf2, resulting in a loss of the restorative effects of RNF11 on TGF-ß signaling. Thus, elevation of RNF11 may be an initial response by a cell to restore proper TGF-ß signaling in response to the onset of malignant transformation. In the event of other cell signaling alterations, such as activated AKT, RNF11 is unable to restore homeostasis and the cell continues to overexpress RNF11 in a failing attempt to cope with ongoing transformation. So, instead of being the protein underlying oncogenic change, RNF11 overexpression may actually serve as an early marker of cell transformation. Detection of RNF11 overexpression may allow for earlier identification of tumor development and improve the efficiency of cancer treatment. Of course, the determination of other ramifications of RNF11 overexpression needs to be firmly established for treatment strategies to be improved.

Another interesting observation is that when RNF11 is transfected into cells with constitutively active AKT, a lower level of RNF11 expression is observed compared with that in cells where AKT is normally active (Fig. 4A). This can be prevented by treatment with the proteasome inhibitor MG132, suggesting that activated AKT leads to ubiquitin-mediated degradation of RNF11. In addition, T135E RNF11 does not exhibit this lowered expression in AKT-active cells. Because T135E RNF11 does not bind 14-3-3, it is likely that AKT-dependent binding of RNF11 to 14-3-3 leads to the degradation of RNF11. Because constitutively active AKT changes the distribution of RNF11 from a predominantly cytoplasmic to mostly nuclear pattern of expression, AKT phosphorylation may selectively induce degradation of cytoplasmic RNF11. Alternatively, 14-3-3 may act to actively transport RNF11 from the cytoplasm to the nucleus following the phosphorylation of RNF11 at threonine 135 by AKT.

We noted that RNF11 was not entirely degraded when AKT was activated. This suggested a non–AKT-dependent mechanism of RNF11 degradation. When RNF11 was cotransfected with both Smad7 and Smurf2, no RNF11 protein was detectable, whereas cotransfection with Smurf2 alone had virtually no effect on RNF11 degradation. Interestingly, cotransfection of RNF11 with Smad7 and Smurf2 also resulted in elimination of Smurf2 protein. Thus, it seems that RNF11 and Smurf2 are degraded together in the presence of Smad7. This suggests a novel mechanism for Smurf2 degradation. It has been shown that Smad7 can bind to Smurf2 even if the Smad7 PY domain is mutated. Thus, it may be that RNF11, whose PY domain is identical to that of Smad7, forms a tripartite complex with Smurf2 and Smad7, resulting in the degradation of RNF11 and Smurf2. This represents a unique self-sacrificing mechanism of Smurf2 degradation mediated by RNF11.

We are beginning to unravel the complexity of RNF11 function. We do know that RNF11 binds a wide range of cellular targets that encompass a vast array of important life functions (6). RNF11 binds to and seems to inhibit the activity of Smurf2, possibly by initiating Smurf2 degradation, thereby alleviating the suppressive effects of Smurf2 on cell signaling. RNF11 binds to 14-3-3 in an AKT-dependent fashion, and this binding may act to sequester RNF11 away from its various cellular activities. More research is necessary to fully understand RNF11 function, but it is becoming more apparent that this protein is an integral part of cellular regulation. Moreover, RNF11 is likely to be regulated by other cell signaling pathways.


   Materials and Methods
Top
 Notes
 Abstract
 Introduction
 Results
 Discussion
 Materials and Methods
Acknowledgements
 References
 
Cell Lines and Luciferase Assays
In order to determine whether AKT phosphorylation affects RNF11, FLAG-RNF11 vectors were transiently transfected into WM35 and WM239 human melanoma cell lines. WM239 cells harbor a PTEN deletion and a resultant elevation in AKT kinase activity compared with the parental WM35 line. In addition, wild-type and mutant RNF11 expressing vectors were transiently transfected into MCF-7 mammary epithelial cells in the absence or presence of cDNAs encoding constitutively active or dominant-negative AKT.

Standard dual luciferase assays were conducted as follows. HepG2 cells were grown at 37°C in 5% CO2, in MEM media supplemented with L-glutamine, sodium pyruvate, nonessential amino acids, Earl's salts, and 10% fetal bovine serum (Sigma, Oakville, Ontario, Canada). HepG2 cells were plated in six-well plates and transiently transfected with the indicated expression vectors, along with 5 ng/well of Renilla luciferase which served as an internal control. HepG2 cells were stimulated 16 hours posttransfection with 10 ng/mL TGFß-1 (R&D Systems, Minneapolis, MN). Reporter gene responsiveness was determined 48 hours posttransfection using the Dual Luciferase Detection system (Promega, Madison, WI) as described by the manufacturer. All luciferase values were normalized to a cotransfected Renilla Luciferase activity as an internal control. All transfections were conducted using LipofectAMINE 2000 (Invitrogen, Burlington, Ontario, Canada).

In vitro Phosphorylation Assays
Bacterial recombinant vectors encoding HIS-tagged RNF11 were expressed and isolated using a ProBond HIS-tag isolation kit (Invitrogen). RNF11 protein purification was verified by Coomassie staining of SDS-PAGE gels and MS/MS-mass spectrometry. Ten nanograms of RNF11 and BCA2, a novel ring finger protein which also contains an AKT phosphorylation site within a 14-3-3 binding domain, were incubated with recombinant active AKT (Cell Signaling Technologies) and 32P{gamma}-ATP for 30 minutes at 37°C. Reactions were also conducted using recombinant GSK3ß as a positive control.

Two-dimensional Phosphopeptide Mapping
WM239 cells, which have constitutively active AKT, were transiently transfected with wild-type FLAG-RNF11 expressing cDNAs. Cells were starved in phosphate-free media containing 5% dialyzed phosphate-depleted fetal bovine serum for 4 hours. Cells were then incubated with 32P-orthophosphate (1 mCi/plate) for 3 hours. Cells were then lysed in 0.5% NP40 buffer containing protease and phosphatase inhibitors. FLAG-RNF11 proteins were immunoprecipitated from lysates using FLAG antibody-conjugated affinity gel (Sigma). Immunoprecipitated proteins were separated on 12% SDS-PAGE gels and radiolabeled proteins visualized by autoradiography. FLAG-RNF11 proteins were excised and subjected to trypsin digestion. The resultant RNF11 fragments were applied to a thin-layer chromatography plate and separated in the first dimension using a Hunter apparatus (Amersham Pharmacia, Montreal, Quebec, Canada). Peptide fragments were separated in the second dimension using thin-layer chromatography. Phosphorylated fragments were visualized using autoradiography. Subsequent experiments were conducted using WM35 cells, which possess normal AKT activity, and WM239 cells treated with the phosphoinositide-3-kinase inhibitor LY2940002 (25 µmol/L, Calbiochem, San Diego, CA). In addition, RNF11 vectors which harbored a threonine 135 to glutamate (T135E) mutation were generated using a site-directed mutagenesis kit (Invitrogen). The effect of this T135E mutation on RNF11 phosphorylation was also evaluated.


   Acknowledgements
Top
 Notes
 Abstract
 Introduction
 Results
 Discussion
 Materials and Methods
 Acknowledgements
References
 
We thank Jeff Wrana and Liliana Attisano for the generous gift of Smad and Smurf vectors, and Joe Testa for AKT plasmids. This is dedicated to Dr. Hoaxia Li who died of breast cancer while doing this study.


   Notes
Top
 Notes
Abstract
 Introduction
 Results
 Discussion
 Materials and Methods
 Acknowledgements
 References
 
Grant support: Canadian Breast Cancer Research Initiative grant #012099 to A. Seth and a postdoctoral fellowship from the U.S. Department of Defense Breast Cancer Research Program (DAMD17-02-1-0573) to M. Connor. Canadian Foundation for Innovation.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

Note: M.K. Connor is currently with the Department of Kinesiology and Health Science, York University, Toronto, Ontario, Canada.

{dagger} {dagger} Deceased Back

Received 9/30/04; revised 6/14/05; accepted 7/14/05.


   References
Top
 Notes
 Abstract
 Introduction
 Results
 Discussion
 Materials and Methods
 Acknowledgements
 References
 

  1. Akiyama S. The mechanisms of the resistance to molecular targeting agents. Nippon Rinsho 2004;62:1297–304.[Medline]
  2. Nishio K. Resistance to target-based therapy and its circumvention. Nippon Rinsho 2004;62:1343–7.[Medline]
  3. Pommier Y, Sordet O, Antony S, Hayward RL, Kohn KW. Apoptosis defects and chemotherapy resistance: molecular interaction maps and networks. Oncogene 2004;23:2934–49.[CrossRef][Medline]
  4. Seki N, Hattori A, Hayashi A, et al. Cloning and expression profile of mouse and human genes, Rnf11/RNF11, encoding a novel RING-H2 finger protein. Biochim Biophys Acta 1999;1489:421–7.[Medline]
  5. Subramaniam V, Li H, Wong M, et al. The RING-H2 protein RNF11 is overexpressed in breast cancer and is a target of Smurf2 E3 ligase. Br J Cancer 2003;89:1538–44.[Medline]
  6. Connor MK, Seth A. A central role for the ring finger protein RNF11 in ubiquitin-mediated proteolysis via interactions with E2s and E3s. Oncogene 2004;23:2089–95.[Medline]
  7. Flores SY, Debonneville C, Staub O. The role of Nedd4/Nedd4-like dependant ubiquitylation in epithelial transport processes. Pflugers Arch 2003;446:334–8.[Medline]
  8. Komuro A, Imamura T, Saitoh M, et al. Negative regulation of transforming growth factor-ß (TGF-ß) signaling by WW domain-containing protein 1 (WWP1). Oncogene 2004;23:6914–23.[CrossRef][Medline]
  9. Kitching R, Wong MJ, Koehler D, et al. The RINGH2 protein RNF11 is differentially expressed in breast tumours and interacts with HECT-type E3 ligases. Biochim Biophys Acta 2003;1639:104–12.[Medline]
  10. Li H, Seth A. An RNF11: Smurf2 complex mediates ubiquitination of the AMSH protein. Oncogene 2004;23:1801–8.[CrossRef][Medline]
  11. Kavsak P, Rasmussen RK, Causing CG, et al. Smad7 binds to Smurf2 to form an E3 ubiquitin ligase that targets the TGF ß receptor for degradation. Mol Cell 2000;6:1365–75.[CrossRef][Medline]
  12. Matthews JM, Sunde M. Zinc fingers–folds for many occasions. IUBMB Life 2002;54:351–5.[Medline]
  13. Ulrich HD. Protein-protein interactions within an E2-RING finger complex. implications for ubiquitin-dependent DNA damage repair. J Biol Chem 2003;278:7051–8.[Abstract/Free Full Text]
  14. Alessi DR, Cohen P. Mechanism of activation and function of protein kinase B. Curr Opin Genet Dev 1998;8:55–62.[CrossRef][Medline]
  15. Datta SR, Brunet A, Greenberg ME. Cellular survival: a play in three akts. Genes Dev 1999;13:2905–27.[Free Full Text]
  16. Brazil DP, Yang ZZ, Hemmings BA. Advances in protein kinase B signalling: AKTion on multiple fronts. Trends Biochem Sci 2004;29:233–42.[CrossRef][Medline]
  17. Liang J, Zubovitz J, Petrocelli T, et al. PKB/Akt phosphorylates p27, impairs nuclear import of p27 and opposes p27-mediated G1 arrest. Nat Med 2002;8:1153–60.[CrossRef][Medline]
  18. Datta SR, Dudek H, Tao X, et al. Akt phosphorylation of BAD couples survival signals to the cell-intrinsic death machinery. Cell 1997;91:231–41.[CrossRef][Medline]
  19. del Peso L, Gonzalez-Garcia M, Page C, Herrera R, Nunez G. Interleukin-3-induced phosphorylation of BAD through the protein kinase akt. Science 1997;278:687–9.[Abstract/Free Full Text]
  20. Wilker E, Yaffe MB. 14-3-3 Proteins—a focus on cancer and human disease. J Mol Cell Cardiol 2004;37:633–42.[CrossRef][Medline]
  21. Ferguson AT, Evron E, Umbricht CB, et al. High frequency of hypermethylation at the 14-3-3 sigma locus leads to gene silencing in breast cancer. Proc Natl Acad Sci U S A 2000;97:6049–54.[Abstract/Free Full Text]
  22. Yaffe MB. How do 14-3-3 proteins work? Gatekeeper phosphorylation and the molecular anvil hypothesis. FEBS Lett 2002;513:53–7.[CrossRef][Medline]
  23. Hermeking H. The 14-3-3 cancer connection. Nat Rev Cancer 2003;3:931–43.[CrossRef][Medline]
  24. Aitken A, Baxter H, Dubois T, et al. Specificity of 14-3-3 isoform dimer interactions and phosphorylation. Biochem Soc Trans 2002;30:351–60.[CrossRef][Medline]
  25. Fresno Vara JA, Casado E, de Castro J, Cejas P, Belda-Iniesta C, Gonzalez-Baron M. PI3K/Akt signalling pathway and cancer. Cancer Treat Rev 2004;30:193–204.[CrossRef][Medline]
  26. Baldin V. 14-3-3 proteins and growth control. Prog Cell Cycle Res 2000;4:49–60.[Medline]
  27. Abe T, Takagi N, Nakano M, Furuya M, Takeo S. Altered bad localization and interaction between bad and bcl-xL in the hippocampus after transient global ischemia. Brain Res 2004;1009:159–68.[Medline]
  28. Connor MK, Kotchetkov R, Cariou S, et al. CRM1/Ran-mediated nuclear export of p27(Kip1) involves a nuclear export signal and links p27 export and proteolysis. Mol Biol Cell 2003;14:201–13.[Abstract/Free Full Text]
  29. Dikic I, Szymkiewicz I, Soubeyran P. Cbl signaling networks in the regulation of cell function. Cell Mol Life Sci 2003;60:1805–27.[CrossRef][Medline]
  30. Duan L, Reddi AL, Ghosh A, Dimri M, Band H. The cbl family and other ubiquitin ligases: destructive forces in control of antigen receptor signaling. Immunity 2004;21:7–17.[CrossRef][Medline]
  31. Deb SP. Cell cycle regulatory functions of the human oncoprotein MDM2. Mol Cancer Res 2003;1:1009–16.[Abstract/Free Full Text]
  32. Iwakuma T, Lozano G. MDM2, an introduction. Mol Cancer Res 2003;1:993–1000.[Abstract/Free Full Text]
  33. Michael D, Oren M. The p53-Mdm2 module and the ubiquitin system. Semin Cancer Biol 2003;13:49–58.[CrossRef][Medline]
  34. Fang S, Lorick KL, Jensen JP, Weissman AM. RING finger ubiquitin protein ligases: implications for tumorigenesis, metastasis and for molecular targets in cancer. Semin Cancer Biol 2003;13:5–14.[CrossRef][Medline]
  35. Slingerland J, Pagano M. Regulation of the cdk inhibitor p27 and its deregulation in cancer. J Cell Physiol 2000;183:10–7.[CrossRef][Medline]
  36. Lutz M, Knaus P. Integration of the TGF-ß pathway into the cellular signalling network. Cell Signal 2002;14:977–88.[CrossRef][Medline]
  37. Mehra A, Wrana JL. TGF-ß and the smad signal transduction pathway. Biochem Cell Biol 2002;80:605–22.[CrossRef][Medline]
  38. Miyazawa K, Shinozaki M, Hara T, Furuya T, Miyazono K. Two major smad pathways in TGF-ß superfamily signalling. Genes Cells 2002;7:1191–204.[Abstract]
  39. Miyazono K, Suzuki H, Imamura T. Regulation of TGF-ß signaling and its roles in progression of tumors. Cancer Sci 2003;94:230–4.[Medline]
  40. Donovan J, Slingerland J. Transforming growth factor-ß and breast cancer: cell cycle arrest by transforming growth factor-ß and its disruption in cancer. Breast Cancer Res 2000;2:116–24.[CrossRef][Medline]
  41. Mentelone G, Pallone F, MacDonald T. Smad7 in TGFß-mediated negative regulation of gut inflammation. Trends Immunol 2004;25:513–7.[CrossRef][Medline]
  42. Liang J, Slingerland JM. Multiple roles of the PI3K/PKB (akt) pathway in cell cycle progression. Cell Cycle 2003;2:339–45.[Medline]




This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Connor, M. K.
Right arrow Articles by Seth, A.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Connor, M. K.
Right arrow Articles by Seth, A.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Cancer Research Clinical Cancer Research
Cancer Epidemiology Biomarkers & Prevention Molecular Cancer Therapeutics
Molecular Cancer Research Cancer Prevention Research
Cancer Prevention Journals Portal Cancer Reviews Online
Annual Meeting Education Book Meeting Abstracts Online